Toward a Predictive Framework for Convergent Evolution: Integrating Natural History, Genetic Mechanisms, and Consequences for the Diversity of Life Anurag A

Home , Calycomyza, Convergent evolution, Daihinibaenetes, Laminas, Poekilocerus, Spilostethus pandurus

AMERICAN NATURALIST Volume 190, Number S1 | August 2017

Convergence, Natural History, and Big Questions in Biology

A symposium organized by Anurag Agrawal

Toward a Predictive Framework for Convergent Evolution: Integrating Natural History, Genetic Mechanisms, and Consequences for the Diversity of Life Anurag A. Agrawal, pp. S1–S12

Pattern and Process in the Comparative Study of Convergent Evolution D. Luke Mahler, Marjorie G. Weber, Catherine E. Wagner, Travis Ingram, pp. S13–S28

Convergently Evolved Toxic Secondary Metabolites in Plants Drive the Parallel Molecular Evolution of Insect Resistance Georg Petschenka, Vera Wagschal, Michael von Tschirnhaus, Alexander Donath, Susanne Dobler, pp. S29–S43

Convergent Phenotypic Evolution despite Contrasting Demographic Histories in the Fauna of White Sands Erica Bree Rosenblum, Christine E. Parent, Eveline T. Diepeveen, Clay Noss, Ke Bi, pp. S44–S56

Convergence and Divergence in a Long-Term Experiment with Bacteria Richard E. Lenski, pp. S57–S68

Evolutionary Scenarios and Primate Natural History Harry W. Greene, pp. S69–S86

Convergence, Consilience, and the Evolution of Temperate Deciduous Forests Erika J. Edwards, David S. Chatelet, Bo-Chang Chen, Jin Yao Ong, Shuichiro Tagane, Hironobu Kanemitsu, Kazuki Tagawa, Kentaro Teramoto, Brian Park, Kuo-Fang Chung, Jer-Ming Hu, Tetsukazu Yahara, Michael J. Donoghue

pp. S87–S104

Geographical Variation in Community Divergence: Insights from Tropical Forest Monodominance by Ectomycorrhizal Trees Tadashi Fukami, Mifuyu Nakajima, Claire Fortunel, Paul V. A. Fine, Christopher Baraloto, Sabrina E. Russo, Kabir G. Peay, pp. S105–S122

vol. 190, supplement the american naturalist august 2017

Symposium Toward a Predictive Framework for Convergent Evolution: Integrating Natural History, Genetic Mechanisms, and Consequences for the Diversity of Life*

Anurag A. Agrawal†

Department of Ecology and Evolutionary Biology, Cornell University, Ithaca, New York 14853; and Department of Entomology, Cornell University, Ithaca, New York 14853 abstract: A charm of biology as a scientific discipline is the diversity ical diversity on our planet. The study of convergence is of life. Although this diversity can make laws of biology challenging to part of a program to discover repeated patterns and general discover, several repeated patterns and general principles govern evolu- principles that govern evolutionary diversification. Indeed, tionary diversification. Convergent evolution, the independent evolu- with the advent of non-model-omics, the study of conver- tion of similar phenotypes, has been at the heart of one approach to un- gent evolution is enjoying a new surge of interest, and this derstand generality in the evolutionary process. Yet understanding when symposium and special issue of the American Naturalist and why organismal traits and strategies repeatedly evolve has been a brings together a superb group to address classic and novel central challenge. These issues were the focus of the American Society questions on the topic. of Naturalists Vice Presidential Symposium in 2016 and are the subject fi of this collection of articles. Although naturalists have long made in- Here I de ne convergence as the independent evolution ferences about convergent evolution and its importance, there has been of similar phenotypes. As has been noted many times, the confusion in the interpretation of the pattern of convergence. Does intrigue of convergence was not lost on one of our greatest convergence primarily indicate adaptation or constraint? How often naturalists, Charles Darwin, when he identified traits such should convergence be expected? Are there general principles that as luminescent organs in seemingly distantly related in- would allow us to predict where and when and by what mechanisms sects and pollen packages in distantly related plants. Dar- convergent evolution should occur? What role does natural history “ play in advancing our understanding of general evolutionary princi- win (1859, p. 193) wrote: although the general appearance ples? In this introductory article, I address these questions, review sev- and function of the organ may be the same . . . some funda- eral generalizations about convergent evolution that have emerged over mental difference can generally be detected . . . Natural se- the past 15 years, and present a framework for advancing the study and lection . . . has sometimes modified in very nearly the same interpretation of convergence. Perhaps the most important emerging manner two parts in two organic beings, which owe but lit- conclusion is that the genetic mechanisms of convergent evolution tle of their structure in common to inheritance from the are phylogenetically conserved; that is, more closely related species tend same ancestor.” Darwin recognized that convergent traits to share the same genetic basis of traits, even when independently evolved. Finally, I highlight how the articles in this special issue further are not necessarily identical in all respects and that their evo- develop concepts, methodologies, and case studies at the frontier of our lution was largely independent, although not completely so, understanding of the causes and consequences of convergent evolution. given that all organisms ultimately share a common ancestor. Most convergent evolution falls under the umbrella of Keywords: adaptation, comparative biology, constraint, evolutionary what may be considered “constrained adaptation”—evolu- ecology, phylogenetic ecology, plant-insect interactions. tion that is limited by the strength of natural selection, genetic architecture, and fitness costs and benefits, all of which nar- Introduction row the number of possible evolutionary outcomes. Although convergence is often interpreted as evidence of both adapta- The search for convergent evolution and its causes is one tion and constraint (sagaciously reviewed by Losos [2011]), way to make sense of the wonderfully bewildering biolog- the two processes are intertwined and thus can be difficult to separate. In this context, constraint has been broadly de- * This issue originated as the 2016 Vice Presidential Symposium presented at fined as “restrictions or limitations on the course or outcome the annual meetings of the American Society of Naturalists. † ” “ E-mail: [email protected]. of evolution (Arnold 1992) or, more generally, the unequal ” Am. Nat. 2017. Vol. 190, pp. S1–S12. q 2017 by The University of Chicago. probability of outcomes in evolution (Schwenk 1994/1995). 0003-0147/2017/190S1-57341$15.00. All rights reserved. Issues of constraint will be important in my discussion of con- DOI: 10.1086/692111 vergent evolution, but because of the varied historical use and

This content downloaded from 132.236.027.111 on July 22, 2017 05:12:23 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S2 The American Naturalist abuse of the term (Futuyma 2010), I will frequently use the ganisms subject to the same selection, a critical approach to term “bias” to simply refer to the pattern of unequal probabil- understanding phylogenetic biases in evolution (Petsch- ity of outcome in evolution (table 1). enka et al. 2017; Rosenblum et al. 2017). As Losos (2011) pointed out, repeated trait-environment Convergence has also been at the center of thinking associations are suggestive of convergent adaptation, and about major ecological patterns and their causes. For exam- measures of natural selection and trait function can eluci- ple, Edwards et al. (2017) seek to understand the causes of date the adaptive nature of convergent traits (box 1). None- deciduousness in woody plants, a highly convergent trait theless, such measures do not address the extent to which that defines several global biomes. In some cases, it has been adaptation may be constrained. Many functional traits can hypothesized that convergence occurs at the community be under selection and yet may be limited in how they can level, driving greater similarity of species within a commu- evolve. As will be discussed later in this article, a pattern of nity than would be predicted from their phylogenetic relat- phylogenetic bias in both the degree of convergence and edness (Schluter 1986). In other cases, dissimilarity evolves the underlying mechanistic basis of convergent traits is due to the convergent evolution of community members reshaping our view of constrained adaptation. The challenge into different niches (Gillespie 2004; Losos 2009). Identify- of studying convergence in our era is thus interpreting pat- ing traits that may be involved in community assembly are terns in natural history on phylogenies, with repeated or critical, as such traits are the link between evolutionary his- divergent genetic mechanisms helping to elucidate general tory and ecological outcomes. Together, these approaches issues in evolution. This was the focus of the American So- represent an exciting frontier in the study of convergent ciety of Naturalists Vice Presidential Symposium in 2016. evolution whereby we are able to study the causes and con- In this introductory article, I aim to (1) introduce general sequences of repeated evolutionary change integrating from issues and concepts in the study of convergence and in par- genes to communities. ticular link the study of convergence to its roots in natural history, (2) provide a framework for addressing modern Natural History: The Roots of the Study questions in convergence and summarize the state of the of Convergent Evolution field in terms of progress and open questions, and (3) highlight themes addressed in the subsequent articles in this Natural history concerns the description of organisms (in- special issue. The current growth in studies of convergence cluding their scientific name), their traits (any phenotype has been fueled by increasingly sophisticated modern ana- or description of their genome), location and distribution, lytical tools of comparative biology, the ability to pinpoint and interactions with the biotic and abiotic environment. genetic mechanisms of convergence using molecular biol- Although natural history serves as a foundation for biology, ogy, and the expansion of observational (as opposed to ex- there continues to be controversy about the role that natural perimental) science in the past few decades. While perhaps history should play in advancing knowledge in modern biol- less common, experimental approaches and novel statisti- ogy (Greene 2005). Yet clearly “reciprocal relationships among cal methods in the study of convergence are helping to ex- the growth of robust theory, experimentation, and accurate plain patterns with process (Lenski 2017; Mahler et al. natural history” (Greene 2005, p. 23) are essential for ad- 2017). We are beginning to study the cellular and molecular vancement in our understanding of evolutionary biology. Nat- mechanisms leading to convergence in distantly related or- ural history continues to be an important source of inspira-

Table 1: Glossary of terms relevant to this article Adaptive phenotype A trait with a current function maintained by natural selection Constraint Restrictions or limitations on the course or outcome of evolution Biased convergence Convergent evolution where the trait consistently has the same mechanistic basis Contingency Chance events that shape evolutionary trajectories Convergent adaptation Repeated independent evolution of similar phenotypes by the same agent of natural selection Convergent evolution Repeated independent evolution of similar phenotypes Environment Abiotic conditions and biotic interactions Mechanism The causal basis of a particular trait, usually several ﬁne-grained traits that underlie a higher-level (coarse-grained) trait Phylogenetically conserved Convergent evolution where the trait consistently has the same mechanistic basis within a clade convergence Unbiased convergence Convergent evolution where the trait has distinct mechanistic bases, even within a clade

This content downloaded from 132.236.027.111 on July 22, 2017 05:12:23 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Convergence and Natural History S3 tion, is critical for knowing and conserving what we have on Another, perhaps more grandiose academic pursuit em- the planet, and, most germane tothis article, is central to the ploying convergence and natural history involved the search conceptual advancement of evolutionary biology in the for general patterns or rules that might govern nature. Alfred broadest sense. Using his classic approach of natural history Russel Wallace (and other contemporaries, including J. W. and phylogenetic thinking, Greene (2017) addresses conver- Slater and E. Haase) posed hypotheses about bright colora- gence in animal behavior, with implications for understand- tion, toxicity, and the acquisition of noxious substances ing our human selves as animals. from the host plants of butterfly caterpillars. Each of these Natural history in the context of comparative biology has naturalists, through their travels, observations, and records, historically been a critical hypothesis generator, which has noted a set of associations that lead to a hypothesis: brightly often yielded general patterns and sometimes pushed the en- colored lepidopteran larvae obtain toxins from their host velope of theory—especially in our understanding of conver- plants and use them in defense, typically against vertebrate gent evolution. Two examples from my own corner of biol- predators. Subsequently, in the 1950s there was a race to dem- ogy, plant-herbivore interactions, illustrate this point. onstrate such sequestration of toxins by caterpillars, which, By the early 1960s, the birth of chemical ecology as a once successful, gave rise to the search for general rules of se- discipline yielded possible explanations for chemical di- questration, aposematism, and mimicry (Reichstein et al. versity in plants and hypothesized consequences for major 1968). The monarch butterfly in particular was the first spe- patterns in biological diversification (Fraenkel 1959; Ehr- cies shown to sequester toxins from its host plant and be- lich and Raven 1964). Key hypotheses about the function came a model to address such general rules (Agrawal 2017). of plant chemistry were informed by careful natural his- In a series of studies led by Miriam Rothschild between tory observations on distantly related groups of organisms 1967 and 1973, the association between aposematism and (Fraenkel 1959). In a classic study on the “raison d’être sequestration was cemented by studies of several distantly of catnip,” Thomas Eisner (1964) reasoned that the chem- related insect groups that feed on related host plants (von ical components (including the terpene nepetalactone) Euw et al. 1967, 1971; Rothschild et al. 1970, 1973), ulti- causing catnip’s effects, feline euphoria, were a defense mately leading her to study 23 aposematic insect species of plants against herbivorous insects. His logic followed from six insect orders that sequestered the same toxic com- not from the incidental effects on cats but from the fact pounds from their host plants in the Apocynaceae. Such that nearly identical compounds were produced conver- findings opened the door to more rigorous statistical anal- gently by several insects, were insecticidal, and in some yses, conceptual development, and theory on defense, se- cases were ejected by the animals in response to risk of pre- questration, aposematism, and mimicry (Ruxton et al. 2004). dation. Eisner was an observer who poked around and fol- Although the frequent association of sequestration, apose- lowed his nose. The ejection of nepetalactone by molested matism, and mimicry now seems obvious, it was the re- beetles suggested an explanation for the function of this ciprocal interaction between natural history observations, compound in plants. studies of convergence, hypothesis testing, and theory de- Since then, many other defensive compounds have been velopment that led to the paradigm. Contemporary work found to be produced by both animals and plants, including addresses the genetic basis of convergent associations such alkaloids and cyanides. The genetic basis of such convergent as specialization and sequestration. In at least some cases, evolution in defenses continues to be revealed (Jensen et al. a diverse set of distantly related organisms utilizes the same 2011; Denoeud et al. 2014), and the hypothesis generation mechanism to achieve adaptation to the same environ- has not stopped with identifying the functional or mechanis- mental challenge (Dobler et al. 2012; Bramer et al. 2015; tic basis of these defensive traits. Conceptual developments Petschenka and Agrawal 2015). starting in the 1970s and advancing with phylogenetic thinking since 2000 have radically improved our understanding of Mechanisms of Convergence: A New Road which plant traits repeatedly evolve together to produce mul- to Assessing the Role of Constraint? tivariate strategies of defense (Feeny 1976; Kursar and Coley 2003; Agrawal and Fishbein 2006; Fine et al. 2006; Mooney Traits are hierarchical in nature, from complex phenotypes et al. 2010; Johnson et al. 2014; Mason et al. 2016). In each (sometimes referred to as coarse-grained traits) down to of these studies, either closely related species (many of which the products of gene expression (i.e., fine-grained traits, inhabit different habitats) or coexisting species (which span a beginning with messenger RNA). Higher-level traits in the broad swath of phylogenetic diversity) have been studied and hierarchy are typically dependent on several traits at lower arrayed along multiple axes of growth and defense. A cluster- levels (Conner and Hartl 2004), and the interpretation of ing of phenotypes has suggested repeatedly evolving syn- convergent evolution has different meanings at these differ- dromes that can be tested for their ecological effects and asso- ent hierarchical scales (Currie 2013). At the highest scale (or ciations with particular environments. coarse grain), an ecological outcome of several traits may

Box 1: A key for categorizing, interpreting, and advancing the study of convergent evolution

1. Has the trait repeatedly evolved? (yes)—the trait exhibits convergent evolution; go to 2 to address why. (no)—the trait is not convergent, and it may be difficult to assess the general importance of the trait except from logic (e.g., photosynthesis is clearly important) or functional studies. The lack of repeated evolution could potentially be addressed experimentally to understand constraints. 2. Is there a pattern of repeated trait-environment association? (yes)—the pattern of convergent evolution is likely due to natural selection; go to 2b (to understand the agents of selection) or 3 (to address biases in convergence). (no)—the trait appears to have evolved in different environmental contexts (not a convergent adaptation), suggesting alternate functions. Go to 2b, which may help refine your hypothesized selective environment if you are convinced that the trait might be a convergent adaptation. 2b. Do functional analyses or other evidence suggest that the convergently evolved trait is the target of repeated natural selection by the same agent? (yes)—the pattern of convergent adaptation is suggested; go to 3 to address selective mechanisms underlying convergence. (no)—the trait may be correlated with traits under selection but may not be directly subject to natural selection (i.e., pleiotropy or hitchhiking). 3. Is the mechanistic basis of the convergently evolved trait identical (or nearly so) in the independent origins of the trait? (yes)—biased convergence is suggested: there were very few possibilities for the way in which this trait could evolve; there is a single optimal solution or alternative solutions either are not as beneficial or occur with some delay and thus are less frequently realized. Go to 3b. (no)—multiple mechanisms of convergence, suggesting that natural selection has favored particular phenotypic space, and the outcome was achieved in distinct ways. Go to 4. 3b. Ask question 3 again if you can go down a mechanistic level. For example, if anthocyanins repeatedly evolved as a sunscreen, did the same genes evolve? Or if the same gene was involved in convergent evolution, was it the same specific site mutation? 4. Are the alternate mechanisms of convergent evolution phylogenetically conserved? (yes)—phylogenetically conserved convergence: history (i.e., phylogenetic signal) and selection are both important, and there is some predictability in what traits evolve by which mechanism in different lineages. Continue to 4b to understand why. (no)—unbiased convergence: selection has resulted in a limited set of phenotypic solutions across lineages, but the means to which we get there have been mechanistically diverse even within lineages. 4b. What is the basis of the phylogenetically conserved convergence in a trait? Does a particular phenotype, gene, genetic architecture (including duplication), or extrinsic factor (such as ecology or life-history strategy) predispose a lineage to evolve the same mechanistic basis for a convergent trait? Answering this question allows us to synthesize the joint role constraint and selection in adaptation. be the trait (or extended phenotype) that exhibits convergent evolution of such indirect defense as a high-level trait gent evolution. For example, flight has evolved at least four involves many distinct fine-grained traits, from housing times, in insects, dinosaurs, birds, and mammals. Addition- structures and food rewards to volatile attractants, traps, ally, the highly similar ecological niches employed by placen- and undiscovered mechanisms. Given that indirect defense tal versus marsupial mammals on different continents also is composed of multiple traits, we can begin to move down exemplifies convergent evolution of coarse-grained traits. If in scale to examine specific traits that generate indirect de- we zoom in, however, the physiological, morphological, and fense. In plants, a mechanism of indirect defense (one step behavioral mechanisms that underlie convergent niches in lower in the hierarchy) may be extrafloral nectaries. We might these radiations may or may not be the same (Luo 2007). next examine the chemical composition of extrafloral nectar Another such coarse-grained trait is the phenomenon of secreted or, more mechanistically, the genes that code for the indirect defense, where protection of an organism from en- nectary and its chemical constituents. These finer-grained emies is achieved through the attraction of, or patrolling by, traits are mechanisms that generate coarse-grained traits. animal bodyguards. The specific traits and conditions un- Understanding underlying mechanisms is key to deter- derlying indirect defense may be manifold, and these occur mining the causes of convergent evolution. For example, across diverse species and systems (e.g., ant-plants [Heil if convergent coarse-grained traits have distinct mecha- 2008], phloem-sucking hemipteran bugs [ant-tending; nisms, especially within a lineage, we may conclude that Styrsky and Eubanks 2007], leaf-chewing caterpillars [also the evolution of those traits was less constrained. In other ant-tending; Pierce et al. 2002], and even some fish and words, even within the backdrop of conserved genetic ar- other vertebrates [Poulin and Grutter 1996]). The conver- chitecture, traits, and ecology common to a clade, the same

This content downloaded from 132.236.027.111 on July 22, 2017 05:12:23 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Convergence and Natural History S5 evolutionary outcome was achieved by distinct means. tively evaluated in the coming decades. Understanding this Conversely, if mechanisms of convergence are themselves bias and its causes (such as a constraining effect of conserved phylogenetically conserved (the mechanisms are shared genetic architecture) will lead to a greater understanding among close relatives that have independently evolved the of convergence across scales. A potential exception to the trait), then convergence likely reflects constraint. I empha- pattern of phylogenetic bias in the mechanisms of conver- size that I am using the term “constraint” here to simply in- gence may be loss of functions, which can presumably occur dicate bias in the outcome of evolution, here driven by through many mechanisms (i.e., mutations) that have the shared traits in a lineage (table 1; box 1). same ultimate consequence (Manceau et al. 2010; Smith et al. 2013). Thus, when the mechanisms of convergence are phylogenetically conserved (i.e., a given mechanism re- Phylogenetic Bias in the Mechanisms of Convergence peatedly evolves within a lineage but distinct mechanisms Several examples illustrate phylogenetic bias in the mecha- evolve between lineages), a biasing effect of genetic architec- nisms of convergent traits. C4 photosynthesis in plants evolved ture, life history, or some other attribute of the lineage is over 60 times, and several different physiological means have suggested. been employed as carbon dioxide–concentrating mechanisms Biases in convergent evolution can also be studied ex- in C4 plants (Sage et al. 2011). Recent evidence demonstrates perimentally. Comparing what has evolved in the natural the convergent recruitment of particular genes within plant world with experimental populations has the potential to lineages, but different lineages employ different genes in con- reveal the extent of constraint (Weinreich et al. 2006; Stern trolling this photosynthetic pathway (Christin et al. 2015). 2013; van Ditmarsch et al. 2013). For example, both muta- Also consider red coloration in flowers, a highly convergent genesis screens for particular phenotypes and experimen- trait that is underpinned by distinct genetic mechanisms. tal evolution studies can reveal a greater number of possi- Evolutionary transitions to red flowers have repeatedly evolved bilities than exist in nature, while their natural counterparts in several lineages, but the prevalence of different mech- typically reveal biased outcomes. A next step will be to re- anisms of red pigment production differs among lineages veal why certain possible mechanisms are not realized. The (Ng and Smith 2016). In addition to these two plant exam- biasing effect of genetic architecture is the tip of the iceberg ples, resistance of animals (insects, amphibians, reptiles, and (Natarajan et al. 2016). mammals) to toxic cardiac glycosides is highly convergent, In some cases, advantageous mutations to groups of genes and although many of the specific genetic changes are con- involved in adaptation may be quite limited, even in highly served across lineages, there is again some phylogenetic bias controlled and benign laboratory conditions. In these cases, (Price et al. 1990; Croyle et al. 1997; Dobler et al. 2012; Ujvari convergence may simply be the product of constrained pos- et al. 2015). sibilities. However, in mutagenesis studies of microbes, plants, The mechanisms by which vertebrates adapt to high- and animals (mostly insects), screens have typically revealed elevation environments, typically involving the evolution of many more potentially functional mutations than those that altered oxygenation properties of hemoglobin, have been are realized through the natural evolutionary process (reviewed widely studied (Storz and Moriyama 2008) and represent an- in Stern 2013). For example, in the case of molecular adap- other case of phylogenetically conserved mechanisms in con- tations of animals to toxic cardiac glycosides (produced by vergent evolution. In a remarkable recent study, Natarajan plants and some animal prey), there are relatively few muta- et al. (2016) studied 28 phylogenetic pairs of bird species and tions that have repeatedly evolved (across many orders of found that the oxygen affinity of hemoglobin in highland insects and in some vertebrates) in the target site (e.g., the bird species was consistently higher than in closely related ubiquitous animal enzyme or the sodium-potassium ATPase; lowland species. Although the altered genes involved were Dobler et al. 2012, 2015; Ujvari et al. 2015). Nonetheless, consistent, the specific amino acid sites of substitution sub- animal cell mutagenesis studies reveal a greater possibility of stantially varied. Thus, evolution was convergent at the level potentially functional mutations in sodium-potassium ATPase, of individual genes but not within genes. Within humming- though this may not be the case when examined in the same birds, which had multiple origins of the highland habit, the genetic background (Price et al. 1990; Croyle et al. 1997). genetic substitutions were nearly identical. Moreover, Natara- Work is just now beginning where uncommon or unrealized jan et al. (2016) demonstrated that those genetic substitu- (but functional) mutations are being introduced to orga- tions were functional only in the hummingbird’s genetic back- nisms (or cell lines) to address why such mutations are not ground, not in that of more basal species. Thus, phylogenetic typically found in nature overall or at least in particular lin- bias in the mechanisms of convergent evolution may be driven eages. Presumably, the uncommon or unrealized mutations by the genetic background of a clade. do not exist because of negative fitness consequences due Phylogenetic bias in the mechanisms of convergence ap- to either pleiotropic effects on some other function or epi- pears to be a general result, one that should be quantita- static effects given the genetic background.

Contingency, Convergence, and Parallel Evolution: allowed us to ask, When there is a lack of convergence, does New Takes on Classic Questions this indicate that contingency dominates? In the LTEE, only one out of 12 replicate populations evolved the ability to utilize Contingency and convergence are sometimes thought to citrate as a carbon source, and this emerged after 30,000 gene- play mutually exclusive roles in evolutionary biology. In rations. Interestingly, the citrate-utilizing line shows repeat- Gould’sinfluential book, Wonderful Life (1990), he made able evolution of citrate use after 20,000 generations. This the case for the importance of contingent or chance events has been tested by restarting the experiment at different time that have had a profound effect on the diversification of life. points using frozen ancestors. The genetic architecture was Yet despite being a hero in Gould’sstory,SimonConway- potentiated by specific mutations after 20,000 generations, Morris subsequently took issue with several of Gould’stheses. and these mutations serve to repeatedly favor subsequent mu- In fact, in two subsequent books, Conway-Morris provided tations that allow for the use of citrate (Blount et al. 2012)— many examples of convergent evolution that he argued pro- but in only one out of 12 populations that started from the vided evidence against Gould’s conclusion of the impor- same clone. In an observational and phylogenetic context, tance of contingency (Conway-Morris 2003, 2015). Although similar patterns have been observed, where a starting char- it might be an attractive proposition that contingency and acter state has a strong influence on the subsequent macro- convergence are alternatives, is that truly the case? No. Con- evolutionary trajectory taken (Smith et al. 2013). vergence can be found at many levels, and yet contingent Various authors have proposed that the repeated evolu- events also shape patterns of life on Earth. The polarization tion of traits may be potentiated by past events. For exam- of the contingency and convergence debate has taken hold ple, an original trait may have a low probability of evolving conceptually but is largely a false dichotomy. (i.e., dependent on contingencies, as in the case of citrate Gould’s contingency was conceptualized at a deep scale use described above), yet what follows is highly repeatable. of chance events that shape the future (e.g., extinction of Interestingly, there is no agreed-upon terminology for such major lineages such as the dinosaurs). If certain traits evolved traits that then cause a bias, predisposition, or inherency only once and those lineages were subject to a contingent (Conway-Morris 2003) toward a particular evolutionary out- catastrophe, then clearly contingency rules. But the impor- come. Traits that enable (Donoghue 2005), channel (Gould tance of such contingent events is not under debate. Rather, 2002), and potentiate (Blount et al. 2008) or that are pre- what is at issue is whether convergence overrides the impor- cursors (Marazzi et al. 2012) to subsequent change have all tance of these chance events, because ultimately organismal been suggested. The extent of convergence and shared mech- traits would return to the convergent state. In all likelihood, anisms of the traits that convergently evolve may frequently the truth lies somewhere in between. While chance events depend on such phenomena. can reset or change the course of evolutionary history, many In experimentally tractable systems, if one has a hypoth- traits have and will continue to exhibit convergent evolution. esis for what causes the predisposition, this can be addressed Contingent events create evolutionary opportunities for con- by introducing specific changes while controlling for other vergent and divergent evolution. The geographic isolation aspects of genetic architecture. In some cases, a particular of large groups of placental and marsupial mammals allowed gene may cause the predisposition (i.e., epistasis), and in for remarkable convergence as these groups diversified. Al- other cases it may be a constellation of genetic factors (in- ternatively, certain other contingencies, such as mass extinc- cluding gene duplication; Riehle et al. 2001; Stern 2013). tion events, have surely shaped the diversity of life on Earth Even the ecology or life history of a group may cause these that we see today. events, and such predispositions may well be evident in phy- Experimental evolution has allowed a reconciliation of logenies (Marazzi et al. 2012). In either case, a somewhat issues and deep insights into the debate over parallel, con- infrequent event may predispose a lineage to convergently vergent, and contingent evolution (Elena and Lenski 2003). evolve subsequent adaptations. As discussed above and in For example, when a single clone is used to found replicate box 1, the same mechanism may repeatedly (convergently) populations, one can quantify whether parallel or conver- evolve within lineages, but different mechanisms may be gent evolution occurs (Arendt and Reznick 2008). Follow- convergently employed in other lineages, all toward the same ing Stayton (2015), here I refer to parallel evolution as two end of adapting to a particular selective agent. The examples lineages starting with the same character state and ultimately of phylogenetically conserved convergences discussed above fl evolving into a different but shared character state. Many (C4 photosynthesis, red ower color in plants, animal adap- traits in Lenski’s replicated long-term evolution experiment tation to high elevation, and animal resistance to cardiac (LTEE; Lenski 2017) have repeatedly evolved (typically in glycosides; Dobler et al. 2012; Christin et al. 2015; Ujvari et al. parallel; e.g., Meyer et al. 2010; Lenski et al. 2015). Alterna- 2015; Natarajan et al. 2016; Ng and Smith 2016) are outstand- tively, other traits have evolved only in single populations ing candidates to study the joint role of biases and selection (Blount et al. 2008). Experimental evolution approaches have in adaptation.

What Is the Expected Level of Convergence? convergences, perhaps depending on both extrinsic attributes such as ecological opportunity and intrinsic attributes such Convergent evolution is inherently intriguing because at some as genetic architecture. level it seems unexpected, especially when convergent traits Given the prevalence of epistasis, gene duplication, and have the same genetic underpinnings. Nonetheless, where effects of genetic background on the fitness advantages of possible, the extent of convergence should be contextual- particular mutations (e.g., Riehle et al. 2001; Weinreich et al. ized against a null model or hypothesis of what might be 2006; Tenaillon et al. 2012; Kryazhimskiy et al. 2014), the expected, both by chance and owing to selective processes. null expectation of convergence should be informed by knowl- Stayton has argued for the importance of genetic drift in de- edge of such genetic effects. Given a particular genetic back- veloping null models for the extent of convergence (Stayton ground, a new mutation may be more compatible or func- 2008). Below I consider ecological opportunity, genetic archi- tional and result in having higher fitness than without that tecture, loss versus gain of functions, and genes of small ver- specific background. As discussed above, we are at just the sus large effect when generating expectations for the level beginning of being able to understand biases in the extent of convergent evolution. of genetic convergence, but given that they exist, we should In a qualitative sense, the likelihood of convergence may work toward a predictive framework that would include null be driven by ecological opportunity, which may be based expectations based on genetic architecture. on what resources are in excess and can be easily employed. It has been suggested that loss of function mutations Viewed through this lens, it is not surprising that plants may on average be less mechanistically convergent than have evolved extrafloral nectaries many times (Weber and gain of functions (Arendt and Reznick 2008; Manceau et al. Keeler 2013). Sugar is often not limiting to plants, pre- 2010; Smith et al. 2013). It is certainly the case that con- dacious ants often live nearby, and plant genomes share vergent loss of pigmentation, which often occurs through many genes. Similarly, as discussed above, many phloem- mutations in the same gene, occurs by distinct mutations sucking hemipteran bugs have evolved indirect defense by at different sites (Protas et al. 2006). The rampant conver- ants who drink their sugary honeydew excretions (Styr- gent loss of the gas (swim) bladder among teleost fishes also sky and Eubanks 2007). Their honeydew is a waste prod- occurs by many distinct mechanisms; even losses within a uct, but nonetheless, there are closely related Hemiptera single species, wild-caught zebra fishes, occurred by over that are ant tended or not, and the evolution of tending typ- 20 distinct mechanisms (McCune and Carlson 2004). In a ically involves specific morphological traits as well as sense, loss of function is analogous to a highly polygenic changes in the amount or composition of honeydew (Völkl trait, since genetic modifications at many locations will et al. 1999; Fischer et al. 2002; Shingleton and Stern 2003; result in the same phenotypic outcome. Shingleton et al. 2005). Thus, the Hemiptera apparently Are convergent phenotypic traits that are controlled by have an ecological resource in excess, which, with some mod- many genes also less likely to have a common mechanistic ba- ification, may be highly sought after by an interaction part- sis? The genetic basis of body size is highly polygenic, and yet ner. geographical clines (e.g., across latitude) in phenotype are In contrast, one group of leaf-chewing insects, the Ly- highly repeatable across species and convergently evolve in caenid butterflies (and some close relatives) have evolved native and introduced populations (e.g., see Lomolino 2005 a nectar-producing gland and recruit ants as defensive body- for a review on vertebrates). Nonetheless, the genetic basis guards as well (Pierce et al. 2002). Within the Lepidoptera this of such convergent phenotypic clines appears to occur by di- maywellbetheonlycaseoftheevolutionofdefensebyants, vergent mechanisms (reviewed for Drosophila in Gilchrist and to my knowledge it has not been reported for beetles and Partridge 1999; Huey et al. 2000; Arendt and Reznick or flies. Some gall wasp lineages have repeatedly evolved the 2008). A polygenic basis for adaptation to high temperature ability to induce nectar production on their galls (produced may also underlie convergence in experimental studies of by their host plants), and they gain protection from patrol- Escherichia coli. For example, experimental adaptation to high ling ants (Inouye and Agrawal 2004; Nicholls et al. 2016). As temperature in over 100 replicate lines revealed that any far as I know, no vertebrate produces nectar to reward tend- given pair of lineages shared relatively few nonsynony- ing ants. Thus, among animals, the high ecological oppor- mous point mutations (2.6%), but at a higher hierarchical tunity for the evolution of defense by ants appears to have level, modified genes and operons were much more likely widely spurred ant-hemipteran mutualism (and the traits to be shared (120%; Tenaillon et al. 2012). In the adaptation that support it), whereas what caused the limited (single?) of fitness in 65 closely related yeast genotypes, Kryazhimskiy evolution of the same strategy in Lepidoptera is more diffi- et al. (2014) demonstrated that although fitness evolved cult to know. Perhaps it was a chance event or circumstance in highly repeatable trajectories, the specific mutations were in the history of the Lycaenids. The challenge for us now is highly variable and dependent on epistasis based on past fit- to somehow quantitatively address the expectation for such ness gains. In contrast to these studies on polygenic trait evo-

This content downloaded from 132.236.027.111 on July 22, 2017 05:12:23 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S8 The American Naturalist lution, in cases of resistance to particular toxins or diseases From an ecological perspective, there is tremendous in- with highly specific target sites (with genes of major effect), terest in understanding the species and traits that may cause much stronger convergence at the molecular level has been predictable outcomes in community assembly and compo- expected and observed (ffrench-Constant et al. 1998; Ash- sition. Such processes may occur over long periods of time field et al. 2004; Dobler et al. 2012; Farhat et al. 2013; Yang (which include evolutionary change) or over shorter pe- et al. 2013; Brodie and Brodie 2015; Ujvari et al. 2015). riods of time (e.g., following disturbance or during succession). Do particular species and their traits, once present in a community, fundamentally change the course of assembly? In the classical cases of adaptive radiations in Beyond Mechanism: Evolutionary and Ecological bounded communities (i.e., lakes and islands), there is of- Consequences of Convergence ten an association between convergently evolved animal Thus far, this article has focused on interpreting patterns ecomorphs and assembly of a fauna of related species of convergent evolution and integrating studies to under- (Losos et al. 1998; Gillespie 2004; Turner 2007). Nonethe- stand the causes of convergence. As discussed above, the less, cause and effect between the convergent traits and as- evolution of particular genes or traits may predispose a sembly of the community can be difficult to discern. The lineage to converge on a particular solution in response consideration of traits that modulate positive and negative to natural selection. In addition, when convergence occurs species interactions is an especially important frontier, as (at whatever mechanistic level), are there predictable out- both can impact community structure, though in different comes for the evolution and ecology of a lineage? ways. Fukami et al. (2017) take on the issue of convergence Perhaps the greatest interest among evolutionary biolo- in plant mycorrhizal associations and how these may have gists in consequences of convergence has come in the form shaped predictable outcomes of community structure in of searching for macroevolutionary key innovations, those diverse tropical forests. traits that evolve and allow organisms to interact with the In a general sense, convergent traits may strongly impact environment in new ways and increase the net lineage di- the process of community assembly through a few different versification rate (Hunter 1998). Although there are now processes. One metric of community assembly is the phyloge- sophisticated analytical tools to detect shifts in diversifica- netic structure of a community, defined as a nonrandom pat- tion rates on phylogenies, even if they occur only once tern of evolutionary relatedness among species (Kraft et al. (Rabosky 2014), single occurrences have little ability to 2007). In the simplest case where habitat filtering is critical, provide generality in terms of how traits impact diversifi- convergence in traits among species will result in a pattern cation. Even though we have little understanding of why of even (sometimes called over-dispersed) communities, those particular traits may facilitate speciation (or retard extinc- where species are less closely related than would be expected tion; Heard and Hauser 1995; Futuyma and Agrawal 2009), by chance (Cavender-Bares et al. 2004). Conversely, if traits recent work on plant-insect interactions suggests that such are phylogenetically conserved (e.g., tolerance of some stress), traits exist (Fine et al. 2004; Kaminski et al. 2010; Forister habitat filtering will result in phylogenetically clustered com- et al. 2011). munities. However, when traits are evolutionarily conserved Indeed, our best cases of traits acting as key innovations, and yet there is selection for dissimilarity within a commu- especially from traits that have repeatedly evolved, come nity (because of competition or other negative species inter- from the defensive traits of plants: latex (Farrell et al. 1991) actions among close relatives), the outcome will result in and extrafloral nectaries (Weber and Agrawal 2014). In both an even community. Finally, for the case where some habitat cases, the traits have convergently evolved in numerous filtering occurs as well as where species interactions are im- plant families, the traits are associated with defense against portant and species’ traits are convergent (likely the most herbivores, and lineages that have evolved these traits have typical scenario), the community outcome is unclear (Kraft higher diversification rates than closely related lineages et al. 2007). Thus, quantitatively parsing out the impact of lacking the traits. Convergent evolution of these defensive convergent evolution on community structure is a critical fron- traits has had profound and predictable consequences for tier in understanding the role of deep evolutionary processes plant diversification. Remarkably, on the coevolutionary in community assembly. flip side, herbivory as a trait in insects is also an iconic case of a convergent trait (feeding strategy) that is widely associated with elevated diversification rates (Mitter et al. Looking Back and Looking Forward 1988; Wiens et al. 2015). Future work will certainly identify other convergent key innovations, but more importantly, There are two related issues for why convergence provides we must address how and why the traits impact speciation such an important and compelling approach to biology, es- or extinction. pecially when describing phenomena or taking a natural

This content downloaded from 132.236.027.111 on July 22, 2017 05:12:23 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Convergence and Natural History S9 history approach. As pointed out by Felsenstein (1985), in- P. Jones, M. Kirkpatrick, R. Lenski, A. McCune, K. Niklas, dependent evolution provides statistical power in compar- R. Petipas, B. Reed, and S. Scheiner. This article was improved ative biology. Correcting for phylogeny, though sometimes by comments from J. Bronstein, J. Conner, J. Losos, J. Maron, having a pejorative connotation, is simply about making ro- A. McCune, J. Thompson, and M. Weber. bust statistical arguments about the association between two phenomena. Whether two traits show correlated evolution, a trait repeatedly evolves in a particular environment, Literature Cited or the repeated evolution of a trait is consistently associated Agrawal, A. A. 2017. Monarchs and milkweed. Princeton University fi with increased diversi cation, convergence in all cases Press, Princeton, NJ. provides power for stronger inference than if the associa- Agrawal, A. A., and M. Fishbein. 2006. Plant defense syndromes. tion occurred only once. This is not to downplay the impor- Ecology 87:S132–S149. tance of contingent events that occurred only once; indeed, Arendt, J., and D. Reznick. 2008. Convergence and parallelism these may be some of the most critical events in evolution- reconsidered: what have we learned about the genetics of adapta- – ary history. Nonetheless, to gain insight into whether there tion? Trends in Ecology and Evolution 23:26 32. Arnold, S. J. 1992. Constraints on phenotypic evolution. American are general rules in biology, there is no substitute for evolu- Naturalist 140(suppl.):S85–S107. tionary replication. Evolutionary replication allows us to Ashfield, T., L. E. Ong, K. Nobuta, C. M. Schneider, and R. W. Innes. come closer to understanding the cause of a particular asso- 2004. Convergent evolution of disease resistance gene specificity ciation because, if repeated in independent lineages, the ge- in two flowering plant families. Plant Cell 16:309–318. netic background, life history, and ecology of the distinct Blount, Z. D., J. E. Barrick, C. J. Davidson, and R. E. Lenski. 2012. lineages are likely to be different, and yet the association Genomic analysis of a key innovation in an experimental Esche- – still stands. richia coli population. Nature 489:513 518. Blount, Z. D., C. Z. Borland, and R. E. Lenski. 2008. Historical con- The current renaissance in descriptive biology has grown tingency and the evolution of a key innovation in an experimental from exploring the natural histories of the genomes of many population of Escherichia coli. Proceedings of the National Acad- organisms (e.g., Parker et al. 2013), engaging in new forms of emy of Sciences of the USA 105:7899–7906. discovery, and comparative analyses. There has been tremen- Bramer, C., S. Dobler, J. Deckert, M. Stemmer, and G. Petschenka. dous growth in the analysis of ecological gradients and under- 2015. Na1/K1-ATPase resistance and cardenolide sequestration: standing the natural pattern of species traits, distributions, basal adaptations to host plant toxins in the milkweed bugs and interactions along latitude, altitude, and other gradients. (Hemiptera: Lygaeidae: Lygaeinae). Proceedings of the Royal Soci- ety B 282:20142346. Part of this revolution has come from a renewed interest in Brodie, E. D., and E. D. Brodie. 2015. Predictably convergent evolu- trait-based ecology, the availability of phylogenetic information of sodium channels in the arms race between predators and tion, and climatic databases. Given this resurgence, I con- prey. Brain, Behavior, and Evolution 86:48–57. clude with four hopeful messages. First, let’s embrace this in- Cavender-Bares, J., D. D. Ackerly, D. A. Baum, and F. A. Bazzaz. terest, as natural history is an important basis of inspiration, 2004. Phylogenetic overdispersion in Floridian oak communities. – discovery, and the conservation of species. Second, the use of American Naturalist 163:823 843. convergence can provide rigor and replication to address Christin, P.-A., M. Arakaki, C. P. Osborne, and E. J. Edwards. 2015. Genetic enablers underlying the clustered evolutionary origins of some of the biggest questions, ranging from understanding C photosynthesis in angiosperms. Molecular Biology and Evolu- fi 4 constraints to diversi cation. Third, there is tremendous po- tion 32: 846–858. tential to understand the underlying drivers of convergence Conner, J. K., and D. L. Hartl. 2004. A primer of ecological genetics. by addressing the extent to which the mechanisms of conver- Sinauer, Sunderland, MA. gent traits are biased by the phylogenetic lineage in which Conway-Morris, S. 2003. Life’s solution: inevitable humans in a lonely they evolve (box 1). Last, the combined use of comparative universe. Cambridge University Press, Cambridge. ——— biology and experimentation (ranging from mutagenesis . 2015. The runes of evolution: how the universe became self- aware. Templeton, West Conshohocken, PA. screens to reciprocal transplant experiments) will be critical Croyle, M. L., A. L. Woo, and J. B. Lingrel. 1997. Extensive random in advancing biology (Weber and Agrawal 2012). mutagenesis analysis of the Na1/K1-ATPase a subunit identifies known and previously unidentified amino acid residues that alter ouabain sensitivity implications for ouabain binding. European Acknowledgments Journal of Biochemistry 248:488–495. Funding and stimulation to think about this topic was pro- Currie, A. 2013. Convergence as evidence. British Journal for the – vided by the John Templeton Foundation and especially Philosophy of Science 64:763 786. Darwin, C. 1859. On the origin of species by means of natural selec- P. Wason. Funding for this symposium and the speakers tion, or the preservation of favoured races in the struggle for life. was provided by the American Society of Naturalists. For Murray, London. discussion, I am grateful to L. Arcila Hernandez, K. Böröczky, Denoeud, F., L. Carretero-Paulet, A. Dereeper, G. Droc, R. Guyot, J. Elias, H. Greene, A. Hastings, K. Holmes, H. Inamine, M. Pietrella, C. Zheng, et al. 2014. The coffee genome provides in-

sight into the convergent evolution of caffeine biosynthesis. Sci- Futuyma, D. J. 2010. Evolutionary constraint and ecological conse- ence 345:1181–1184. quences. Evolution 64:1865–1884. Dobler, S., S. Dalla, V. Wagschal, and A. A. Agrawal. 2012. Futuyma, D. J., and A. A. Agrawal. 2009. Macroevolution and the bi- Community-wide convergent evolution in insect adaptation to toxic ological diversity of plants and herbivores. Proceedings of the Na- cardenolides by substitutions in the Na,K-ATPase. Proceedings of tional Academy of Sciences of the USA 106:18054–18061. the National Academy of Sciences of the USA 109:13040–13045. Gilchrist, A. S., and L. Partridge. 1999. A comparison of the genetic Dobler, S., G. Petschenka, V. Wagschal, and L. Flacht. 2015. Conver- basis of wing size divergence in three parallel body size clines of gent adaptive evolution—how insects master the challenge of car- Drosophila melanogaster. Genetics 153:1775–1787. diac glycoside-containing host plants. Entomologia Experimentalis Gillespie, R. 2004. Community assembly through adaptive radiation et Applicata 157:30–39. in Hawaiian spiders. Science 303:356–359. Donoghue, M. J. 2005. Key innovations, convergence, and success: Gould, S. J. 1990. Wonderful life: the Burgess Shale and the nature of macroevolutionary lessons from plant phylogeny. Paleobiology 31: history. Norton, New York. 77–93. ———. 2002. The structure of evolutionary theory. Harvard Uni- Edwards, E. J., D. S. Chatelet, B.-C. Chen, J. Y. Ong, S. Tagane, H. versity Press, Cambridge, MA. Kanemitsu, K. Tagawa, et al. 2017. Convergence, consilience, and Greene, H. W. 2005. Organisms in nature as a central focus for bi- the evolution of temperate deciduous forests. American Naturalist ology. Trends in Ecology and Evolution 20:23–27. 190(suppl.):S87–S104. ———. 2017. Evolutionary scenarios and primate natural history. Ehrlich, P. R., and P. H. Raven. 1964. Butterflies and plants: a study American Naturalist 190(suppl.):S69–S86. in coevolution. Evolution 18:586–608. Heard, S. B., and D. L. Hauser. 1995. Key evolutionary innovations Eisner, T. 1964. Catnip: its raison d’être. Science 146:1318–1320. and their ecological mechanisms. Historical Biology 10:151–173. Elena, S. F., and R. E. Lenski. 2003. Evolution experiments with Heil, M. 2008. Indirect defence via tritrophic interactions. New Phy- microorganisms: the dynamics and genetic bases of adaptation. tologist 178:41–61. Nature Reviews Genetics 4:457–469. Huey, R. B., G. W. Gilchrist, M. L. Carlson, D. Berrigan, and L. S. Farhat, M. R., B. J. Shapiro, K. J. Kieser, R. Sultana, K. R. Jacobson, Serra. 2000. Rapid evolution of a geographic cline in size in an in- T. C. Victor, R. M. Warren, et al. 2013. Genomic analysis identi- troduced fly. Science 287:308–309. fies targets of convergent positive selection in drug-resistant My- Hunter, J. P. 1998. Key innovations and the ecology of macroevolu- cobacterium tuberculosis. Nature Genetics 45:1183–1189. tion. Trends in Ecology and Evolution 13:31–36. Farrell, B. D., D. E. Dussourd, and C. Mitter. 1991. Escalation of Inouye, B. D., and A. A. Agrawal. 2004. Ant mutualists alter the plant defense: do latex and resin canals spur plant diversification? composition and attack rate of the parasitoid community for American Naturalist 138:881–900. the gall wasp Disholcaspis eldoradensis (Cynipidae). Ecological Feeny, P. P. 1976. Plant apparency and chemical defense. Pages 1–40 Entomology 29:692–696. in J. W. Wallace and R. L. Mansell, eds. Biochemical interaction Jensen, N. B., M. Zagrobelny, K. Hjernø, C. E. Olsen, J. Houghton- between plants and insects. Plenum, New York. Larsen, J. Borch, B. L. Møller, et al. 2011. Convergent evolution in Felsenstein, J. 1985. Phylogenies and the comparative method. Ameri- biosynthesis of cyanogenic defence compounds in plants and can Naturalist 125:1–15. insects. Nature Communications 2:273. ffrench-Constant, R. H., B. Pittendrigh, A. Vaughan, and N. An- Johnson, M. T., A. R. Ives, J. Ahern, and J. P. Salminen. 2014. Mac- thony. 1998. Why are there so few resistance-associated mutations roevolution of plant defenses against herbivores in the evening in insecticide target genes? Philosophical Transactions of the primroses. New Phytologist 203:267–279. Royal Society B 353:1685–1693. Kaminski, L. A., A. V. Freitas, and P. S. Oliveira. 2010. Interaction Fine, P. V. A., I. Mesones, and P. D. Coley. 2004. Herbivores pro- between mutualisms: ant-tended butterflies exploit enemy-free mote habitat specialization by trees in amazonian forests. Science space provided by ant-treehopper associations. American Natural- 305:663–665. ist 176:322–334. Fine, P. V. A., Z. J. Miller, I. Mesones, S. Irazuzta, H. M. Appel, Kraft, N. J., W. K. Cornwell, C. O. Webb, and D. D. Ackerly. 2007. M. H. H. Stevens, I. Saaksjarvi, et al. 2006. The growth-defense Trait evolution, community assembly, and the phylogenetic struc- trade-off and habitat specialization by plants in Amazonian ture of ecological communities. American Naturalist 170:271–283. forests. Ecology 87:S150–S162. Kryazhimskiy, S., D. P. Rice, E. R. Jerison, and M. M. Desai. 2014. Fischer, M. K., W. Völkl, R. Schopf, and K. H. Hoffmann. 2002. Age- Global epistasis makes adaptation predictable despite sequence- specific patterns in honeydew production and honeydew compo- level stochasticity. Science 344:1519–1522. sition in the aphid Metopeurum fuscoviride: implications for ant- Kursar, T. A., and P. D. Coley. 2003. Convergence in defense syn- attendance. Journal of Insect Physiology 48:319–326. dromes of young leaves in tropical rainforests. Biochemical Sys- Forister, M. L., Z. Gompert, C. C. Nice, G. W. Forister, and J. A. tematics and Ecology 31:929–949. Fordyce. 2011. Ant association facilitates the evolution of diet Lenski, R. E. 2017. Convergence and divergence in a long-term breadth in a lycaenid butterfly. Proceedings of the Royal Society experiment with bacteria. American Naturalist 190(suppl.):S57– B 278:1539–1547. S68. Fraenkel, G. 1959. The raison d’etre of secondary plant substances. Lenski, R. E., M. J. Wiser, N. Ribeck, Z. D. Blount, J. R. Nahum, J. J. Science 129:1466–1470. Morris, L. Zaman, et al. 2015. Sustained fitness gains and variabil- Fukami, T., M. Nakajima, C. Fortunel, P. V. A. Fine, C. Baraloto, ity in fitness trajectories in the long-term evolution experiment S. E. Russo, and K. G. Peay. 2017. Geographical variation in com- with Escherichia coli. Proceedings of the Royal Society B 282: munity divergence: insights from tropical forest monodominance 20152292. by ectomycorrhizal trees. American Naturalist 190(suppl.):S105– Lomolino, M. V. 2005. Body size evolution in insular vertebrates: gen- S122. erality of the island rule. Journal of Biogeography 32:1683–1699.

Losos, J. 2009. Lizards in an evolutionary tree: ecology and adaptive Poulin, R., and A. S. Grutter. 1996. Cleaning symbioses: proximate radiation of anoles. University of California Press, Oakland. and adaptive explanations. Bioscience 46:512–. 517 Losos, J. B. 2011. Convergence, adaptation, and constraint. Evolution Price, E., D. Rice, and J. Lingrel. 1990. Structure-function studies of 65:1827–1840. Na,K-ATPase: site-directed mutagenesis of the border residues Losos, J. B., T. R. Jackman, A. Larson, K. de Queiroz, and L. Rodriguez- from the H1-H2 extracellular domain of the a subunit. Journal Schettino. 1998. Contingency and determinism in replicated adaptive of Biological Chemistry 265:6638–6641. radiations of island lizards. Science 279:2115–2118. Protas, M. E., C. Hersey, D. Kochanek, Y. Zhou, H. Wilkens, W. R. Luo, Z.-X. 2007. Transformation and diversification in early mam- Jeffery, L. I. Zon, et al. 2006. Genetic analysis of cavefish reveals mal evolution. Nature 450:1011–1019. molecular convergence in the evolution of albinism. Nature Ge- Mahler, D. L., M. G. Weber, C. E. Wagner, and T. Ingram. 2017. Pat- netics 38:107–111. tern and process in the comparative study of convergent evolu- Rabosky, D. L. 2014. Automatic detection of key innovations, rate shifts, tion. American Naturalist 190(suppl.):S13–S28. and diversity-dependence on phylogenetic trees. PLoS ONE 9:e89543. Manceau, M., V. S. Domingues, C. R. Linnen, E. B. Rosenblum, and Reichstein, T., J. Voneuw, J. A. Parsons, and M. Rothschild. 1968. H. E. Hoekstra. 2010. Convergence in pigmentation at multiple Heart poisons in monarch butterfly. Science 161:861–866. levels: mutations, genes and function. Philosophical Transactions Riehle, M. M., A. F. Bennett, and A. D. Long. 2001. Genetic architec- of the Royal Society B 365:2439–2450. ture of thermal adaptation in Escherichia coli. Proceedings of the Marazzi, B., C. Ané, M. F. Simon, A. Delgado-Salinas, M. Luckow, National Academy of Sciences of the USA 98:525–530. and M. J. Sanderson. 2012. Locating evolutionary precursors on Rosenblum, E. B., C. E. Parent, E. T. Diepeveen, and K. Bi. 2017. a phylogenetic tree. Evolution 66:3918–3930. Convergent phenotypic evolution despite contrasting demo- Mason, C. M., A. W. Bowsher, B. L. Crowell, R. M. Celoy, C. J. Tsai, graphic histories in the fauna of White Sands. American Natural- and L. A. Donovan. 2016. Macroevolution of leaf defenses and ist 190(suppl.):S44–S56. secondary metabolites across the genus Helianthus. New Phy- Rothschild, M., J. von Euw, and T. Reichstein. 1970. Cardiac glyco- tologist 209:1720–1733. sides in oleander aphid, Aphis nerii. Journal of Insect Physiology McCune, A. R., and R. L. Carlson. 2004. Twenty ways to lose your 16:1141–1145. bladder: common natural mutants in zebrafish and widespread ———. 1973. Cardiac glycosides (heart poisons) in the polka-dot convergence of swim bladder loss among teleost fishes. Evolution moth Syntomeida epilais Walk. (Ctenuchidae: Lep.) with some ob- and Development 6:246–259. servations on the toxic qualities of Amata (p Syntomis) phegea Meyer, J. R., A. A. Agrawal, R. T. Quick, D. T. Dobias, D. Schneider, (L.). Proceedings of the Royal Society B 183:227–247. and R. E. Lenski. 2010. Parallel changes in host resistance to viral Ruxton, G. D., T. N. Sherratt, and M. P. Speed. 2004. Avoiding attack: infection during 45,000 generations of relaxed selection. Evolu- the evolutionary ecology of crypsis, warning signals and mimicry. tion 64:3024–3034. Oxford University Press, Oxford.

Mitter, C., B. Farrell, and B. Wiegmann. 1988. The phylogenetic Sage, R. F., P.-A. Christin, and E. J. Edwards. 2011. The C4 plant lineages study of adaptive zones: has phytophagy promoted insect diversi- of planet Earth. Journal of Experimental Botany 62:3155–3169. fication? American Naturalist 132:107–128. Schluter, D. 1986. Tests for similarity and convergence of finch com- Mooney, K. A., R. Halitschke, A. Kessler, and A. A. Agrawal. 2010. munities. Ecology 67:1073–1085. Evolutionary trade-offs in plants mediate the strength of trophic Schwenk, K. 1994/1995. A utilitarian approach to evolutionary con- cascades. Science 327:1642–1644. straint. Zoology 98:251–262. Natarajan, C., F. G. Hoffmann, R. E. Weber, A. Fago, C. C. Witt, and Shingleton, A. W., and D. L. Stern. 2003. Molecular phylogenetic ev- J. F. Storz. 2016. Predictable convergence in hemoglobin function idence for multiple gains or losses of ant mutualism within the has unpredictable molecular underpinnings. Science 354:336–339. aphid genus Chaitophorus. Molecular Phylogenetics and Evolu- Ng, J., and S. D. Smith. 2016. Widespread flower color convergence in tion 26:26–35. Solanaceae via alternate biochemical pathways. New Phytologist Shingleton, A. W., D. L. Stern, and W. A. Foster. 2005. The origin of 209:407–417. a mutualism: a morphological trait promoting the evolution of Nicholls, J. A., G. Melika, G. N. Stone, and M. A. McPeek. 2016. ant-aphid mutualisms. Evolution 59:921–926. Sweet tetra-trophic interactions: multiple evolution of nectar se- Smith, S. D., S. Wang, and M. D. Rausher. 2013. Functional evolu- cretion, a defensive extended phenotype in Cynipid gall wasps. tion of an anthocyanin pathway enzyme during a flower color American Naturalist 189:67–77. transition. Molecular Biology and Evolution 30:602–612. Parker, J., G. Tsagkogeorga, J. A. Cotton, Y. Liu, P. Provero, E. Stupka, Stayton, C. T. 2008. Is convergence surprising? an examination of and S. J. Rossiter. 2013. Genome-wide signatures of convergent the frequency of convergence in simulated datasets. Journal of evolution in echolocating mammals. Nature 502:228–231. Theoretical Biology 252:1–14. Petschenka, G., and A. A. Agrawal. 2015. Milkweed butterfly resis- ———. 2015. What does convergent evolution mean? the interpre- tance to plant toxins is linked to sequestration, not coping with tation of convergence and its implications in the search for limits a toxic diet. Proceedings of the Royal Society B 282:20151865. to evolution. Interface Focus 5:20150039. Petschenka, G., V. Wagschal, M. von Tschirnhaus, A. Donath, and Stern, D. L. 2013. The genetic causes of convergent evolution. Nature S. Dobler. 2017. Convergently evolved toxic secondary metabolites Reviews Genetics 14:751–764. in plants drive the parallel molecular evolution of insect resis- Storz, J. F., and H. Moriyama. 2008. Mechanisms of hemoglobin ad- tance. American Naturalist 190(suppl.):S29–S43. aptation to high altitude hypoxia. High Altitude Medicine and Bi- Pierce, N. E., M. F. Braby, A. Heath, D. J. Lohman, J. Mathew, D. B. ology 9:148–157. Rand, and M. A. Travassos. 2002. The ecology and evolution of Styrsky, J. D., and M. D. Eubanks. 2007. Ecological consequences of ant association in the Lycaenidae (Lepidoptera). Annual Review interactions between ants and honeydew-producing insects. Pro- of Entomology 47:733–771. ceedings of the Royal Society B 274:151–164.

Tenaillon, O., A. Rodríguez-Verdugo, R. L. Gaut, P. McDonald, A. F. Weber, M. G., and A. A. Agrawal. 2012. Phylogeny, ecology, and the Bennett, A. D. Long, and B. S. Gaut. 2012. The molecular diversity coupling of comparative and experimental approaches. Trends in of adaptive convergence. Science 335:457–461. Ecology and Evolution 27:394–403. Turner, G. F. 2007. Adaptive radiation of cichlid fish. Current Biol- ———. 2014. Defense mutualisms enhance plant diversification. ogy 17:R827–R831. Proceedings of the National Academy of Sciences of the USA Ujvari, B., N. R. Casewell, K. Sunagar, K. Arbuckle, W. Wüster, N. Lo, 111:16442–16447. D. O’Meally, et al. 2015. Widespread convergence in toxin resistance Weber, M. G., and K. H. Keeler. 2013. The phylogenetic distribution by predictable molecular evolution. Proceedings of the National of extrafloral nectaries in plants. Annals of Botany 111:1251– Academy of Sciences of the USA 112:11911–11916. 1261. van Ditmarsch, D., K. E. Boyle, H. Sakhtah, J. E. Oyler, C. D. Nadell, Weinreich, D. M., N. F. Delaney, M. A. DePristo, and D. L. Hartl. É. Déziel, L. E. Dietrich, et al. 2013. Convergent evolution of hy- 2006. Darwinian evolution can follow only very few mutational perswarming leads to impaired biofilm formation in pathogenic paths to fitter proteins. Science 312:111–114. bacteria. Cell Reports 4:697–708. Wiens, J. J., R. T. Lapoint, and N. K. Whiteman. 2015. Herbivory in- Völkl, W., J. Woodring, M. Fischer, M. W. Lorenz, and K. H. creases diversification across insect clades. Nature Communications Hoffmann. 1999. Ant-aphid mutualisms: the impact of honeydew 6:8370. production and honeydew sugar composition on ant preferences. Yang,S.,J.Li,X.Zhang,Q.Zhang,J.Huang,J.-Q.Chen,D.L.Hartl, Oecologia 118:483–491. et al. 2013. Rapidly evolving R genes in diverse grass species con- von Euw, J., L. Fishelson, J. A. Parsons, T. Reichstein, and M. fer resistance to rice blast disease. Proceedings of the National Rothschild. 1967. Cardenolides (heart poisons) in a grasshopper Academy of Sciences of the USA 110:18572–18577. feeding on milkweeds. Nature 214:35–39. von Euw, J., T. Reichstein, and M. Rothschild. 1971. Heart poisons (cardiac glycosides) in the lygaeid bugs Caenocoris nerii and Spilostethus pandurus. Insect Biochemistry 1:373–384. Editor: Judith L. Bronstein

“Phases of Kangaroo Life. At Peace.” From “General Notes: Zoölogy” (The American Naturalist, 1884, 12:1271–1279).

Symposium Pattern and Process in the Comparative Study of Convergent Evolution*

D. Luke Mahler,1,† Marjorie G. Weber,2 Catherine E. Wagner,3 and Travis Ingram4

1. Department of Ecology and Evolutionary Biology, University of Toronto, Toronto, Ontario M5S 3B2, Canada; 2. Department of Plant Biology, Michigan State University, East Lansing, Michigan 48824; 3. Biodiversity Institute and Department of Botany, University of Wyoming, Laramie, Wyoming 82071; 4. Department of Zoology, University of Otago, Dunedin, Otago 9016, New Zealand

abstract: Understanding processes that have shaped broad-scale Introduction biodiversity patterns is a fundamental goal in evolutionary biology. The phenomenon of phenotypic convergence plays a fun- The development of phylogenetic comparative methods has yielded damental role in the study of organic evolution. Although a tool kit for analyzing contemporary patterns by explicitly modeling processes of change in the past, providing neontologists tools for ask- convergence itself is not necessarily indicative of any par- ing questions previously accessible only for select taxa via the fossil rec- ticular evolutionary process (Losos 2011a; Speed and Ar- ord or laboratory experimentation. The comparative approach, how- buckle 2016), the repeated appearance of similar forms in ever, differs operationally from alternative approaches to studying disparate lineages stands in apparent contrast to the ex- convergence in that, for studies of only extant species, convergence pected pattern of divergence over time during evolution must be inferred using evolutionary process models rather than being and thus demands an explanation (Wake et al. 2011). Con- directly measured. As a result, investigation of evolutionary pattern vergent evolution has been attributed to a great diversity of and process cannot be decoupled in comparative studies of convergence, even though such a decoupling could in theory guard against causes, at times being invoked as evidence for the impor- adaptationist bias. Assumptions about evolutionary process underly- tance of multiple and sometimes opposing evolutionary ing comparative tools can shape the inference of convergent pattern in processes. As such, confusion persists around the connec- sometimes profound ways and can color interpretation of such pat- tion between patterns of convergence, mechanisms of evo- terns. We discuss these issues and other limitations common to most lution, and modeled processes in comparative methods (see phylogenetic comparative approaches and suggest ways that they can box 1 for deﬁnitions of these terms). be avoided in practice. We conclude by promoting a multipronged ap- Our aim is to clarify these concepts and show how they proach to studying convergence that integrates comparative methods with complementary tests of evolutionary mechanisms and includes eco- relate to common assumptions in the comparative study logical and biogeographical perspectives. Carefully employed, the com- of convergence, recommending best practices and advocat- parative method remains a powerful tool for enriching our understanding integrative ways to link convergent patterns with mech- ing of convergence in macroevolution, especially for investigation of why anistic hypotheses. We begin by reviewing the comparative convergence occurs in some settings but not others. study of phenotypic convergence in continuously valued Keywords: convergence, phylogenetic comparative methods, adap- traits and the factors that make it uniquely challenging to tive radiation, evolutionary process, adaptation. study deductively. We then discuss the relationship between pattern and process in the comparative study of convergence, giving special attention to the facts that (1) all comparative tools for studying convergence make inductive inferences about convergence and thus assume an underlying model of evolution and (2) despite the assumption of a modeled pro- * This issue originated as the 2016 Vice Presidential Symposium presented at cess, there can often be a many-to-one mapping of real evo- the annual meetings of the American Society of Naturalists. † lutionary processes to modeled processes. Finally, we argue Corresponding author; e-mail: [email protected]. that because these limitations can hinder interpretation, in- ORCIDs: Mahler, http://orcid.org/0000-0001-6483-3667; Weber, http://orcid tegration of comparative phylogenetic models of conver- .org/0000-0001-8629-6284; Wagner, http://orcid.org/0000-0001-8585-6120; In- gram, http://orcid.org/0000-0003-0709-5260. gence with other forms of inference is needed to increase ﬁ Am. Nat. 2017. Vol. 190, pp. S13–S28. q 2017 by The University of Chicago. con dence in links between pattern and process. We sug- 0003-0147/2017/190S1-57359$15.00. All rights reserved. gest several research directions to improve future prospects DOI: 10.1086/692648 for gaining meaningful insights about evolutionary pro-

Box 1: Deﬁnitions

Convergence We define convergent phenotypic evolution (or “convergence”) as the pattern of evolution in which species in two independently evolving lineages become phenotypically similar. Importantly, such a pattern is defined as convergence regardless of the evolutionary process that gave rise to it (Stayton 2015a). In this article, we will discuss convergent patterns in which lineages evolve greater phenotypic similarity than was exhibited by their ancestors. We do not discount convergence definitions that include patterns in which two or more lineages independently evolve similar traits even when their ancestors were also similar (i.e., parallelism; Arendt and Reznick 2008a, 2008b; Leander 2008; Scotland 2011; Wake et al. 2011; Rosenblum et al. 2014); nonetheless, for clarity and because most comparative convergence tools are specialized for the study of the evolution of greater similarity among de- scendants than ancestors, we do not discuss parallelism in this article. As any statement about convergence requires information about ancestral phenotypes, our discussion of the role of evolutionary process models in inferring ancestral phenotypes using phylogenetic comparative methods (see the main text) should be relevant regardless of the specificdefinition of convergence assumed.

Evolutionary Process/Evolutionary Mechanism Evolutionary processes, which we use interchangeably with the term “evolutionary mechanisms,” refer to speciﬁc underlying agents of evolutionary change, such as genetic drift or natural selection, that give rise to observed patterns of organismal diversity (Eldredge and Cracraft 1980; Chapleau et al. 1988). A fundamental goal in evolutionary biology is to test hypotheses about the evolutionary mechanisms that shape patterns of biological diversity through time.

Evolutionary Process Model In this article, the term “evolutionary process model” refers to a model of the evolutionary process assumed by a phylogenetic comparative method. Most such methods assume phenomenological evolutionary models, which means that they are thought to represent macroevolutionary expectations arising from a given evolutionary process but do not explicitly model the mechanism of evolutionary change itself. For this reason, some evolutionary process models may be consistent with more than one mechanism of evolutionary change (see the main text).

Ecological Mechanism We refer to “ecological mechanism” as an ecological interaction that is an agent of natural selection on an organism. Ecological mechanisms describe how an organism interacts with its environment or with other species, and include competition, mutualism, and abiotic tolerances, among others. There has been relatively little investigation of how different ecological mechanisms are expected to shape patterns of convergent evolution.

cesses from the comparative study of convergent patterns, interesting as evidence for adaptation or for the importance including extended model development, pairing compara- of ecology in the evolution of phenotypic diversity. For ex- tive study with other approaches for studying the evolution- ample, the evolution of pale dorsal coloration in numerous ary process, and better incorporating fossil information. vertebrates and invertebrates inhabiting White Sands Na- tional Monument indisputably reﬂects adaptation for increased crypsis (Rosenblum et al. 2010), and the repeated evo- The Potential Causes of Macroevolutionary Convergence lution of cold tolerance in distantly related conifers is clearly attributable to adaptation to similar environmental condi- The investigation of convergence plays an important role in tions (Yeaman et al. 2016). Large-scale convergence between evolutionary study. To many, instances of convergence are whole faunas occurring in similar ecological communities

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:28 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Pattern, Process, and Convergence S15 and environments suggests that evolution can in some cir- as when the fossil record preserves unambiguous ancestor- cumstances exhibit a surprising degree of predictability and descendant sequences (e.g., Bell 1987; McCune 1987), when that ecological factors can repeatedly and predictably shape source populations that gave rise to convergent daughter macroevolutionary diversification (Nevo 1979; Conway Mor- populations are still extant (e.g., Hoekstra et al. 2006; Ro- ris 2010; Mahler et al. 2013; Esquerré and Keogh 2016; Moen senblum et al. 2010), and when convergence is particularly et al. 2016). rapid (e.g., Pascoal et al. 2014), including in laboratory ex- Alternatively, convergence has been taken as evidence for periments on organisms with very short generation times constraints on the production of variation (Haldane 1932; (e.g., Meyer et al. 2012; Spor et al. 2014). Studies documenting Maynard Smith et al. 1985; Schluter 1996), manifested at convergence in this way represent some of the best and most one or more hierarchical levels of biological organization cherished evidence for convergent evolution, but they are un- (Wake and Larson 1987; Wake et al. 2011). Such constraints common and are limited in what they can tell us about the can arise from biased mutation, pleiotropic gene networks, evolution of convergent patterns across the tree of life. structural limitations, or limits on phenotypic variation im- Convergence has a long history of study despite the lim- posed by ontogenetically nested developmental sequences ited opportunity for deductive inference, but aside from (Gould 1980; Alberch 1982; Oster and Alberch 1982; Ar- exceptional cases such as those described above, much his- nold 1992; McCune and Carlson 2004; Brakefield 2011; torical study of convergence has been qualitative in nature. Streisfeld and Rausher 2011; Stern 2013). For example, Most of the canonical examples of convergence described Bright et al. (2016) concluded that the vast majority of cranio- in introductory biology textbooks, such as the streamlined facial variation in predatory birds was attributable to con- profiles of fast-moving pelagic vertebrates or the winter pel- served patterns of allometry and covariation between traits age of Arctic foxes and snowshoe hares, are so visually strik- (phenotypic integration), with only a small amount of resid- ing and occur in such distant relatives that there can be little ual variation explained by feeding ecology. Convergence question about their convergent origins. What was histori- resulting from such factors suggests that genetic or develop- cally lacking was a cohesive quantitative framework for in- mental constraints play an important and perhaps dominant ferring the trajectories of convergence; the lack of such a role in shaping the evolution of phenotypic diversity (Gould structure imposed formidable limits on the study of the 1980, 2002; Alberch 1982; Wake and Larson 1987). ultimate drivers of convergent evolution. This framework Finally, some degree of evolutionary convergence may be emerged with the union of the comparative method with an an expected outcome of chance, especially for traits with low explicitly phylogenetic perspective in the 1980s and 1990s dimensionality (Wagner 2000; Stayton 2008). These possibil- (Felsenstein 1985; Harvey and Pagel 1991). ities are not mutually exclusive, of course, and convergence due to chance may be most likely in scenarios in which Phylogenetic Approaches to Studying Convergence constraints on the generation of variation restrict evolutionary outcomes to a small set of phenotypes with compa- The advent of phylogenetic comparative approaches to study- rable fitness (Losos 2011a; Spor et al. 2014). Likewise, some ing trait evolution expanded the scope of convergence studies, authors have uncovered phylogenetic patterns suggesting making it possible to test quantitative hypotheses about con- that certain convergent adaptations are hierarchically convergent evolution in any group with sufficient phylogenetic strained by the presence or absence of preadaptations (Ma- and phenotypic information. This development effectively razzi et al. 2012; Beaulieu et al. 2013). For example, the con- opened the quantitative study of convergence, previously vergent evolution of arboreal adaptations in oribatid mites limited to exceptional cases, to the entire tree of life. When appears contingent on the prior evolution of sexual repro- viewed within a phylogenetic comparative framework, re- duction and strong sclerotization (Maraun et al. 2009). peated convergence of any kind provides researchers with a degree of statistical replication rarely afforded to students of the explicitly historical science of evolution. Regardless Challenges to Measuring and Studying of the question of interest, the repeated evolution of similar Convergent Evolution phenotypes in disparate lineages provides independent rep- Despite its importance in evolutionary inquiry, conver- licates in a grand, unplanned evolutionary experiment. The gence is difficult to directly identify, and once identified, ability to repeatedly query putative cause and evolutionary ef- it is difficult to ascribe to underlying mechanisms with cer- fect allows investigators to overcome the risks of “just so” sto- tainty (Maynard Smith et al. 1985). A principal challenge in rytelling (Gould and Lewontin 1979) in the study of adapta- studying convergence is that it is rarely possible to study us- tion, structural constraint, and even chance (Maynard Smith ing deductive inference due to the difficulty of observing et al. 1985; Harvey and Pagel 1991; Losos 2011a). both ancestor and descendant phenotypes. Empirically, this The development of tools for investigating convergence is generally possible only in exceptional circumstances, such accelerated especially rapidly during the last decade (Speed

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:28 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S16 The American Naturalist and Arbuckle 2016). New inductive tools provide a wide va- cause they have adapted to similar arboreal substrates riety of methods for quantifying convergence—including (Williams 1983; Losos 2009), not because they have been se- its occurrence, frequency, extent, and historical trajectory. lected to resemble one another. There are a few exceptions Such tools come in a variety of forms, the details of which where species are directly selected to converge with one an- are described in depth elsewhere (see Mahler and Ingram other, such as mimicry complexes (Endler 1981) or charac- 2014; Stayton 2015a; Arbuckle and Speed 2016; Speed ter convergence driven by competition for nonsubstitutable and Arbuckle 2016 for more in-depth descriptions). Most, resources (MacArthur and Levins 1967; Abrams 1987; Fox however, can be classed into one of three categories: (1) sta- and Vasseur 2008), but otherwise the evolutionary processes tistical indices, which measure an expected emergent fea- responsible for increasing similarity in a converging lineage ture of convergent phenotypic evolution on a phylogeny; are blind to the phenotype of the lineage to which it is con- (2) ancestor reconstruction methods, in which ancestral verging. As standard evolutionary theory can explain the phenotypes are estimated under some assumed evolutionary processes by which independent lineages evolve to be more model and then used to identify and quantify convergence similar, there is no need for a special theory of convergence patterns; and (3) model-fitting approaches, in which evolu- (Speed and Arbuckle 2016), and convergence is thus best tionary models explicitly incorporating processes expected defined as a pattern. to cause convergence are parameterized and compared to Although we agree with several recent reviews that con- models in which any convergence occurs by chance. vergence should be defined as a pattern, we argue that when All existing comparative methods for studying conver- using the comparative approach, convergence must be gence have particular limitations and weaknesses, as we will studied with the evolutionary process in mind. We there- discuss below. More importantly, however, is that all of these fore reject arguments that comparative analyses of conver- tools make inductive inferences about convergence, an ap- gence should proceed in a two-step manner—first testing proach that has practical consequences for the design and infor convergent pattern, and then investigating potential terpretation of comparative studies. Regardless of the method evolutionary processes responsible for this pattern (Stayton used, the results must be interpreted in light of an assumed 2015a, 2015b; Speed and Arbuckle 2016). In a recent re- model of the evolutionary process; as a result, it is not possible view, Stayton (2015a) made a case for this two-step ap- to decouple the quantification of convergent pattern from the proach, specifi cally criticizing the use of process-based com- study of evolutionary process using comparative data, as has parative tools for identifying convergence. Stayton’sconcern been recently recommended (Stayton 2015a). This issue is is defensible—in applying an evolutionary model to com- compounded by the fact that many widely used evolutionary parative data (e.g., multiple-optimum Ornstein-Uhlenbeck process models may plausibly represent multiple underlying [OU]; Butler and King 2004), the investigator assumes that evolutionary mechanisms—a potential many-to-one map- the process being modeled is an appropriate representation ping of mechanism to model. At the heart of these issues is of trait evolution in lineages of interest. In the absence of the complex relationship between pattern and process in appropriate model comparison, this could bias an investi- comparative biology. gation toward a particular evolutionary explanation for convergence. The risk of bias is arguably greatest for adaptive explanations (sensu Gould and Lewontin 1979), and Pattern and Process Stayton marshaled evidence for adaptationist bias in con- An important property of any evolutionary phenomenon vergence definitions provided in many prominent biology is the extent to which it represents a pattern versus a pro- texts (Stayton 2015a). To safeguard against adaptationist cess (box 1). For example, the term “adaptation” is mean- biases, Stayton recommended that comparative studies first ingfully defined as both a process (e.g., adaptation occurs employ process-neutral statistical tools to identify and mea- when a population evolves greater fitness via natural selec- sure macroevolutionary convergence and then, patterns in tion) and a pattern (e.g., an adaptation is a trait that in- hand, test alternative hypotheses about the evolutionary pro- creases an individual’s fitness compared to individuals cesses that may have given rise to these patterns. without that trait; Gould and Vrba 1982; Futuyma 2005). While the goal to investigate convergence without mak- In the case of evolutionary convergence, we feel that in al- ing assumptions about process is a worthy one, in phylo- most all scenarios convergence is best defined as a pattern genetic comparative biology it is impossible to study evo- (Stayton 2015a). Convergence is less meaningful as a pro- lutionary pattern and process independently (Pagel and cess, because convergent evolution is nearly always an emer- Harvey 1989; Harvey and Pagel 1991; Freckleton et al. gent outcome of evolutionary processes operating indepen- 2011; Hunt 2012). The purpose of phylogenetic compara- dently in multiple lineages rather than any intrinsically tive methods is to study patterns that inherently arise as a convergent processes. For example, short-limbed twig spe- consequence of evolutionary processes, both to under- cialist anoles on different Antillean islands are similar be- stand how history has shaped these patterns and to infer

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:28 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Pattern, Process, and Convergence S17 the processes associated with this history. As referenced approaches. Index methods implicitly use process models to above, phylogenetic comparative biology is largely an in- generate a frame of reference with which to compare puta- ductive science, due to the usual lack of direct observa- tively convergent evolutionary patterns. For example, the tional evidence of historical processes and patterns. The Wheatsheaf index (Arbuckle et al. 2014) uses pairwise phy- comparative method provides a framework for testing hy- logenetic distances as a yardstick for evaluating and then potheses about these phenomena, but an essential compo- weighting observed pairwise trait differences, to be con- nent of this framework is the assumption of evolutionary trasted with the correspondence between pairwise phy- process models. logenetic and trait differences expected under Brownian motion. Index tools have been described as “process-neutral” or “process-free” (Stayton 2015a; Speed and Arbuckle 2016), Evolutionary Process Models Underlying Tests which is accurate in the sense that convergent patterns are not of Convergent Pattern assumed to have evolved under any given evolutionary mech- In the case of convergence, the assumption of an evolu- anism. However, these measures are useful only in reference tionary process model is required at some step (often im- to expectations under a particular evolutionary process model, plicitly) by all available comparative tools. The most widely which is sometimes unspecified but most often Brownian assumed model is Brownian motion, which is used to model motion. Furthermore, because no underlying historical pro- trait evolution in a variety of contexts. Brownian motion is a cess model is applied to the data themselves, these tools can very simple model that represents the expectations of con- be limited by an inability to distinguish convergence from tinuous phenotypic evolution under neutral genetic drift other causes of evolutionary similarity between distant rel- (Lande 1976; Felsenstein 1988). Like most models used in atives, such as a simple lack of divergence (Stayton 2015a; phylogenetic comparative methods, Brownian motion is a Speed and Arbuckle 2016). We suggest that additional in- phenomenological model of the evolution of mean species- sights may be gained by comparing statistical indices to dis- level characters, reflecting macroevolutionary expectations tributions simulated under alternative models of the under- but not incorporating microevolutionary mechanisms. It is lying process (sensu Slater and Pennell 2014). often used to represent a hypothesis of neutral evolution Ancestral state reconstruction (ASR) methods critically (especially as a null model), but some have argued for its rely on an assumed model for quantification of both the utility in representing other evolutionary mechanisms such frequency and the strength of convergence. Although sev- as fluctuating directional selection or adaptive radiation on eral kinds of ASR methods have been used to assess con- a dynamic adaptive landscape (although with important vergence, they all model a historical trajectory of evolution caveats; Felsenstein 1988; O’Meara et al. 2006). In studies under an assumed model (almost always Brownian mo- of adaptation, a popular generalization of Brownian motion tion) and then analyze estimated ancestral phenotypes to is the Ornstein-Uhlenbeck model (Hansen 1997; Butler and detect or quantify convergence (reviewed in Stayton 2015a, King 2004; O’Meara and Beaulieu 2014). The OU model 2015b; Arbuckle and Speed 2016; Speed and Arbuckle includes a Brownian drift term as well as a parameter de- 2016). ASR methods have been used to study convergence scribing the strength of attraction to some optimum value. since the dawn of the comparative methods era (e.g., Dono- Extensions allow different lineages to be attracted to differ- ghue 1989; Brooks and McLennan 1991; Losos 1992) but ent optima, which may be interpreted as peaks on an adaptive have gained renewed popularity with the recent development landscape. Unlike Brownian motion, specific OU models can of phylomorphospace tools for visualizing evolutionary tra- model processes consistent with deterministic evolutionary jectories and quantifying the frequency and strength of con- convergence, though it is important to note that both are evo- vergence (Sidlauskas 2008; Stayton 2011, 2015b). Stayton lutionary process models (box 1). Recently developed Lévy (2015b) and Speed and Arbuckle (2016) classified available process models represent an alternative generalization of ASR methods as process-free on the grounds that they do Brownian motion in which a Brownian drift process is punc- not assume that convergent evolutionary patterns were the tuated by large, instantaneous shifts in trait value (i.e., evolu- result of adaptive mechanisms. These approaches are not truly tionary jumps; Eastman et al. 2013; Landis et al. 2013). Lévy process-free, however—they rely on parameter estimates process models lack parameters specifically expected to pro- from a model that assumes the observed patterns evolved duce convergence but can be used to test hypotheses about under a process consistent with Brownian motion, such the frequency of convergent jumps in groups for which the as genetic drift (as well as some, but certainly not all, alter- phenotypic similarity of certain species has already been es- native evolutionary mechanisms; Felsenstein 1988; Hansen tablished (Eastman et al. 2013). and Martins 1996; O’Meara et al. 2006). ASR methods can Although all comparative methods for studying conver- employ alternative macroevolutionary process models, in- gence employ evolutionary models in one fashion or an- cluding models with explicitly adaptive processes, so long other, the role and potential impact of the model vary across as it is possible to reconstruct ancestral states under such

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:28 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S18 The American Naturalist models (e.g., Elliot and Mooers 2014; Uyeda and Harmon convergent evolution critically depend on the assumed 2014). However, this is rarely done, perhaps because tools model of the underlying evolutionary process. Quantita- for carrying out ASR have not kept pace with the rapid de- tive measures of convergent pattern from a given method velopment of methods for fitting alternative models of trait may differ markedly if an alternative model of the evolution- evolution. ary process is assumed. This can be particularly problem- Naturally, evolutionary process models play a promi- atic for Brownian motion–based ASR methods described as nent role in methods for studying convergence that are ex- process-free which may overestimate or underestimate the plicitly based on model fitting. Such tools take one of two frequency of convergence in groups evolving on rugged adap- approaches. In the first, an investigator parameterizes an tive landscapes and yield inaccurate estimates of the strength evolutionary process model in a way that explicitly incorpo- or extent of convergence in any circumstance in which Brown- rates hypothesized convergence events, fits the model to ian motion is a poor model of the true evolutionary process. data, and then compares the fit to that of an alternative model To illustrate, we consider a clade diversifying on an adaptive lacking convergence. This is most commonly done using landscape in which several subclades have undergone peak multiple-optimum OU models to represent the evolution shifts to much larger phenotype values but without conver- of a clade on an adaptive landscape, with occasional peak gence to the same optima (fig. 1). The reconstruction of an- shifts in which a lineage escapes the influence of its histor- cestral states assuming Brownian motion imposes an aver- ical adaptive peak and is attracted to another (Hansen 1997; aging effect on ancestral phenotype estimates that is most Butler and King 2004; Bartoszek et al. 2012; Beaulieu et al. pronounced at the root of the tree. This reconstruction sub- 2012; O’Meara and Beaulieu 2014). Because it is straightfor- stitutes the true pattern of iterated divergence from small to ward to design OU models that permit independent line- large phenotype values with a pattern in which at least four ages to evolve toward a shared adaptive peak, they provide major lineages converge from intermediate to small values. a natural framework for the investigation of adaptive con- Several ASR-based convergence metrics suggest substantial vergence. Generalized OU models permit a great deal of convergence in this clade (fig. 1A). In contrast, if the true flexibility in parameterization, including multiple attraction (i.e., generating) Ornstein-Uhlenbeck model is instead as- strengths or rates of Brownian drift in addition to multiple sumed when carrying out ASR, the same pattern-based met- optima (Beaulieu et al. 2012) and multivariate OU (Bartoszek rics accurately capture the lack of true convergence (fig. 1B). et al. 2012), although highly complex OU models can suffer This is a particularly striking example but we suspect not an from parameter identifiability issues (Ho and Ané 2014; unrepresentative one, due to the well-documented tendency O’Meara and Beaulieu 2014; Cressler et al. 2015; Cooper of Brownian motion–based ASR methods to infer increas- et al. 2016b; Khabbazian et al. 2016). Most methods using ingly intermediate ancestral states for deeper nodes (Sch- OU models require a prior hypothesis for the phylogenetic luter et al. 1997; Oakley and Cunningham 2000). Similar placement of adaptive peak shifts, which limits their utility problems can be anticipated any time ASR methods are used in estimating the frequency of convergence and precludes to investigate a group for which Brownian motion is not a tests for convergence in clades where putatively convergent reasonably good representation of the evolutionary process, taxa have not been identified. These limitations can be and model misspecification can likewise lead to failure to avoided by using a second type of modeling approach in identify convergence events or grossly inaccurate estimates of which the number or rate of evolutionary peak shifts is es- the strength or extent of convergence. timated as a model parameter. To achieve this, some such The issue we discuss here is shared across phylogenetic tools simply extend the multiple-peak OU modeling frame- comparative biology. Hunt (2012) made similar points work by automating the evaluation of candidate peak shift about the measurement of evolutionary rate, showing that configurations (Ingram and Mahler 2013; Uyeda and Harmon measures based on process models were more meaningful 2014; Khabbazian et al. 2016), while alternative methods across evolutionary timescales than traditional interval- assume a Lévy process model of punctuated evolution based (and process-free) rate measures. However, these (Eastman et al. 2013; Landis et al. 2013). Both techniques rate estimates were accurate only if the investigator as- allow assessment of the frequency of evolutionary shifts, sumed the correct model of evolution, due to the complex including convergence events (Eastman et al. 2013; Ingram and model-specific relationship between evolution’s tempo and Mahler 2013). (i.e., change over time) and mode (the process underlying this change). Due to the inseparability of tempo and mode, Hunt argued that the two must be considered in concert in Why Treating Comparative Approaches as Process-Free studies of the evolutionary rate. A similar consideration Can Lead to Problems in the Study of Convergence applies to comparative studies of lineage diversification For each of the comparative approaches to studying con- rates among clades, which are more accurately estimated vergence outlined above, the resulting inferences about using methods that assume an underlying diversification

ABt t s s

C2 = 1.12 r C2 = 0.10 r C5 = 6 q C5 = 0 q

p p o o n n m m l l phenotype k k j j i i h h g g f f e e d d c c

012345 b b a a

0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 time time

Figure 1: Models assumed when reconstructing ancestral states can dramatically affect inferences about the frequency and strength of convergent evolution. Here, data for 20 species were simulated using a known phylogeny under a three-optimum Ornstein-Uhlenbeck (OU) process with strong selection and no convergent evolution (a p 4; j2 p 1; v p 0, 3, 5; total tree length p 1; colors represent correspondence to phenotypic optima). We reconstructed ancestral states assuming a standard Brownian motion model (A) and a three-optimum OU model (B) that closely represents the true evolutionary process under which the data were generated (the number of optima and phylogenetic locations of shifts between optima were known a priori; all other parameters were estimated). We then used ancestral state reconstruction–based comparative methods from Stayton (2015a) to estimate the frequency (C5) and magnitude (C2) of convergent evolution for the set of species with small phenotypes (species a–j; several related measures of the magnitude of convergence yield similar results but are not shown). C5 tallies the number of independent lineages that cross into the phenotype space occupied by the focal set of species (here this space is simply defined as the range of extant phenotype values for this set). C2 indicates the phenotypic distance closed by evolution for this set of species and is cal- 2 culated as the average value of (Dmax Dtips) for all pairs of species in the focal set, where Dmax is the maximum phenotypic difference between a pair of species since their divergence and Dtips is the phenotypic difference between the same pair of species in the present. Note that assuming Brownian motion in this example (A) leads to an overestimate of the frequency and strength of convergence events in this subset of species (from intermediate to small phenotype values). If we assume the true OU model (B), we correctly infer no convergences (C5 p 0) and two divergences from small to large phenotype. Because assuming a different evolutionary model can fundamentally alter comparative inference about convergent evolution, we contend that there is no such thing as a process-free comparative measure of convergence. model than with process-free methods that simply control risk that can be managed in part through conscientious for elapsed time (Rabosky 2012). These examples reflect a consideration of alternative evolutionary models during general truth—pattern and process are inseparable in the analysis and careful interpretation of results (see box 2 study of phylogenetic comparative data, and it is not possi- for discussion of best practices). The fit of adaptive models ble to make inferences about evolutionary patterns without should always be compared to nonadaptive alternatives, assuming something about the evolutionary process (Pagel and interpretation should favor results obtained under and Harvey 1989; Freckleton et al. 2011; Hunt 2012). Not all the best-performing model or, in the absence of a single comparative inferences are equal, of course, and some may be best model, results obtained under all plausible candidate more robust than others to violations of model assumptions. models. However, applying these best practices can only However, convergence metrics that explicitly incorporate take one so far in avoiding adaptationist pitfalls, due to model-based reconstruction of ancestral phenotypes are the issue that evolutionary process models may effectively likely to be particularly sensitive to violations of assumptions model more than one evolutionary mechanism. about the underlying evolutionary model (Oakley and Cun- ningham 2000). Many-to-One Mapping of Real Evolutionary Given the need to assume an evolutionary process to Processes to Modeled Processes study convergence, how can one avoid the potential for adaptationist bias? The potential for such bias is irrefut- Most available phylogenetic comparative methods employ able (Hansen 2014; Stayton 2015a,2015b), but this is a phenomenological models of the evolutionary process that

The comparative method has important limitations that must be taken into account in any study (Losos 2011b; Maddison and FitzJohn 2015; Rabosky and Goldberg 2015; Cooper et al. 2016a), including studies that focus on convergence. Here we recommend several considerations that we think are essential to making strong inferences about historical patterns and processes of convergence.

Study Design The comparative approach investigates patterns that result from uncontrolled natural processes rather than experimental manipulation (Freckleton et al. 2011). Nonetheless, the considerations that guide experimental design equally apply to comparative analyses. Statistical replication is necessary for addressing most questions about convergence, although the relevant form and degree of replication can depend on the question being asked. For example, if one simply wishes to test whether two taxa have in fact converged, this can be tested with a simple model comparison or ancestral state reconstruction (ASR)–based test, although the scope of inference will be limited. Other study objectives will require relatively diverse clades to have any statistical power. For example, testing whether a shift to a new habitat is consistently associated with convergence will require a system containing numerous habitat shifts. It will almost never be possible to determine the cause of single convergence events using comparative methods alone because of the inability to rule out the possibility that an observed correlation is spurious (Gould and Lewontin 1979; Maddison and FitzJohn 2015). Study design should also involve consideration of phylogenetic scale. Many evolutionary models can be useful at some scales but inadequate at others (Estes and Arnold 2007; Hunt 2012). For example, larger clades are more likely to have been shaped by a more heterogeneous mixture of evolutionary processes (Beaulieu et al. 2013). At the very smallest phylogenetic scales, it may not be possible to distinguish complex evolutionary models from simpler alternatives (Boettiger et al. 2012), even if the former better reﬂect reality.

Model Comparison Alternative evolutionary process models can differ profoundly in how they reconstruct historical patterns (fig. 1), making it essential to compare models in any phylogenetic study of convergence. While model comparison has become routine in many areas of comparative biology (Posada and Crandall 1998; Harmon et al. 2010; Morlon 2014; Pennell et al. 2015), it is often neglected in phylogenetic studies of trait convergence. This may be because common tools for measuring convergence assume Brownian motion evolution; investigators interested in assuming alternative models must customize existing tools to do so. This is especially true for ASR methods (but see Elliot and Mooers 2014; Uyeda and Harmon 2014). Because ancestral state estimates can differ so profoundly under alternative evolutionary models, we suspect that a greater appreciation for the importance of model comparison might result from the development of more flexible ASR tools. Although model comparison is essential for studying convergence, we caution against discussing results from alternative models on equal footing when some models clearly outperform others. Comparative studies commonly report and interpret the results of several alternative methods, with results from different methods regarded as complementary. This is to be encouraged when methods are internally consistent with one another but can be misleading when they assume different models of evolution, especially if these differences lead to meaningful differences in the quantification of convergent evolution. For example, if Brownian motion is found to yield a much worse fit to data than a multiple-peak Ornstein-Uhlenbeck (OU) model (such as in the toy example in fig. 1), the use of comparative methods that assume Brownian motion, such as most ASR and index methods, does not meaningfully contribute to our understanding of convergence and may even undermine it. Care should be taken that results that rely on alternative evolutionary process models are themselves regarded as fundamentally distinct (and potentially incompatible) rather than complementary per se.

Model Parameters and Model Adequacy The parameters of fitted evolutionary models can be richly informative with respect to convergence, and many of the interesting features of evolutionary convergence that have inspired pattern-based tools may be effectively cap- tured by the parameters of evolutionary models. For example, the strength of attraction in an OU model can be interpreted as a rate of adaptation in lineages that converge on a shared adaptive peak and may represent the balance between historical constraint and adaptation (Hansen 1997, 2012; Beaulieu et al. 2012; Collar et al. 2014). Inspection of model parameters can also be used to identify when models provide a poor fit to data. For example, multi-optimum OU models frequently return estimates for some optima that fall outside the observed range of species trait data. This may reflect ongoing adaptation toward an extreme phenotype (Hansen 1997) or a mismatch between model assumptions and reality (Mahler and Ingram 2014). Some data sets cannot be fitwellbymulti- optimum OU models, highlighting the importance of testing whether a model can adequately reproduce key patterns in the data rather than simply assessing which model from a set of candidates fits best (Pennell et al. 2015). Simulation-based approaches can help ensure robust inference in virtually any scenario, including empirical conditions for which model performance may yet be unknown (Boettiger et al. 2012; Mahler et al. 2013; Elliot and Mooers 2014; Slater and Pennell 2014; Pennell et al. 2015; Clarke et al. 2017).

may plausibly represent more than one kind of evolution- or shape its course (i.e., the why). Replicated convergence ary mechanism—that is, a many-to-one mapping of true provides a powerful framework for both avenues of inquiry. evolutionary mechanism to modeled macroevolutionary The power of this replication has been harnessed in combina- process (Hansen and Martins 1996; O’Meara et al. 2006; tion with high-throughput sequencing technologies in the Revell et al. 2008; Hansen 2012; Pennell 2014). For exam- last decade to greatly increase our understanding of the mo- ple, a single-peak OU model may represent adaptive evo- lecular mechanisms behind convergent phenotypic change lution in a clade that has already reached a phenotypic op- (Elmer and Meyer 2011; Stern 2013; Rosenblum et al. 2014). timum (Hansen 1997), or it could represent evolutionary By comparison, somewhat less progress has been made in un- stasis due to a constraint on the production of variation derstanding the ecological and phylogenetic context in which (e.g., Harmon et al. 2010). Many-to-one mapping is possi- convergence occurs. ble for a diversity of evolutionary process models, from The phylogenetic comparative method can be especially Brownian motion to early burst and saltational macroevo- useful for addressing questions about causes of convergent lutionary models (Freckleton and Harvey 2006; O’Meara evolution because it is well suited for investigation at the et al. 2006; Mahler et al. 2010; Venditti et al. 2011; Pennell large spatial and temporal scales at which such factors et al. 2014). Although many such models were introduced shape the evolution of biodiversity. In many cases, though, with speciﬁc microevolutionary mechanisms in mind, they comparative models on their own may fail to distinguish describe variation at a comparatively coarse macroevolu- among alternative hypotheses about the causes of conver- tionary scale and contain no direct link to such ﬁne-scale gent evolution, even when carefully applied. This is an in- mechanisms. The lack of mechanistic detail in these models trinsic feature of the comparative approach that results from presents another formidable limitation to the interpretation the many-to-one mapping of process to pattern in macroevo- of macroevolutionary convergence. The investigator must lution (Hansen and Martins 1996; Pennell 2014). Thus, the consider such possibilities in any analysis of convergence— comparative approach will often be much more powerful as we will argue below, such considerations can be aided by when integrated with complementary avenues of investiga- investigation of the study group using complementary ap- tion. Here we discuss ways in which comparative inferences proaches that allow more direct tests of mechanism and by about the processes underlying convergent evolution can be considering the natural history of the organisms and their strengthened by the incorporation of (1) more diverse causal environments. mechanisms, (2) biogeography, and (3) fossil data into comparative approaches to studying convergence. Integrative Approaches to the Study of Macroevolutionary Convergence Incorporating a Greater Diversity of Causal Mechanisms into Evolutionary Process Models A complete understanding of any evolutionary phenomenon requires knowledge of both the detailed mechanisms A key goal in the study of evolution is to understand how by which evolutionary change occurs (i.e., the how) and microevolutionary mechanisms shape macroevolution, but the circumstances that ultimately bring about such change elucidating this link has proved challenging (Uyeda et al.

2011; Rosindell et al. 2015). The comparative study of con- fully diversify the models available in our macroevolution- vergence can help to evaluate this relationship to the extent ary tool kit (e.g., Elliot and Mooers 2014; Clarke et al. 2017). that we can make meaningful connections between poten- Even the development of improved models may not tial causative factors and convergent pattern. Candidate overcome the issue of many-to-one mapping of mecha- factors that may result in macroevolutionary convergence nism to evolutionary process model. However, hypotheses include developmental mode (e.g., Wake 1982), genome ar- and models are not the same thing, and the utility of the chitecture (e.g., Stern 2013), changes in climate (e.g., Yea- comparative method depends on the ability of the investi- man et al. 2016), mutualistic interactions that involve pheno- gator to use natural history knowledge and comparative type matching (e.g., Hoyal Cuthill and Charleston 2015), tools together to craft specific mechanism-inspired hy- repeated antagonistic interactions (e.g., Siepielski and Benk- potheses that can be tested using comparative data. In ad- man 2007), and competition for nonsubstitutable resources dition to improvements associated with integrating more (e.g., Abrams 1987; Scheffer and van Nes 2006). However, diverse causal mechanisms into comparative methods, despite ongoing research interest in these areas, we still the study of convergence will be aided by designing studies lack answers to basic questions such as, Do evolutionary that creatively use ecological experiments to test for specific shifts in reproductive system change the likelihood that ecological mechanisms acting to produce patterns observed clades will exhibit convergence? Is convergence due to abi- at the clade level (Weber and Agrawal 2012). Ecological hy- otic selection more or less common than convergence due potheses about the drivers of convergence generally contain to biotic selection? and How likely is convergence to occur, predictions about the relationship between trait similarity, persist, or break down under antagonistic or mutualistic abundance, and the relative fitness of species in a given selection? community (table 1). For example, in instances where con- One source of improvement may come from renewed vergence is hypothesized to result from selection for Mül- attention to model development. As the scope of phyloge- lerian mimicry (i.e., selection for greater phenotype match- netic comparative methods has grown in recent years, ing among aposematic species), experiments manipulating models that focus on constrained or bounded evolution trait similarity in contemporary communities can be paired have received somewhat less attention than those inspired with phylogenetic tests of convergence (in relation to timing by explicitly adaptive mechanisms. New work in this area of sympatry). The prediction in this case is that (1) the re- could help to diversify the scope of comparative inquiry duction of phenotypic similarity decreases species fitness (e.g., Boucher and Démery 2016). Future developments by increasing predation and (2) convergence occurred when in the field should also work to clarify what kinds of mac- species were in sympatry, not before. This integrative frame- roevolutionary patterns we expect to arise from different work can be applied to antagonistic hypotheses as well and ecological processes. Recently developed comparative models represents a powerful approach to testing adaptive hypoth- based on simple species interactions provide a welcome eses about the ecological drivers of convergence. first step in this direction (Yoder and Nuismer 2010; Pennell and Harmon 2013; Nuismer and Harmon 2015; Integrating Biogeographic Approaches into Comparative Drury et al. 2016; Clarke et al. 2017). Although these Studies of Convergence methods do not explicitly model convergence, the ecological processes underlying them may result in convergent Integrating an understanding of species and clade bioge- phenotypes. In addition, recent years have seen the rapid ography can greatly enhance attempts to link ecological pro- development of a more sophisticated theory of species co- cess to macroevolutionary pattern. Evolving lineages can di- existence and community ecology (Chesson 2000; Hubbell rectly interact only when they co-occur, and accounting for 2001; Pennell and Harmon 2013; Vellend 2016), and fu- co-occurrence patterns will be essential if we are to distin- ture modeling efforts would do well to identify a set of guish the influence of species interactions on convergence expected evolutionary outcomes that reflects current eco- from alternative factors. Furthermore, biogeographic per- logical thinking. Combining modern ecological theory with spectives can disentangle the influence of species range overlap the phylogenetic replication made possible in comparative from abiotic factors such as climate or soil type. The biogeo- studies of convergence will make for a powerful approach graphic approach is ripe for application to convergence gen- to studying the macroevolutionary signature of ecological erally, and in table 1 we outline several ways in which phylo- mechanisms. We speculate that a principal roadblock to genetic comparative methods may be combined with such an the development of more diverse and detailed macroevolu- approach to augment their resolving power when investigat- tionary models has been the difficulty (or impossibility) of ing the ecological and evolutionary processes underlying con- representing such models as closed-form likelihood expres- vergent patterns. sions. Simulation-based approaches (including approxi- The incorporation of a biogeographic perspective has mate Bayesian computation methods) may be required to yielded new insights in recent studies of replicated adap-

Table 1: Examples of hypothesized mechanisms driving patterns of convergence and predictions from integrated analysis Predictions for patterns Hypothesized underlying of coexistence of mechanism Biogeographic predictions convergent forms Ecological predictions Convergence due to Convergence is independent of Probability of convergence is Manipulating the density of a trait chance biogeography independent of community in a community does not change context the selective value of the trait Convergence due to Convergence is correlated with Probability of convergence is Manipulating the density of a trait selection driven by shared physical conditions independent of range in a community does not change physical environment overlap with convergent the selective value of the trait (e.g., climate, light) species Convergence due to Convergence may or may not Convergent forms evolve in Negative density-fitness relationship: competition resulting in be correlated with physical allopatry but only where the abundance of a phenotype in niche partitioning and environment they are sympatric with a the community decreases the character displacement competitor selective value of that phenotype. Convergent communities should exhibit similar patterns of niche partitioning Convergence due to Convergence may be correlated Convergent forms evolve in Resource limitation leads to selection competition for with particular abiotic sympatry for greater similarity in the shared nonsubstitutable conditions across phylogeny phenotype: supplementing resources (e.g., an essential nutrient) resource decreases selection on this trait Convergence due to Convergence may or may not Probability of convergence is Average fitness is a positive facilitation/mutualism be correlated with physical independent of range function of the abundance of environment overlap with convergent the mutualist and a negative species function of the abundance of similar competitors Convergence due to Convergence may or may not Probability of convergence is Average fitness is a positive function commensalism be correlated with physical independent of range of the abundance of the environment overlap with convergent commensal host species Convergence due to Convergence may or may not Convergent forms evolve in Average fitness is a negative predation/parasitism be correlated with physical allopatry or sympatry but function of the abundance of the environment only where they are sym- antagonist patric with an antagonist Note: Linking pattern to process is a central challenge in comparative biology. However, in some cases, integrating multiple forms of inference can help researchers narrow possible pattern-to-process links. Here we provide several examples of how this framework could be applied to phylogenetic comparative studies of convergence. It is important to note that while inferring past causation with absolute certainty is impossible, explicitly considering the predictions of alternative hypotheses can help researchers identify mechanisms consistent with observed patterns. We provide several examples of mechanistic hypotheses, which are not necessarily mutually exclusive (a fact that should be accounted for in the design of a comparative study). tive radiations in which entire well-structured ecological via mechanisms other than interspecific competition. Here, guilds have evolved convergently (Schluter 2000; Mahler information about range overlap can be informative. If in- and Ingram 2014). The existence of numerous replicated terspecific competition played a role in the replicated evo- adaptive radiations suggests an important and determinis- lution of ecological specialists, we would expect the conver- tic role for interspecific competition and subsequent char- gent species to occur allopatrically and to have evolved only acter displacement as a cause of convergence into the same a single time in a given region, but we would have no such set of niches (e.g., Frédérich et al. 2013; Grundler and expectation if competition were not important in this diver- Rabosky 2014; Esquerré and Keogh 2016; Moen et al. 2016). gence. This pattern is observed in replicated Greater Antil- An alternative possibility, however, is that such convergence lean Anolis radiations and in concert with experimental results more from biomechanical trade-offs involved in spe- studies of both competition and character displacement cializing on particular resources than from competitive in- (Pacala and Roughgarden 1982; Leal et al. 1998; Stuart et al. teractions per se and that such specialists may have emerged 2014), strongly suggests a role for competition in contributing

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:28 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S24 The American Naturalist to the repeated evolution of similar ecomorphs on different gent pattern from process and allow more powerful hy- islands (Mahler et al. 2013). pothesis tests than can be achieved using information Other ecological mechanisms will leave distinct patterns solely about extant species. The integration of fossils into in biogeographic patterns of convergent taxa. Cichlids in molecular phylogenetic frameworks remains far from rou- African lakes are well known for replicated radiations in tine, however, and a key hurdle is the often fragmentary different lakes (Wagner et al. 2012), but Muschick et al. nature of fossil data and the considerable uncertainty of- (2012) showed that within Lake Tanganyika, numerous con- ten associated with phylogenetic placement of such fossils. vergent species co-occur within the lake. The pattern of sym- Future efforts to address these challenges should yield patric convergent taxa (see also Kozak et al. 2009; Ingram and large dividends for the comparative study of convergence. Kai 2014) challenges the hypothesis of competition-driven character displacement and might instead suggest compet- Conclusions itive character convergence (MacArthur and Levins 1967; Abrams 1987; Scheffer and van Nes 2006). While a great The phylogenetic comparative method provides a rich set deal of work is needed to validate the hypothesis that com- of tools for answering questions about convergence, in- petition can drive convergence between coexisting taxa cluding many that are otherwise inaccessible to biologists. across entire clades, the possibility highlights the need for Models of the evolutionary process are at the core of all an increased understanding of the expected biogeographic comparative methods, however, and these models can pro- and macroevolutionary consequences of a broader range of foundly influence the outcomes of comparative studies of ecological processes. convergence. This intrinsic link between pattern and process in phylogenetic comparative methods can be a liability if ig- nored but can be a powerful asset when models are explicitly Revisiting the Fossil Record used to test hypotheses about the ecological and evolution- The overwhelming majority of comparative phylogenetic ary processes that give rise to phenotypic patterns. Future studies are conducted using only extant species. Inferences progress in the comparative study of convergence should re- from comparative analyses therefore suffer an unfortunate sult from development of more realistic models of ecological temporal asymmetry, whereby estimates of both historical and evolutionary processes, better integration of compara- pattern and process are associated with increasing levels of tive study with complementary research on biogeography uncertainty as one looks further back in time (Schluter et al. and ecology, and the growing incorporation of fossil infor- 1997; Cunningham et al. 1998; Oakley and Cunningham mation into phylogenetic investigations currently dominated 2000; Losos 2011b). For this reason alone, fossil data can by extant taxa. make very large marginal improvements to the accuracy of comparative inference, and the incorporation of fossil infor- Acknowledgments mation into comparative studies of convergence promises to help distinguish among alternative evolutionary models and We thank A. Agrawal for inviting us to contribute to this refine parameter estimates of key evolutionary processes. Re- special issue and for providing thoughtful feedback on early cent years have witnessed encouraging progress in the merg- drafts of our manuscript. We also wish to thank T. Stayton ing of paleontology and comparative phylogenetic methods, as well as the presenters and many attendees of the 2016 both with the development of integrative new models for phy- American Society of Naturalists Vice Presidential Sympo- logenetic inference and divergence dating (e.g., Ronquist et al. sium for discussing these matters with us following our meet- 2012; Heath et al. 2014; Drummond and Stadler 2016; Zhang ing presentation. Finally, we thank B. O’Meara, an anonymous et al. 2016) and for the fitting of comparative models of con- reviewer, and members of the Mahler lab for insightful com- tinuous trait evolution (Slater et al. 2012; Slater and Harmon ments on our manuscript. 2013; Hunt and Slater 2016). There has been enough progress to demonstrate that Literature Cited fossil data can dramatically improve comparative inference and in some cases shift the weight of evidence to al- Abrams, P. A. 1987. Alternative models of character displacement and ternative hypotheses (Slater et al. 2012; Mitchell 2015; niche shift. I. Adaptive shifts in resource use when there is competition Slater 2015; Hunt and Slater 2016). In studies of conver- for nutritionally nonsubstitutable resources. Evolution 41:651–661. gence, even one or a few fossil data points indicating trait Alberch, P. 1982. Developmental constraints in evolutionary processes. Pages 313–332 in J. T. Bonner, ed. Evolution and develop- values of ancestors may be critical in distinguishing be- ment. Springer, Berlin. tween alternative scenarios (e.g., the histories depicted in Arbuckle, K., C. M. Bennett, and M. P. Speed. 2014. A simple mea- fig. 1A,1B). By anchoring ancestors in trait space, the in- sure of the strength of convergent evolution. Methods in Ecology corporation of fossil data can potentially separate conver- and Evolution 5:685–693.

Arbuckle, K., and M. P. Speed. 2016. Analyzing convergent evolu- Cooper, N., G. H. Thomas, C. Venditti, A. Meade, and R. P. Freckleton. tion: a practical guide to methods. Pages 23–36 in P. Pontarotti, 2016b. A cautionary note on the use of Ornstein-Uhlenbeck models in ed. Evolutionary biology. Springer, Switzerland. macroevolutionary studies. Biological Journal of the Linnean Society Arendt, J., and D. Reznick. 2008a. Convergence and parallelism 118:64–77. reconsidered: what have we learned about the genetics of adapta- Cressler, C. E., M. A. Butler, and A. A. King. 2015. Detecting adap- tion? Trends in Ecology and Evolution 23:26–32. tive evolution in phylogenetic comparative analysis using the ———. 2008b. Moving beyond phylogenetic assumptions about Ornstein-Uhlenbeck model. Systematic Biology 64:953–968. evolutionary convergence: response to Leander. Trends in Ecology Cunningham, C. W., K. E. Omland, and T. H. Oakley. 1998. and Evolution 23:483–484. Reconstructing ancestral character states: a critical reappraisal. Arnold, S. J. 1992. Constraints on phenotypic evolution. American Trends in Ecology and Evolution 13:361–366. Naturalist 140(suppl.):S85–S107. Donoghue, M. J. 1989. Phylogenies and the analysis of evolutionary Bartoszek, K., J. Pienaar, P. Mostad, S. Andersson, and T. F. Hansen. sequences, with examples from seed plants. Evolution 43:1137– 2012. A phylogenetic comparative method for studying multivar- 1156. iate adaptation. Journal of Theoretical Biology 314:204–215. Drummond, A. J., and T. Stadler. 2016. Bayesian phylogenetic esti- Beaulieu, J. M., D.-C. Jhwueng, C. Boettiger, and B. C. O’Meara. 2012. mation of fossil ages. Philosophical Transactions of the Royal So- Modeling stabilizing selection: expanding the Ornstein-Uhlenbeck ciety B 371:20150129. model of adaptive evolution. Evolution 66:2369–2383. Drury, J., J. Clavel, M. Manceau, and H. Morlon. 2016. Estimating Beaulieu, J. M., B. C. O’Meara, and M. J. Donoghue. 2013. Identifying the effect of competition on trait evolution using maximum like- hidden rate changes in the evolution of a binary morphological lihood inference. Systematic Biology 65:700–710. character: the evolution of plant habit in campanulid angiosperms. Eastman, J. M., D. Wegmann, C. Leuenberger, and L. J. Harmon. Systematic Biology 62:725–737. 2013. Simpsonian “evolution by jumps” in an adaptive radiation Bell, M. A. 1987. Interacting evolutionary constraints in pelvic reduc- of Anolis lizards. arXiv:1305.4216. tion of threespine sticklebacks, Gasterosteus aculeatus (Pisces, Gaste- Eldredge, N., and J. Cracraft. 1980. Phylogenetic patterns and the rosteidae). Biological Journal of the Linnean Society 31:347–382. evolutionary process: method and theory in comparative biology. Boettiger, C., G. Coop, and P. Ralph. 2012. Is your phylogeny infor- Columbia University Press, New York. mative? measuring the power of comparative methods. Evolution Elliot, M. G., and A. Ø. Mooers. 2014. Inferring ancestral states without 66:2240–2251. assuming neutrality or gradualism using a stable model of continuous Boucher, F. C., and V. Démery. 2016. Inferring bounded evolution in character evolution. BMC Evolutionary Biology 14:226. phenotypic characters from phylogenetic comparative data. Sys- Elmer, K. R., and A. Meyer. 2011. Adaptation in the age of ecological tematic Biology 65:651–661. genomics: insights from parallelism and convergence. Trends in Brakefield, P. M. 2011. Evo-devo and accounting for Darwin’send- Ecology and Evolution 26:298–306. less forms. Philosophical Transactions of the Royal Society B Endler, J. A. 1981. An overview of the relationships between mimicry 366:2069–2075. and crypsis. Biological Journal of the Linnean Society 16:25–31. Bright, J. A., J. Marugán-Lobón, S. N. Cobb, and E. J. Rayfield. 2016. Esquerré, D., and J. S. Keogh. 2016. Parallel selective pressures drive The shapes of bird beaks are highly controlled by nondietary convergent diversification of phenotypes in pythons and boas. factors. Proceedings of the National Academy of Sciences of the Ecology Letters 19:800–809. USA 113:5352–5357. Estes, S., and S. J. Arnold. 2007. Resolving the paradox of stasis: Brooks, D. R., and D. A. McLennan. 1991. Phylogeny, ecology, and models with stabilizing selection explain evolutionary divergence behavior: a research program in comparative biology. University on all timescales. American Naturalist 169:227–244. of Chicago Press, Chicago. Felsenstein, J. 1985. Phylogenies and the comparative method. Amer- Butler, M. A., and A. A. King. 2004. Phylogenetic comparative anal- ican Naturalist 125:1–15. ysis: a modeling approach for adaptive evolution. American Nat- ———. 1988. Phylogenies and quantitative characters. Annual Re- uralist 164:683–695. view of Ecology and Systematics 19:445–471. Chapleau, F., P. H. Johansen, and M. Williamson. 1988. The distinc- Fox, J. W., and D. A. Vasseur. 2008. Character convergence under tion between pattern and process in evolutionary biology: the use competition for nutritionally essential resources. American Natu- and abuse of the term “strategy.” Oikos 53:136–138. ralist 172:667–680. Chesson, P. 2000. Mechanisms of maintenance of species diversity. Freckleton, R. P., N. Cooper, and W. Jetz. 2011. Comparative meth- Annual Review of Ecology and Systematics 31:343–366. ods as a statistical fix: the dangers of ignoring an evolutionary Clarke, M., G. H. Thomas, and R. P. Freckleton. 2017. Trait evolu- model. American Naturalist 178:E10–E17. tion in adaptive radiations: modeling and measuring interspecific Freckleton, R. P., and P. H. Harvey. 2006. Detecting non-Brownian competition on phylogenies. American Naturalist 189:121–137. trait evolution in adaptive radiations. PLoS Biology 4:e373. Collar, D. C., J. S. Reece, M. E. Alfaro, P. C. Wainwright, and R. S. Mehta. Frédérich, B., L. Sorenson, F. Santini, G. J. Slater, and M. E. Alfaro. 2014. Imperfect morphological convergence: variable changes in cra- 2013. Iterative ecological radiation and convergence during the nial structures underlie transitions to durophagy in moray eels. Amer- evolutionary history of damselfishes (Pomacentridae). American ican Naturalist 183:E168–E184. Naturalist 181:94–113. Conway Morris, S. 2010. Evolution: like any other science it is predict- Futuyma, D. J. 2005. Evolution. Sinauer, Sunderland, MA. able. Philosophical Transactions of the Royal Society B 365:133–145. Gould, S. J. 1980. The evolutionary biology of constraint. Daedalus Cooper, N., G. H. Thomas, and R. G. FitzJohn. 2016a. Shedding light 1980:39–52. on the “dark side” of phylogenetic comparative methods. Methods ———. 2002. The structure of evolutionary theory. Harvard Uni- in Ecology and Evolution 7:693–699. versity Press, Cambridge, MA.

Gould, S. J., and R. C. Lewontin. 1979. The spandrels of San Marco evolution in coexisting clades of North American salamanders. and the Panglossian paradigm: a critique of the adaptationist Evolution 63:1769–1784. programme. Proceedings of the Royal Society B 205:581–598. Lande, R. 1976. Natural selection and random genetic drift in phe- Gould, S. J., and E. S. Vrba. 1982. Exaptation—a missing term in the notypic evolution. Evolution 30:314–334. science of form. Paleobiology 8:4–15. Landis, M. J., J. G. Schraiber, and M. Liang. 2013. Phylogenetic anal- Grundler, M. C., and D. L. Rabosky. 2014. Trophic divergence de- ysis using Lévy processes: finding jumps in the evolution of con- spite morphological convergence in a continental radiation of tinuous traits. Systematic Biology 62:193–204. snakes. Proceedings of the Royal Society B 281:20140413. Leal, M., J. A. Rodríguez-Robles, and J. B. Losos. 1998. An experi- Haldane, J. B. S. 1932. The causes of evolution. Longman, London. mental study of interspecific interactions between two Puerto Rican Hansen, T. F. 1997. Stabilizing selection and the comparative analy- Anolis lizards. Oecologia 117:273–278. sis of adaptation. Evolution 51:1341–1351. Leander, B. S. 2008. Different modes of convergent evolution reflect ———. 2012. Adaptive landscapes and macroevolutionary dynam- phylogenetic distances: a reply to Arendt and Reznick. Trends in ics. Pages 205–226 in E. Svensson and R. Calsbeek, eds. The adaptive Ecology and Evolution 23:481–482. landscape in evolutionary biology. Oxford University Press, Oxford. Losos, J. B. 1992. The evolution of convergent structure in Caribbean ———. 2014. Use and misuse of comparative methods in the study Anolis communities. Systematic Biology 41:403–420. of adaptation. Pages 351–379 in L. Z. Garamszegi, ed. Modern ———. 2009. Lizards in an evolutionary tree: ecology and adaptive phylogenetic comparative methods and their application in evolu- radiation of anoles. University of California Press, Berkeley. tionary biology. Springer, Berlin. ———. 2011a. Convergence, adaptation and constraint. Evolution Hansen, T. F., and E. P. Martins. 1996. Translating between micro- 65:1827–1840. evolutionary process and macroevolutionary patterns: the correla- ———. 2011b. Seeing the forest for the trees: the limitations of phy- tion structure of interspecific data. Evolution 50:1404–1417. logenies in comparative biology. American Naturalist 177:709–727. Harmon, L. J., J. B. Losos, T. J. Davies, R. G. Gillespie, J. L. Gittleman, MacArthur, R., and R. Levins. 1967. The limiting similarity, convergence, W.B.Jennings,K.H.Kozak,etal.2010.Earlyburstsofbodysizeand and divergence of coexisting species. American Naturalist 101:377– shape evolution are rare in comparative data. Evolution 64:2385–2396. 385. Harvey, P. H., and M. D. Pagel. 1991. The comparative method in Maddison, W. P., and R. G. FitzJohn. 2015. The unsolved challenge evolutionary biology. Oxford University Press, Oxford. to phylogenetic correlation tests for categorical characters. Sys- Heath, T. A., J. P. Huelsenbeck, and T. Stadler. 2014. The fossilized tematic Biology 64:127–136. birth-death process for coherent calibration of divergence-time Mahler, D. L., and T. Ingram. 2014. Phylogenetic comparative meth- estimates. Proceedings of the National Academy of Sciences of the ods for studying clade-wide convergence. Pages 425–450 in L. Z. USA 111:E2957–E2966. Garamszegi, ed. Modern phylogenetic comparative methods and Ho, L. S. T., and C. Ané. 2014. Intrinsic inference difficulties for trait their application in evolutionary biology. Springer, Berlin. evolution with Ornstein-Uhlenbeck models. Methods in Ecology Mahler, D. L., T. Ingram, L. J. Revell, and J. B. Losos. 2013. Excep- and Evolution 5:1133–1146. tional convergence on the macroevolutionary landscape in island Hoekstra, H. E., R. J. Hirschmann, R. A. Bundey, P. A. Insel, and J. P. lizard radiations. Science 341:292–295. Crossland. 2006. A single amino acid mutation contributes to Mahler, D. L., L. J. Revell, R. E. Glor, and J. B. Losos. 2010. Ecological op- adaptive beach mouse color pattern. Science 313:101–104. portunity and the rate of morphological evolution in the diversifica- Hoyal Cuthill, J. F., and M. Charleston. 2015. Wing patterning genes tion of Greater Antillean anoles. Evolution 64:2731–2745. and coevolution of Müllerian mimicry in Heliconius butterflies: Maraun, M., G. Erdmann, G. Schulz, R. A. Norton, S. Scheu, and K. support from phylogeography, cophylogeny, and divergence times. Domes. 2009. Multiple convergent evolution of arboreal life in Evolution 69:3082–3096. oribatid mites indicates the primacy of ecology. Proceedings of Hubbell, S. P. 2001. The unified neutral theory of biodiversity and the Royal Society B 276:3219–3227. biogeography. Monographs in Population Biology 32. Princeton Marazzi, B., C. Ané, M. F. Simon, A. Delgado-Salinas, M. Luckow, University Press, Princeton, NJ. and M. J. Sanderson. 2012. Locating evolutionary precursors on Hunt, G. 2012. Measuring rates of phenotypic evolution and the in- a phylogenetic tree. Evolution 66:3918–3930. separability of tempo and mode. Paleobiology 38:351–373. Maynard Smith, J., R. Burian, S. Kauffman, P. Alberch, J. Campbell, B. Hunt, G., and G. Slater. 2016. Integrating paleontological and phy- Goodwin, R. Lande, et al. 1985. Developmental constraints and evolu- logenetic approaches to macroevolution. Annual Review of Ecol- tion: a perspective from the Mountain Lake conference on develop- ogy, Evolution, and Systematics 47:189–213. ment and evolution. Quarterly Review of Biology 60:265–287. Ingram, T., and Y. Kai. 2014. The geography of morphological con- McCune, A. R. 1987. Lakes as laboratories of evolution: endemic vergence in the radiations of Pacific Sebastes rockfishes. American fishes and environmental cyclicity. Palaios 2:446–454. Naturalist 184:E115–E131. McCune, A. R., and R. L. Carlson. 2004. Twenty ways to lose your Ingram, T., and D. L. Mahler. 2013. SURFACE: detecting convergent bladder: common natural mutants in zebrafish and widespread evolution from comparative data by fitting Ornstein-Uhlenbeck convergence of swim bladder loss among teleost fishes. Evolution models with stepwise Akaike information criterion. Methods in and Development 6:246–259. Ecology and Evolution 4:416–. 425 Meyer, J. R., D. T. Dobias, J. S. Weitz, J. E. Barrick, R. T. Quick, and Khabbazian, M., R. Kriebel, K. Rohe, and C. Ané. 2016. Fast and ac- R. E. Lenski. 2012. Repeatability and contingency in the evolution curate detection of evolutionary shifts in Ornstein-Uhlenbeck of a key innovation in phage lambda. Science 335:428–432. models. Methods in Ecology and Evolution 7:811–824. Mitchell, J. S. 2015. Extant-only comparative methods fail to recover Kozak, K. H., R. W. Mendyk, and J. J. Wiens. 2009. Can parallel di- the disparity preserved in the bird fossil record. Evolution 69:2414– versification occur in sympatry? repeated patterns of body-size 2424.

Moen, D. S., H. Morlon, and J. J. Wiens. 2016. Testing convergence Rosenblum, E. B., C. E. Parent, and E. E. Brandt. 2014. The molec- versus history: convergence dominates phenotypic evolution for ular basis of phenotypic convergence. Annual Review of Ecology, over 150 million years in frogs. Systematic Biology 65:146–160. Evolution, and Systematics 45:203–226. Morlon, H. 2014. Phylogenetic approaches for studying diversifica- Rosenblum, E. B., H. Römpler, T. Schöneberg, and H. E. Hoekstra. tion. Ecology Letters 17:508–525. 2010. Molecular and functional basis of phenotypic convergence Muschick, M., A. Indermaur, and W. Salzburger. 2012. Convergent in white lizards at White Sands. Proceedings of the National evolution within an adaptive radiation of cichlid fishes. Current Academy of Sciences of the USA 107:2113–2117. Biology 22:2362–2368. Rosindell, J., L. J. Harmon, and R. S. Etienne. 2015. Unifying ecology Nevo, E. 1979. Adaptive convergence and divergence of subterranean and macroevolution with individual-based theory. Ecology Letters mammals. Annual Review of Ecology and Systematics 10:269–308. 18:472–482. Nuismer, S. L., and L. J. Harmon. 2015. Predicting rates of interspecific Scheffer, M., and E. H. van Nes. 2006. Self-organized similarity, the interaction from phylogenetic trees. Ecology Letters 18:17–27. evolutionary emergence of groups of similar species. Proceedings of Oakley, T. H., and C. W. Cunningham. 2000. Independent contrasts the National Academy of Sciences of the USA 103:6230–6235. succeed where ancestor reconstruction fails in a known bacterio- Schluter, D. 1996. Adaptive radiation along genetic lines of least re- phage phylogeny. Evolution 54:397–405. sistance. Evolution 50:1766–1774. O’Meara, B. C., C. Ané, M. J. Sanderson, and P. C. Wainwright. ———. 2000. The ecology of adaptive radiation. Oxford University 2006. Testing for different rates of continuous trait evolution us- Press, Oxford. ing likelihood. Evolution 60:922–933. Schluter, D., T. Price, A. Ø. Mooers, and D. Ludwig. 1997. Likelihood O’Meara, B. C., and J. M. Beaulieu. 2014. Modelling stabilizing selec- of ancestor states in adaptive radiation. Evolution 51:1699–1711. tion: the attraction of Ornstein-Uhlenbeck models. Pages 381–393 Scotland, R. W. 2011. What is parallelism? Evolution and Develop- in L. Z. Garamszegi, ed. Modern phylogenetic comparative methods ment 13:214–227. and their application in evolutionary biology. Springer, Berlin. Sidlauskas, B. 2008. Continuous and arrested morphological diversi- Oster, G., and P. Alberch. 1982. Evolution and bifurcation of devel- fication in sister clades of characiform fishes: a phylomorphospace opmental programs. Evolution 1982:444–459. approach. Evolution 62:3135–3156. Pacala, S., and J. Roughgarden. 1982. Resource partitioning and in- Siepielski, A. M., and C. W. Benkman. 2007. Convergent patterns in terspecific competition in two two-species insular Anolis lizard the selection mosaic for two North American bird-dispersed communities. Science 217:444–446. pines. Ecological Monographs 77:203–220. Pagel, M. D., and P. H. Harvey. 1989. Comparative methods for examin- Slater, G. J. 2015. Iterative adaptive radiations of fossil canids show ing adaptation depend on evolutionary models. Folia Primatologica no evidence for diversity-dependent trait evolution. Proceedings 53:203–220. of the National Academy of Sciences of the USA 112:4897–4902. Pascoal, S., T. Cezard, A. Eik-Nes, K. Gharbi, J. Majewska, E. Payne, Slater, G. J., and L. J. Harmon. 2013. Unifying fossils and phyloge- M. G. Ritchie, et al. 2014. Rapid convergent evolution in wild nies for comparative analyses of diversification and trait evolution. crickets. Current Biology 24:1369–1374. Methods in Ecology and Evolution 4:699–702. Pennell, M. W. 2014. Modern phylogenetic comparative methods Slater, G. J., L. J. Harmon, and M. E. Alfaro. 2012. Integrating fossils and their application in evolutionary biology: concepts and prac- with molecular phylogenies improves inferences of trait evolution. tice. Systematic Biology 64:161–163. Evolution 66:3931–3944. Pennell, M. W., R. G. FitzJohn, W. K. Cornwell, and L. J. Harmon. Slater, G. J., and M. W. Pennell. 2014. Robust regression and poste- 2015. Model adequacy and the macroevolution of angiosperm rior predictive simulation increase power to detect early bursts of functional traits. American Naturalist 186:E33–E50. trait evolution. Systematic Biology 63:293–308. Pennell, M. W., and L. J. Harmon. 2013. An integrative view of phy- Speed, M. P., and K. Arbuckle. 2016. Quantification provides a con- logenetic comparative methods: connections to population genet- ceptual basis for convergent evolution. Biological Reviews 92:815– ics, community ecology, and paleobiology. Annals of the New 829. doi:10.1111/brv.12257. York Academy of Sciences 1289:90–105. Spor, A., D. J. Kvitek, T. Nidelet, J. Martin, J. Legrand, C. Dillmann, Pennell, M. W., L. J. Harmon, and J. C. Uyeda. 2014. Is there room A. Bourgais, et al. 2014. Phenotypic and genotypic convergences for punctuated equilibrium in macroevolution? Trends in Ecology are influenced by historical contingency and environment in yeast. and Evolution 29:23–32. Evolution 68:772–790. Posada, D., and K. A. Crandall. 1998. Modeltest: testing the model of Stayton, C. T. 2008. Is convergence surprising? an examination of DNA substitution. Bioinformatics 14:817–818. the frequency of convergence in simulated datasets. Journal of Rabosky, D. L. 2012. Testing the time-for-speciation effect in the as- Theoretical Biology 252:1–14. sembly of regional biotas. Methods in Ecology and Evolution ———. 2011. Biomechanics on the half shell: functional perfor- 3:224–233. mance influences patterns of morphological variation in the emydid Rabosky, D. L., and E. E. Goldberg. 2015. Model inadequacy and turtle carapace. Zoology 114:213–223. mistaken inferences of trait-dependent speciation. Systematic Bi- ———.2015a.Thedefinition, recognition, and interpretation of con- ology 64:340–355. vergent evolution, and two new measures for quantifying and as- Revell, L. J., L. J. Harmon, and D. C. Collar. 2008. Phylogenetic signal, sessing the significance of convergence. Evolution 69:2140–2153. evolutionary process, and rate. Systematic Biology 57:591–601. ———. 2015b. What does convergent evolution mean? the interpre- Ronquist, F., S. Klopfstein, L. Vilhelmsen, S. Schulmeister, D. L. tation of convergence and its implications in the search for limits Murray, and A. P. Rasnitsyn. 2012. A total-evidence approach to evolution. Interface Focus 5:20150039. to dating with fossils, applied to the early radiation of the Hyme- Stern, D. L. 2013. The genetic causes of convergent evolution. Nature noptera. Systematic Biology 61:973–999. Reviews Genetics 14:751–764.

Streisfeld, M. A., and M. D. Rausher. 2011. Population genetics, plei- Pages 51–66 in D. Mossakowski and G. Roth, eds. Environmental otropy, and the preferential ﬁxation of mutations during adaptive adaptation and evolution. Fischer, New York. evolution. Evolution 65:629–642. Wake, D. B., and A. Larson. 1987. Multidimensional analysis of an Stuart, Y. E., T. S. Campbell, P. A. Hohenlohe, R. G. Reynolds, L. J. evolving lineage. Science 238:42–48. Revell, and J. B. Losos. 2014. Rapid evolution of a native species Wake, D. B., M. H. Wake, and C. D. Specht. 2011. Homoplasy: from following invasion by a congener. Science 346:463–466. detecting pattern to determining process and mechanism of evo- Uyeda, J. C., T. F. Hansen, S. J. Arnold, and J. Pienaar. 2011. The lution. Science 331:1032–1035. million-year wait for macroevolutionary bursts. Proceedings of Weber, M. G., and A. A. Agrawal. 2012. Phylogeny, ecology, and the the National Academy of Sciences of the USA 108:15908–15913. coupling of comparative and experimental approaches. Trends in Uyeda, J. C., and L. J. Harmon. 2014. A novel Bayesian method for Ecology and Evolution 27:394–403. inferring and interpreting the dynamics of adaptive landscapes Williams, E. E. 1983. Ecomorphs, faunas, island size, and diverse end from phylogenetic comparative data. Systematic Biology 63:902– points in island radiations of Anolis. Pages 326–370 in R. B. Huey, 918. E. R. Pianka, and T. W. Schoener, eds. Lizard ecology: studies of a Vellend, M. 2016. The theory of ecological communities. Princeton model organism. Harvard University Press, Cambridge, MA. University Press, Princeton, NJ. Yeaman, S., K. A. Hodgins, K. E. Lotterhos, H. Suren, S. Nadeau, J. C. Venditti, C., A. Meade, and M. Pagel. 2011. Multiple routes to mam- Degner, K. A. Nurkowski, et al. 2016. Convergent local adaptation malian diversity. Nature 479:393–396. to climate in distantly related conifers. Science 353:1431–1433. Wagner, C. E., L. J. Harmon, and O. Seehausen. 2012. Ecological op- Yoder, J. B., and S. L. Nuismer. 2010. When does coevolution pro- portunity and sexual selection together predict adaptive radiation. mote diversiﬁcation? American Naturalist 176:802–817. Nature 487:366–369. Zhang, C., T. Stadler, S. Klopfstein, T. A. Heath, and F. Ronquist. Wagner, P. J. 2000. Exhaustion of morphologic character states 2016. Total-evidence dating under the fossilized birth-death pro- among fossil taxa. Evolution 54:365–386. cess. Systematic Biology 65:228–249. Wake, D. B. 1982. Functional and developmental constraints and opportunities in the evolution of feeding systems in urodeles. Symposium Editor: Anurag A. Agrawal

“There are in all about twenty species of Draco, inhabiting the East Indies. The power of ﬂightisnotverygreat,butprobablyexceedsthatof Ptychozoön.” Figured: Draco volans.From“Volant Adaptation in Vertebrates” by Richard S. Lull (The American Naturalist, 1906, 40:537–566).

Symposium Convergently Evolved Toxic Secondary Metabolites in Plants Drive the Parallel Molecular Evolution of Insect Resistance*

Georg Petschenka,1 Vera Wagschal,2 Michael von Tschirnhaus,3 Alexander Donath,4 and Susanne Dobler2,†

1. Institute for Insect Biotechnology, Justus-Liebig-Universität Giessen, Giessen, Germany; 2. Molecular Evolutionary Biology, Institute of Zoology, Universität Hamburg, Hamburg, Germany; 3. Fakultät Biologie, Universität Bielefeld, Bielefeld, Germany; 4. Zoologisches Forschungsmuseum Alexander Koenig, Zentrum für Molekulare Biodiversitätsforschung, Bonn, Germany Online enhancements: appendix. Dryad data: http://dx.doi.org/10.5061/dryad.20191. abstract: Natural selection imposed by natural toxins has led to tations, a major question in evolutionary genetics (Storz striking levels of convergent evolution at the molecular level. Cardiac 2016). Recent studies of species that acquired resistance glycosides represent a group of plant toxins that block the Na,K- against toxins in their diet or the environment revealed ATPase, a vital membrane protein in animals. Several herbivorous striking levels of convergent molecular adaptations. Here insects have convergently evolved resistant Na,K-ATPases, and in the interaction of a noxious substance with a well-defined some species, convergent gene duplications have also arisen, likely target site of a specific protein might facilitate the occur- to cope with pleiotropic costs of resistance. To understand the genetic rence of molecular convergence, as specific amino acid sub- basis and predictability of these adaptations, we studied five independent lineages of leaf-mining flies (Diptera: Agromyzidae). These flies stitutions can prevent docking of the toxin. Accordingly, have colonized host plants in four botanical families that convergently across several classes of toxins, it has been shown that re- evolved cardiac glycosides of two structural types: cardenolides and sistance was repeatedly achieved by substitutions of a few bufadienolides. We compared each of six fly species feeding on such amino acids at specific positions within the target site of plants to a phylogenetically related but nonadapted species. Irre- the receptor. Examples include predatory snakes that con- spective of the type of cardiac glycoside in the host plant, five out of sume toxic amphibians and have convergently evolved re- – six exposed species displayed substitutions in the cardiac glycoside sistance to tetrodotoxin (Feldman et al. 2012), poison frogs binding site of the Na,K-ATPase that were previously described in other insect orders; in only one species was the gene duplicated. In that acquire toxic alkaloids from arthropods to defend them- vitro assays of nervous tissue extractions confirmed that the substitu- selves against predators (Tarvin et al. 2016), and diverse tions lead to increased resistance of the Na,K-ATPase. Our results insects that have evolved resistance to cyclodiene insecticides demonstrate that target site insensitivity of Na,K-ATPase is a common (Andreev et al. 1999) or pyrethroids (Rinkevich et al. 2013). response to dietary cardiac glycosides leading to highly predictable amino An emerging model system for convergent evolution that acid changes; nonetheless, convergent evolution of gene duplication for has been studied in unprecedented detail is resistance to this multifunctional enzyme appears more constrained. cardiac glycosides in insects as well as in vertebrates. Car- Keywords: cardiac glycosides, Na,K-ATPase resistance, target site in- diac glycosides are potent toxins that inhibit the ubiquitous sensitivity, Agromyzidae, phylogeny, gene duplication. animal cation carrier Na,K-ATPase. They typically occur in plants but are also produced by some animals (e.g., toads, leaf beetles, and fireflies; Dobler et al. 2011; Agrawal et al. Introduction 2012). Cardiac glycosides are composed of a steroidal skel- fi Evolutionary responses to common selective pressures are eton with a ve-membered lactone ring at C17 in carde- ideally suited to study the predictability of molecular adap- nolides and a six-membered lactone ring in bufadienolides (fig. 1). Both types either occur as glycosides with one or * This issue originated as the 2016 Vice Presidential Symposium presented at several sugars attached to C3 or as genins (Malcolm 1991; the annual meetings of the American Society of Naturalists. Agrawal et al. 2012). By binding to the Na,K-ATPase, they † Corresponding author; e-mail: [email protected]. block the enzyme in the phosphorylated state and thus im- ORCIDs: von Tschirnhaus, http://orcid.org/0000-0002-1903-4767; Donath, pair various physiological functions such as maintenance of http://orcid.org/0000-0001-5618-0547; Dobler, http://orcid.org/0000-0002-0635 -7719. an electrochemical gradient across the cell membrane and generation of neuronal action potentials (Horisberger 2004). Am. Nat. 2017. Vol. 190, pp. S29–S43. q 2017 by The University of Chicago. 0003-0147/2017/190S1-57252$15.00. All rights reserved. Extensive mutagenesis experiments with the mammalian Na, DOI: 10.1086/691711 K-ATPase and recent crystal structures revealed the specific

Figure 1: A, Structure of the pig Na,K-ATPase with bound ouabain (Yatime et al. 2011; Protein Data Bank ID 3N23). Here, a, b,andg refer to the subunits of the Na,K-ATPase, of which g has so far been described only for vertebrates. M p transmembrane domains, A p actuator domain, N p nucleotide-binding domain, P p phosphorylation domain. B, Enlarged and rotated view of the ouabain-binding cavity with relevant amino acids. A p alanine, E p glutamic acid, F p phenylalanine, G p glycine, I p isoleucine, N p asparagine, Q p glutamine, T p threonine, V p valine. C, Structure of cardiac glycosides. Top, hellebrin, a bufadienolide from Helleborus species; middle, ouabain, a standard cardenolide used for the physiological assays; bottom, calotropin, a common cardenolide from Asclepias species. amino acid residues that are necessary for cardiac glycoside sequesters toxic cardenolides from its host plants (Asclepias binding (Price and Lingrel 1988; Croyle et al. 1997; Qiu et al. species) as a defense against predators. Later, the molecular 2005; Laursen et al. 2013, 2015). These data point to the prom- basis was unraveled (Holzinger et al. 1992), and more re- inent role of several amino acids in the transmembrane seg- cently it was shown that target site insensitivity in the mon- ments of the enzyme’s a-subunit that directly interact with arch butterfly is caused by two amino acid substitutions thecardiacglycoside moleculeand stabilizeitintheextracellu- (Aardema et al. 2012; Dalla et al. 2013) that emerged con- lar binding pocket (fig. 1). secutively across the phylogeny of milkweed butterflies Resistance to cardiac glycosides often relies on substitu- leading to a stepwise increase of resistance (Petschenka et al. tions of amino acids in the binding pocket of the Na,K- 2013a). Target site insensitivity of the Na,K-ATPase has now ATPase. The first evidence was found in Danaus plexippus, been reported in at least six orders of insects with multiple the monarch butterfly (Vaughan and Jungreis 1977), which evolutionary origins (Al-Robai et al. 1993; Dobler et al. 2012,

2015; Zhen et al. 2012). While resistance can involve substi- cated Na,K-ATPase genes using a transcriptomic approach. tutions at least at five amino acid positions, the identical sub- Specifically, we wanted to investigate how strongly substitu- stitution of histidine for asparagine at position 122 occurred tions are associated with gene duplications. For three fly at least six times independently (Dobler et al. 2015). A com- species that independently adapted to cardiac glycoside plants parable level of molecular convergence was found in cardiac and two outgroup species, we additionally assessed resistance glycoside–producingtoads andtheir vertebrate predators (Uj- of their Na,K-ATPase in vitro and thereby confirmed the vari et al. 2015). functionality of the observed molecular substitutions. In insects, it was furthermore shown that resistance- conferring amino acid substitutions may be associated with Material and Methods duplications of the Na,K-ATPase gene (Zhen et al. 2012). Natural History of Flies Such duplications might alleviate negative pleiotropic effects associated with increased resistance to cardiac gly- We compared six species of agromyzid taxa feeding on car- cosides. While resistance-conferring mutations may com- diac glycoside plants to related species whose host plants promise the pumping activity of the Na,K-ATPase (Dalla are not known to contain these toxins. The species pairs and Dobler 2016), other important functions of this en- were chosen according to a priori assumptions about phy- zyme might not be impaired by the observed amino acid logenetic relationships based on morphology. substitutions. Increasing evidence points to important roles Phytomyza digitalis Hering, 1925, versus Phytomyza cras- of the Na,K-ATPase as a signal transducer and morphoge- siseta Zetterstedt, 1860.—We compared P. digitalis, whose netic trigger, not only in vertebrates but also in insects host plants are restricted to the genus Digitalis (Plantagina- (Liang et al. 2007; Paul et al. 2007; Li and Xie 2009). A bal- ceae), including aD. grandiflor , D. lutea,andD. lanata (Hering ance between more or less substituted copies of the Na,K- 1927, 1951; www.bladmineerders.nl), which all produce carde- ATPase and tissue-specific expression patterns may allow nolides (Luckner and Wichtl 2000), to P. crassiseta, a specialist adjusting between effective ion transport and these other on various species of Veronica (Plantaginaceae; Hering 1927; functions. Therefore, duplications of the Na,K-ATPase gene www.bladmineerders.nl) not known to produce cardenolides. in cardiac glycoside–adapted insects have been interpreted Phytomyza hellebori Kaltenbach, 1872, versus Phytomyza as an evolutionary strategy to reduce negative pleiotropic ef- fallaciosa Brischke, 1880.—We also investigated P. hellebori, fects (Zhen et al. 2012; Dalla and Dobler 2016). which exclusively uses Helleborus (Ranunculaceae) species as Although cardiac glycosides are known to occur in 18 plant hosts (Hering 1951). Several species of Helleborus are known families (Krenn and Kopp 1998; Agrawal et al. 2012), almost to produce bufadienolides (e.g., hellebrin; Wissner and Ka- all work on resistance has focused on insect herbivores feed- ting 1974). While H. foetidus seems to be the preferred host ing on plants in the family Apocynaceae. Here we studied plant at some locations (Ludwig 1907; www.bladmineerders.nl) the parallel adaptation of five independent lineages of leaf- and was reported to lack bufadienolides, this fly species also mining flies (Agromyzidae) adapted to cardiac glycoside– uses other Helleborus species as hosts (e.g., H. niger, H. vi- producing plants (six fly species adapted to cardiac glycoside ridis; Mortelmans et al. 2014; table 1) known to contain plants, including one generalist species) compared with five bufadienolides (Wissner and Kating 1974; Glombitza et al. closely related fly species feeding on plants without cardiac 1989). The lack of cardiac glycosides in H. foetidus is ques- glycosides. The host plants of these flies include species from tionable, as we have a clear indication of the presence of such four botanical families (Apocynaceae, Brassicaceae, Planta- compounds based on inhibition of Na,K-ATPase (G. Pet- ginaceae, and Ranunculaceae) in which cardiac glycosides schenka, unpublished data). As a comparison for P. helle- have themselves independently evolved. Thus, our data set bori, the Ranunculus feeder P. fallaciosa (Spencer 1990; comprises multiple independent comparisons of animal host www.bladmineerders.nl) was included. use and plant toxin production, a convergently evolved inter- Napomyza scrophulariae Spencer, 1966, versus Napomyza action from the perspective of each partner. lateralis Fallén, 1823.—With respect to the other species we Using our five independent contrasts of fly species, we investigated, knowledge on the natural history of N. scro- tested how often amino acid substitutions evolved that have phulariae is very limited. The species seems to be specialized previously been assumed to-confer resistance of the Na,K- on Digitalis purpurea but may also use other hosts (Spencer ATPase. We were especially interested to test for repeated 1966; www.ukflymines.co.uk). Nonetheless, the only speci- evolution of target site insensitivity by identical amino acid mens that were unambiguously determined as N. scrophu- substitutions. To screen for such substitutions, we first an- lariae were all from D. purpurea (as were the specimens in- alyzed Na,K-ATPase gene sequences amplified by reverse vestigated here; table 1); accordingly, we refer to this species transcription polymerase chain reaction (RT-PCR). In spe- as a specialist. As a phylogenetic comparison, we employed cies possessing putatively resistance-conferring amino acid N. lateralis, a polyphagous species that has a preference for substitutions, we further tested for the existence of dupli- Asteraceae (Spencer 1990; www.bladmineerders.nl).

Table 1: Dipteran species used in this study and plants from which specimens were collected GenBank accession no.

This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:35AM Fly species Host plant Collection site, year Use Na,K-ATPase COI 28S CAD Chromatomyia horticola Erysimum cheiranthoides Pevestorf, Germany, 2011 RT-PCR (1) LT795081 LT795089 LT795087 LT795098 Liriomyza asclepiadis Asclepias syriaca Ithaca, NY, 2014, 2015 In vitro assay, transcriptome HE956747 HE862403 LT795104 LT795100 Liriomyza eupatorii Eupatorium cannabinum Giessen, Germany, 2015 Transcriptome LT795109 LT795097 LT795105 LT795101 Napomyza scrophulariae Digitalis purpurea Taunusstein, Waldkatzenbach, RT-PCR (2) LT795080 LT795091 LT795086 LT795099 Germany, 2011 Napomyza lateralis Asteraceae sp. Tangstedt, Germany, 2011 RT-PCR (2) LT795079 EF104710 EF104884 EF104796 Phytomyza crassiseta Veronica sp. Waldkatzenbach, Germany, RT-PCR (2) LT795078 EU367549 LT795085 EU367638 2011 Phytomyza digitalis Digitalis grandiﬂora Mehrstetten, Germany, 2011 RT-PCR (2), in vitro assay, LT795083 LT795093 LT795106 LT795103 transcriptome Phytomyza fallaciosa Ranunculus sp. Taunusstein, Germany, RT-PCR (2), transcriptome LT795082 LT795094 LT795107 . . . 2011, 2015 Phytomyza hellebori Helleborus foetidus (Mosbach: Hamburg,A Hochheim/Main,A RT-PCR (5), in vitro assay, LT795110, LT795095 LT795108 LT795102 native population; Hamburg, Jena,A,C Mosbach,A,B transcriptome LT795111 Jena: ornamental), Helleborus Germany, 2011, 2012 niger (Hochheim: ornamental) Phytomyza ilicis Ilex aquifolia Giessen,B Hamburg,A,B RT-PCR (1), in vitro assay LT795077 EU367541 LT795084 EU367630 Germany, 2011, 2016 Drosophila melanogaster . . . Laboratory culture In vitro assay AF044974 ...... (oregonR) Note: When specimens were obtained at different locations, letters in superscript indicate where specimens used for reverse transcription polymerase chain reaction (RT-PCR; a), Na,K-ATPase in vitro assays (b), or transcriptomes (c) originated from. Numbers of biological replicates for reverse transcription polymerase chain reaction–based sequencing of Na,K-ATPase genes are given in parentheses. Parallel Evolution of Insect Resistance S33

Liriomyza asclepiadis Spencer, 1969, versus Liriomyza rated by large intronic sequences (e.g., in D. melanogaster, eupatorii Kaltenbach, 1873.—Another comparison involved the coding sequence of the commonest isoform of the Na, L. asclepiadis, a specialized feeder on Asclepias species (Spen- K-ATPase a stretches over 10.195 bp and 8 exons). For cer 1990), and an unidentified Liriomyza species (referred to RT-PCR, flies were decapitated and only heads were used, as L. sp. Mexico) collected from Asclepias angustifolia that is as their brains represent a rich source of Na,K-ATPase differentiated in its cytochrome oxidase I (COI) sequences (Lebovitz et al. 1989). RNA was extracted with the RNeasy from L. asclepiadis by 12.5% sequence divergence (Dobler kit (Qiagen, Hilden, Germany) and reverse transcribed (Su- et al. 2012). The Asclepias host plants of both species contain perscript III, Invitrogen, Carlsbad, CA). Na,K-ATPase se- cardenolides (Agrawal et al. 2012). While L. asclepiadis is quences were amplified in standard PCR using the prim- specialized on Asclepias, we have no information about the ers given in Dobler et al. (2012). This yielded a fragment of host range of the unidentified Liriomyza species. As it seems the Na,K-ATPase covering the amino acids 89–805. The to be closely related to L. asclepiadis and no generalist leaf numbers refer to the mature protein of the pig (Sus scrofa) miners are known to use Asclepias species (Spencer 1990), to enable comparison with the extensive literature on verte- we refer to it as a specialist. We compared these two Lirio- brate Na,K-ATPase including crystal structures of the pro- myza species to L. eupatorii, which uses several genera of tein (Yatime et al. 2011; Laursen et al. 2013, 2015). In addi- Asteraceae as hosts (www.bladmineerders.nl). tion, gene sequences for COI, 28S, and CAD were obtained Chromatomyia horticola Goureau, 1851, versus Phytomyza for C. horticola, L. sp. (Mexico), N. lateralis, N. scrophulariae, ilicis Curtis, 1846.—Last, we sequenced the Na,K-ATPase and P. ilicis (see table 1) using primers given in Scheffer et al. gene of C. horticola, a highly polyphagous species (www (2007) and additional primers for COI given in Maus et al. .bladmineerders.nl), which we have found on the cardenolide- (2001). producing Brassicaceae Erysimum cheiranthoides (Lei et al. 1996). While all pairwise comparisons listed above were based Transcriptome Analyses on an a priori assumed phylogenetic relationship between fly species adapted to cardiac glycosides and nonadapted spe- Transcriptomes based on mRNA preparations were obtained cies, here we arbitrarily selected the holly leaf miner, P. ilicis, for L. asclepiadis, L. eupatorii, P. digitalis, P. fallaciosa, a specialist on Ilex aquifolium, as a comparison for C. horticola. and P. hellebori. Since Ranunculus sp. and Eupatorium cannabinum are hosts for several species of leaf-mining flies, identification of L. eupatorii and P. fallaciosa based on host Collection of Flies and Determination plant and mine morphology was verified by barcoding in- Species of Agromyzidae were determined by means of the dividual flies that were later pooled for transcriptome anal- morphology of mines and host plant species (www.blad yses. For this purpose, DNA of 2 –3 legs of each individual mineerders.nl, www.leafmines.co.uk; Hering 1927, 1935). In was extracted following the method detailed in Carvalho addition, species identities were confirmed based on adult et al. (2009). A PCR product for COI was generated with characters and verified by sequence data compared to refer- primers S1634 and A2188 and sequenced to confirm species ence sequences in GenBank. Host plant leaves containing identity before transcriptome sequencing. RNA extractions mines were collected in the field (table 1) and kept under of roughly 10 flies of L. eupatorii, P. digitalis, P. fallaciosa, room conditions until the emergence of flies. Upon hatching and P. hellebori and of a single individual of L. asclepiadis and wing hardening, flies were frozen and stored at 2807C. were obtained as above. mRNA-specific cDNA library con- In some cases, field-collected flies were also used for rearing struction and paired-end sequencing on an Illumina HiSeq to propagate specimens. Rearing was carried out in mesh 2000 (Illumina, San Diego, CA) were carried out by StarSeq cages under ambient conditions on potted host plants or (Mainz, Germany) and yielded roughly 50 million reads cut branches in water. Drosophila melanogaster (Drosophilidae, for each species. All transcriptomes, except for P. helle- strain oregonR) was kindly provided by the Max Planck Insti- bori, were assembled with Trinity 2.1.1 (Grabherr et al. 2011) tute for Developmental Biology, Tübingen, Germany. using internal trimming with modified trimming options (SLIDINGWINDOW:4:30 LEADING:30 TRAILING:30 MINLEN:50). Reads of P. hellebori were first trimmed with RNA Extraction, RT-PCR, and Sequencing sickle, version 1.29 (Joshi and Fass 2011; min-length: 50 bp), For all fly species mentioned above, sequences were obtained and then assembled using trinityrnaseq_r20140413p1 (Grab- for the mitochondrial COI gene and the nuclear 28S and herr et al. 2011). Local Blast searches (Blastn 2.2.271 [Zhang cinnamyl alcohol dehydrogenase (CAD) genes to include et al. 2000] or Tblastn 2.2.271 [Altschul et al. 1997]) were them in phylogenetic analysis and for their Na,K-ATPase used to identify gene copies of the Na,K-ATPase a-subunit genes. All sequencing was based on RNA extractions, as and the genes needed for the phylogenetic analysis. Na,K- the Na,K-ATPase 1a gene consists of multiple exons sepa- ATPase sequences were also checked for alternative gene cop-

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:35 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S34 The American Naturalist ies with CLC Genomics (Qiagen) by mapping all trimmed In Vitro Assays of Na,K-ATPase reads against reference sequences identified in the Trinity Resistance of the Na,K-ATPase can be monitored in vitro assemblies. Paralogs of the Na,K-ATPase are usually easy by measuring the enzyme’s activity through spectrophoto- to recognize since variation between gene copies by far ex- metric quantification of inorganic phosphate released from ceeds allelic variation observed between individuals. Intra- adenosine triphosphate (ATP) hydrolysis under increasing individual variation accounts for only a few silent substi- concentrations of cardiac glycosides. We assessed in vitro tutions, and we extremely rarely observed nonsynonymous resistance of Na,K-ATPase to ouabain (a water-soluble stan- substitutions between individuals, the few cases being re- dard cardenolide) in all specialist agromyzid species feeding stricted to the large intracellular loop of the enzyme. Never- on cardiac glycoside–containing plants except for N. scro- theless, we acknowledge that we can recognize only gene phulariae and L. sp. (not enough material available). For copies that differ by at least a few substitutions and are ex- comparison, we included the nonadapted agromyzid spe- pressed in the adult flies. cies P. ilicis and D. melanogaster as an outgroup. For Na, K-ATPase in vitro assays, flies were decapitated, and pooled Phylogenetic Analyses heads (between 17–45) were homogenized in an all-glass grinder (Wheaton) in 200–500 mL deionized water. Homog- To test whether the chosen species pairs of Agromyzidae enates were frozen at 2807Candfreeze-dried.Lyophilisates were suitable phylogenetic comparisons, we used COI, 28S, were stored at 2807C until use. Na,K-ATPase assays were and CAD sequences from Scheffer et al. (2007) and Winkler carried out as described previously (Petschenka et al. 2013a). et al. (2009) and extended the data set with sequences of Briefly, head extractions reconstituted with 300–400 mLde- the flies investigated here (see table 1 for GenBank acces- 2 2 ionized water were incubated at a concentration of 10 8–10 3 sion numbers). From the original very extensive data set of M ouabain, and Na,K-ATPase activity was measured as the Scheffer et al. (2007), we included all species of Liriomyza, amount of inorganic phosphate generated from ATP hy- Napomyza,andPhytomyza and three species of all major drolysis compared to a noninhibited control. We used Origin- genera resolved between Agromyza and Phytomyza.Specif- Pro 2016 (OriginLab, Northampton, MA) for nonlinear curve ically, we chose three species of Agromyza, Aulagromyza, fitting. Top and bottom asymptotes were set to 100 and 0, Calycomyza, Cerodontha, Ophiomya,andPhytoliriomyza arc- respectively. For graphic display, replicate data (2–4 biolog- tica. In addition, we used COI and CAD sequence data for ical replicates per fly species) were fitted using the “concat- Phytomyza ilicis and Phytomyza crassiseta from Winkler enate fit” option. Pairwise statistical comparisons of dose- et al. (2009). In all cases where sequences from our own response curves were carried out using the “compare data RT-PCR or transcriptome analyses were available as well as sets” option in OriginPro. For this function of the program, sequences from Scheffer et al. (2007) or Winkler et al. (2009), curves fitted through the means of the data have to be used. the sequences were identical or the differences negligible We evaluated P values of pairwise statistical tests after Holm- (few silent substitutions in the protein-coding genes). Bonferroni corrections. Due to the unusual shape of the curve Sequence alignment for COI (1,404 bp) and CAD (704 bp) obtained for P. hellebori, curve fitting using the means was was straight-forward and immediately congruent with the not possible, and thus we excluded P. hellebori from statisti- alignment by Scheffer et al. (2007) archived in TreeBASE cal analysis. Raw data are deposited in the Dryad Digital Re- (www.treebase.org, accession no. SN3150). In contrast, 28S pository: http://dx.doi.org/10.5061/dryad.20191 (Petschenka proved difficult to align so that we used the alignment of et al. 2017). Scheffer and coauthors and manually added our sequences to this alignment in MEGA7 (Kumar et al. 2016). This resulted in a total length of 823 bp and the exclusion of 168 bases distributed in small groups along the gene that Scheffer and Results colleagues had judged impossible to align unambiguously. Phylogenetic Analyses The aligned concatenated sequences were subjected to maximum likelihood analysis with RAxML (Stamatakis 2014) fit- Analysis of a comprehensive sample of agromyzid flies in- ting a GTRGAMMA model of sequence evolution for each of cluding our focal species corroborated the suitability of the the three gene partitions separately. Na,K-ATPase sequences chosen phylogenetic comparisons. The three molecular mark- of the focal species were aligned with MEGA7 and subjected ers yielded a concatenated data set of 2,931 bp of 28S, COI, to maximum likelihood analysis assuming a GTRGAMMA and CAD sequences and resulted in a well-supported tree model of sequence evolution. For both data sets, bootstrap sup- that is in good agreement with the previous analysis of Schef- port values were generated in 500 replicated analyses. Align- fer et al. (2007; fig. 2). All genera as represented here are re- ments are archived in TreeBASE (www.treebase.org, acces- solved as monophyletic groups, with the exception of the sion no. 20548). placement of Chromatomyia horticola within the genus Phy-

Figure 2: Maximum likelihood analysis of COI, CAD, and 28S gene sequences of leaf-mining flies with GTRGAMMA models fitted to each gene partition. The data set was compiled from Scheffer et al. (2007) and enlarged by sequences from transcriptomes (asterisk), reverse transcription polymerase chain reaction (RT-PCR; double asterisk), or partly from Winkler et al. (2009), with additional sequences from RT-PCR (triple asterisk). Species representing our phylogenetic comparisons are boxed and connected by boldface lines, those feeding on cardiac glycoside–containing plants are indicated “1CG,” and those from plants devoid of cardiac glycosides are indicated “2CG.” Plant names indicate the hosts from which the specimens were collected. Branches are drawn according to scale; numbers along branches indicate bootstrap values in 500 replicated analyses. tomyza. The lack of clear differentiation between Chromato- Sequence Analysis of Na,K-ATPase Genes myia and Phytomyza species has been extensively demonstrated and discussed in Scheffer et al. (2007) and Winkler Our molecular screening for target site insensitivity mediated et al. (2009). More importantly, for the present study, our by mutations in the cardiac glycoside–binding pocket of the phylogenetic analysis corroborated the five independent col- Na,K-ATPase gene revealed multiple convergently acquired onizations of different plant lineages with cardiac glycosides, amino acid substitutions. No such substitutions were detected including the cardenolide type (in Asclepias, Digitalis,and in four out of five species that are not exposed to dietary car- Erysimum) and the bufadienolide type (in Helleborus; fig. 2). diac glycosides, rather these species maintained the ancestral In particular, the chosen species pairs, Phytomyza digitalis amino acid residues present in Drosophila melanogaster at all versus Phytomyza crassiseta, Phytomyza hellebori versus Phy- positions known to be part of the cardiac glycoside–binding tomyza fallaciosa, Napomyza scrophulariae versus Napomyza pocket (fig. A1, available online). In contrast, multiple substi- lateralis, Liriomyza asclepiadis/Liriomyza sp. Mexico versus tutions of amino acid residues were observed at these posi- Liriomyza eupatorii,andC. horticola versus Phytomyza ilicis tions in five out of six species feeding on cardiac glycosides were confirmed as appropriate phylogenetic comparisons plants. Specifically, two residues at the border of the first ex- that contrast species adapted to cardiac glycosides to non- tracellular loop, 111 and 122, which are well known to be crit- adapted relatives. ical for cardiac glycoside resistance in other insects, were re-

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:35 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S36 The American Naturalist peatedly exchanged (fig. 3). An exchange of glutamine 111 for lebori, our transcriptome analyses and RT-PCR revealed ev- leucine (Q111L), which we previously reported for the un- idence for a second gene copy. These two copies have a P dis- identified Liriomyza sp. collected from Asclepias angustifolia tance of 0.009, with seven nonsynonymous substitutions of in Mexico (Dobler et al. 2012) was observed in two further 21 total. In addition, all five individuals analyzed by PCR species, P. digitalis and N. scrophulariae. The same exchange were polymorphic at the same sites. One of the two gene is also present in the polyphagous N. lateralis and must have copies features the conserved glutamine at position 111 occurred in the common ancestor of both species, N. later- and the conserved asparagine at position 122, while the sec- alis and N. scrophulariae. The substitution of histidine at po- ond gene copy has a histidine at both positions (figs. 3, A1). sition 122 for the conserved asparagine detected earlier by An analysis of the read frequencies in the transcriptome direct RT-PCR amplification in L. asclepiadis (Dobler et al. showed that the double-substituted gene copy is roughly 2012) was confirmed here by our transcriptome analysis. The fourfold higher expressed than the one with the ancestral same substitution was once more discovered in P. hellebori condition. Other than these differences, the transcriptomes and coincides with an additional exchange of glutamine 111 provided evidence for only splice variants at the N-terminus for histidine, a substitution that has not been previously re- oftheenzymeandformutuallyexclusiveexons(exon6)imme- ported in insects or vertebrates. diately downstream of the region analyzed here. These splice A duplication of the Na,K-ATPase gene could be detected variants correspond to the well-described situation in D. mela- in only one of the agromyzid flies analyzed here. In P. hel- nogaster (Palladino et al. 2003).

Figure 3: Maximum likelihood analysis of Na,K-ATPase sequences of Agromyzidae (GTRGAMMA model). The amino acid residues at positions 111 and 122 are indicated along the branches and at the terminal nodes (numbering according to the mammalian Na,K-ATPase); the derived conditions are indicated in boldface and italic. Species on cardiac glycoside plants are boxed. The insert shows the relevant amino acids in the enlarged cardiac glycoside–binding cavity of the Na,K-ATPase. H p histidine, L p leucine, N p asparagine, Q p glutamine. Sequences are from transcriptomes (asterisk) or reverse transcription polymerase chain reaction (double asterisk). Branch lengths are drawn according to scale; numbers along branches indicate bootstrap values in 500 replicated analyses.

Na,K-ATPase Assays (Blanco and Mercer 1998). For fly-head preparations from P. hellebori and P. digitalis,weverified that ouabain is not Our in vitro assays of tissue-extracted fly Na,K-ATPases metabolized under the conditions of our assay, demon- revealed different degrees of cardenolide resistance across strating that the observed resistance is based on the bio- the fly species tested (fig. 4). Na,K-ATPase of the non- chemical properties of Na,K-ATPase and is not due to agromyzid outgroup species D. melanogaster was strongly degradation of the inhibitor. Moreover, ouabain is known inhibited by ouabain (half maximal inhibitory concentra- 2 not to be metabolized by the Malpighian tubules of D. tion [IC ] p 4:6 # 10 7 M). The Na,K-ATPase of the 50 melanogaster (Torrie et al. 2004). agromyzid species P. ilicis, which is not adapted to cardiac p glycosides, showed a similar response to ouabain (IC50 4:86 # 1027 M), and statistical comparison revealed no dif- Discussion ference between inhibition of the P. ilicis and D. melanogaster p : p : Na,K-ATPase (F2, 8 3 802, P 069). In contrast, Na,K- The strategies of herbivorous insects to cope with toxic car- ATPases of P. digitalis and L. asclepiadis displayed 3 and diac glycosides represent a striking case of convergent mo- 30 times increased cardenolide resistance, respectively (P. lecular evolution (Dobler et al. 2012, 2015; Zhen et al. 2012) p : # 26 : # digitalis:IC50 1 47 10 M; L. asclepiadis:143 that allows insights into the origin and underlying rules of 1025 M). Comparison of both dose-response curves to the adaptive amino acid substitutions (Storz 2016). The most curve obtained from P. ilicis and comparisons among each striking observation is the high frequency of occurrence other revealed significant differences (P. digitalis vs. P. ilicis: of one specific amino acid substitution in the Na,K-ATPase, p : ! : p F2, 8 100 664, P 001; L. asclepiadis vs. P. ilicis: F2, 8 N122H, that has evolved independently in six orders of phy- : ! : p 732 122, P 001; L. asclepiadis vs. P. digitalis: F2, 8 tophagous insects exposed to cardiac glycosides: in the Caeli- 504:26, P ! :001). Phytomyza hellebori was a special case fera (S. Dobler, G. Petschenka, and V. Wagschal, unpublished as ouabain inhibition of the P. hellebori Na,K-ATPase de- data), Coleoptera (at least twice), Diptera, Hemiptera, Hyme- viated from the monophasic sigmoidal shape of the other noptera, and Lepidoptera (Holzinger et al. 1992; Dobler et al. dose-response curves. The differing form of this curve re- 2012, 2015; Zhen et al. 2012). The present study increases the sembled a two-component curve caused by two coexisting number of independent origins of the N122H exchange, as we forms of Na,KATPase with different sensitivities to ouabain show that this substitution also evolved twice independently

Figure 4: Sensitivity of dipteran Na,K-ATPases to ouabain in vitro. Data points represent the mean of 2–4 biological replicates (5SE); the number of actual replicates per species is given in parentheses after species names. Solid lines indicate species that are exposed to cardiac glycosides in their diets; dashed lines represent species that are not naturally exposed to the toxins. Dark blue p Phytomyza ilicis (4), green p Drosophila melanogaster (4), purple p Phytomyza digitalis (4), light blue p Phytomyza hellebori (4), orange p Liriomyza asclepiadis (2). Note the deviating characteristics of P. hellebori.

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:35 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S38 The American Naturalist in the Agromyzidae (fig. 5). As we will discuss below, further of the Na,K-ATPase a gene (Aardema et al. 2012; Pierce potentially resistance-conferring amino acid substitutions have et al. 2016; S. Dobler and V. Wagschal, unpublished data) repeatedly evolved in the cardiac glycoside–binding pocket of nor in vitro enzyme assays testing for the resistance of the the Na,K-ATPase in insects as well as in vertebrate predators corresponding protein to cardiac glycosides (fig. 4; Pets- exposed to cardiac glycosides in their diet. chenka et al. 2013a; Bramer et al. 2015) so far revealed evi- As far as we can tell, a remarkable difference from our dence for noticeable intraspecific variability. This contrasts insects to other cases of convergent molecular evolution with other well-investigated systems where the molecular consists of the lack of intraspecific variability in resistance foundation of adaptive traits has been unraveled, such as to dietary cardiac glycosides. Neither sequence analyses the blanched coloration displayed by lizards at White Sands

Figure 5: Compilation of available data on Na,K-ATPase sequences in insects adapted and nonadapted to cardiac glycosides. The overall tree and node ages correspond to the best-supported insect phylogeny to date (Misof et al. 2014) that has been complemented by phylogenetic hypotheses presented in the Tree of Life Web Project (Maddison and Schulz 2007) and our own data. Species in black are not exposed to cardiac glycosides, species in orange are known to sequester cardiac glycosides from their host plants, and species in blue feed on cardiac glycosides plants but either are not known or not able to sequester cardiac glycosides. To the right are amino acid residues known to form the cardiac glycoside–binding pocket of the Na,K-ATPase (see ﬁg. 1), with potential resistance-conferring exchanges highlighted by gray boxes. The exchange of E312D (glutamic acid for aspartic acid) between vertebrates and all insects is considered negligible. Stars indicate gene duplications. In Lygaeus kalmii and Oncopeltus fasciatus, both indicated copies are present plus a third less heavily substituted one featuring N122H and F786S exchanges. In Rhyssomatus lineaticollis, a second copy has the ancestral Q111 but a N122Y substitution. In Chrysochus auratus, the second copy is ancestral except for a Q111L exchange (Zhen et al. 2012). In Phytomyza hellebori,thesecondgene copy has the ancestral state. For GenBank accession numbers, see table 1.

(Rosenblum et al. 2010; this issue) and beach mice (Hoekstra the binding cavity in a similar fashion (Laursen et al. 2015). et al. 2006; Mullen and Hoekstra 2008) or insect resistance to Our data further support that the optimal adaptation does pyrethroids (Rinkevich et al. 2013). In these cases, heritable not depend on the type of cardiac glycoside (or other aspects variability persists since the selective pressures vary between of host plant variation), as a histidine at position 122 was habitats. In the case of adaptations to dietary cardiac gly- found in Phytomyza hellebori exposed to bufadienolides (and cosides, adoption of these feeding habits lies in the distant the same applies to the sawfly Monophadnus latus; Dobler past and the favorable mutations have apparently been driven et al. 2015), while Liriomyza asclepiadis, like the other species to fixation (Aardema et al. 2012). A screen for intraspecific featuring the N122H exchange, is exposed to cardenolides in variability might be most promising in generalist insect spe- its Asclepias host plants (fig. 5). cies if populations vary in host use, feeding locally on plants In many cases, in insects and vertebrates alike, substitu- with or without cardiac glycosides. tions at position 122 are accompanied by exchanges at po- While we detected a striking amount of molecular con- sition 111. The two residues Q111 and N122 are situated at the vergence in resistance-conferring substitutions in the Na, entrance of the cardiac glycoside–binding cavity and have K-ATPase, this strategy is not the only possible adaptation long been known to be of special importance for cardiac to avoid the toxic effect of dietary cardiac glycosides. Imper- glycoside–binding affinity (Price and Lingrel 1988; Price meable guts that prevent the toxins from entering the body, et al. 1990; Croyle et al. 1997). As the crystal structures or fast and efficient elimination from the hemolymph via the show, binding of cardiac glycosides leads to a conforma- Malpighian tubules, represent possible alternatives and have tional change bringing the aM1–M2 loop closer to the li- been detected in nonadapted as well as cardiac glycoside– gand, and these two residues get close enough for hydrogen adapted insects (Scudder and Meredith 1982; Torrie et al. bonding (Yatime et al. 2011; Laursen et al. 2015). The re- 2004; Petschenka and Dobler 2009; Petschenka and Agrawal placement of Q111 and N122 by residues not capable of hy- 2015). In addition, a protection of the sensitive neural tis- drogen bonding or having bulkier and charged side chains sue expressing high levels of Na,K-ATPase by active carriers like histidine apparently prohibit high-affinity cardiac gly- in the insects’ blood-brain barrier plays an important role and coside binding. Experimental evidence suggests that substi- is possibly sufficient to cope with small concentrations of he- tutions at positions 111 and 122 have synergistic functional molymph cardiac glycosides (Petschenka and Dobler 2009; effects on resistance (Price et al. 1990; Dobler et al. 2012; Petschenka et al. 2013b). Thus, alternative evolutionary tra- Dalla et al. 2013; Dalla and Dobler 2016). jectories are possible to cope with dietary cardiac glycosides. Compared to position 122, where we observed only ex- Remarkably, convergent molecular evolution of cardiac changes of N122H, position 111 is more variable as to which glycoside resistance by amino acid substitutions has also amino acid substitutions cause resistance to cardiac glyco- evolved in several vertebrates feeding on bufadienolide- sides. At this position, leucine, valine, threonine, and glu- producing toads (Jaisser et al. 1992; Ujvari et al. 2013, tamic acid have been repeatedly observed in cardiac glycoside– 2015; Mohammadi et al. 2016). As in insects, amino acid adapted insects (Labeyrie and Dobler 2004; Dobler et al. substitutions in the first extracellular loop of the Na,K- 2012, 2015; Zhen et al. 2012), and here for the first time ATPase mediate resistance to these toxins in their diet. It an exchange of Q111 for histidine is reported. Position 111 is, however, noteworthy that in vertebrates adapted to car- is of importance in vertebrates adapted to cardiac glycosides diac glycosides, the substitution of N122H observed so of- as well, and the observed exchanges partially match those ten in insects has never been described. Rather, other amino present in insects. In snakes and lizards known to feed on acid replacements are present at position 122 (Price and toads, the Q111L exchange has evolved four times conver- Lingrel 1988; Jaisser et al. 1992; Ujvari et al. 2013, 2015). gently, while toads, rodents, the European hedgehog, and Two reasons could account for these differences: First, the the frog Leptodactylus latrans display a substitution of argi- genetic background of the remaining enzyme may differ be- nine for Q111 (Price and Lingrel 1988; Jaisser et al. 1992; tween insects and vertebrates, and thereby, the effect of spe- Ujvari et al. 2013; Ujvari et al. 2015). Though in vertebrates cific substitutions may also vary (e.g., Poelwijk et al. 2007; the exchanges of Q111 are also coupled with substitutions at Storz 2016). Mutagenesis experiments that introduced the the C-terminal end of the first extracellular loop of the Na, N122H exchange into the human and the Drosophila Na, K-ATPase, these exchanges (at positions 119, 120, or 122) dif- K-ATPase in cell culture corroborate this idea (S. Dobler, fer from those observed in insects. Overall, this demonstrates S. Dalla, and F. List, unpublished data). Second, the toxins an astonishingly high degree of molecular convergence across the two groups are exposed to may require different adap- animal phyla, but the use of different (presumably optimal) tations. While the investigated vertebrates have evolved re- substitutions at the same time provides evidence for con- sistance to toad bufadienolides, most insects analyzed so straints (or phylogenetic bias) that may be due to either the far are exposed to plant cardenolides. Recent crystallogra- differences in the genetic background of the vertebrate versus phy, however, suggests that both types of toxins occupy insect Na,K-ATPase or different physiological conditions.

Although Q111L seems to be associated with resistance et al. 2013; Petschenka et al. 2013a; Bramer et al. 2015). In in vertebrates, the functional effect of this substitution accordance with these results obtained from genetically en- has not been tested in isolation, that is, without the accom- gineered Drosophila Na,K-ATPase, the inhibition curve ob- panying substitutions at the positions mentioned above. tained for L. asclepiadis shows that the single exchange of While the observed resistance in Phytomyza digitalis coin- N122H in an enzyme that is otherwise highly similar to the cideswithQ111L,thereisnoevidencethatthissubstitu- sensitive Drosophila Na,K-ATPase is sufficient to increase re- tion actually increases resistance of Na,K-ATPase in several sistance by 31-fold. lepidopteran insects, including nymphalid butterflies, arctiid The heterogeneity of Na,K-ATPase sequences across the moths, and the oleander hawk moth (Petschenka et al. 2012, fly species adapted to cardiac glycosides is mirrored by the 2013a; Dobler et al. 2015). Compared to butterflies and different sensitivities of their Na,K-ATPases to ouabain. moths, the situation in leaf-mining flies might be different, Drosophila melanogaster and P. ilicis display similar sen- and the effect of Q111L needs to be investigated in more de- sitivities to ouabain, suggesting that the Na,K-ATPase of tail. Along the same lines, it is currently unclear why the Na, P. ilicis likely represents the ancestral condition in the Agro- K-ATPase of the generalist Napomyza lateralis carries this myzidae. The dose-response curve of P. digitalis Na,K- substitution, although the species might not be exposed to ATPase in comparison displays a threefold-higher resistance cardiac glycosides in its diet. to ouabain, thus supporting adaptation to its host plant’s In insects as well as in vertebrates, the available data sug- cardenolides. The highest in vitro resistance in our data set gest that amino acid substitutions in and at the border of was observed for L. asclepiadis. However, P. hellebori may the first extracellular loop (M1–M2) are much more likely have an even more resistant form of the Na,K-ATPase, but to evolve than exchanges in the remainder of the cardiac its interpretation is confounded by the coexistence of two glycoside–binding cavity. According to the Na,K-ATPase different forms of Na,K-ATPase. In P. hellebori, the dose- crystal structures, bulky and hydrophobic side chains of response curve can be interpreted as displaying the be- the transmembrane helices M4–M6 form a general docking havior of two forms with differing sensitivity: while inhi- platform for the steroid core of all cardiac glycosides, while bition of the sensitive copy (with the ancestral Q111 and N122 their lactone rings fit in a hydrophobic funnel of these amino acids) leads to a rapid activity decline at relatively transmembrane helices that is leading to the cation binding low concentrations of ouabain, the resistant copy (with H111 sites of the enzyme (Laursen et al. 2015). The proximity to and H122) maintains a level of activity at millimolar ouabain the cation binding sites seems to create functional con- concentrations that is higher than for L. asclepiadis. This straints that may explain why amino acid substitutions in resembles the situation described for rat testis, where both these parts of the enzyme are so rare. So far they have been the a1anda4 isoforms, with strongly differing ouabain sen- observed only in lygaeid bugs that possess three copies of sitivities, are expressed together (Blanco and Mercer 1998). the Na,K-ATPase a1 gene and in aphids where retrotrans- Gene duplications as observed here for P. hellebori could position seems to have created a second gene copy with al- buffer the detrimental effects of substitutions by counter- tered properties (Zhen et al. 2012). Functional data support acting purifying selection on individual proteins and could the notion that the amino acid exchanges observed in trans- have a direct advantage due to expression of higher gene membrane domains M5 and M6 of Oncopeltus fasciatus se- doses (Soskine and Tawfik 2010; Kondrashov 2012). Other riously compromise the enzyme’s activity as an ion trans- models, in contrast, suggest that gene duplication is a neu- porter (Dalla and Dobler 2016). tral event that may not be under selective pressure (Soskine Testing the functional effects of the observed amino acid and Tawfik 2010). For insects adapted to cardiac glycosides, substitutions is critical for understanding the value of single previous analyses suggested that resistance-conferring mu- replacements versus combinations as well as potential pleio- tations may go along with duplications of the Na,K-ATPase tropic constraints. Two approaches can be followed: en- a1 gene (Zhen et al. 2012; indicated by stars in fig. 5). This is zyme assays, as used here, to test for the sensitivity of the especially prominent in lygaeid bugs where two duplica- native enzyme that is expressed at high levels in the nervous tions gave rise to three gene copies that differ in the number tissue of animals (Vaughan and Jungreis 1977; Moore and of substitutions, including unusual exchanges toward the Scudder 1986; Al-Robai 1993; Petschenka et al. 2012) or C-terminus of the enzyme (Zhen et al. 2012). The Agro- genetic modification of Na,K-ATPase genes followed by myzidae P. hellebori provides a further example for a dupli- evaluation of the physiological effects of the introduced mu- cation of the Na,K-ATPase gene that resulted in an unal- tations in enzymes expressed in cell culture (Price et al. 1990; tered, ancestral form and a fourfold-higher expressed form Holzinger et al. 1992; Qiu et al. 2005; Dalla et al. 2013; Ujvari with two resistance-conferring mutations in the M1–M2 et al. 2013; Dalla and Dobler 2016). Both lines of evidence domain (Q111H, N122H). As for the debate whether substi- demonstrate that the exchange of N122H causes a high level tutions precede duplications or only follow later when the of resistance to cardiac glycosides (Holzinger et al. 1992; Dalla duplicate is freed from purifying selection (Soskine and Taw-

fik 2010), P. hellebori provides an example for an accu- Agrawal, A. A., G. Petschenka, R. A. Bingham, M. G. Weber, and S. mulation of mutations post duplication, as the ancestral Rasman. 2012. Toxic cardenolides: chemical ecology and coevolu- copy that is shared by the closely related P. fallaciosa is still tion of specialized plant-herbivore interactions. New Phytologist 194:28–45. present. The duplication may have been driven by the ad- Al-Robai, A. A. 1993. Different ouabain sensitivities of Na1/K1- vantage of increased gene dosage in the face of toxins com- ATPase from Poekilocerus bufonius tissues and a possible physio- promising gene function and subfunctionalization of the du- logical cost. Comparative Biochemistry and Physiology B: Bio- plicates ensued later (Kondrashov 2012). chemistry and Molecular Biology 106:805–812. Duplications of the Na,K-ATPase gene can alleviate neg- Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. ative pleiotropy imposed by the dual role of the Na,K-ATPase Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a for ion transport and morphogenesis (Xie and Askari 2002; new generation of protein database search programs. Nucleic Acids Research 25:3389–3402. Palladino et al. 2003; Horisberger 2004; Paul et al. 2007) and Andreev, D., M. Kreitmann, T. W. Phillips, R. W. Beeman, and R. H. allow for resistance-conferring substitutions that compromise ffrench-Constant. 1999. Multiple origins of cyclodiene insecticide ion transport (Dalla and Dobler 2016). Nevertheless, in our resistance in Tribolium castaneum (Coleoptera: Tenebrionidae). analysis of Agromyzidae on cardiac glycoside plants, we de- Journal of Molecular Evolution 48:615–624. tected four incidents of substitutions in the cardiac glycoside– Blanco, G., and R. W. Mercer. 1998. Isozymes of the Na-K-ATPase: binding pocket but only a single gene duplication event, de- heterogeneity in structure, diversity in function. American Journal – spite the fact that all five independent lineages started from of Physiology-Renal Physiology 275:F633 F650. Bramer, C., S. Dobler, J. Deckert, M. Stemmer, and G. Petschenka. the same ancestral state. This suggests that in the case of 2015. Na1/K1-ATPase resistance and cardenolide sequestration: the Na,K-ATPase, constraints on gene duplications seem basal adaptations to host plant toxins in the milkweed bugs (He- to be stronger than constraints on amino acid substitutions, miptera: Lygaeidae: Lygaeinae). Proceedings of the Royal Society at least as long as substitutions occur only at the favored res- B 282, http://dx.doi.org/10.1098/rspb.2014.2346. idues of the first extracellular loop and the bordering trans- Carvalho, G. B., W. W. Ja, and S. Benzer. 2009. Non-lethal PCR membrane domains (M1–M2). We still know too little about genotyping of single Drosophila. Biotechniques 46:312–314. additional roles of insect Na,K-ATPase genes beyond ion Croyle, M. L., A. L. Woo, and J. B. Lingrel. 1997. Extensive random mutagenesis analysis of the Na1/K1-ATPase a subunit identifies transport (Paul et al. 2007), but the evidence accumulating known and previously unidentified amino acid residues that alter for vertebrate Na,K-ATPases suggests that pleiotropic con- ouabain sensitivity—implications for ouabain binding. European straints by Na,K-ATPase–initiated signaling cascades may Journal of Biochemistry 248:488–495. be immense (Xie and Askari 2002; Schoner and Scheiner- Dalla, S., and S. Dobler. 2016. Gene duplications circumvent trade- Bobis 2007; Li and Xie 2009). The functional constraints of offs in enzyme function: insect adaptation to toxic host plants. this enzyme imposed by its dual role thus seem to enforce Evolution 70:2767–2777. a strongly canalized evolutionary path. Dalla, S., H. G. P. Swarts, J. B. Koenderink, and S. Dobler. 2013. Amino acid substitutions of Na,K-ATPase–conferring decreased sensitivity to cardenolides in insects compared to mammals. In- Acknowledgments sect Biochemistry and Molecular Biology 43:1109–1115. Dobler, S., S. Dalla, V. Wagschal, and A. A. Agrawal. 2012. Community- We thank A. Agrawal for the opportunity to contribute wide convergent evolution in insect adaptation to toxic cardenolides to this symposium, for helpful discussions, and for his crit- by substitutions in the Na,K-ATPase. Proceedings of the National ical reading of the manuscript. He, A. Hastings, and T. Züst Academy of Sciences of the USA 109:13040–13045. helped with the collection of Liriomyza asclepiadis. M. Fal- Dobler, S., G. Petschenka, and H. C. Pankoke. 2011. Coping with toxic ’ kenberg and R. Trusch supported the rediscovery of Phy- plant compounds: the insect s perspective on iridoid glycosides and cardenolides. Phytochemistry 72:1591–1602. tomyza digitalis in Germany. L. M. Arcila Hernandez, A. Dobler, S., G. Petschenka, V. Wagschal, and L. Flacht. 2015. Conver- Baudach, R. Halitschke, M. Schott, and S. Stiehler helped gent adaptive evolution: how insects master the challenge of cardiac with collecting leaf-mining flies and data. A. Vilcinskas gen- glycoside–containing host plants. Entomologia Experimentalis et erously provided materials for some parts of this study to Applicata 157:30–39. G.P. The project was funded by a grant from the Templeton Feldman, C. R., E. D. Brodie Jr., E. D. Brodie 3rd, and M. E. Foundation and Deutsche Forschungsgemeinschaft (DFG; Pfrender. 2012. Constraint shapes convergence in tetrodotoxin- Do 527/5-3) to S.D. G.P. was financially supported by the resistant sodium channels of snakes. Proceedings of the National Academy of Sciences of the USA 109:4556–4561. DFG (PE 2059/1, PE 2059/2). Glombitza, K.-W., C. Kucera-Waldmann, and U. Fricke. 1989. Do roots of Helleborus niger contain cardioactive substances? Planta Literature Cited Medica 55:107. Grabherr, M. G., B. J. Haas, M. Yassour, J. Z. Levin, D. A. Thompson, Aardema, M. L., Y. Zhen, and P. Andolfatto. 2012. The evolution of I. Amit, X. Adiconis, et al. 2011. Trinity: reconstructing a full-length cardenolide-resistant forms of Na1,K1-ATPase in Danainae but- transcriptome without a genome from RNA-Seq data. Nature Bio- terflies. Molecular Ecology 21:340–349. technology 29:644–652.

Hering, E. M. 1927. I. Agromyzidae (80. Familie). Die Tierwelt Maddison, D. R., and K. S. Schulz. 2007. The Tree of Life Web Proj- Deutschlands und der angrenzenden Meeresteile nach ihren Merk- ect. http://tolweb.org/tree. Accessed August 30, 2016. malen und nach ihrer Lebensweise. Fischer, Jena. Malcolm, S. B. 1991. Cardenolide-mediated interactions between ———. 1935. Die Blattminen Mittel- und Nord-Europas einschliesslich plants and herbivores. Pages 251–295 in G. A. Rosenthal and Englands; Bestimmungstabellen aller von Insektenlarven der verschie- M. R. Berenbaum, eds. Herbivores: their interactions with second- denen Ordnungen erzeugten Minen. G. Feller, Neubrandenburg. ary plant metabolites. Academic Press, San Diego, CA. ———. 1951. Biology of the leaf miners. Dr. W. Junk, s-Gravenhage. Maus, C., K. Peschke, and S. Dobler. 2001. Phylogeny of Aleochara Hoekstra, H. E., R. J. Hirschmann, R. A. Bundey, P. A. Insel, and J. P. inferred from mitochondrial cytochrome oxidase sequences (Co- Crossland. 2006. A single amino acid mutation contributes to leoptera: Staphylinidae). Molecular Phylogenetics and Evolution adaptive beach mouse color pattern. Science 313:101–104. 18:202–216. Holzinger, F., C. Frick, and M. Wink. 1992. Molecular basis for the Misof, B., S. L. Liu, K. Meusemann, R. S. Peters, A. Donath, C. insensitivity of the monarch (Danaus plexippus) to cardiac glyco- Mayer, P. B. Frandsen, et al. 2014. Phylogenomics resolves the sides. FEBS Letters 314:477–480. timing and pattern of insect evolution. Science 346:763–767. Horisberger, J. D. 2004. Recent insights into the structure and mech- Mohammadi, S., E. D. Brodie Jr., L. A. Neuman-Lee, and A. H. anism of the sodium pump. Physiology 19:377–387. Savitzky. 2016. Mutations to the cardiotonic steroid binding site Jaisser, F., C. M. Canessa, J. D. Horisberger, and B. C. Rossier. 1992. of Na1/K1-ATPase are associated with high level of resistance to Primary sequence and functional expression of a novel ouabain- gamabufotalin in a natricine snake. Toxicon 114:13–15. resistant Na,K-ATPase—the beta-subunit modulates potassium Moore, L. V., and G. G. E. Scudder. 1986. Ouabain-resistant Na,K- activation of the Na,K-pump. Journal of Biological Chemistry 267: ATPases and cardenolide tolerance in the large milkweed bug, 16895–16903. Oncopeltus fasciatus. Journal of Insect Physiology 32:27–33. Joshi, N. A., and J. N. Fass. 2011. Sickle: a sliding-window, adaptive, Mortelmans, J., M. Boeraeve, W. Tamsyn, W. Proesmans, and D. quality-based trimming tool for FastQ files. Version 1.29. https:// Dekeukeleire. 2014. Thirteen new Agromyzidae for Belgium. Bul- github.com/najoshi/sickle. letin de la Societé Royale Belge d’Entomologie 150:141–148. Kondrashov, F. A. 2012. Gene duplication as a mechanism of genomic Mullen, L. M., and H. E. Hoekstra. 2008. Natural selection along an adaptation to a changing environment. Proceedings of the Royal environmental gradient: a classic cline in mouse pigmentation. Evo- Society B 279:5048–5057. lution 62:1555–1570. Krenn, L., and B. Kopp. 1998. Bufadienolides from animal and plant Palladino, M. J., J. E. Bower, R. Kreber, and B. Ganetzky. 2003. Neural sources. Phytochemistry 48:1–29. dysfunction and neurodegeneration in Drosophila Na1/K1 ATPase Kumar, S., G. Stecher, and K. Tamura. 2016. MEGA7: Molecular Evo- alpha subunit mutants. Journal of Neuroscience 23:1276–1286. lutionary Genetics Analysis. Version 7.0 for bigger datasets. Molec- Paul, S. M., M. J. Palladino, and G. J. Beitel. 2007. A pump-independent ular Biology and Evolution 33:1870–1874. function of the Na,K-ATPase is required for epithelial junction func- Labeyrie, E., and S. Dobler. 2004. Molecular adaptation of Chry- tion and tracheal tube-size control. Development 134:147–155. sochus leaf beetles to toxic compounds in their food plants. Molec- Petschenka, G., and A. A. Agrawal. 2015. Milkweed butterflyresis- ular Biology and Evolution 21:218–221. tance to plant toxins is linked to sequestration, not coping with a Laursen, M., J. L. Gregersen, L. Yatime, P. Nissen, and N. U. Fedosova. toxic diet. Proceedings of the Royal Society B 282:20151865. 2015. Structures and characterization of digoxin- and bufalin- Petschenka, G., and S. Dobler. 2009. Target-site sensitivity in a spe- bound Na1,K1-ATPase compared with the ouabain-bound com- cialized herbivore towards major toxic compounds of its host plant: plex. Proceedings of the National Academy of Sciences of the the Na1K1-ATPase of the oleander hawk moth (Daphnis nerii)is USA 112:1755–1760. highly susceptible to cardenolides. Chemoecology 19:235–239. Laursen, M., L. Yatime, P. Nissen, and N. U. Fedosova. 2013. Crystal Petschenka, G., S. Fandrich, N. Sander, V. Wagschal, M. Boppré, and structure of the high-affinity Na1K1-ATPase-ouabain complex S. Dobler. 2013a. Stepwise evolution of resistance to toxic carde- with Mg21 bound in the cation binding site. Proceedings of the nolides via genetic substitutions in the Na1,K1-ATPase of milk- National Academy of Sciences of the USA 110:10958–10963. weed butterflies (Lepidoptera: Danaini). Evolution 67:2753–2761. Lebovitz, R. M., K. Takeyasu, and D. M. Fambrough. 1989. Mole- Petschenka, G., J. Offe, and S. Dobler. 2012. Physiological screening cular characterization and expression of the (Na1 1 K1)-ATPase for target site insensitivity and localization of Na1/K1-ATPase in alpha-subunit in Drosophila melanogaster. Embo Journal 8:193– cardenolide-adapted Lepidoptera. Journal of Insect Physiology 202. 58:607–612. Lei, Z.-H., S. Yahara, T. Nohara, T.-B. Shan, and J.-Z. Xiong. 1996. Petschenka, G., C. Pick, V. Wagschal, and S. Dobler. 2013b.Func- Cardenolides from Erysimum cheiranthoides. Phytochemistry 41: tional evidence for physiological mechanisms to circumvent neu- 1187–1189. rotoxicity of cardenolides in an adapted and a non-adapted hawk- Li, Z., and Z. Xie. 2009. The Na/K-ATPase/Src complex and cardio- moth species. Proceedings of the Royal Society B 280:20123089. tonic steroid-activated protein kinase cascades. Pflugers Archive Petschenka, G., V. Wagschal, M. von Tschirnhaus, A. Donath, and S. 457:635–644. Dobler. 2017. Data from: Convergently evolved toxic secondary Liang, M., J. Tian, L. Liu, S. Pierre, J. Liu, J. Shapiro, and Z. J. Xie. metabolites in plants drive the parallel molecular evolution of insect 2007. Identification of a pool of non-pumping Na/K-ATPase. resistance. American Naturalist, Dryad Digital Repository, http:// Journal of Biological Chemistry 282:10585–10593. dx.doi.org/10.5061/dryad.20191. Luckner, M., and M. Wichtl. 2000. Digitalis. Wissenschaftliche, Pierce, A. A., J. C. De Roode, and L. Tao. 2016. Comparative genetics Stuttgart. of Na1/K1-ATPase in monarch butterfly populations with varying Ludwig, F. 1907. Weiteres zur Biologie von Helleborus foetidus. host plant toxicity. Biological Journal of the Linnean Society 119: Zeitschrift für wissenschaftliche Insektenbiologie 3:45–50. 194–200.

Poelwijk, F. J., D. J. Kiviet, D. M. Weinreich, and S. J. Tans. 2007. Storz, J. F. 2016. Causes of molecular convergence and parallelism in Empirical fitness landscapes reveal accessible evolutionary paths. protein evolution. Nature Reviews Genetics 17:239–250. Nature 445:383–386. Tarvin, R. D., J. C. Santos, L. A. O’Connell, H. H. Zakon, and D. C. Price, E. M., and J. B. Lingrel. 1988. Structure-function relationships Cannatella. 2016. Convergent substitutions in a sodium channel in the Na, K-ATPase alpha subunit: site-directed mutagenesis of suggest multiple origins of toxin resistance in poison frogs. Molec- glutamine-111 to arginine and asparagine-122 to aspartic acid ular Biology and Evolution 33:1068–1081. generates a ouabain-resistant enzyme. Biochemistry 27:8400–8408. Torrie, L. S., J. C. Radford, T. D. Southall, L. Kean, A. J. Dinsmore, Price, E. M., D. A. Rice, and J. B. Lingrel. 1990. Structure-function S. A. Davies, and J. A. Dow. 2004. Resolution of the insect ouabain studies of Na,K-ATPase: site-directed mutagenesis of the border paradox. Proceedings of the National Academy of Sciences of the residues from the H1-H2 extracellular domain of the a subunit. USA 101:13689–13693. Journal of Biological Chemistry 265:6638–6641. Ujvari, B., N. R. Casewell, K. Sunagar, K. Arbuckle, W. Wuster, N. Qiu, L. Y., E. Krieger, G. Schaftenaar, H. G. P. Swarts, P. H. G. M. Lo, D. O’Meally, et al. 2015. Widespread convergence in toxin re- Willems, J. L. H. H. M. De Pont, and J. B. Koenderink. 2005. Re- sistance by predictable molecular evolution. Proceedings of the construction of the complete ouabain-binding pocket of Na,K- National Academy of Sciences of the USA 112:11911–11916. ATPase in gastric H,K-ATPase by substitution of only seven amino Ujvari, B., H. C. Mun, A. D. Conigrave, A. Bray, J. Osterkamp, acids. Journal of Biological Chemistry 280:32349–32355. P. Halling, and T. Madsen. 2013. Isolation breeds naivety: island Rinkevich, F. D., Y. Du, and K. Dong. 2013. Diversity and convergence living robs Australian varanid lizards of toad-toxin immunity via of sodium channel mutations involved in resistance to pyrethroids. four-base-pair mutation. Evolution 67:289–294. Pesticide Biochemistry and Physiology 106:93–100. Vaughan, G. L., and A. M. Jungreis. 1977. Insensitivity of lepidopteran Rosenblum, E. B., H. Rompler, T. Schoneberg, and H. E. Hoekstra. tissues to ouabain: physiological mechanisms for protection from 2010. Molecular and functional basis of phenotypic convergence cardiac glycosides. Journal of Insect Physiology 23:585–589. in white lizards at White Sands. Proceedings of the National Acad- Winkler, I. S., S. J. Scheffer, and C. Mitter. 2009. Molecular phylog- emy of Sciences of the USA 107:2113–2117. eny and systematics of leaf-mining flies (Diptera: Agromyzidae): Scheffer, S. J., I. S. Winkler, and B. M. Wiegmann. 2007. Phylogenetic delimitation of Phytomyza Fallén sensu lato and included species relationships within the leaf-mining flies (Diptera: Agromyzidae) groups, with new insights on morphological and host-use evolu- inferred from sequence data from multiple genes. Molecular tion. Systematic Entomology 34:260–292. Phylogenetics and Evolution 42:756–775. Wissner, W., and H. Kating. 1974. Botanische und Phytochemische Schoner, W., and G. Scheiner-Bobis. 2007. Endogenous and exoge- Untersuchungen an den europäischen und kleinasiatischen Arten nous cardiac glycosides: their roles in hypertension, salt metabo- der Gattung Helleborus. Planta Medica 26:228–249. lism, and cell growth. American Journal of Physiology–Cell Phys- Xie, Z., and A. Askari. 2002. Na1/K1-ATPase as a signal transducer. iology 293:C509–C536. European Journal of Biochemistry 269:2434–2439. Scudder, G. G. E., and J. Meredith. 1982. The permeability of the Yatime, L., M. Laursen, J. P. Morth, M. Esman, P. Nissen, and N. U. midgut of three insects to cardiac glycosides. Journal of Insect Fedosova. 2011. Structural insights into the high affinity binding Physiology 28:689–694. of cardiotonic steroids to the Na1,K1-ATPase. Journal of Structural Soskine, M., and D. S. Tawfik. 2010. Mutational effects and the evolu- Biology 174:296–306. tion of new protein functions. Nature Reviews Genetics 11:572–582. Zhang, Z., S. Schwartz, L. Wagner, and W. Miller. 2000. A greedy Spencer, K. A. 1966. A clarification of the genus Napomyza West- algorithm for aligning DNA sequences. Journal of Computational wood (Diptera: Agromyzidae). Proceedings of the Royal Entomo- Biology 7:203–214. logical Society B 35:29–40. Zhen, Y., M. L. Aardema, E. M. Medina, M. Schumer, and P. Andolfatto. ———. 1990. Host specialization in the world Agromyzidae (Dip- 2012. Parallel molecular evolution in an herbivore community. Sci- tera). Kluwer Academic, Dordrecht. ence 337:1634–1637. Stamatakis, A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30: 1312–1313. Symposium Editor: Anurag A. Agrawal

Symposium Convergent Phenotypic Evolution despite Contrasting Demographic Histories in the Fauna of White Sands*

Erica Bree Rosenblum,1,2,† Christine E. Parent,1,‡,§ Eveline T. Diepeveen,1,‡,∥ Clay Noss,1 andKeBi2,3

1. Department of Environmental Science, Policy, and Management, University of California, Berkeley, California 94720; 2. Museum of Vertebrate Zoology, University of California, Berkeley, California 94720; 3. Computational Genomics Resource Laboratory, University of California, Berkeley, California 94720 Online enhancements: appendixes. abstract: When are evolutionary outcomes predictable? Cases of Understanding the factors that predict evolutionary response convergent evolution can shed light on when, why, and how different to natural selection is a central goal of evolutionary biology. species exhibit shared evolutionary trajectories. In particular, study- Instances of convergent evolution—the independent evolu- ing diverse species in a common environment can illuminate how dif- tion of phenotypic similarity in different lineages—provide ferent factors facilitate or constrain adaptive evolution. Here we inte- unparalleled opportunities to study the mechanisms of adap- grate studies of pattern and process in the fauna at White Sands (New tive trait evolution. Ultimately, whether different species ex- Mexico) to understand the determinants of convergent evolution. hibit a similar evolutionary response depends critically on Numerous animal species at White Sands exhibit phenotypic convergence in response to a novel—and shared—selective environment: three interacting determinants (reviewed in Rosenblum et al. geologically young gypsum dunes. We synthesize 15 years of research 2014): (1) natural selection (e.g., whether the mode, strength, on White Sands lizards to assess the contribution of natural selection, and dimensionality of selection is similar across species), genetic architecture, and population demography to patterns of phe- (2) genetic architecture of adaptive traits (e.g., whether her- notypic evolution. We also present new data for two species of White itability, mutational effect sizes, and patterns of epistasis are Sands arthropods, Ammobaenetes arenicolus and Habronattus ustula- parallel across species), and (3) population demography (e.g., fi tus. Overall, we nd dramatic phenotypic convergence across diverse whether population size, patterns of population structure, species at White Sands. Although the direction of phenotypic response fl is parallel, the magnitude of phenotypic response varies among species. and dynamics of gene ow are similar across species). We also find that species exhibit strikingly different demographic A key challenge for understanding evolutionary predict- patterns across the ecotone. The species with the most genetic structure ability—even for cases of convergent evolution—is disen- between White Sands and dark-soil populations generally exhibit the tangling the factors that can promote or hinder adaptive trait least phenotypic divergence, suggesting population demography as a evolution. When different lineages in different environments key modulator of adaptation. Comparative studies are particularly im- are compared, patterns of phenotypic evolution can seem id- portant for understanding the determinants of convergence in natural iosyncratic, and it can be difficult to distinguish the contri- systems. butions of selection, demography, and genomic architecture Keywords: White Sands, convergent evolution, jumping spiders, to observed patterns. In contrast, cases of repeated evolution sand-treader crickets, restriction site–associated DNA sequencing. within a single community provide exciting opportunities to differentiate among the determinants of adaptive evolution. When distantly related species exhibit similar phenotypic re- * This issue originated as the 2016 Vice Presidential Symposium presented at sponses in a shared environment, we can address more nu- the annual meetings of the American Society of Naturalists. † anced hypotheses about evolutionary predictability. Corresponding author; e-mail: [email protected]. ‡ These authors contributed equally. Here, we integrate studies of pattern and process in the § Present address: Department of Biological Sciences, University of Idaho, White Sands system to understand the factors that pro- Moscow, Idaho 83844. mote and hinder convergent evolution. We first synthesize ∥ Present address: Department of Bionanoscience, Delft University of Tech- what we have learned about the patterns of convergent evo- nology, Delft, Netherlands. lution—and the mechanisms influencing that pattern— ORCIDs: Parent, http://orcid.org/0000-0002-4378-6715; Diepeveen, http:// orcid.org/0000-0002-0768-8528; Noss, http://orcid.org/0000-0001-5114-6155. from 15 years of work on White Sands lizards. We then pre- Am. Nat. 2017. Vol. 190, pp. S44–S56. q 2017 by The University of Chicago. sent new data on two invertebrate species as a next step toward 0003-0147/2017/190S1-57238$15.00. All rights reserved. a community-scale synthesis. Finally, we highlight outstand- DOI: 10.1086/692138 ing questions about convergent evolution at White Sands,

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:44 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Convergent Evolution at White Sands S45 with particular attention to the interplay among natural se- Degenhardt et al. 1996). Differences in sexual-signaling color lection, population demography, and genomic architecture. patches between dark and light color morphs could contrib- Throughout, we use a pattern-based definition of convergent ute to observed preference for local mates in White Sands S. evolution: the independent evolution of phenotypic similar- cowlesi and H. maculata (Rosenblum 2008; Hardwick et al. ity in different lineages. Defining convergence as a pattern— 2013), but the potential for incipient speciation in White observed at the phenotypic level—allows us to be more ex- Sands lizards is outside the scope of our focus here on con- plicit as we endeavor to link process to pattern, mechanism vergent evolution. to outcome, and cause to consequence (reviewed in Rosen- In addition to convergent evolution of coloration in blum et al. 2014 and Stayton 2015). White Sands lizards, the three lizard species also exhibit shifts in a variety of other phenotypes. Although detailed studies of heritability for these traits have not been White Sands as a Stage for Convergent Evolution conducted, we observe striking differences between dark- White Sands is a striking and geologically recent forma- soil and White Sands populations in a myriad of traits, in- tion in south-central New Mexico. The expansive gyp- cluding morphology (e.g., body size, limb length; Rosen- sum dune system formed after the Last Glacial Maximum, blum and Harmon 2011), performance (e.g., sprint speed, with the bulk of the sedimentation deposition in the past bite force; Des Roches et al. 2013, 2016), resource use 2,000–7,000 years (Langford 2003; Kocurek et al. 2007). (e.g., diet, microhabitat; Des Roches et al. 2011, 2015, At ∼650 km2, White Sands is the largest gypsum dune field 2016), and behavior (e.g., predator wariness, mate choice; in the world. The gypsum sands contrast dramatically with Rosenblum 2008; Robertson et al. 2011; Hardwick et al. the brown soils of the surrounding Chihuahuan Desert, cre- 2013). For example, in all three species, White Sands ating a divergent selective environment. Many diverse ani- lizards tend to have relatively larger heads size than dark- mal species have colonized White Sands and have con- soil lizards (Rosenblum and Harmon 2011; Des Roches vergently evolved blanched coloration in the dune habitat et al. 2016). Larger head size, in turn, correlates with stronger (fig. 1). bite force, and bite force is associated with dietary differences The best-studied example of phenotypic convergence at between light and dark populations (Des Roches et al. 2016). White Sands is in the lizard fauna. The three lizard species Specifically, for all three species, White Sands lizards con- that inhabit the heart of the dunes (Sceloporus cowlesi [East- sume a more varied diet than dark-soil lizards (i.e., with ern fence lizard], Holbrookia maculata [lesser earless liz- stronger bite and wider gape, larger and harder-bodied prey ard], and Aspidoscelis inornata [little striped whiptail]) all can be included in the diet; e.g., Herrel et al. 2001; Des Roches exhibit dramatically blanched dorsal coloration. In contrast, et al. 2015, 2016). populations of all three species exhibit darker dorsal coloration in the rest of their ranges, typically similar to local substrate colors (Degenhardt et al. 1996). The White Sands Factors Affecting Convergence in White Sands Lizards lizards exhibit similar general patterns of color evolution. In all three species, the change in coloration is in the same Despite strong evidence for convergent evolution of color- direction and is explained primarily by the brightness as- ation and evidence for similar patterns of change for many pect of color (i.e., changes in brightness explain 180% of other phenotypes in the three White Sands lizard species, the interpopulation variation in color; Rosenblum 2006). patterns are never completely identical across species. We Evolution of dorsal brightness is not only in the same di- often find that the direction of phenotypic change is con- rection across species but also of roughly similar magnitude cordant but the magnitude of change is different. For ex- across species (Rosenblum 2006; Rosenblum and Harmon ample, the lesser earless lizard (Holbrookia maculata)often 2011). While explaining less of total variation in dorsal color, exhibits the most pronounced differences between White changes in the hue and chroma aspects of coloration also Sands and dark-soil habitats, while Sceloporus cowlesi often trend in the same direction for all three species (Rosenblum exhibits the least (e.g., Rosenblum and Harmon 2011 for 2006; Robertson and Rosenblum 2009). dorsal coloration; Des Roches et al. 2015 for diet; Des Roches In addition to dorsal color variation, sexual-signaling patches et al. 2013 for sprint speed). are also significantly different in color between dark and light What explains similarities and differences among species lizards in all three species (Robertson and Rosenblum 2009). in the magnitude of divergence across the White Sands eco- These differences are not consistent across species, which is tone? Here we evaluate the contribution of three key deter- unsurprising, given that the location, color, and use of sexual- minants that can influence the direction and magnitude of signaling color patches vary across species (e.g., blue ventral phenotypic response across species: (1) natural selection, color in male S. cowlesi, seasonal orange throat color in fe- (2) genetic architecture of adaptive traits, and (3) demo- male H. maculata, and blue head color of male A. inornata; graphic context.

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:44 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:44AM

Figure 1: Convergent evolution of blanched coloration in White Sands fauna. The top and bottom rows show representative phenotypes from White Sands and dark-soil habitats, respectively. Focal species for this study shown (from left to right): Sceloporus cowlesi, Aspidoscleis inornata, Holbrookia maculata, Habronattus ustulatus,andAmmobaenetes arenicolus. Convergent Evolution at White Sands S47

Natural Selection White Sands population and the dark-soil population. We released painted lizards into the enclosures and scored survival The White Sands environment clearly provides a dramatic of substrate-matched versus that of substrate-unmatched backdrop for natural selection. Classic observational work lizards. These experiments helped confirm the activity of on reptile color variation has long posited the adaptive value avian predation at White Sands but also revealed differences of substrate matching, especially for small diurnal lizards in lizard survival rates across relatively fine spatial and tem- (e.g., Cott 1940; Norris and Lowe 1964). Experimental stud- poral scales and also different survival patterns for males and ies have also demonstrated the functional importance of females. An alternative promising approach is the use of substrate matching to avoid predation in a variety of taxa mark-recapture studies to assess differences in strength of se- (e.g., Dice 1947; Reed and Janzen 1999). Several experimen- lection across species. We conducted a multiyear effort to tal studies on lizards (e.g., Luke 1989) and small mammals mark lizards with elastomer tags and assess traits correlated (e.g., Kaufman 1973) specifically used avian predators that with survival in S. cowlesi (Des Roches et al. 2017), and com- are known to prey on lizards at White Sands, such as the parison with a similar H. maculata data set (S. Des Roches loggerhead shrike and greater roadrunner (E.B.R., personal and E. B. Rosenblum, unpublished data) promises to shed observation). Our early common-garden rearing experiments further light on differences in dynamics of selection among also demonstrated that color variation in the White Sands species. Ultimately, the observational and experimental ap- lizards was not explained by ontogenetic or physiological proaches will complement each other in the effort to under- plasticity (Rosenblum 2005). Moreover, there is evidence that stand whether White Sands species experience different selec- all three species experienced ecological release associated tion pressures despite sharing a common environment. with colonizing the gypsum dunes (Des Roches et al. 2011). Specifically, fewer predators and competitors are found in Genetic Architecture White Sands, relative to dark-soil habitats, and all three focal lizard species correspondingly exhibit higher abundances in Difference among species in the genetic architecture of func- the gypsum habitat (i.e., density compensation; Des Roches tionally relevant traits can influence species’ response to se- et al. 2011). Thus, the context for selection at White Sands lection. Many aspects of genetic architecture can influence appears simple: heritable phenotypes, clear optimality crite- the probability of convergent evolution at the phenotypic ria, an obvious agent of selection, and decreased predation level and the probability that similar genetic mechanisms un- and interspecific competition. derlie phenotypic convergence (reviewed in Rosenblum et al. However, the dynamics of selection might be different for 2014). different species, even if they inhabit a common macroenvi- For White Sands lizards, convergent evolution of blanched ronment. For example, differences in microhabitat use or coloration is the most obvious and tractable trait for genetic behavior can modulate exposure to predation and ultimately dissection. Our early candidate gene studies revealed muta- the strength of natural selection. To date, we know little tions associated with blanched color in all three lizard species about differences among species in the dynamics of selection in the melanocortin-1 receptor (Mc1r) gene (Rosenblum et al. at White Sands. However, differences among the three lizard 2004). The protein produced by Mc1r is a key player in the species in foraging mode, microhabitat use, and other traits vertebrate melanin synthesis pathway, and mutations in could affect dynamics of natural selection (e.g., Dixon 1967). Mc1r are known to be associated with color variation in a For example, Aspidoscelis inornata is an active forager, while variety of other systems (e.g., Barsh 1996; Manceau et al. H. maculata and S. cowlesi are sit-and-wait predators (e.g., 2010). Our early work showed a strong statistical association Degenhardt et al. 1996). Further, H. maculata typically uses between blanched coloration and a single Mc1r mutation in open, unvegetated microhabitat, while S. cowlesi is more com- each of the three White Sands lizards. All three mutations monly found in vegetated microhabitat (e.g., Hager 2001). lead to amino acid substitutions, and all three substitutions Thus, it is possible, for example, that selection has been occur in a transmembrane region of the protein, which is stronger for optimal substrate matching in H. maculata,which important for ligand binding, signal transduction, and struc- would be consistent with the stronger phenotypic response tural integrity of the receptor. Our subsequent functional in this species. assays confirmed that the Mc1r mutations in two of the three Explicit tests are required to understand whether dynamics species (S. cowlesi and A. inornata) have important func- of selection differ among species at White Sands. Our initial tional effects leading to decreased melanin production efforts to quantify selection at White Sands involved con- (i.e., by reducing receptor integration in S. cowlesi and re- ducting large-scale enclosure experiments. We constructed ducing receptor signaling in A. inornata; Rosenblum et al. replicated 100-m2 enclosures in the natural gypsum habitat 2010). Although data are less conclusive for the third species (Hardwick et al. 2015). We then painted dozens of White (H. maculata), we have not ruled out a role for Mc1r in color Sands H. maculata to match the average dorsal color of the variation in this species.

The molecular signature of selection at and around the Sands. Moreover, differences among species in population Mc1r gene is also striking—and similar—for S. cowlesi demography may help explain differences among species in and A. inornata. We recently obtained population-level se- magnitude of evolutionary change across the ecotone. quence capture data for a ∼40-kb window of the chromo- Even our earliest molecular data suggested dramatic dif- somal region around the Mc1r gene (the gene itself is a ferences in underlying population demography among the single exon !1 kb in length). We found no evidence for se- three White Sands lizard species. Early mitochondrial data lection around Mc1r in dark-soil populations of S. cowlesi for White Sands and dark-soil populations demonstrated and A. inornata, but we found strong evidence for selection that patterns of population structure were different across around Mc1r in White Sands populations of these species species (Rosenblum 2006). For example, an analysis of mo- (Laurent et al. 2016). Moreover, estimates of the age of lecular variance showed the three lizard species on a spec- the blanched allele at White Sands were remarkably young trum from nearly complete genetic separation between dark F p : and similar across species (i.e., 1,200 and 900 years for S. and light populations of H. maculata ( ST 0 82) to nearly F p : cowlesi and A. inornata, respectively; Laurent et al. 2016). complete panmixis in A. inornata ( ST 0 09), with inter- F p : Thus, there are noteworthy mechanistic similarities under- mediate structure observed for S. cowlesi ( ST 0 54). In the lying convergent color evolution in at least two of the lizard past decade, we sampled additional populations and addi- species. Mutations at the same gene appear to have swept tional regions of the genome. Our first analyses with multi- through the populations during a similar time period, con- locus nuclear data (∼200 single-nucleotide polymorphisms sistent with the age of the White Sands formation itself. [SNPs] for S. cowlesi and ∼50 amplified fragment length In addition to similarities in genetic architecture of color- polymorphism bands for H. maculata and A. inornata)largely ation in White Sands lizards, we find several important dif- supported the mitochondrial patterns (Rosenblum et al. 2007; ferences across species. One intriguing difference is the dom- Rosenblum and Harmon 2011), again suggesting that blanched inance of the blanched Mc1r allele in S. cowlesi but not in coloration could evolve under very different demographic A. inornata. Allele frequencies in natural populations, pat- scenarios. Our most recent data sets that rely on broader terns of association between color phenotype and different genomic sampling (i.e., 120,000 SNPs for S. cowlesi and genotypic classes, and functional studies all suggest that the 113,000 SNPs for A. inornata) show that adding data im- blanched allele is dominant in S. cowlesi but recessive in A. proves our ability to discriminate genetically among popula- inornata (Rosenblum et al. 2010). Although both dominant tions but that differences in underlying population demogra- and recessive alleles contribute to adaptation in natural sys- phy remain among species (Laurent et al. 2016). Specifically, tems, allelic dominance can affect the visibility of adaptive the inferred split time between light and dark populations alleles to selection, the likelihood of maladaptive gene swamp- was younger for A. inornata than for S. cowlesi, and there ing via gene flow, and ultimately the probability of fixation was stronger evidence for ongoing migration after coloniza- (e.g., Orr and Betancourt 2001; Nuismer et al. 2012). The tion for S. cowlesi (Laurent et al. 2016). consequences of differences in allelic dominance on the adap- Our data point to the possibility that gene flow may con- tive trajectories of S. cowlesi and A. inornata require further strain local adaptation at White Sands. For example, we work, as does understanding other genes that contribute to have the most consistent and conclusive evidence for ongo- convergent phenotypic evolution in the novel gypsum habi- ing gene flow across the White Sands ecotone in S. cowlesi tat. Variation at Mc1r cannot explain all observed color var- (Rosenblum et al. 2007; Rosenblum and Harmon 2011; iation in the White Sands system, suggesting that other genes Laurent et al. 2016). Sceloporus cowlesi is also the species are also involved in color variation. Moreover, we are still that tends to exhibit the least phenotypic divergence across working to understand the molecular basis of convergent the ecotone (e.g., for color, body size, performance, and diet; traits other than color. Ultimately, identifying the genes and Rosenblum 2006; Rosenblum and Harmon 2011; Des Roches gene interactions that underlie convergence at White Sands et al. 2013, 2015, 2016). Ultimately, understanding the rela- will allow us to understand how the genetic architecture tionship between gene flow and local adaptation will be fa- of adaptive traits influences the direction and magnitude of cilitated by continued sampling of dark-soil populations to evolutionary change across species. refine our understanding of the ancestry, time of colonization, and dynamics of ongoing gene flow for the White Sands populations. Population Demography Differences in underlying population demography can also New Insights from New Data: White Sands Arthropods contribute to different evolutionary outcomes across species. A number of demographic factors—such as population size, Thus far we have focused on convergent evolution in the time since colonization, and rates of gene flow—have likely White Sands lizard fauna, but several additional—and more influenced species response to natural selection at White distantly related—species also exhibit blanched phenotypes

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:44 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Convergent Evolution at White Sands S49 in the gypsum habitat. Here we present new data on two other most commonly found in low vegetation or on sandy sub- dramatic examples of phenotypic convergence associated with strate near vegetative cover at White Sands. Sand-treader crick- the colonization of White Sands in the terrestrial inverte- ets burrow in the daytime and emerge to forage at night brate fauna: the sand-treader cricket Ammobaenetes areni- (Weissman 1997), and they are commonly found near their colus (Strohecker 1947) and the jumping spider Habronattus burrows at the base of the sand dunes, typically in areas with ustulatus (Griswold 1979). Both species are characterized sparse vegetation cover (E.B.R. and C.E.P., personal obser- by pale forms at White Sands and darker forms in the sur- vation). Despite differences in activity period, both species rounding Chihuahuan Desert, convergent with the pattern are likely targeted by a range of visually oriented predators. observed for the lizards (fig. 1). By increasing the phyloge- Therefore, dorsal color variation could be linked to selection netic breadth of our work, we can add generality to under- for crypsis in A. arenicolus and H. ustulatus, as has been standing the factors that influence the direction and mag- found in numerous other animal species (e.g., Vignieri et al. nitude of evolutionary response across a shared ecotone. 2010). Alternatively, dorsal color variation can serve roles in We compared sand-treader crickets and jumping spiders thermoregulation, mate choice, aggression, and immunity from White Sands to those from nearby dark-soil sites. To (e.g., Horth 2003; Fedorka et al. 2013; Roulin 2016). How- understand phenotypic response across the ecotone in these ever, most of these alternatives have little intuitive sup- species, we quantified coloration of the dorsal body surface, port (e.g., because ambient and substrate temperatures are using spectrophotometry and photographic analysis. To un- typically lower at White Sands than in the surrounding Chi- derstand the demographic history for these populations, we huahuan Desert [Hager 2000], a thermoregulatory hypothe- collected genetic data, including thousands of loci obtained sis would predict the opposite patterns, where darker body with a double-digest restriction site–associated DNA sequenc- color would be favored in the gypsum habitat). Although ing (RADseq) approach and the mitochondrial cytochrome c natural selection for crypsis likely explains color differences oxidase subunit I (COI) gene. Sampling and methodological across populations, sexual selection may play a role within details are presented in appendix A, and detailed results are populations, particularly for the spiders, which—like the presented in appendix B (apps. A, B available online). Here White Sands lizards—exhibit some sexual dichromatism. we highlight key findings that relate to and extend our pre- If dynamics of selection differ between the two inverte- vious work on White Sands lizards. brate species, there could be an adaptive explanation for differences in phenotypic patterns. However, additional observational and experimental studies will be required to assess Arthropod Phenotypic Convergence whether differences between the species (or between the sexes) Consistent with patterns we have previously described for in habitat use, life history, and exposure to predation may White Sands lizards (Rosenblum 2006; Rosenblum and Har- have contributed to differences in phenotypic patterns across mon 2011), we found strong evidence for phenotypic con- the ecotone. Experimental studies will also be particularly vergence in the two lineages of terrestrial invertebrates. Pop- useful for assessing the potential for phenotypic plasticity to ulations of the sand-treader cricket A. arenicolus and the contribute to color variation in A. arenicolus and H. ustulatus jumping spider H. ustulatus at White Sands exhibit signifi- and for identifying genes contributing to color variation in cantly lighter dorsal body coloration than populations in the these species. Although both species are relatively difficult surrounding Chihuahuan Desert (fig. 2). Although both in- to breed in the lab, anecdotally we found that baby spiders vertebrate species exhibit the same direction of phenotypic born in the lab from White Sands mothers were blanched change, the magnitude of color divergence between White in color. Sands and dark-soil individuals differs. Specifically, the spiders exhibit less divergence in dorsal coloration than the Arthropod Demographic History crickets in comparisons of White Sands to dark-soil populations (fig. 2). The spiders also exhibit sexual dichromatism, Another possible explanation for differences in phenotypic while the crickets do not. Specifically, both males and females response across the White Sands ecotone is differences in are lighter in coloration at White Sands for both species, but underlying demography (e.g., population size, colonization male spiders are significantly darker than female spiders in history, contemporary patterns of gene flow). Convergence both habitats (fig. 2). in dorsal coloration is accompanied by conspicuously differ- Similartothelizardsinthissystem, natural selection for ent demographic histories in A. arenicolus and H. ustulatus. substrate matching is the most likely explanation for blanched The sand-treader cricket A. arenicolus exhibits strong pop- coloration in White Sands invertebrates. Both A. arenicolus ulation structure, while the jumping spider H. ustulatus ex- and H. ustulatus spend time exposed on the dunes. Jump- hibits little genetic differentiation across the White Sands ing spiders are diurnal and active predators of other terres- ecotone. The cricket mitochondrial gene tree shows a well- trial invertebrates (Foelix 1982; Griswold 1987), and they are supported monophyletic White Sands clade (fig. 3), and the

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:44 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:44AM

Figure 2: Dorsal color variation for Ammobaenetes arenicolus (A)andHabronattus ustulatus (B): mean dorsal color (and standard error of the mean) for White Sands (open symbols) and dark-soil (closed symbols) samples. Circles, squares, and triangles represent female, male, and substrate samples, respectively. Note that Y-axes differ because color was quantiﬁed differently for the two species (see “Color Quantiﬁcation” in app. A, available online). PC1 p principal component 1. Convergent Evolution at White Sands S51

C. gracilipes H. oregonensis

Figure 3: Contrasting demographic patterns based on mitochondrial cytochrome c oxidase subunit I gene sequences for Ammobaenetes arenicolus (A)andHabronattus ustulatus (B): Bayesian inference, with numbers at nodes representing posterior probabilities. The branches for the outgroup species are shortened by half, indicated by the “//” symbol. Open and ﬁlled symbols represent samples collected in White Sands and dark-soil habitats, respectively.

RADseq data show strong population structure, with White and Watterson’s v are more than an order of magnitude Sands and dark-soil individuals clearly in separate genetic smaller for the White Sands population than for the dark- clusters (fig. 4). In contrast, the spider mitochondrial gene soil population (table 1), whereas for the spiders, p and v tree shows no clear clustering by habitat (fig. 3). Although are high and comparable for the White Sands and dark-soil there is support for multiple clades in the spider tree, these populations (table 1). Mitochondrial nucleotide diversity is clades contain both White Sands and dark-soil individuals more than two orders of magnitude higher in the White (with no substructure based on dark-soil collecting locality). Sands spiders than in the White Sands crickets. Mitochon- Consistent with the mitochondrial data, the spider RADseq drial and RADseq data give concordant results in the spiders data show largely overlapping clusters of White Sands and but show slightly different patterns in the crickets. Specifi- dark-soil individuals, with little structuring of genetic varia- cally, for the crickets, p and v are substantially higher and tion by habitat (fig. 4). FST is substantially lower for the RADseq data set than for Metrics of population differentiation also show strong dif- the mitochondrial data set, but both data sets indicate much ferences in genetic patterns between the spider and cricket stronger population structure for the crickets than for the data sets. For the mitochondrial data set, FST is more than spiders. p : an order of magnitude higher for the crickets (FST 0 45) Our molecular results are consistent with different de- p : than for the spiders (FST 0 03; table 1). Moreover, the mographic scenarios in the two focal invertebrate species. mitochondrial data suggest dramatically reduced diversity For the crickets, our results are consistent with a simple in the White Sands cricket population but not in the spider colonization history (likely with a reduced population size), population. For the crickets, both nucleotide diversity (p) limited ongoing gene flow, and strong genetic divergence

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:44 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:44AM

Figure 4: Contrasting patterns of population structure based on restriction site–associated DNA sequencing (RADseq) data for Ammobaenetes arenicolus (A)andHabronattus ustulatus (B). A, For A. arenicolus, the population structure plot (top)forK p 2 shows strong population structure. Each individual is represented by a vertical bar and labeled as dark soil (black horizontal bar) or White Sands (white horizontal bar). The genetic principal-component analysis (PCA) plot (bottom left) shows no overlap between dark-soil (ﬁlled circles) and White Sands (open circles) individuals. Ellipses represent 95% conﬁdence intervals for the dark-soil and White Sands populations. The joint site frequency spectrum (bottom right) shows only loosely correlated allele frequencies. B, For H. ustulatus, the population structure plot (top) shows less structure and more admixture between the two populations for K p 2. There is substantial overlap between the dark-soil and White Sands individuals in the genetic PCA plot (bottom left), and the joint site frequency spectrum shows strongly correlated allele frequencies (bottom right). Convergent Evolution at White Sands S53

p ’ v Table 1: Estimates of the nucleotide diversity ( ), Watterson s theta ( ), and FST for the mitochondrial and nuclear data sets Ammobaenetes arenicolus Habronattus ustulatus Data set, measure Dark soil White Sands Dark soil White Sands COI: p .0030 .0002 .0294 .0261 v .0048 .0003 .0177 .0154

FST .4541 .0295 RAD sites: p .0213 .0198 .0210 .0220 v .0316 .0215 .0336 .0347

FST .0312 .0140 Note: COI p cytochrome c oxidase subunit I gene; RAD p restriction site–associated DNA sequencing.

across habitats. For the spiders, our results suggest admix- posed to a common ecotone. While the direction of color ture across the White Sands ecotone, and different scenarios evolution is typically convergent in the White Sands fauna, could underlie the lack of population structure, including the magnitude of phenotypic change varies across species. multiple colonization events, a large recent founding popula- Moreover, differences in phenotypic response appear to cor- tion with substantial genetic variation, and high levels of on- relate with variation in underlying demographic patterns. going gene flow. Our results are consistent with patterns Specifically, species with more genetic differentiation across observed in other groups of jumping spiders. Hybridization, the White Sands ecotone (e.g., Holbrookia maculata and Am- introgression, interspecificgeneflow, and incomplete line- mobaenetes arenicolus) typically exhibit more phenotypic dif- age sorting are not uncommon in the Habronattus clade, ferentiation. Thus, population demography appears to be an and gene trees often fail to resolve geographic or phenotypic important modulator of phenotypic evolution in this sys- groups (e.g., Maddison and McMahon 2000; Masta 2000; tem. Our research also suggests that subtle variation in the Hedin and Lowder 2009). To refine our understanding of dynamics of natural selection and differences in genomic the relationship between initial colonization history and on- architecture across species can influence patterns of pheno- going gene flow in these species, we will need more thorough typic convergence. sampling throughout the Tularosa Basin (i.e., within species Moving forward, the White Sands system can be used to replication) and more explicit modeling of alternative demo- address important outstanding questions about the interplay graphic scenarios. among natural selection, genetic architecture, and population demography. For example, what is the role of phenotypic plasticity in promoting or hindering phenotypic divergence? Convergence across the Community at White Sands How do different species experience the same environment? Cases of convergent evolution have long been used as evi- How similar are underlying molecular and functional mech- dence that species can exhibit similar evolutionary responses anisms of adaptation in closely versus distantly related spe- when exposed to comparable selection pressures (e.g., Arendt cies? How do levels of gene flow and strength of selection and Reznick 2008; Losos 2011; McGhee 2011; Wake et al. interact to determine phenotypic outcome? What null expec- 2011; Conte et al. 2012). However, myriad factors influence tations are most appropriate to use when studying conver- whether species will adapt along parallel trajectories. Study- gent evolution? ing evolutionary outcomes across diverse lineages provides Developing a general understanding of the factors that fa- replicated variation in factors that can influence adaptive con- cilitate and constrain convergent evolution at White Sands vergence (e.g., trait heritability, population size, gene flow, will be promoted by studying additional species. Other ani- behavior). Further, comparisons across species in a single en- mal species exhibit blanched forms on the dunes and can vironment (e.g., where colonization time is geologically con- be integrated into a broader community-scale study. For ex- strained and abiotic context is shared) can reduce complexity ample, the moth Euxoa misturata and the pocket mouse and analytical noise. Thus, we can better understand whether Perognathus flavescens also have blanched forms in the gyp- adaptation is constrained in similar ways across diverse spe- sum habitat that contrast with nearby dark forms. In addi- cies and which key factors modulate evolutionary response. tion to these dramatic examples, White Sands is home to Our work thus far on White Sands lizards and arthopods multiple endemic species of Lepidoptera that are pale com- shows both shared and unique patterns across species ex- pared to close relatives (Metzler 2014).

Although White Sands is a relatively depauperate eco- Degenhardt, W. G., C. W. Painter, and A. H. Price. 1996. Amphibians system, the species that have survived on the dunes typi- and reptiles of New Mexico. University of New Mexico Press, Al- cally exhibit thriving populations. For example in the her- buquerque. Des Roches, S., M. S. Brinkmeyer, L. J. Harmon, and E. B. Rosenblum. petofauna, 28 species of reptiles and amphibians are found 2015. Ecological release and directional change in White Sands lizard in the immediate vicinity (Prival and Goode 2005), but only trophic ecomorphology. Evolutionary Ecology 29:1–16. three of these species are found commonly in the heart of the Des Roches, S., L. J. Harmon, and E. B. Rosenblum. 2016. Coloniza- dunes. These three species exhibit dramatic density compen- tion of a novel depauperate habitat leads to trophic niche shifts in sation, with much higher local population densities at White three desert lizard species. Oikos 125:343–353. Sands than in nearby dark-soil environments (Des Roches Des Roches, S., J. M. Robertson, L. J. Harmon, and E. B. Rosenblum. et al. 2011). Relatively high densities allow for a diversity of 2011. Ecological release in White Sands lizards. Ecology and Evo- lution 1:571–578. research approaches to be applied at White Sands. Integrat- fi Des Roches, S., R. Sollman, K. Calhoun, A. Rothstein, and E. B. Ro- ing natural history studies, manipulative lab and eld exper- senblum. 2017. Survival by genotype: patterns at Mc1r are not black iments, detailed phenotypic analysis, genomic sequencing, and white at the White Sands ecotone. Molecular Ecology 26:320– and demographic modeling across the entire White Sands 329. community will ultimately shed light on the factors govern- Des Roches, S., J. Torresdal, T. W. Morgan, L. J. Harmon, and E. B. ing the probability—and the degree—of convergent evolu- Rosenblum. 2013. Beyond black and white: divergent behavior and tion in natural systems. performance in three rapidly evolving lizard species at White Sands. Biological Journal of the Linnean Society 111:169–182. Dice, L. R. 1947. Effectiveness of selection by owls of deer mice Acknowledgments (Peromyscus maniculatus) which contrast in color with their background. Contributions from the Laboratory of Vertebrate Biology We thank White Sands National Monument, White Sands of the University of Michigan 34:1–20. Missile Range, and Jornada Long-Term Ecological Research Dixon, J. R. 1967. Aspects of the biology of the lizards of the White Station for access to field sites and research facilitation over Sands, New Mexico. Los Angeles County Museum Contributions the past 15 years. In particular we are grateful to J. Anderson, in Science 129:1–22. D. Burkett, D. Bustos, B. Conrod, and P. Cutler. We thank Fedorka, K. M., V. Lee, and W. E. Winterhalter. 2013. Thermal en- M. Girard for help in the field and K. Mathews for help with vironment shapes cuticle melanism and melanin-based immunity in the ground cricket Allonemobius socius. Evolutionary Ecology care of spiders in the lab. We also thank D. Elias, M. Girard, 27:521–531. D. Lightfoot, E. Metzler, and the Rosenblum lab for pro- Foelix, R. 1982. Biology of spiders. Harvard University Press, Cam- ductive conversations about White Sands invertebrates. We bridge, MA. thank H. Hoekstra for sharing RADseq adapters, and we Griswold, C. E. 1979. New species of Pellenes from California thank A. Agrawal, L. Mahler, K. Zamudio, and members of (Araneae: Salticidae). Journal of Arachnology 7:129–138. the Rosenblum lab for helpful comments on the manuscript. ———. 1987. A revision of the jumping spider genus Habronattus This work was supported by a National Science Foundation F. O. P.-Cambridge (Araneae: Salticidae), with phenetic and cladistic analyses. University of California Publications in Entomol- CAREER grant to E.B.R. (DEB-1054062) and a research grant ogy 107. University of California Press, Berkeley. from the Basler Stiftung für Biologische Forschung to E.T.D. Hager, S. B. 2000. Variation in body temperature and thermoregula- Sequencing was performed by the Vincent J. Coates Geno- tory behavior between two populations of the lesser earless lizard, mics Sequencing Laboratory at the University of California, Holbrookia maculata. Contemporary Herpetology 2000:1. http:// Berkeley, which is supported by National Institutes of Health jnah.cnah.org/ch/ch/2000/1/index.htm. S10 Instrumentation Grants S10RR029668 and S10RR027303. ———. 2001. Microhabitat use and activity patterns of Holbrookia Animal work was approved by the Animal Care and Use maculata and Sceloporus undulatus at White Sands National Mon- ument, New Mexico. Journal of Herpetology 35:326–330. Committee at the University of California, Berkeley (protocol Hardwick, K. M., L. Harmon, S. D. Hardwick, and E. B. Rosenblum. 2014-11-6857), or the University of Idaho (protocol 2009-37). 2015. When field experiments yield unexpected results: lessons learned from measuring selection in White Sands lizards. PLoS Literature Cited ONE 10(2):e0118560. doi:10.1371/journal.pone.0118560. Hardwick, K. M., J. M. Robertson, and E. B. Rosenblum. 2013. Asym- Arendt, J., and D. Reznick. 2008. Convergence and parallelism recon- metrical mate preference in recently adapted White Sands and black sidered: what have we learned about the genetics of adaptation? Trends lava populations of Sceloporus undulatus. Current Zoology 59:20–30. in Ecology and Evolution 23:26–32. Hedin, M., and M. C. Lowder. 2009. Phylogeography of the Habronat- Barsh, G. S. 1996. The genetics of pigmentation: from fancy genes to tus amicus species complex (Araneae: Salticidae) of western North complex traits. Trends in Genetics 12:299–305. America, with evidence for localized asymmetrical mitochondrial Conte, G. L., M. E. Arnegard, C. L. Peichel, and D. Schluter. 2012. introgression. Zootaxa 2307:39–60. The probability of genetic parallelism and convergence in natural Herrel, A., R. Van Damme, B. Vanhooydonck, and F. De Vree. 2001. populations. Proceedings of the Royal Society B 279:5039–5047. The implications of bite performance for diet in two species of Cott, H. B. 1940. Adaptive coloration in animals. Methuen, London. lacertid lizards. Canadian Journal of Zoology 79:662–670.

Horth, L. 2003. Melanic body colour and aggressive mating behaviour Rosenblum, E. B. 2005. The role of phenotypic plasticity in color are correlated traits in male mosquitofish (Gambusia hotbrooki). variation in Tularosa Basin lizards. Copeia 2005:586–596. Proceedings of the Royal Society B 270(1519):1033–1040. ———. 2006. Convergent evolution and divergent selection: lizards Kaufman, D. W. 1973. Shrike prey selection: color or conspicuous- at the White Sands ecotone. American Naturalist 167:1–15. ness? Auk 90:204–206. ———. 2008. Preference for local mates in a recently diverged pop- Kocurek, G., M. Carr, R. Ewing, K. G. Havholm, Y. C. Nagar, and ulation of the lesser earless lizard (Holbrookia maculata)atWhite A. K. Singhvi. 2007. White Sands Dune Field, New Mexico: age, Sands. Journal of Herpetology 42:572–583. dune dynamics and recent accumulations. Sedimentary Geology Rosenblum, E. B., and L. J. Harmon. 2011. “Same same but differ- 197:313–331. ent”: replicated ecological speciation at White Sands. Evolution Langford, R. P. 2003. The Holocene history of the White Sands dune 65:946–960. field and influences on eolian deflation and playa lakes. Quaternary Rosenblum, E. B., M. Hickerson, and C. Moritz. 2007. A multilocus International 104:31–39. perspective on colonization accompanied by selection and gene Laurent, S., S. P. Pfeifer, M. L. Settles, S. S. Hunter, K. M. Hardwick, flow. Evolution 61:2971–2985. L. Ormond, V. Sousa, J. J. Jensen, and E. B. Rosenblum. 2016. The Rosenblum, E. B., H. E. Hoekstra, and M. W. Nachman. 2004. Adap- population genomics of rapid adaptation: disentangling signatures tive reptile color variation and the evolution of the Mc1r gene. of selection and demography in White Sands lizards. Molecular Evolution 58:1794–1808. Ecology 25:306–323. Rosenblum, E. B., C. E. Parent, and E. Brandt. 2014. The molecular Losos, J. B. 2011. Convergence, adaptation, and constraint. Evolution basis of phenotypic convergence. Annual Reviews of Ecology, 65:1827–1840. Evolution, and Systematics 45:203–226. Luke, C. A. 1989. Color as a phenotypically plastic character in the Rosenblum, E. B., H. Römpler, T. Schöneberg, and H. E. Hoekstra. side-blotched lizard, Uta stansburiana. PhD diss. University of 2010. Molecular and functional basis of phenotypic convergence California, Berkeley. in white lizards at White Sands. Proceedings of the National Acad- Maddison, W., and M. McMahon. 2000. Divergence and reticulation emy of Sciences of the USA 107:2113–2117. among montane populations of a jumping spider (Habronattus Roulin, A. 2016. Condition-dependence, pleiotropy and the handi- pugillis Griswold). Systematic Biology 49:400–421. cap principle of sexual selection in melanin-based colouration. Bi- Manceau, M. C., V. Domingues, C. R. Linnen, E. B. Rosenblum, and ological Reviews 91:328–348. H. E. Hoekstra. 2010. The genetic basis of phenotypic conver- Stayton, C. T. 2015. The definition, recognition, and interpretation of gence: lessons from pigmentation. Philosophical Transactions of convergent evolution, and two new measures for quantifying and the Royal Society B 365:2439–2450. assessing the significance of convergence. Evolution 69:2140–2153. Masta, S. E. 2000. Phylogeography of the jumping spider Habronattus Strohecker, H. F. 1947. Some southwestern Gryllacrididae (Orthop- pugillis (Araneae: Salticidae): recent vicariance of sky island popu- tera). Annals of the Entomological Society of America 40:241–246. lations? Evolution 54:1699–1711. Vignieri, S. N., J. G. Larson, and H. E. Hoekstra. 2010. The selective McGhee, G. R. 2011. Convergent evolution: limited forms most beauti- advantage of crypsis in mice. Evolution 64:2153–2158. ful. Vienna Series in Theoretical Biology. MIT Press, Cambridge, MA. Wake, D. B., M. H. Wake, and C. D. Specht. 2011. Homoplasy: from Metzler, E. H. 2014. The remarkable endemism of moths at White detecting pattern to determining process and mechanism of evo- Sands National Monument in New Mexico, USA, with special em- lution. Science 331:1032–1035. phasis on Gelechioidea (Lepidoptera). Journal of Asia-Pacific Bio- Weissman, M. J. 1997. Natural history of the giant sand treader camel diversity 7:e1–e5. cricket, Daihinibaenetes giganteus Tinkham (Orthoptera: Rhaphi- Norris, K., and C. H. Lowe. 1964. An analysis of background color- dophoridae). Journal of Orthoptera Research 6:33–48. matching in amphibians and reptiles. Ecology 45:565–580. Nuismer, S. L., A. MacPherson, and E. B. Rosenblum. 2012. Crossing fl the threshold: gene ow, dominance, and the critical level of References Cited Only in the Online Appendixes standing genetic variation required for adaptation to novel environments. Journal of Evolutionary Biology 25:2665–2671. Bi, K., T. Linderoth, D. Vanderpool, J. M. Good, R. Nielsen, and Orr, H. A., and A. J. Betancourt. 2001. Haldane’s sieve and adapta- C. Moritz. 2013. Unlocking the vault: next generation museum pop- tion from the standing genetic variation. Genetics 157:875–884. ulation genomics. Molecular Ecology 22:6018–6032. Prival, D., and M. Goode. 2005. Chihuahuan Desert national parks rep- Darriba, D., G. L. Taboada, R. Doallo, and D. Posada. 2012. jModelTest tile and amphibian inventory. Final report. University of Arizona, 2: more models, new heuristics and parallel computing. Nature Meth- Tucson. ods 9:772. Reed, W. L., and F. J. Janzen. 1999. Natural selection by avian pred- Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number ators on size and colour of a freshwater snail (Pomacea flagellata). of clusters of individuals using the software structure: a simulation Biological Journal of the Linnean Society 67:331–342. study. Molecular Ecology 14:2611–2620. Robertson, J., K. Hoversten, M. Grundler, T. Poorten, D. Hews, and Folmer, O., M. Black, W. Hoeh, R. Lutz, and R. Vrijenhoek. 1994. E. B. Rosenblum. 2011. Colonization of novel White Sands habitat DNA primers for amplification of mitochondrial cytochrome c associated with changes in lizard anti-predator behavior. Biolog- oxidase subunit I from diverse metazoan invertebrates. Molecular ical Journal of the Linnean Society 103:657–667. Marine Biology and Biotechnology 3:294–299. Robertson, J., and E. B. Rosenblum. 2009. Rapid divergence of social Fumagalli, M., F. G. Vieira, T. S. Korneliussen, T. Linderoth, E. Huerta- signal coloration across the White Sands ecotone for three lizard Sánchez, A. Albrechtsen, and R. Nielsen. 2013. Quantifying popula- species under strong natural selection. Biological Journal of the tion genetic differentiation from next-generation sequencing data. Linnean Society 98:243–255. Genetics 195:979–992.

Gregory, T. R., and D. P. Shorthouse. 2003. Genome sizes of spiders. SNP discovery and genotyping in model and non-model species. Journal of Heredity 94:285–290. PLoS ONE 7(5):e37135. doi:10.1371/journal.pone.0037135. Guindon, S., and O. Gascuel. 2003. A simple, fast, and accurate algo- Posada, D. 2008. jModelTest: phylogenetic model averaging. Molec- rithm to estimate large phylogenies by maximum likelihood. Sys- ular Biology and Evolution 25:1253–1256. tematic Biology 52:696–704. Posnien, N., V. Zeng, E. E. Schwager, M. Pechmann, M. Hilbrant, Hedin, M. C. 1997. Speciational history in a diverse clade of habitat- J. D. Keefe, W. G. M. Damen, N.-M. Prpic, A. P. McGregor, and specialized spiders (Araneae: Nesticidae: Nesticus): inferences from C. G. Extavour. 2014. A comprehensive reference transcriptome geographic-based sampling. Evolution 51:1929–1945. resource for the common house spider Parasteatoda tepidariorum. Huelsenbeck, J. P., and F. Ronquist. 2001. MrBayes: Bayesian infer- PLoS ONE 9(8):e104885. doi:10.1371/journal.pone.0104885. ence of phylogenetic trees. Bioinformatics 17:754–755. R Development Core Team. 2014. R: a language and environment Korneliussen, T. S., A. Albrechtsen, and R. Nielsen. 2014. ANGSD: for statistical computing. R Foundation for Statistical Computing, analysis of next generation sequencing data. BMC Bioinformatics Vienna, Austria. http://www.R-project.org/. 15:356. doi:10.1186/s12859-014-0356-4. Ronquist, F., and J. P. Huelsenbeck. 2003. MrBayes 3: Bayesian phy- Li, H., B. Handsaker, A. Wysoker, T. Fennell, J. Ruan, N. Homer, logenetic inference under mixed models. Bioinformatics 19:1572– G. Marth, G. Abecasis, R. Durbin, and 1000 Genome Project Data 1574. Processing Subgroup. 2009. The sequence alignment/map (SAM) Sanggaard, K. W., J. S. Bechsgaard, X. Fang, J. Duan, T. F. Dyrlund, format and SAMtools. Bioinformatics 25:2078–2079. V. Gupta, X. Jiang, et al. 2014. Spider genomes provide insight into Librado, P., and J. Rozas. 2009. DnaSP v5: a software for comprehen- composition and evolution of venom and silk. Nature Communica- sive analysis of DNA polymorphism data. Bioinformatics 25:1451– tions 5:3765. doi:10.1038/ncomms4765. 1452. Schneider, C. A., W. S. Rasband, and K. W. Eliceiri. 2012. NIH Image McKenna, A., M. Hanna, E. Banks, A. Sivachenko, K. Cibulskis, to ImageJ: 25 years of image analysis. Nature Methods 9:671–675. A. Kernytsky, K. Garimella, et al. 2010. The Genome Analysis Tool- Simon, C., F. Frati, A. Beckenbach, B. Crespi, H. Liu, and P. Flook. kit: a MapReduce framework for analyzing next-generation DNA 1994. Evolution, weighting, and phylogenetic utility of mitochon- sequencing data. Genome Research 20:1297–1303. drial gene sequences and a compilation of conserved polymerase Nielsen, R., T. Korneliussen, A. Albrechtsen, Y. Li, and J. Wang. 2012. chain reaction primers. Annals of the Entomological Society of SNP calling, genotype calling, and sample allele frequency estima- America 87:651–701. tion from new-generation sequencing data. PLoS ONE 7(7):e37558. Skotte, L., T. S. Korneliussen, and A. Albrechtsen. 2013. Estimating doi:10.1371/journal.pone.0037558. individual admixture proportions from next generation sequenc- Nielsen, R., J. S. Paul, A. Albrechtsen, and Y. S. Song. 2011. Geno- ing data. Genetics 195:693–702. type and SNP calling from next-generation sequencing data. Na- Zhang, J.-X., and W. P. Maddison. 2013. Molecular phylogeny, diver- ture Reviews Genetics 12:443–451. gence times and biogeography of spiders of the subfamily Euoph- O’Grady, P. M., J. B. Clark, and M. G. Kidwell. 1998. Phylogeny of ryinae (Araneae: Salticidae). Molecular Phylogenetics and Evolution the Drosophila saltans species group based on combined analysis 68:81–92. of nuclear and mitochondrial DNA sequences. Molecular Biology and Evolution 15:656–664. Peterson, B. K., J. N. Weber, E. H. Kay, H. S. Fisher, and H. E. Hoekstra. 2012. Double digest RADseq: an inexpensive method for de novo Symposium Editor: Anurag A. Agrawal

“The farmer should know his true insect friends as well as his insect foes. We introduce to our readers a large family of ground-beetles (Carabidæ, from Carabus, the name of the typical genus) which prey on those insects largely injurious to crops.” From “Entomological Cal- endar” (The American Naturalist, 1868, 2:110–111).

Symposium Convergence and Divergence in a Long-Term Experiment with Bacteria*

Richard E. Lenski†

Department of Microbiology and Molecular Genetics, Michigan State University, East Lansing, Michigan 48824 abstract: strikingly lack a recognizable brain (Conway Morris 2003, Suitably designed experiments offer the possibility of – quantifying evolutionary convergence because the fraction of repli- pp. 151 158). But in how many lineages did camera-like eyes cate populations that converge is known. Here I review an experiment fail to evolve? Even if we had that number, how should we ac- with Escherichia coli, in which 12 populations were founded from the count for differences in, say, the sizes and ages of lineages same ancestral strain and have evolved for almost 30 years and more when trying to formulate and test hypotheses about ecological than 65,000 generations under the same conditions. The tension be- or genetic factors that might have predisposed a lineage to tween divergence and convergence has been a major focus of this ex- evolve camera-like eyes? And in those cases where such fea- periment. I summarize analyses of competitive fitness, correlated responses to different environments, cell morphology, the capacity to tures did not evolve, despite seemingly comparable opportu- use a previously untapped resource, mutation rates, genomic changes, nities, did some other difference in circumstances promote and within-population polymorphisms. These analyses reveal conver- divergence instead of convergence? gence, divergence, and often a complicated mix thereof. Complica- Although lacking the grandeur of nature, laboratory ex- tions include concordance in the direction of evolutionary change periments offer the possibility of quantifying convergence with sustained quantitative variation among populations, and the po- because the denominator—the number of replicate popula- tential for a given trait to exhibit divergence on one timescale and con- tions that do or do not evolve the same way—is known. vergence on another. Despite these complications, which also occur in nature, experiments provide a powerful way to study evolutionary Moreover, the sizes and ages of lineages are known in most convergence based on analyzing replicate lineages that experience the experiments. Therefore, describing the repeatability of evo- same environment. lution seems simple in principle, although in practice complications may arise. For example, divergent outcomes over fi Keywords: adaptation, Escherichia coli, experimental evolution, t- the short run could become convergent over the long run, ness, mutation rate, parallel evolution. or vice versa, depending on the rates of phenotypic transition and their sensitivity to prior events that promote or im- Introduction pede the transitions. Also, one might well see a mixture of convergent and divergent responses if one examines many Against the backdrop of the diversity of life, nature is re- different traits. plete with examples of parallel and convergent phenotypic In recent years, evolution experiments have been used to evolution. Beyond compiling examples of convergence, how- address a wide range of questions (Elena and Lenski 2003; fi ever, it is dif cult to devise a framework for quantifying the Garland and Rose 2009; Kawecki et al. 2012; Barrick and extent of convergence or predicting when it is likely, or un- Lenski 2013), and the issue of evolutionary repeatability likely, to arise. One complication is the possibility of report- has been a focus of many of these studies. I will not attempt “ ” ing bias, or what might be called the denominator problem. to review the entire field of experimental evolution here, or For example, camera-like eyes have evolved independently even those studies focused on repeatability, but I will begin in many lineages, including not only vertebrates and cepha- by briefly mentioning three studies that call attention to im- lopod mollusks but also some gastropod mollusks, marine portant issues. First, Burke et al. (2010) found evidence of fi — annelids, spiders, and cubozoan jelly sh the last of which repeatable genomic evolution across replicate Drosophila melanogaster populations that had been selected for accel- * This issue originated as the 2016 Vice Presidential Symposium presented at erated development. The replicates all derived from a single the annual meetings of the American Society of Naturalists. † ancestral population that harbored substantial variation. E-mail: [email protected]. ORCIDs: Lenski, http://orcid.org/0000-0002-1064-8375. Thus, the observed repeatability of the genomic changes — – q probably stemmed from standing genetic variation that Am. Nat. 2017. Vol. 190, pp. S57 S68. 2017 by The University of Chicago. — 0003-0147/2017/190S1-57225$15.00. All rights reserved. is, collateral evolution in the sense of Stern (2013) and DOI: 10.1086/691209 not parallel evolution based on independent mutational

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:46 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S58 The American Naturalist events. Second, in a study of the bacteriophage fX174, addressed by evolution experiments to better understand Wichman et al. (1999) found that almost half of the inde- the fascinating tension between chance and necessity in pendent mutations that reached high frequency in two rep- evolving systems. licate populations were identical, indicating extensive par- Before moving to the experiment, I should brieflydiscuss allelism at the level of the DNA sequence itself. In other a semantic issue that I hope will not cause confusion. In pre- studies, however, repeated phenotypic evolution usually vious presentations on the LTEE, I have usually referred to results from different mutations in the same gene or dif- repeated evolutionary changes as examples of parallelism, ferent genes that affect the same trait. Third, using a high- rather than convergence. I have done so because the repli- resolution random–bar code approach, Levy et al. (2015) cate populations started with the same ancestral state, and monitored thousands of lineages defined by new beneficial so the changes are, geometrically speaking, parallel and mutations within a single population of Saccharomyces cere- not convergent. This distinction is clear when the relevant visiae. Some of these mutations undoubtedly affected the states are discrete (e.g., the identity of a particular nucleo- same gene, and perhaps even the same nucleotide, even tide in a DNA sequence) and the history of changes is though they were independent events, as indicated by differ- known precisely. But for traits that might change repeatedly, ent bar codes. Thus, parallel evolution can occur not only and especially for quantitative traits (e.g., size or fitness), we across replicate populations but also across multiple lineages typically measure the net effect of many changes, some of within a single population. which might be parallel, others divergent, and yet others con- Instead of reviewing the entire field, I will focus on one vergent following prior divergence. In general, our knowl- experiment that I have directed for nearly three decades edge of the steps along the way becomes less certain as the and where the repeatability of evolution—the tension be- time elapsed since the identical ancestral states gets longer. tween divergence and convergence—has been at the fore- Nonetheless, I think it is fair to say that many of us are most front since it began. In fact, one reason I called this an “evo- interested in the push and pull of evolutionary forces that lution experiment,” as opposed to the then-more-familiar promote divergence versus convergence. Therefore, in this “selection experiment,” was to emphasize the fact that any review, I often refer broadly to divergence and convergence, changes that occurred during this experiment encompassed with the understanding that convergence may, depending the origin, as well as the fate, of genetic variation—in essence, on context, include some changes that are, in a strict sense, highlighting the distinction between parallel and collateral parallel. evolution that Stern (2013) also emphasized. In addition to presenting what I think are interesting examples of evolution Experimental Overview in action, I hope to make clear three facets of convergence and divergence that emerge from this experiment. First, we have The LTEE has 12 populations, each founded from the same observed striking examples of phenotypic and genomic con- ancestral strain of Escherichia coli B except for a genetic vergence across the replicate populations. Second, we have marker that distinguishes two sets of six populations each also seen striking cases of divergence, despite the fact that (Lenski et al. 1991). The marker serves two purposes. First, the populations started from the same genotype and have it provides an indication of possible cross contamination, been evolving in identical physical environments. Third, this which would compromise the independence of the replicate mixture of convergence and divergence, when played out populations. Second, it allows one to compete a sample over time and across many traits, defies easy summation from an evolved population against the ancestral strain that and instead requires a more nuanced description. As a con- possesses the alternative marker, or to compete samples sequence, assessing the repeatability of evolution is both from two reciprocally marked populations against one an- more interesting and more difficult than I imagined when other. Each population started from a single haploid cell, I began the experiment. And if things become this compli- which was achieved by plating colonies (with each one cated in a set of small, closed, identical flasks, how much the outgrowth of a single cell) and then using a different col- more challenging it must be to study the repeatability of ony to found each population. As a result, all of the genetic evolution in the natural world. variation available for evolution was generated by new mu- In the next section, I briefly describe the basic structure tations, and no variation was identical by descent across and methods of the long-term evolution experiment (LTEE) replicate populations, thus excluding collateral evolution with Escherichia coli. I then summarize some of the pheno- (Stern 2013) as a potential source of convergence. Each typic and genetic analyses that reveal convergence, diver- population exists in a small Erlenmeyer flask that is held gence, and the complicated mix thereof. I conclude by in a shaking incubator at 377C, where it grows in 10 mL discussing some of the challenges that arise in interpreting of a medium with glucose as the limiting resource. Every the data that result from even such a simple experiment as day, someone removes 1% of the volume from each flask the LTEE, and by suggesting a few questions that might be and transfers it to 10 mL of fresh medium. The 100-fold

≈ : dilution and regrowth allow log2100 6 64 doublings, or 2.0 generations, per day, which we round to 100 generations 1.8 for every 15 transfers. Each day, the bacteria experience a lag phase followed by exponential growth, depletion of the 1.6 limiting glucose, and a stationary phase where they sit until the next transfer. Even as the bacteria have evolved to grow 1.4 faster than their progenitors, the number of generations 1.2 each day is set by the dilution and regrowth. The population size initially fluctuated between ∼5 # 106 and ∼5 # 108 cells; 1.0 those numbers have declined somewhat as the bacteria evolved to ancestor Fitness relative 0 1020304050 larger individual cells. Time (thousands of generations) The LTEE started in 1988, and it has now been running for over 65,000 generations. There have been occasional Figure 1: Trajectories of mean fitness for the 12 independently interruptions along the way, such as moving the experiment evolving populations of the long-term evolution experiment. Each across the country and various accidents. Even more im- curve shows the best fit of a power-law model to the data from portant than their rapid growth, which allows an experi- one population; three trajectories are truncated because of difficul- ment to encompass so many generations, bacteria can be ties in performing the competitions used to estimate fitness. Data frozen and then revived. Every 100 generations in the early are from Wiser et al. (2013). history of the LTEE, and then every 500 generations since, samples have been stored at 2807C, providing a frozen fos- gains. The among-population variance in fitness shows no sil record. These samples serve two critical functions. First, sign of declining with time. The square root of that vari- in the event of an interruption or accident, the LTEE can be ance—in essence, the genetic standard deviation—has hov- restarted without going back to the beginning. Second, one ered around a few percent, which is substantially less than can directly compete or compare samples that lived at dif- one-tenth of the average fitness increase relative to the an- ferent points in time—in essence, time travel. The long du- cestor. Some, but not all, of the persistent variation is asso- ration of the LTEE not only provides an impressive number ciated with evolved differences in mutation rates (Lenski of generations but also means that new technologies, includ- et al. 2015), as described below. Also, one unlucky popu- ing the ability to sequence entire genomes, have emerged lation, called Ara11, appears to have gotten stuck in a less since it began. productive region of the adaptive landscape than all of the others, having both the lowest fitness and the slowest rate of improvement from 40,000 to 60,000 generations (Lenski Convergence and Divergence et al. 2015). Fitness in the LTEE Environment Correlated Responses to Other Environments All of the populations have become more fit, according to assays in which the evolved bacteria compete against the re- One can also run competition assays between the evolved ciprocally marked ancestral strain in the same environment populations and their ancestor in environments that differ as the LTEE has run (Lenski and Travisano 1994; Wiser from the LTEE environment. In contrast to the high and et al. 2013). Before every fitness assay, both competitors relatively homogeneous fitness values measured in the LTEE have been frozen, revived, and acclimated to that environ- environment, fitness levels tend to be lower and more vari- ment. After 50,000 generations, a typical population has a able in other environments (Travisano et al. 1995b). For ex- fitness of ∼1.7 relative to the ancestor, meaning that the ample, when clones sampled at generation 2,000 competed evolved bacteria grow about 70% faster during head-to- against the ancestor in an environment where maltose re- head competition (Wiser et al. 2013). Moreover, the fitness placed glucose, on average they were no more fit than the trajectories are similar in their overall form, with much ancestor. However, the evolved bacteria were extremely het- greater improvements in the early generations than in later erogeneous, with some almost as fit in maltose as in glucose ones (fig. 1). and others much less fit than the ancestor when competing On the other hand, there is significant among-population for maltose. With lactose replacing glucose, the evolved lines variation in fitness—that is, heterogeneity greater than one were on average as fit as they were in glucose, but the genetic would expect from measurement error (Lenski et al. 2015). variation in fitness was even greater than it was in maltose. Moreover, from 40,000 to 60,000 generations the relative- In short, the correlated responses to alternative environ- fitness ranks of the populations were strongly conserved, ments have tended to be more variable than the direct re- even as they all continued to experience significant fitness sponse to selection.

But this tendency toward greater divergence in novel en- predicted they should become smaller. All else being equal, vironments has some wrinkles. For example, all 12 popula- smaller cells have more surface area relative to volume, tions completely lost their ability to grow on another sugar, which seems advantageous in a resource-limited environ- ribose, over the first 2,000 generations of the LTEE (Coo- ment. However, that is not what has happened. Instead, per et al. 2001). In this case, the parallel losses resulted all 12 populations produce much larger individual cells from deletion mutations that were demonstrably beneficial than the ancestral strain (Lenski and Travisano 1994; Vasi in the glucose-based medium and, moreover, occurred spon- et al. 1994; Lenski and Mongold 2000). The adaptive signif- taneously at an exceptionally high rate. The high mutation icance, if any, of the larger cells remains unclear. One po- rate was caused by a transposable element adjacent to the tential explanation is that faster-growing cells tend to be ribose operon that transposed into the operon and then larger, with each cell having more chromosomal copies, ri- underwent homologous recombination, leading to a dele- bosomes, and so on. The evolved cells grow faster than the tion. These losses were so consistent across the replicate ancestor, so perhaps they are larger simply because they are populations as to generate convergent, not divergent, out- growing faster. We tested this hypothesis by forcing evolved comes. and ancestral clones to grow at the same rate in separate There is also a sort of gray area that exists between diver- chemostats, where we could control the growth rate by setting gence and convergence. Over time, many of the LTEE pop- the dilution rate (Mongold and Lenski 1996). We confirmed ulations have lost their ability to grow on maltose; after that faster-growing cells are larger for both the ancestral and 20,000 generations, clones from over half of them could the evolved bacteria. However, the evolved bacteria exhibit a not use that sugar (Pelosi et al. 2006). So one could say that steeper increase in cell size with growth rate, such that they the populations underwent divergent responses to maltose produce larger cells than the ancestral strain at all growth early in the LTEE but are now undergoing convergent evo- rates. Thus, this hypothesis correctly predicts the direction lution toward the maltose-negative state. Thus, both the of the change in cell size, but it does not account for the mag- metric we use—quantitative variation in performance ver- nitude of the change. Another possibility—one that remains sus discrete losses of function—and the timescale over which hypothetical—is that larger cells have an advantage in the evolution takes place can affect whether we describe certain “feast and famine” transfer regime of the LTEE because they patterns as divergent or convergent. are, in effect, larger sponges and thus able to acquire glucose One of the unexpected responses to an alternative envi- faster than smaller ones. The larger cells might thereby se- ronment is that many of the populations have gained resis- quester that limiting resource and pass it (or its by-products) tance to infection by the phage Lambda (Meyer et al. 2010). along as they divide. In any case, all of the populations pro- That resistance is surprising because the bacteria in the duce much larger cells than their common ancestor, so that LTEE environment are never exposed to Lambda or any they have evolved in the same direction, although their re- other virus. Conventional wisdom about trade-offs is that sulting cell volumes are quite variable (Lenski and Travisano resistance has a fitness cost, which would lead one to expect 1994). that evolution in the absence of parasites should lead to There has also been substantial divergence in cell shape, increased susceptibility, not resistance. In fact, the evolved with most lineages clearly rod shaped but others somewhat resistance to Lambda has the same genetic basis as the al- more spherical (Lenski and Mongold 2000). Also, as the tered performance on maltose. It so happens that mutations cells have evolved to become larger, their numerical yield in a gene called malT result in reduced expression of a gene at the end of the daily growth cycle has decreased (Vasi that encodes a porin protein through which maltose crosses et al. 1994; Lenski and Mongold 2000). That is, there are the outer membrane of the Escherichia coli cell. That same fewer, but larger, cells than there were at the outset of the protein, called LamB, is also the receptor to which Lambda LTEE. Nonetheless, the product of cell number and average binds to initiate an infection. Mutations in malT provide a cell volume has increased in all 12 populations, which demonstrable fitness advantage in the glucose-based LTEE means that the biovolume yield increased in parallel as well. environment (Pelosi et al. 2006), probably because they re- Moreover, there was a near-perfect inverse relation between duce the costs of unnecessary expression of LamB and other the numerical yield and the average cell size of the popula- proteins regulated by malT, and the resistance to phage tions at 10,000 generations (Lenski and Mongold 2000). Lambda is simply a pleiotropic effect of these cost-saving The constancy of that product indicates that the changes mutations. in yield were not only qualitatively but also quantitatively convergent. Once again, however, different analyses of cell morphology—the particular traits measured as well as the Cell Size and Yield direction versus the magnitude of evolved changes—may Had I thought in advance about how the size of the bacte- affect whether one is more impressed by the convergence rial cells would change during the LTEE, I would have or by the divergence of the replicate populations.

A New Metabolic Capability At first, we struggled to understand what had changed genetically that enabled the cells to grow on citrate. How- The LTEE culture medium contains glucose as the limiting ever, we ran what we call “replay” experiments to distin- resource, but there is another large pool of carbon and en- guish between two hypotheses. According to one, the new ergy in the medium that the cells cannot consume. In par- Cit1 function resulted from some rare mutation—for ex- ticular, the medium contains citrate as a chelating agent, ample, an inversion that had to occur at two exact points and its concentration is much higher than that of glucose. in the genome—that could have happened at any time in However, a near-universal feature of E. coli as a species is that population’s history and would have yielded the same that it cannot grow on citrate in oxygen-containing envi- ability to grow on citrate. Alternatively, the origin of this ronments, because the cells are unable to take up citrate function was contingent on one or more earlier “potentiat- from the environment. So every day, after the bacteria con- ing” mutations that, by themselves, conferred no capacity sumed all of the available glucose, another source of carbon to use citrate but, with the right subsequent mutation, en- and energy sat there untapped. That situation continued for abled its consumption. Several large replay experiments sup- over 30,000 generations, but then one population, called ported the hypothesis of historical contingency: neither the Ara23, began to use the citrate (Blount et al. 2008), re- ancestor nor any clone tested from population Ara23 before sulting in a culture that was much more turbid than the 20,000 generations produced any Cit1 mutants, but later others (fig. 2A). At first, I thought that a citrate-using con- clones generated many such mutants (Blount et al. 2008). taminant of another species was responsible, and so the Soon it became affordable to sequence whole genomes population was restarted from its most recent frozen sam- and identify the mutations responsible for the new Cit1 ple. When the citrate users reemerged, we analyzed genetic function (Blount et al. 2012). This analysis revealed that markers that confirmed that they were indeed E. coli cells the mutation that produced the first Cit1 cell in population descended from the strain used to start the LTEE. Ara23 was a tandem duplication of a segment that includes

Figure 2: Evolution of a new function in population Ara23. A, Increase in optical density (i.e., turbidity) when some cells evolved the capacity to use citrate. Redrawn from Blount et al. (2008). B, Tandem duplication leading to two copies of the citT gene, with the second copy coming under the control of one of the promoter elements, shown as arrows, downstream of the ancestral copy. Redrawn from Blount et al. (2012).

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:46 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S62 The American Naturalist the citT gene, which encodes a protein used to transport cit- (glucose) to two (glucose and citrate) as a consequence of rate and some other carboxylic acids under anoxic condi- how the new function works. In particular, the citT-encoded tions. However, the effect of the duplication is not a dou- protein is an antiporter: in the course of importing one mol- bling of the basal expression level; in fact, there was no ecule into the cell, it exports another molecule into the en- detectable expression of the ancestral copy of citT (Blount vironment. Thus, as the citrate is consumed, other, less valu- et al. 2012). Instead, the duplication brought together the able but still usable sources of carbon and energy—succinate, new copy of that gene with a promoter region that was fumarate, and malate—accumulate in the medium, leading to downstream of the original copy (fig. 2B), which led to the a system with not one, not two, but several distinct resources. expression of the citT-encoded transporter during aerobic Both the Cit1 and Cit2 lineages further evolved in ways that growth just as the glucose was depleted. All of the Cit1 mu- improved their growth on these by-products of citrate trans- tants found in the replay experiments also had rearrange- port (Turner et al. 2015b). ments of their DNA that appeared to bring together the citT As of this writing, at 65,000 generations, no other LTEE gene and either the same or other promoters, although in population has evolved the ability to grow on the citrate. none of the replays was the rearrangement identical to the The evolution of this function is therefore a conspicuous ex- one in the original Ara23 population. The genomic data ample of divergence under identical conditions from the also showed that the new module with the combined regu- same initial genetic state. It will be interesting to see whether latory domain and structural gene underwent further du- any of the other 11 populations make this transition in the plications that improved growth on citrate. By cloning the years, decades, and centuries ahead. If some or all of them module into a plasmid that could be moved into other ge- do, then what was once seen as divergence might eventually netic backgrounds, we found strong epistasis between those be called convergence. Having replicate populations evolve backgrounds and the module. The new module conferred under identical conditions solves the denominator problem some capacity to grow on citrate even in the ancestral strain; with respect to the number of lineages at risk, but it does however, growth was much weaker in that background and not address the open-ended nature of the time at risk for several others than it was in the background of a late- some particular change to occur. generation clone closely related to the lineage in which the new function evolved (Blount et al. 2012). These results thus Mutation Rates confirmed the importance of potentiating mutations in setting the stage for the subsequent evolution of the Cit1 phe- Six of the 12 LTEE populations evolved point-mutation rates notype. Later genetic analyses identified a specific mutation roughly 100-fold higher than that of their ancestor (Snie- in another gene that played an important role in that poten- gowski et al. 1997; Tenaillon et al. 2016). In four cases the tiation (Quandt et al. 2015). hypermutability resulted from mutations that affect DNA After the bacteria began to use the citrate, it caused im- mismatch repair, and in two cases it was caused by muta- portant changes to the ecology of the system. Because of tions affecting enzymes that remove oxidized nucleotides. the high concentration of citrate in the medium, the popu- Hence, the set of populations as a whole have diverged in lation density increased several-fold. Also, the Cit1 bacteria their mutation rates. However, one could also say that six did not drive the Cit2 bacteria to extinction; instead, the populations converged on higher mutation rates while di- two lineages coexisted for over 10,000 generations (Blount verging in terms of the biochemical basis for their hyper- et al. 2012; Turner et al. 2015a). The Cit2 cells could persist mutability. In any case, these changes have had important because they had an advantage in competition for the glu- consequences. For example, the hypermutable lineages have cose; in particular, they transitioned from stationary phase accumulated many more mutations, and a higher propor- to renewed growth after transfer into fresh medium faster tion of synonymous substitutions, than those that retained than did the Cit1 cells (Blount et al. 2008). This difference the low ancestral mutation rate (Tenaillon et al. 2016). Also, probably occurred because the Cit1 cells enter stationary the populations that evolved hypermutability early in the phase after depleting the citrate and then switch over to LTEE have reached higher fitness levels than the other pop- the more valuable glucose in the fresh medium, whereas ulations (Wiser et al. 2013; Lenski et al. 2015), although the the Cit2 cells used glucose both before they entered station- fitness differences are far smaller than the differences in ary phase and when they recommenced growth. Eventually, mutation rate owing to clonal interference (i.e., competition however, the Cit2 lineage became extinct. The cause of that among lineages with different beneficial mutations) and to extinction is unclear, but it appears to have been caused the increased load of deleterious mutations. Moreover, as by some transient perturbation to the experiment (Turner a consequence of the increased mutational load and the de- et al. 2015a). clining opportunity for further adaptation, five of the hyper- In fact, the Ara23 ecosystem became even more compli- mutable populations subsequently evolved lower mutation cated than one might expect in going from one resource rates (Wielgoss et al. 2013; Tenaillon et al. 2016). Thus, we

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:46 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). A Long-Term Experiment with Bacteria S63 see evidence of convergence not only in the mutation-rate 100 phenotype but also in its temporal trajectory. 80 At this point, one might be wondering whether the lineage in population Ara23 that gained the ability to grow on 60 citrate had previously evolved hypermutability. In fact, it 40 had not. However, it evolved a mutator phenotype not long 20 after it acquired that function. Moreover, Ara23 is the only mutations Total population that became hypermutable but, so far, has not at 0 least partially compensated for that change. As a conse- 01020304050 quence of its new ability to use citrate, this population has Time (thousands of generations) additional opportunities for adaptation, apparently overrid- ing the benefit that would accrue from a reduced load of deleterious mutations. Figure 3: Trajectories for the total number of mutations in genomes sampled from five long-term evolution experiment populations that In addition to the six populations that evolved higher fi — 1 did not evolve any type of hypermutability. Each curve shows the t rates of point mutation, another population Ara 1, the of a two-parameter model, with terms representing beneficial and laggard with respect to its fitness trajectory—evolved a dif- neutral mutations, to numbers obtained from two clones at each ferent type of hypermutability (Lenski et al. 2015; Tenaillon of 11 time points. Data are from Tenaillon et al. (2016). et al. 2016). In this case, one of the several transposable elements present in the ancestral genome evolved heightened terms. Second, nonsynonymous and intergenic point muta- activity that caused a substantial increase in the number of tions were found in substantial excess relative to the expec- insertion mutations. tation based on synonymous substitutions, after adjust- ment for the number of sites at risk for each class. Third, insertions and deletions, as well as nonsynonymous and Genomic Changes intergenic point mutations, were seen more often than ex- When the LTEE began in 1988, not a single bacterial ge- pected in comparison to a mutation-accumulation experi- nome had been fully sequenced. In one recent paper, 264 ge- ment, in which passage through single-cell bottlenecks nomes sampled from the LTEE were fully sequenced and largely eliminated the effects of natural selection. analyzed, including two representatives from all 12 popu- Moreover, there was strong convergent evolution at the lations at 11 time points through generation 50,000 (Te- level of genes. Fifty-seven protein-coding genes (out of naillon et al. 2016). As discussed above, some of the LTEE 14,000) had two or more independent nonsynonymous populations evolved hypermutability, and those changes mutations across all of the nonmutator and premutator obviously affect genomic evolution. If we exclude the pop- lineages combined (Tenaillon et al. 2016). These 57 genes ulations that became hypermutable (including the one with comprise only 2% of the total length of protein-coding increased insertions), the genomes sampled from the other genes, but they harbored fully 50% of the nonsynonymous populations averaged ∼76 total mutations after 50,000 gen- mutations in those genomes. That degree of convergence erations (fig. 3). Of that total, ∼56% were point mutations; is extremely unlikely (P ! 102143), according to simulations various types of insertions and deletions account for most of in which mutations were randomly distributed among the the rest. protein-coding genes according to length (Tenaillon et al. Most mutations that occurred during the LTEE were pre- 2016). Some of the genes showing the strongest convergence sumably neutral or deleterious, but those that gave some encode proteins with core regulatory and metabolic func- competitive advantage were more likely to reach high fre- tions. Two such genes, spoT and malT, were first discovered quency and thus be present in the genomes that were se- from changes in gene-expression profiles, through tran- quenced. Even so, neutral and deleterious mutations can scriptomics and proteomics, respectively, and then tracing hitchhike with beneficial driver mutations; they may also back to the causal mutations (Cooper et al. 2003; Pelosi et al. be present in a sampled genome as the result of a recent 2006). Another gene, topA, was found after changes in DNA mutation or random drift. However, several lines of evi- superhelicity were observed (Crozat et al. 2010). Several dence indicate that most of the mutations observed in the other genes showing convergent evolution encode proteins nonmutator populations, and in the others before they be- involved in peptidoglycan synthesis, and mutations there came hypermutable, were beneficial (Tenaillon et al. 2016). contribute to the evolved changes in cell size and shape First, the trajectory for the number of mutations over time (Philippe et al. 2009). There was also a significant, though was better fit by a model with both neutral and beneficial weaker, signal of convergent evolution in the lineages that mutations, which are expected to increase linearly and at a evolved hypermutability (Tenaillon et al. 2016). That sig- declining rate, respectively, than by models without both nal was weaker because the beneficial mutations in those

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:46 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S64 The American Naturalist lineages were diluted in a sea of neutral or weakly deleteri- to screen and sort mutations (http://barricklab.org/shiny ous hitchhikers and drifters. /LTEE-Ecoli/), I asked, for each population at each of the Deletion and insertion mutations also exhibited substan- 11 time points, how many genes had one or more nonsynony- tial convergence in the LTEE (Tenaillon et al. 2016). How- mous substitutions in at least one of the two sequenced ever, it is more difficult to interpret these events, for two clones. That number rose from ∼2, on average, at 500 gen- reasons. First, many of the parallel indels potentially affect erations to ∼32 at 50,000 generations. Then, for every pair of multiple genes. Second, some of them seem to reflect hyper- populations, I asked how many genes had nonsynonymous mutability caused by transposable elements and homolo- mutations in both of the samples from the same generation. gous recombination involving repeated sequences, rather That number increased steadily, from ∼0.3 to ∼5.8, on aver- than convergence caused by natural selection. In any case, age, after 500 and 50,000 generations, respectively. I then cal- many of the gene regions most often affected by indels in culated the proportional overlap in the affected genes for each = 1 the LTEE appear to have been acquired by horizontal trans- population pair at each time point as 2nab (na nb), where fer in the distant past. The disruption or deletion of such na and nb are the total number of affected genes for each pop- genes may be effectively neutral if their products are unused ulation and nab is the number they share. This index of par- in the LTEE environment, or those mutations might be allelism was ∼15% after 500 generations, it rose to ∼33% at beneficial if the genes and their products impose some met- 2,000 and 5,000 generations, and it then declined gradually abolic burden on the cells. to ∼18% by 50,000 generations. Conte et al. (2012) suggest two possible explanations for the declining probability of reusing the same genes at longer Trajectory of Genetic Parallelism divergence times. First, more recently diverged lineages are While the statistical signal for genetic convergence is ex- more likely to share genetic variants that could fuel parallel tremely strong, what can be said about its magnitude and evolution (i.e., collateral evolution, in the sense of Stern how it changes over time? In a review of published studies, 2013). This issue is not relevant to the LTEE, however, be- Conte et al. (2012) sought to quantify the probability that cause there was no shared variation at the start. Second, the the same genes were reused in cases of parallel and conver- genetic backgrounds in which new alleles must function di- gent evolution of various phenotypes in nature. Depending verge over time, introducing the potential for epistatic in- on the methods used in the underlying studies, they found teractions to alter the spectrum of mutations that produce that the same genes were involved in about one-third to a particular beneficial effect and thereby generate historical one-half of the cases they examined. However, this proba- contingencies. This explanation is certainly plausible for bility varied, depending on how long ago the lineages had the LTEE as well, and indeed it is implicated in the evolu- diverged—there was a greater chance that the same genes tion of citrate utilization (Blount et al. 2008). were responsible for parallel changes in the most recently Two other explanations are also relevant to the LTEE, es- diverged lineages. pecially for such an integrative trait as fitness. First, there As noted in the introduction, one of the advantages of an are probably more genes that can yield mutations that con- experiment like the LTEE is that we know the denominator fer small fitness gains than there are genes that can produce (i.e., the number of populations evolving under a given set mutations that confer large gains. Because of the dynamics of conditions) when it comes to analyzing convergence. We of selection, including clonal interference between asexual also know precisely how long populations have been sepa- lineages that possess different beneficial mutations, there rated since their common ancestor. Therefore, it is feasible is a tendency for beneficial mutations of large effect to fix to address the same issues examined by Conte et al. (2012), before those that confer smaller benefits (Gerrish and Lenski albeit in a very different system. However, there are em- 1998; Wiser et al. 2013). As a consequence, adaptation may pirical challenges: the evolution of hypermutable lineages follow more similar paths—involving those few genes with dilutes and thus obscures the driver mutations, and some the largest benefits—early in the LTEE, before fanning out mutational events affect multiple genes. Moreover, fitness along more diverse paths in later generations. Second, the is an extremely broad and integrative phenotype relative to, fraction of accumulated mutations that are beneficial declines say, floral pigmentation or resistance to some toxin. None- over time, with a corresponding increase in the proportion theless, I sought to address the issue as follows. I focused on of neutral or weakly deleterious mutations (Tenaillon et al. nonsynonymous mutations in protein-coding genes, using 2016). While most mutations that affect the same gene in a the data in Tenaillon et al. (2016). I used all 12 populations pair of populations are beneficial, those parallel changes be- through 2,000 generations, because none had evolved hyper- come a smaller fraction of the total mutations over time. This mutability to that point; I used only the six populations that effect might well account for much of the apparent increase in retained the ancestral point-mutation rate through 50,000 the diversity of genetic pathways over time. The accumula- generations for the later samples. Using a Web-based tool tion of neutral or weakly deleterious mutations could also

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:46 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). A Long-Term Experiment with Bacteria S65 generate historical contingencies if some of them provide 1 “stepping stones” that enable adaptations that would other- 0.9 wise not be accessible by a strictly hill-climbing process (Co- 0.8 vert et al. 2013). 0.7 0.6 Stable Polymorphisms 0.5 0.4 Bacteria reproduce asexually, and those in the LTEE lack 0.3 any means for horizontal gene transfer. Unlike some bacte- 0.2 rial species, E. coli is not naturally transformable, and the Frequency of S morph 0.1 strain used to found the LTEE does not carry any plasmids 0 or functional phages that could move genes between cells 0 5000 10,000 15,000 20,000 by conjugation or transduction, respectively. Bacteria are Generation also haploid, and so polymorphisms maintained by hetero- zygote advantage are not possible. Moreover, the LTEE — Figure 4: Fluctuations in the frequency of the S type in population environment is simple, thereby limiting but not eliminat- 2 — Ara 2. Two ecotypes, L and S, have coexisted for many thousands ing the opportunity for distinct ecotypes to evolve and sta- of generations, but their relative abundances have changed substan- bly coexist. In short, one might expect evolution in the LTEE tially as beneficial mutations arose in their respective lineages. Re- to occur by sequential selective sweeps of beneficial muta- printed from Rozen and Lenski (2000). tions, either singly or as cohorts (Lang et al. 2013; Maddam- setti et al. 2015), such that any polymorphisms are transient, not persistent. ages—one lineage gets a beneficial mutation that drives the I already mentioned one exception to this expectation, equilibrium ratio in its favor, and then the other does, and namely, the coexistence between the Cit1 and Cit2 lineages so on (Le Gac et al. 2012). However, neither lineage has ac- in population Ara23 after the former gained the new abil- quired a beneficial mutation that is sufficient to overcome ity to use the exogenously supplied citrate. Although the the advantage when rare that keeps the other lineage from evolution of that ability was surprising, the fact that two becoming extinct. That is, a sustained polymorphism de- lineages could stably coexist on two resources—glucose pends not only on the effect sizes of the beneficial mutations and citrate—is not. However, another population, Ara22, that arise in the contending lineages but also on the strength produced a polymorphism that arose much earlier and of the frequency-dependent interaction that gives each eco- has persisted far longer (Rozen and Lenski 2000; Le Gac type its advantage when rare (Maddamsetti et al. 2015). et al. 2012; Tenaillon et al. 2016). The polymorphism was In population Ara21, two genetically distinct lineages first detected through heritable differences in colony mor- coexisted from about 7,000 to 15,000 generations, and com- phology, with two morphotypes dubbed “L” and “S” for their petitions between representatives of the two lineages indi- large and small colonies, respectively. Experiments in which cate a negative frequency-dependent interaction (Maddam- the two types from the same generation were mixed at dif- setti et al. 2015). There were also dramatic back-and-forth ferent initial frequencies demonstrate a negative frequency- fluctuations in the relative abundance of the two lineages, dependent interaction that supports coexistence (Rozen and similar to those seen in Ara22. However, in Ara21, one Lenski 2000; Le Gac et al. 2012). Other experiments have lineage eventually accumulated beneficial mutations that shown that the L type grows faster on glucose, although both delivered a knockout blow, driving the other to extinction types grow substantially faster than the ancestral strain. How- (Maddamsetti et al. 2015). Genome sequencing has also un- ever, the S type can also grow in spent medium where L had covered distinct lineages that coexisted in other populations previously grown but can grow no further (Rozen and Lenski for over 10,000 generations (Blount et al. 2012; Tenaillon 2000). Recent work indicates that acetate, which is excreted et al. 2016). These deeply diverged lineages suggest that neg- by both types during growth on glucose, is used more effec- ative frequency-dependent interactions, whether transient tively by and supports the S population (Großkopf et al. or sustained, have evolved in at least several of the LTEE 2016). populations. Although the two ecotypes converge on a stable equilibrium over the course of several weeks, over longer periods Overview and Outlook their relative abundance has shifted dramatically (Rozen and Lenski 2000), with their ratio varying by almost two I began this review by suggesting that evolution experiments orders of magnitude over the first 20,000 generations (fig. 4). offer the potential to quantify rigorously the extent of con- These shifts reflect the ongoing evolution of the two line- vergence and divergence, because the number of lineages

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:46 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S66 The American Naturalist that might or might not change in a particular way is known gists working with plants and animals. The mere handful of and because the conditions can be made essentially identical genes that underlie similarly specific traits that have changed with respect to both the environment and the initial genetic during the LTEE, such as DNA superhelicity (Crozat et al. state. Using the Escherichia coli LTEE as an example, I have 2010) and hypermutability (Sniegowski et al. 1997; Wiel- shown how this approach has allowed my team to quantify goss et al. 2013; Tenaillon et al. 2016), might provide more the among-population variation in fitness, test for parallel relevant comparisons with typical studies of evolutionary mutations in genes, and characterize the patterns of change convergence. in many other traits. When I look across the sweep of these In any case, other evolution experiments will be neces- analyses—from genetics and genomics through physiology sary to illuminate important factors that could influence and morphology to performance and ecology—I am fasci- the likelihood of convergent versus divergent responses. nated by both the convergences and the divergences of the Does recombination accelerate adaptive evolution, and replicate populations. While we have seen many striking does it promote convergence above and beyond the effect examples of convergence, the more traits we study and the of any shared variants present at the start of an experiment more deeply we examine them, we see that each population (Azevedo et al. 2006; Cooper 2007; McDonald et al. 2016)? is unique and follows its own evolutionary path. For exam- Does environmental complexity, such as that produced by ple, all populations evolved to produce much larger cells spatial structure, multiple resources, or coevolutionary in- than the common ancestor, and yet the extent of the size teractions, promote not only diversity within populations changes and the resulting shapes of the cells show consider- but also divergence across replicate lineages (Korona et al. able variation. 1994; Rainey and Travisano 1998; Cooper and Lenski 2010; Another complication arises if we examine the extent of Paterson et al. 2010; Nahum et al. 2015)? How do popula- convergence and divergence over time. For example, six of tion size and initial genotypes, whether sampled from na- the LTEE populations evolved hypermutator phenotypes, ture or from other experiments, affect the propensity for and six did not. But perhaps all of them will eventually be- subsequent divergence (Travisano et al. 1995a; Burch and come mutators. Or perhaps the mutator lineages will re- Chao 1999; Moore and Woods 2006; Weinreich et al. 2006; evolve lower mutation rates, as some of them have already Perfeito et al. 2007; Woods et al. 2011; Lindsey et al. 2013; done. Similarly, only one population has evolved the ability Kryazhimskiy et al. 2014)? In short, what genetic and envi- to grow on citrate even after 60,000 generations, but per- ronmental factors determine whether fitness landscapes are haps all of them will have gained this ability after 600,000 smooth or rugged, and how do they interact to promote di- or 6,000,000 generations. Or maybe some of them will vergent or convergent solutions to a given set of challenges evolve other new abilities, such as horizontal gene transfer (De Visser and Krug 2014; Wang et al. 2016)? With so many or predation on other cells. In short, we know that the pop- questions to answer, the tension between evolutionary con- ulations started out identical, and we know that they now vergence and divergence—between the repeatable and the differ from one another despite having undergone many unique—should be a source of fascination and discovery similar changes, but it is unclear whether the divergences far into the future. will persist indefinitely or might yet prove to be transient. My hope is that the LTEE will continue for many scientific generations (Fox and Lenski 2015), providing further Acknowledgments insights into the core tension between chance and necessity in evolution. Of course, these E. coli populations are differ- I thank everyone who has worked on the LTEE over the ent in important respects from the systems that most evo- course of nearly 30 years, including especially N. Hajela, lutionary biologists study. Their asexual reproduction leads who has been the lab manager and technician for almost to clonal interference, which might promote convergence 20 years; D. Schneider, who has worked to understand the because the relatively few mutations with the largest bene- genetic changes in the LTEE lines for nearly two decades; ficial effects are even more likely to fix first than they are in and Z. Blount, who has studied the causes and consequences recombining populations. On the other hand, asexual re- of citrate utilization for over a decade. I thank A. Agrawal for production may promote divergence if it allows syner- inviting me to write this review; D. Schluter for a suggestion gistically beneficial combinations of mutations to fixmore that led to the section on the trajectory of genetic parallel- readily. Also, analyses of the LTEE populations have often fo- ism; J. Barrick for developing the tool that facilitated the cused on extremely broad and integrative phenotypic traits, analyses in that section; and Z. Blount, O. Tenaillon, and including cell size and competitive fitness, where the number M. Wiser for preparing figures. The LTEE is currently sup- of genes with relevant mutations is likely to be much larger ported by the National Science Foundation (DEB-1451740), than that for specific traits, such as pigmentation and toxin the BEACON Center for the Study of Evolution in Action resistance, that are typically studied by evolutionary biolo- (DBI-0939454), and Michigan State University.

Literature Cited Gerrish, P. J., and R. E. Lenski. 1998. The fate of competing beneficial mutations in an asexual population. Genetica 102/103:127– Azevedo, R. B. R., R. Lohau, S. Srinivasan, K. K. Dang, and C. L. 144. Burch. 2006. Sexual reproduction selects for robustness and neg- Großkopf, T., J. Consuegra, J. Gaffé, J. Willison, R. E. Lenski, O. S. ative epistasis in artificial gene networks. Nature 440:87–90. Soyer, and D. Schneider. 2016. Metabolic modelling in a dynamic Barrick, J. E., and R. E. Lenski. 2013. Genome dynamics during ex- evolutionary framework predicts adaptive diversification of bacte- perimental evolution. Nature Reviews Genetics 14:827–839. ria in a long-term evolution experiment. BMC Evolutionary Biol- Blount, Z. D., J. E. Barrick, C. J. Davidson, and R. E. Lenski. 2012. ogy 16:163. doi:10.1186/s12862-016-0733-x. Genomic analysis of a key innovation in an experimental Esche- Kawecki, T. J., R. E. Lenski, D. Ebert, B. Hollis, I. Olivieri, and M. C. richia coli population. Nature 489:513–518. Whitlock. 2012. Experimental evolution. Trends in Ecology and Blount, Z. D., C. Z. Borland, and R. E. Lenski. 2008. Historical con- Evolution 27:547–560. tingency and the evolution of a key innovation in an experimental Korona, R., C. H. Nakatsu, L. J. Forney, and R. E. Lenski. 1994. Evi- population of Escherichia coli. Proceedings of the National Acad- dence for multiple adaptive peaks from populations of bacteria emy of Sciences of the USA 105:7899–7906. evolving in a structured habitat. Proceedings of the National Acad- Burch, C. L., and L. Chao. 1999. Evolution by small steps and rugged emy of Sciences of the USA 91:9037–9041. landscapes in the RNA virus f6. Genetics 151:921–927. Kryazhimskiy, S., D. P. Rice, E. R. Jerison, and M. M. Desai. 2014. Burke, M. K., J. P. Dunham, P. Shahrestani, K. R. Thornton, M. R. Global epistasis makes adaptation predictable despite sequence- Rose, and A. D. Long. 2010. Genome-wide analysis of a long-term level stochasticity. Science 344:1519–1522. evolution experiment with Drosophila. Nature 467:587–590. Lang, G. I., D. P. Rice, M. J. Hickman, E. Sodergren, G. M. Wein- Conte, G. L., M. E. Arnegard, C. L. Peichel, and D. Schluter. 2012. stock, D. Botstein, and M. M. Desai. 2013. Pervasive genetic hitch- The probability of genetic parallelism and convergence in natural hiking and clonal interference in forty evolving yeast populations. populations. Proceedings of the Royal Society B 279:5039–5047. Nature 500:571–574. Conway Morris, S. 2003. Life’s solution: inevitable humans in a lonely Le Gac, M., J. Plucain, T. Hindré, R. E. Lenski, and D. Schneider. universe. Cambridge University Press, Cambridge. 2012. Ecological and evolutionary dynamics of coexisting lineages Cooper, T. F. 2007. Recombination speeds adaptation by reducing during a long-term experiment with Escherichia coli.Proceedings competition between beneficial mutations in populations of Esch- of the National Academy of Sciences of the USA 109:9487–9492. erichia coli. PLoS Biology 5(9):e225. doi:10.1371/journal.pbio.0050 Lenski, R. E., and J. A. Mongold. 2000. Cell size, shape, and fitness in 225. evolving populations of bacteria. Pages 221–235 in J. H. Brown Cooper, T. F., and R. E. Lenski. 2010. Experimental evolution with E. and G. B. West, eds. Scaling in biology. Oxford University Press, coli in diverse resource environments. I. Fluctuating environments Oxford. promote divergence of replicate populations. BMC Evolutionary Lenski, R. E., M. R. Rose, S. C. Simpson, and S. C. Tadler. 1991. Biology 10:11. doi:10.1186/1471-2148-10-11. Long-term experimental evolution in Escherichia coli.I.Adapta- Cooper, T. F., D. E. Rozen, and R. E. Lenski. 2003. Parallel changes tion and divergence during 2,000 generations. American Natural- in gene expression after 20,000 generations of evolution in Esche- ist 138:1315–1341. richia coli. Proceedings of the National Academy of Sciences of the Lenski, R. E., and M. Travisano. 1994. Dynamics of adaptation and di- USA 100:1072–1077. versification: a 10,000-generation experiment with bacterial popu- Cooper, V. S., D. Schneider, M. Blot, and R. E. Lenski. 2001. Mech- lations. Proceedings of the National Academy of Sciences of the anisms causing rapid and parallel losses of ribose catabolism in USA 91:6808–6814. evolving populations of E. coli B. Journal of Bacteriology 183:2834– Lenski, R. E., M. J. Wiser, N. Ribeck, Z. D. Blount, J. R. Nahum, J. J. 2841. Morris, L. Zaman, et al. 2015. Sustained fitness gains and variability Covert, A. W., III, R. E. Lenski, C. O. Wilke, and C. Ofria. 2013. Ex- in fitness trajectories in the long-term evolution experiment with periments on the role of deleterious mutations as stepping stones Escherichia coli. Proceedings of the Royal Society B 282:20152292. in adaptive evolution. Proceedings of the National Academy of doi:10.1098/rspb.2015.2292. Sciences of the USA 110:E3171–E3178. Levy, S. F., J. R. Blundell, S. Venkataram, D. A. Petrov, D. S. Fisher, Crozat, E., C. Winkworth, J. Gaffé, P. F. Hallin, M. A. Riley, R. E. and G. Sherlock. 2015. Quantitative evolutionary dynamics using Lenski, and D. Schneider. 2010. Parallel genetic and phenotypic high-resolution lineage tracking. Nature 519:181–186. evolution of DNA superhelicity in experimental populations of Lindsey, H. A., J. Gallie, S. Taylor, and B. Kerr. 2013. Evolutionary Escherichia coli. Molecular Biology and Evolution 27:2113–2128. rescue from extinction is contingent on a lower rate of environ- De Visser, J. A. G. M., and J. Krug. 2014. Empirical fitness land- mental change. Nature 494:463–467. scapes and the predictability of evolution. Nature Reviews Genet- Maddamsetti, R., R. E. Lenski, and J. E. Barrick. 2015. Adaptation, ics 15:480–490. clonal interference, and frequency-dependent interactions in a long- Elena, S. F., and R. E. Lenski. 2003. Evolution experiments with micro- term evolution experiment with Escherichia coli. Genetics 200:619– organisms: the dynamics and genetic bases of adaptation. Nature 631. Reviews Genetics 4:457–469. McDonald, M. J., D. P. Rice, and M. M. Desai. 2016. Sex speeds ad- Fox, J. W., and R. E. Lenski. 2015. From here to eternity—the theory aptation by altering the dynamics of molecular evolution. Nature and practice of a really long experiment. PLoS Biology 13(6): 531:233–236. e1002185. doi:10.1371/journal.pbio.1002185. Meyer, J. R., A. A. Agrawal, R. T. Quick, D. T. Dobias, D. Schneider, Garland, T., Jr., and M. R. Rose, eds. 2009. Experimental evolution: and R. E. Lenski. 2010. Parallel changes in host resistance to viral concepts, methods, and applications of selection experiments. infection during 45,000 generations of relaxed selection. Evolution University of California Press, Berkeley. 64:3024–3034.

Mongold, J. A., and R. E. Lenski. 1996. Experimental rejection of a Travisano, M., J. A. Mongold, A. F. Bennett, and R. E. Lenski. 1995a. nonadaptive explanation for increased cell size in Escherichia coli. Experimental tests of the roles of adaptation, chance, and history Journal of Bacteriology 178:5333–5334. in evolution. Science 267:87–90. Moore, F. B.-G., and R. Woods. 2006. Tempo and constraint of adaptive Travisano, M., F. Vasi, and R. E. Lenski. 1995b. Long-term experi- evolution in Escherichia coli (Enterobacteriaceae, Enterobacteriales). mental evolution in Escherichia coli. III. Variation among replicate Biological Journal of the Linnean Society 88:403–411. populations in correlated responses to novel environments. Evolu- Nahum, J. R., P. Godfrey-Smith, B. N. Harding, J. H. Marcus, J. tion 49:189–200. Carlson-Stevermer, and B. Kerr. 2015. A tortoise-hare pattern Turner, C. B., Z. D. Blount, and R. E. Lenski. 2015a.Replayingevo- seen in adapting structured and unstructured populations suggests lution to test the cause of extinction of one ecotype in an experi- a rugged fitness landscape in bacteria. Proceedings of the National mentally evolved population. PLoS ONE 10(11):e0142050. doi:10 Academy of Sciences of the USA 112:7530–7535. .1371/journal.pone.0142050. Paterson, S., T. Vogwill, A. Buckling, R. Benmayor, A. J. Spiers, N. R. Turner, C. B., Z. D. Blount, D. H. Mitchell, and R. E. Lenski. 2015b. Thomson, M. Quail, et al. 2010. Antagonistic coevolution accelerates Evolution and coexistence in response to a key innovation in a molecular evolution. Nature 464:275–278. long-term evolution experiment with Escherichia coli. bioRxiv Pelosi, L., L. Kühn, D. Guetta, J. Garin, J. Geiselmann, R. E. Lenski, 020958. doi:10.1101/020958. and D. Schneider. 2006. Parallel changes in global protein profiles Vasi, F., M. Travisano, and R. E. Lenski. 1994. Long-term experimen- during long-term experimental evolution in Escherichia coli.Ge- tal evolution in Escherichia coli. II. Changes in life-history traits netics 173:1851–1869. during adaptation to a seasonal environment. American Naturalist Perfeito, L., L. Fernandes, C. Mota, and I. Gordo. 2007. Adaptive mu- 144:432–456. tations in bacteria: high rate and small effects. Science 317:813–815. Wang, Y., C. D. Arenas, D. M. Stoebel, K. Flynn, E. Knapp, M. M. Philippe, N., L. Pelosi, R. E. Lenski, and D. Schneider. 2009. Evolu- Dillon, A. Wünsche, et al. 2016. Benefit of transferred mutations tion of penicillin-binding protein 2 concentration and cell shape is better predicted by the fitness of recipients than by their ecolog- during a long-term experiment with Escherichia coli.Journalof ical or genetic relatedness. Proceedings of the National Academy Bacteriology 191:909–921. of Sciences of the USA 113:5047–5052. Quandt, E. M., J. Gollihar, Z. D. Blount, A. D. Ellington, G. Geor- Weinreich, D. M., N. F. Delaney, M. A. DePristo, and D. L. Hartl. giou, and J. E. Barrick. 2015. Fine-tuning citrate synthase flux po- 2006. Darwinian evolution can follow only very few mutational tentiates and refines metabolic innovation in the Lenski evolution paths to fitter proteins. Science 312:111–114. experiment. eLife 4:e09696. doi:10.7554/eLife.09696. Wichman, H. A., M. R. Badgett, L. A. Scott, C. M. Boulianne, and Rainey, P. B., and M. Travisano. 1998. Adaptive radiation in a het- J. J. Bull. 1999. Different trajectories of parallel evolution during erogeneous environment. Nature 394:69:72. viral adaptation. Science 285:422–424. Rozen, D. E., and R. E. Lenski. 2000. Long-term experimental evolu- Wielgoss, S., J. E. Barrick, O. Tenaillon, M. J. Wiser, W. J. Dittmar, S. tion in Escherichia coli. VIII. Dynamics of a balanced polymor- Cruveiller, B. Chane-Woon-Ming, C. Médigue, R. E. Lenski, and phism. American Naturalist 155:24–35. D. Schneider. 2013. Mutation rate dynamics in a bacterial popu- Sniegowski, P. D., P. J. Gerrish, and R. E. Lenski. 1997. Evolution of lation balance evolvability and genetic load. Proceedings of the high mutation rates in experimental populations of Escherichia National Academy of Sciences of the USA 110:222–227. coli. Nature 387:703–705. Wiser, M. J., N. Ribeck, and R. E. Lenski. 2013. Long-term dynamics Stern, D. L. 2013. The genetic causes of convergent evolution. Nature of adaptation in asexual populations. Science 342:1364–1367. Reviews Genetics 14:751–764. Woods, R. J., J. E. Barrick, T. F. Cooper, U. Shrestha, M. R. Kauth, Tenaillon, O., J. E. Barrick, N. Ribeck, D. E. Deatherage, J. L. and R. E. Lenski. 2011. Second-order selection for evolvability Blanchard, A. Dasgupta, G. C. Wu, et al. 2016. Tempo and mode in a large Escherichia coli population. Science 331:1433–1436. of genome evolution in a 50,000-generation experiment. Nature 536:165–170. Symposium Editor: Anurag Agrawal

“The Bill-ﬁsh (Scomberesox Storerii), . . . which but ﬁfteen years since I saw stranded on the shore by the thousands, driven in by its devouring pursuers, has gradually decreased, till at the present time it has nearly, if not quite, been driven away, and I think that during the past year there was not one specimen seen at Provincetown.” From “The Habits and Migrations of Some of the Marine Fishes of Mas- sachusetts” by James H. Blake (The American Naturalist, 1870, 4:513–521).

Symposium Evolutionary Scenarios and Primate Natural History*

Harry W. Greene†

Department of Ecology and Evolutionary Biology, Cornell University, Ithaca, New York 14853 abstract: Scenarios summarize evolutionary patterns and processes topic, what roles might be played by phylogenetic system- by interpreting organismal traits and their natural history correlates in atics and natural history? a phylogenetic context. They are constructed by (1) describing pheno- As cognitive and cultural organisms, who we are is sub- types (including physiology and behavior), ideally with attention to for- stantially contingent on what we do and with whom we as- mative roles of development, experience, and culture; (2) inferring homol- sociate, rather than simply what we look like, inside and out. ogies, homoplasies, ancestral character states, and their transformations How humans came to be as we are is a matter of member- with phylogenetic analyses; and (3) integrating those components with ships in families, societies, and more ancient cultural and ge- ecological and other ancillary data. At their best, evolutionary scenarios nealogical lineages; why we have certain characteristics thus are factually dense narratives that entail no known falsehoods; their em- fl pirical and methodological shortcomings are transparent, they might re ects, at least partly, causal relationships among those com- be rejected based on new discoveries, and their potential ideological pit- plex intrinsic and extrinsic influences. All that said, for evo- falls are flagged for scrutiny. They are exemplified here by homoplastic lutionary and organismal biologists, understanding the whos, foraging with percussive tools by humans, chimpanzees, capuchins, and hows, and whys of other species, including of our closest kin macaques; homoplastic hunting with spears by humans and chimpan- among other primates, involves comparing similar attributes zees; and private experiences (e.g., sense of fairness, grief) among diverse among related taxa—assessing homologies and homoplasies animals, the homologous or homoplastic status of which often remains unexplored. Although scenarios are problematic when used to bolster for morphological, physiological, and behavioral traits in the political agendas, if constructed carefully and regarded skeptically, they context of ecological correlates (for a pioneering synthesis, can synthesize knowledge, inspire research, engender public understand- see Brooks and McLennan 1991; among many taxon-focused ing of evolution, enrich ethical debates, and provide a deeper historical examples: Kappeler and Heymann 1996, for primates; Losos context for conservation, including nature appreciation. 2009, for lizards, Cavender-Bares 2016, for oaks; for broader Keywords: homology, homoplasy, scenarios, tools, weapons, private perspectives, see Autumn et al. 2002; Losos 2011). The result- experiences. ing summaries are narrative in structure and thereby vulnerable to abridgement, selectivity, simplification, and other potential distortions (e.g., Landau 1984; O’Hara 1992; Cartmill Introduction 2002); consequently, because primate evolution involves our own stories, we should not be surprised that evidentiary and This paper stems from the 2016 American Society of Natu- ideological controversies plague them (e.g., Caplan 1978; Se- ralists Vice Presidential Symposium, for which speakers were gerstråle 2000; Pinker 2002; Borofsky 2004; Prindle 2009; asked to consider homology and homoplasy in the context of Marks 2015). big questions in biology. Taking that charge literally, and in- Accordingly, this paper explores the content, legitimacy, spired by recent discoveries in anthropology and primatol- and relevance of what Eldredge (1979) termed phylogenetic ogy, I first chose: Who are we, and over the long haul, how scenarios, with special reference to primates (for their phylog- and why have we come to be this way? These are timelessly eny and classification, see fig. 1). My treatment emphasizes intriguing, often controversial puzzles for anthropologists and comparative analyses for inferring historical changes in mor- evolutionary biologists, as well as for philosophers and other phology, physiology, and behavior (e.g., Greene 1986, p. 3; Au- scholars in the humanities. But how far can we go in pursu- tumn et al. 2002; Ereshefsky 2007, pp. 668–670; Losos 2011), ing answers, and by way of framing a more manageable rather than denying a place for game-theoretical and other complementary approaches (e.g., Dugatkin and Reeve 2000; * This issue originated as the 2016 Vice Presidential Symposium presented at Borgerhoff Mulder and Beheim 2011). I first recount the con- the annual meetings of the American Society of Naturalists. † ceptualization and critical early reception of what soon came E-mail: [email protected]. ORCIDs: Greene, http://orcid.org/0000-0002-2160-9355. to be called evolutionary scenarios; second, examine the sta- Am. Nat. 2017. Vol. 190, pp. S69–S86. q 2017 by The University of Chicago. tus of behavioral, ecological, archaeological, and other data in 0003-0147/2017/190S1-57348$15.00. All rights reserved. their construction; third, exemplify some of their potential DOI: 10.1086/692830 phylogenetic and evidentiary breadth with three case stu-

Strepsirrhini (lemurs, lorises, bushbabies)

Tarsiiformes (tarsiers)

Platyrrhini (New World monkeys)

Cercopithecoidea (Old World monkeys)

Hylobatidae (gibbons)

Pongidae (orangutans)

Gorillinae (gorillas) Primates

Pan Haplorhini

(chimpanzees, bonobos) Anthropoidea Catarrhini Ardipithecus ramidus

Australopithecus afarensis Hominidae Homininae

other australopithecines Hominini

Homo (humans)

Figure 1: Phylogenetic relationships (left) and classification (right) of primates, based on Wilson and Reeder (2005), Cartmill and Smith (2011), and Fleagle (2013). dies; and finally, delve into their implications for broader 1978; Gould and Lewontin 1979; Cracraft 1981) and coun- concerns. terclaims (e.g., Mayr 1983; Wanntorp 1983; Baum and Larson 1991; Larson and Losos 1996; Alcock 1998) that adaptation is poorly substantiated, overemphasized as an evolutionary force, Evolutionary Scenarios Defined, and at best difficult to study from macroevolutionary per- Condemned, and Defended spectives (also discussed in the next section); and (4) the rise I use excerpts to reconstruct the origin and early reception of of phylogenetic systematics (e.g., Eldredge and Cracraft 1980; scenario as a term in evolutionary biology because decades- Hull 1988), which beyond morphology soon encompassed old references are likely obscure to many readers, some sub- physiology, behavior, and ecology (e.g., Wanntorp 1983; Brooks sequent coverage is misleading, and the original authors’ word- 1985; Donoghue 1989; Brooks and McLennan 1991)—and ings are instructive. As it happened, this innovation occurred was itself a manifestation of the still-ongoing nineteenth- amid the following contexts: (1) claims (Atz 1970; Klopfer century historicization of biology (de Queiroz 1988; O’Hara 1973) and counterclaims (e.g., Greene and Burghardt 1978; 1988; Quinn 2016a). Greene 1994, 1999) that behavior cannot be homologized and These four threads were scarcely entwined, at least overtly, does not evolve (addressed in the next section); (2) Wilson’s in the 1970s and early 1980s. However, Gould’s and Lewon- (1975) book Sociobiology: The New Synthesis, in which the tin’s pejorative framing of adaptationism and the adaptation- chapter on humans was only quasi-phylogenetic (see pp. 550– ist program occurred in the context of their structuralist 551) and spawned a highly public controversy (e.g., Caplan perspectives (see below) and disregard for ecology, strongly 1978; Segerstråle 2000; Prindle 2009); (3) claims (Lewontin motivated as well by Sociobiology’s chapter on humans and

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:50 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Scenarios and Natural History S71 perhaps earlier concerns about ethological themes in Nazi-era tion that everything exists for a purpose). Scientists know publications (Alcock 1998; Klopfer 1999; Segerstråle 2000; that these tales are stories; unfortunately, they are pre- Gould 2002, pp. 42–43; Burkhardt 2005; Allmon 2009; All- sented in the professional literature where they are taken mon et al. 2009; Prindle 2009). Eldredge’s(1979)originalcon- too seriously and literally. Then they become “facts” and ceptualization of scenarios cited neither Wilson’s (1975) book enter the popular literature, often in such socially dubious nor his critics, but it echoed the latter’s reservations: forms as the ancestral killer ape who absolves us from re- sponsibility for our current nastiness, or as the “innate” [A] phylogenetic scenario [is]. . . a [phylogenetic] tree male dominance that justifies cultural sexism as the mark with an overlay of adaptationist narrative. . . . [A] clad- of nature. (Gould 1980, pp. xv–xvi) ogram must precede the construction of a tree . . . a tree before a scenario. . . . [T]o be debated are (1) the proce- Lewontin, Gould’s antiadaptationist coauthor, was even dures for cladogram, tree, and scenario construction and more cynical about understanding, let alone portraying, (2) whether or not the additional assumptions . . . [for the deep history of behavior, later opining, “It might be inter- the most] complex level of analysis are worth it. (Eldredge esting to know how cognition (whatever that is) arose and 1979, p. 168) spread and changed, but we cannot know. Tough luck” (Lewontin 1998, p. 130; also see Burghardt 2004). Criticisms Scenarios are inductive narratives (the best are also of adaptationism, however, were soon countered with concep- seductive) concocted to explain how some particular con- tual and methodological advances (e.g., Wanntorp et al. 1990; figuration of events . . . took place. . . . [Their] hallmark Baum and Larson 1991; Brooks and McLennan 1991; Brooks [is] . . . the analysis of the adaptive significance of evolu- 1996; Larson and Losos 1996; Alcock 1998; Autumn et al. tionary changes in size, form, and structure. . . . [They] 2002; see below), while taxon-focused biologists elaborated are engrossing and concern themselves with the applica- the notion of scenarios with nonhuman examples, emphasiz- tion of what we think we know to the real world, as pre- ing explicit phylogenetic analyses, empirical support, and served in the fossil record . . . [but they are] mostly fairy falsification of alternative versions (e.g., Greene 1983, 1992; tales . . . a maze of untestable propositions concerning se- Wake 1992; Desutter-Grandcolas 1995). The term scenario lection, function, niche utilization, and community inte- now is used routinely in diverse contexts (e.g., Sessions et al. gration, and alas, do not generally represent good science. 2016), including anthropology (e.g., Dominy 2015). (Eldredge 1979, pp. 192–193) Nonetheless, almost four decades after Eldredge defined scenarios within the context of phylogenetic systematics, [Scenarios] are easily capable of testing and refuta- these quotations from a prominent anthropologist’sbook, tion. . . . [T]he problem here is that we expect scenar- Tales of the Ex-Apes: How We Think about Human Evolution, ios, even more than trees, to be wrong in detail. . . . This illustrate contempt that I suspect is common in some circles: ad hoc changing of hypothesis content rightly infuriates strict adherents of hypothetico-deductive methodology in We can study what a feature does, and we can study how science and is the feature of scenarios that most horrifies it got there, but to ask what is it for is to decorate the them. (Eldredge 1979, p. 194) scientific question with a lot of metaphysical accessories that it just doesn’t need . . . to assume that there is a rea- A year later, Eldredge’s coauthor of punctuated equilib- son for it—a deterministic, selective regime for the fea- rium theory (Eldredge and Gould 1972), after six pages of ture; a particular optimal solution to a problem. But ac- praise for a paleontologist’s fictionalized account of Ice tually, there may be no reasons for some things, just Age humans, offered this stinging critique: naturalistic causes and uses, and a lot of random noise; life may be more like clothes than like saws. (Marks Kurtén’s [1980] novel is a more appropriate place than 2015, p. 62) the professional literature itself for discussing many of the truly scientific issues that swirl about the Neandertal- [T]hat we can do something does not mean that we Cro-Magnon debate. Evolutionary biology has been se- evolved to do it, a well-known fallacy known as adapta- verely hampered by a speculative style of argument that tionism. (Marks 2015, p. 99) records anatomy and ecology and then tries to construct historical or adaptive explanations for why this bone Yet even for such a fundamental feature as bipedality, looks like that or this creature lived here. These specu- we know far more about how it evolved than why it lations have been charitably called “scenarios”; they are evolved . . . what parts of the body changed, and how often more contemptuously, and rightly, labeled “stories” they changed. . . that’s boring, because it’s anatomy. (or “just-so stories” if they rely on the fallacious assump- What we can’t tell you is why it happened. That is his-

tory, that is origin myth, that is interesting. . . . The fact cestrally occupied by hominins (e.g., Harrison 2011; O’Brien that apes sometimes stand erect when threatening each et al. 2013; Werdelin and Lewis 2013; Domínguez-Rodrigo other . . . that humans can run longer distances than 2014). apes . . . that having your eyes higher off the ground Even Eldredge, in his original formulation, offered a guarded allows you to see farther—and many others—have been reprieve and suggestions for progress: suggested as scenarios for the evolution of bipedalism. I no longer oppose . . . scenarios (which are, after all, That is to say, it must have been good for something. the most fun). . . . As long as we understand precisely What these propositions all share is the property of futil- what we are doing . . . [and have] an adequate grasp of ’ ity. If there was an advantage to bipedalism, we can ttell the probability we are wrong and what assumptions we what it was, from our vantage point of 5 or 6 million have added along the way, there no longer seems to me “ years later. . . . We have to bracket the question Why to be any reason for anyone to tell anyone else what not ” did we become bipedal? and set it off from more empiri- to do. (Eldredge 1979, p. 169) cally based scientific discourse. (Marks 2015, pp. 117–118) [We can] improve scenarios: (1) by basing them ’ Marks s concerns are evidently imbedded in a more gen- more explicitly on [phylogenetic] trees and (2) by elimi- eral disdain for approaches that to him, as with Gould and nating some of the more purely speculative elements. . . . Lewontin (e.g., Segerstråle 2000; Allmon 2009; Prindle 2009), [A]ny explicit statement about mode or rate of selection smack of dangerous biological determinism: would be idle and fatuous. . . . Adaptation and especially natural selection have no real value in the elaboration of “ ” The pop science origin myth of human evolution ob- most scenarios and are two examples of excess conceptual serves the genetic intimacy of humans and apes, applies baggage. (Eldredge 1979, p. 193) the cultural assumption that genetic relationships are the most important relationships, and concludes our identity [Scenarios give] us ideas—lower level hypotheses we can easily be established from our ancestors. But . . . the might very well be able to test (and which we might not fact that your ancestors may have been peasants or slaves have formulated. . .)—which force us to stretch our imag- fi does not make you a peasant or slave. We nd the cultural inations. . . . They are certainly more fun to construct idea of reducing identity to ancestry to be morally repug- than a mere tree or a dry cladogram . . . as long as we nant. The reason is simple, that we are different from explicitly realize how we build scenarios and what their our ancestors, and our identity is established dialectically, status is as scientific propositions, we should continue recognizing that we are simultaneously both composed to build them, hoping to find them scientifically, as well of their DNA and yet different from them. (Marks 2015, as spiritually, uplifting and rewarding. (Eldredge 1979, p. 110) pp. 194–195) ’ fi But, of course, clothes often do have design properties, Gould s critique likewise ended with a justi cation for — their advantages contingent on particular environmental scenarios but apparently only when they are constructed and cultural conditions—try exploring the Arctic in a sunga— by evolutionary biologists with opportunities to publish and that people are not now slaves does not guarantee their novels: ancestors were not, nor that knowing they once were would Yet these stories have a role in science. They probe the not prove useful for, say, socially progressive political advocacy. range of alternatives; they channel thought into the con- Moreover, although we might never know precisely why bi- struction of testable hypotheses; they serve as tentative pedalism evolved in our lineage, in terms of particular selec- frameworks for the ordering of observations. But they tive forces, data do constrain the plausibility of alternative are stories. So why not treat them as such [by presenting evolutionary scenarios for hominins. Ardi (Ardipithecus them in fiction], get all the benefits and pleasures, and ramidus) and Lucy (Australopithecus afarensis) are exciting avoid the wrangles that arise from their usual, inappro- fossils because (assuming the former is a hominin) they con- priate placement? (Gould 1980, p. xvi) firm that our lineage, after diverging from one leading to bonobos (Pan paniscus) and chimpanzees (Pan troglodytes), I conclude that phylogenetic scenario arose under a cloud evolved bipedal skeletal attributes before we got big brains, of skepticism, owing to concerns about evidence, scientific not vice versa (White et al. 2009; Wood and Harrison 2011; status, and political misuse—the former two including allega- Fleagle 2013). And as for the alleged futility of scenarios, we tions that we never can know much about ancestral behaviors do have rich paleoecological and archaeological contexts for and ecological contexts, the latter with respect to modern interpreting primate evolution, including the inference that human affairs. Nonetheless, building on earlier discussions our lineage diversified in more open habitats than those an- (e.g., Greene 1983, 1992; Wake 1992), as well as Eldredge’s

(1979) and Gould’s (1980) own musings, the concept can ogy has since been widely abandoned as a term in evolution- be expanded as follows: Evolutionary scenarios summarize ary biology (e.g., Eldredge and Cracraft 1980; Sanderson and inferred patterns and processes by overlaying phenotypes Hufford 1996; Wake et al. 2011; Quinn 2016b; but see Ere- and their ecological correlates on phylogenetic trees and then shefsky 2012). Moreover, some researchers prefer that ho- interpreting the results in narrative format. At their best, moplasy not be defined with respect to selection, such that scenarios are factually dense, and unlike Kurtén’s (1980) pre- from this structuralist (or formalist) standpoint, processes historical fiction, they entail no unidentified speculation, let favoring or constraining independent evolution of similar alone known falsehoods; their methodological and empirical phenotypes are a matter for separate investigations and in- shortcomings are obvious, they might be rejected based on ference, rather than unsupported assumptions (e.g., Cracraft new discoveries, and their ideological pitfalls are flagged for 1981; Wake 1996; Gould 2002, pp. 1076–1081). scrutiny. They are constructed by (1) describing phenotypes Evaluating resemblances as homology or homoplasy— (including physiology and behavior), with attention, when elucidating phenotypic transformations during lineage diver- possible and applicable, to formative roles of developmental sification—initially entailed parsimonious congruence of char- genetics, experience, and culture; (2) distinguishing homol- acter state data in the context of a cladogram (e.g., Eldredge ogy from homoplasy in the context of a phylogenetic tree, and Cracraft 1980; Hall 1994; Sanderson and Hufford 1996); such that ancestral phenotypic states and subsequent trans- ancestral-state reconstruction now also relies on likelihood formations can be inferred; and (3) integrating the forgoing analyses and hypothesis-testing frameworks, statistical aspects components with ancillary data (e.g., from paleoecology, ar- of which will continue to evolve (e.g., Griffiths et al. 2015). chaeology) in a manner that is transparent in terms of facts, Regardless of analytic techniques, a core task is delimitation uncertainties, and ideological implications. of characters and, ideally, their generative mechanisms, such As such, evolutionary scenarios can range from explicitly that when possible we analyze the phylogeny of ontogenies speculative (e.g., canid-human relationships; Treves and Bo- (de Queiroz 1985)—thus encompassing developmental geno- nancic 2016) to extensively corroborated, from a few para- mics, individual experience, phenotypic plasticity, and, when graphs in length (e.g., alcohol consumption by primates; Do- applicable, culture (e.g., Greene 1994; Ereshefsky 2007; Mc- miny 2015) to monographic, at which point they may be Cune and Schimenti 2012; Foster 2013; Hall 2013; for the labeled theories (e.g., play: Burghardt 2005; snakes and pri- ontogeny of prehensile-tail use in two platyrrhines, see Bezan- mates: Isbell 2009; fire and cooking: Wrangham 2009; mother- son 2012; of locomotion and postures in wild chimpanzees, hood: Hrdy 2011; aggression and hunting: Pickering 2013; see Sarringhaus et al. 2014). Another core challenge is ade- weaponry and battle: Emlen 2014). I also reckon that care- quatephylogeneticsampling,such that component lineages fully constructed and skeptically regarded, scenarios can play are neither under- nor overrepresented; in particular, we should positive roles in anthropology, biology, and society more be cautious of incorporating only one member (e.g., chim- generally—whether or not they are regarded as scientific, panzees) of a multispecies sister group as representative of testable, or falsifiable (e.g., Eldredge 1979; Cartmill 2002; Fitz- character states for the ancestor of both lineages (e.g., Bate- hugh 2016) and despite risks of political pitfalls (e.g., Gould man 1996; for hominids, Cartmill and Smith 2011, p. 104; 1980; Klopfer 1999; Burkhardt 2005; Marks 2015). As for Wood and Harrison 2011; Pickering 2014). careful construction and skeptical reception, important con- An important topic of ongoing exploration, conceptually cerns include the nature of resemblances, windows on an- and analytically, is phylogenetic analyses of culture and its cient behavior and ecology, and macroevolutionary perspec- tangible indicators (material culture; e.g., Gray et al 2007; Tëm- tives on adaptation. kin and Eldredge 2007; Collard et al. 2007, 2008, 2011; O’Brien et al. 2012), as well as placing that research within broader frameworks of gene-culture evolution (e.g., Richerson and Elaborating on Three Core Issues Boyd 2005). Of course, individual and cultural experience shape much of primate behavior, complicating comparative Resemblance, Homology, and Homoplasy assessments—toothbrushes are recent inventions, so, parsi- Homology, granting some important conceptual nuances, is mony notwithstanding, observations of geographically dispa- resemblance due to common ancestry (e.g., Hall 1994; Mc- rate and ethnically diverse people cleaning their teeth with sim- Cune and Schimenti 2012). In older literature, it was con- ilar motor patterns need not imply homologous resemblance trasted with convergence, the latter earlier called analogy, to the behavior of ancestral Homo sapiens 1200,000 years and often defined as independently evolved similarities due ago. On the other hand, tooth picking with twigs and floss- to similar functional or ecological roles. The rise of phyloge- ing with hair appear sporadically among Old and New World netic systematics included preference for homoplasy as an nonhuman anthropoids (McGrew and Tutin 1973; Watanabe overarching term for independently evolved resemblances, et al. 2007; Haslam and Falótico 2015), thus ancient procliv- whether due to convergence, parallelism, or reversals; anal- ities for dental tool use might underlie modern oral hygiene

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:50 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S74 The American Naturalist practices. And in any case, the emergence of nonhuman ar- ior of extinct animals (Seilacher 2007). Classic examples of chaeology raises prospects that we might someday study tran- the former include Jurassic social dinosaur trackways (Ostrom sitional material cultures among primate lineages (Haslam et al. 1972) and the 3.7-million-year-old Laetoli footprints, evi- 2009; see also below). dently made by two adult Australopithecus afarensis with a child in tow (Leakey 1981; White and Suwa 1987). Anatom- ical indicators of ancient behavior are exemplified by associ- Behavior and Ecology as Phenotypic ation of an enlarged gyrus in 10-million-year-old otter brain Attributes with Histories? endocasts with that structure and tactile hand sensitivity in Behavior is what animals do and do not do, visibly and oth- extant lutrines (Radinsky 1968); of bipedal skeletal features erwise; besides external manifestations (e.g., postures, move- among hominins (Fleagle 2013, pp. 364–365); and of innova- ments, vocalizations, immobility), it can include surveillance, tion, social learning, and tool use with brain size among cognition, emotions, and other private experiences (Burg- primates (Reader and Laland 2002). So-called fossilized be- hardt 1997; see below). As such, behavioral systems reflect havior deserves skeptical appraisal, however, because animals neurosensory, neuromuscular, neurohormonal, integrative, usually do not die during routine activities—granted, the pos- and mental processes, themselves influenced—as are mor- ture of a Cretaceous Oviraptor dinosaur over eggs so closely phology and physiology—in complex ways by genetics, de- resembles that of incubating modern birds as likely to reflect velopment, experience, and, in some species, culture (Burg- homology (Norell et al. 1995); on the other hand, a fossil early hardt and Bowers 2017). Although early ethologists treated amniote with a congener in its throat, rather than indicative motor patterns comparatively in the same way as anatomy of cannibalism, probably choked attempting atypical tail-first (e.g., Lorenz 1941; Daanje 1950), Atz (1970) asserted that ingestion (Greene 1994). Other records of ancient behavior even those overt, often stereotyped behaviors are too difficult come from prehistoric artist-ethologists (e.g., Guthrie 2005; to describe, too variable, too subject to selection and conver- Greene 2013, p. 236), evidenced by 130,000-year-old repre- gence, and unsubstantiated by fossils; thus, he argued, behav- sentations of courtship in cave lions (Panthera atrox; Yama- ioral homologies are at best difficult to detect and should not guchi et al. 2004), and for hominines, archaeological sources be expected above the level of genera. In fact, none of these (for an extended example, see Pickering 2013). criticisms can be generalized, unsupported claims to the con- With respect to activities that are internal and not ob- trary notwithstanding (e.g., Klopfer 1999, pp. 124–125), and servable directly, recall that Tinbergen (1963) codified four well-corroborated behavioral homologies, judged by the same questions for ethologists: How does behavior develop, how sorts of evidence used for morphology and physiology, char- is it controlled, what is its function or ecological role, and what acterize even some large, ancient clades (e.g., Greene and is its evolutionary history? In posing a fifth question—What Burghardt 1978; Wenzel 1992; de Queiroz and Wimberger are the private experiences of animals?—Burghardt (1997) 1993; Greene 1994, 1999; Proctor 1996; Ereshefsky 2007). called for investigations of cognition, emotion, intention, con- The most comprehensive source for inferences about an- sciousness, awareness, and other internal behavior within Tin- cient behavior likely always will be phylogenetic analyses and bergen’s framework. As such, private experiences might be ancestral-state reconstructions, based on comparison among studied across taxa by combining critical anthropomorphism living taxa (e.g., oviposition in frogs: Zamudio et al. 2016; so- (Burghardt 1991, 2016) with laboratory (e.g., Brosnan and de cial systems in primates: Di Fiore and Rendall 1994; Shultz Waal 2003) and field (e.g., Holekamp et al. 2007) experi- et al. 2011) and subject to the same limitations and inno- ments, as well as with brain imaging, genomics, and other vations as for other phenotypic attributes (e.g., Cooper et al. emerging technologies (Preuss 2012). Private experiences 2016; for hominins, see, e.g., Wood and Harrison 2011). As thereby can be subjected to phylogenetic analysis, and indeed with morphology and physiology, character delineation, anal- de Waal (2016a) recently has summarized a career’s worth of yses of variation within and among taxa, and phylogenetic primate research in that context. Core problems, once again, sampling breadth are core challenges—namely, detecting reg- include adequate phylogenetic sampling and defining char- ularities and modularity in behavior, avoiding functionally acters (e.g., for insight, see Shettleworth 2012; for psycho- and ideologically laden terminology, and redressing the scar- logical homologies, see Ereshefsky 2007)—the latter perhaps city for many taxa of detailed, high-quality descriptions (e.g., entailing trade-offs between, on the one hand, stringent cri- Drummond 1981; Gowaty 1981; Proctor 1996; Greene 1999, teria and “killjoy explanations” (Shettleworth 2010), and 2005; Scholes 2008). Important related issues include sources on the other, more open-minded skepticism and optimistic of information for ancient and internal behavior, and the sta- prospects for detecting trans-species mental homologies (e.g., tus of ecological data in phylogenetic analyses and scenario Cartmill 2000; Rivas and Burghardt 2002). construction. Ecological correlates of behavior, whether biotic (e.g., pred- Trace fossils and morphological correlates based on living ators, competitors) or abiotic (e.g., climate, soil type), are taxa also provide relatively common evidence for the behav- not themselves attributes of organisms, passed from genera-

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:50 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Scenarios and Natural History S75 tion to generation. They thus arguably should not be ana- Third, irrespective of whether we infer that selection shaped lyzed phylogenetically (Grandcolas et al. 2011), although pro- ancient adaptations, widespread correlations of homoplasies pensities to choose particular prey and microhabitats, for with shifts in functional and ecological contexts imply law- example, might be transmitted genetically and/or culturally, like evolutionary patterns (e.g., Brooks 1996; Losos 2011; Ag- such that they can be treated as behavioral traits (e.g., rawal 2017)—the expensive tissue hypothesis, for example, Burghardt 1967). Likewise, a phylogenetic analysis of mam- first proposed for hominins, is supported by a correlation be- mals in which incidence of lethal violence, a population-level tween large brains and small guts in amphibians (Liao et al. trait, served as proxy for aggression (Gómez et al. 2016) might 2016). Moreover, phylogenetic analyses can reveal whether prove controversial, the more so because that word is subject novel forms, functions, and ecological roles arose simulta- to diverse, sometimes ambiguous usage. In any case, avail- neously or sequentially, thus distinguishing between adap- ability of ecological data for interpreting human evolution tive and exaptive explanations (Gould and Vrba 1982; Greene continues to grow, encompassing not only more traditional 1986). Conversely, absence of homoplasy might reflect lineage- archaeological and paleontological methods (e.g., Harrison wide constraints (e.g., life history and socioecological traits 2011; Pickering 2013) but also stable isotope analyses and of primates; Kappeler and Heymann 1996). And although other technological innovations (e.g., Cerling et al. 2013; Cartmill (2002) argued that prospects are slim for homo- Yeaves et al. 2013; Katoh et al. 2016). plasies revealing much about humans, some astonishing windows on behavioral evolution in primates have opened during the past decade, as illustrated below. Adaptationist “Stories” as Macroevolutionary Hypotheses Even today some might argue, paraphrasing Brown (1982, p. 886), that most evolutionary biologists are interested in Evolutionary Scenarios and (Mainly) Primate Behavior natural populations instead of fossils, in variation among in- Percussive Tools and Extractive Foraging dividuals, rather than among species and higher taxa. I believe instead that many anthropologists and biologists, as well Bonobos and chimpanzees compose our living sister group as members of the public, are fascinated by big questions like within hominine catarrhines, and their use of tools has been those that began this essay and, consequently, will be inter- increasingly appreciated over the past half century (e.g., ested in evolutionary scenarios. A key issue, then, is just what Goodall 1964; McGrew 1992). Recent discoveries, however, can we infer about adaptations as historical phenomena, as reveal that although bonobos rarely use stone hammers and opposed to reflecting natural selection within populations to- anvils for extractive foraging (Roffman et al. 2015), chimps day? Three points bear emphasis. show complexity and cultural variation in that behavior (e.g., First, behavioral character states can be characteristic of Haslam et al. 2009; Lycett et al. 2009). Moreover, percussive species and higher taxa, rather than variable within popula- tool use also occurs in a cercopithecoid catarrhine, the Bur- tions (e.g., Greene 1994, 1999; Proctor 1996); they thus can mese island subspecies of long-tailed macaque (Macaca be corroborated as homologous or homoplastic at deeper fascicularis aurea), which feeds on nuts and shellfish (e.g., levels in evolutionary history and used to construct scenarios Carpenter 1887; Gumert et al. 2009), and in a platyrrhine, (e.g., Greene 1986, 1992; Coddington 1988; for social behav- the bearded capuchin (Sapajus libidinosus), which harvests ior in primates, see Shultz et al. 2011). Moreover, as with fruits and nuts (fig. 2A; e.g., Fragaszy et al. 2004; Falótico other phenotypic attributes, hypotheses of homology and and Ottoni 2016; Luncz et al. 2016). Similar behavior thus homoplasy are falsifiable if their constituent claims prove has originated four times among primates, including within inconsistent with new evidence from paleontology, archaeol- hominins; most intriguingly, gracile (Cebus) and robust (Sa- ogy, and other ancillary sources (e.g., Greene 1986; Larson pajus) capuchins diverged several million years ago, and the and Losos 1996; Wood and Harrison 2011). latter sometimes inhabit savanna woodlands, have enlarged Second, inferring natural selection from phylogenetic pat- brains, and excel at tool use, reminiscent of the Pan-Homo terns depends on evidence for its general prevalence as a split (Haslam et al. 2009; Lynch Alfaro et al. 2012). Each of force in evolution and resulting assumptions about past the three nonhuman species is subject to ongoing research, population-level processes. Some evolutionary biologists be- yielding rich comparative details about ontogeny, control lieve selection must have caused most if not all phyloge- mechanisms, ecological roles, and cultural transmission of netic correlations between phenotypes and ecological circum- percussive tool use (e.g., Visalberghi et al. 2015; Cardoso and stances (e.g., Mayr 1983; Coddington 1988; Baum and Larson Ottoni 2016). 1991; Losos 2011), while others are more minimalistic (e.g., Material culture now confirms percussive tools for an un- Cracraft 1981; Greene 1986; Autumn et al. 2002)—in any identified (likely pre-Homo) hominin 3.3 million years ago case, selection evidently underlies numerous well-studied ex- (Harmand et al. 2015), chimps ∼4,300 years ago (Mercador amples of homoplasy (e.g., Losos 2011; Langin et al. 2015). et al. 2007), Burmese macaques 1100 years ago (Haslam et al.

A This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:50AM B

Figure 2: A, Bearded capuchin with stone hammer, poised to crack a nut. B, Nonhuman primate archaeological site in Brazil, with stone hammer in situ (arrow), used 1600 years ago by a bearded capuchin. Photo credits: P. Oxford (A); M. Haslam (B). Scenarios and Natural History S77

2016a), and bearded capuchins 1600 years ago (fig. 2B;Has- prey, lesser bushbabies (Galago senegalensis), in tree cavities lam et al. 2016b). A deeper, denser archaeological record might (fig. 3B; Roach 2008; Pruetz et al. 2015). Also intriguingly, reduce independent origins to three if the common ancestor bearded capuchins break off and thin the tips of sticks, with of Pan and Homo used percussive tools (Hoevers 2015), which they probe for honey and small prey animals, as well to two if ancestral catarrhines were implicated, or even—as as jab at dangerous snakes and their mimics (e.g., a false jara- seems unlikely given the phylogenetic distance among those raca pit viper [Xenodon merremi]; fig. 4; Falótico and Ottoni taxa—to one if percussive tools characterized the ancestor 2014). of catarrhines and platyrrhines. Moreover, developmental Once again, character descriptions and nomenclature are studies of that behavior might reveal homologous precur- central to phylogenetic analyses and interpretation, and ac- sors across anthropoids or an even larger clade (e.g., stone cording to Roach (2008) and Pickering (2013), some other handling: Huffman et al. 2010; object manipulation and/or researchers were critical of Pruetz and Bertolani’s (2007) cognitive abilities: Hoevers 2015), such that percussive tool calling the savanna-woodland chimps’ modified tools spears use could have arisen as multiple parallelisms, perhaps each and their behavior hunting (perhaps because bushbabies of- time encouraged by particular ecological conditions (e.g., ten are killed by females, and prey are small compared to the terrestriality; Meulman et al. 2012). colobus monkeys [Piliocolobus sp.] taken by forest-dwelling In any case, when bearded capuchins pound stones to- chimps). Like Roach (2008) and Pickering (2013), I regard gether, perhaps for nutritional powder, they produce sharp- both words, employed conventionally, as appropriate for the edged flakes that are indistinguishable from early hominin weapons made by early humans and by chimps, whereas the tools (Proffitt et al. 2016). This discovery suggests that stones thinned sticks with which bearded capuchins poke (but evi- inadvertently flaked by nonhuman primates might be erro- dently do not pierce) snakes can be termed protolances. Other neously interpreted as tools produced by hominins, as well plausible antecedents of defensive weapons in hominins, de- as that early hominins could have adopted similar incidental serving of further study in that context (see Crabb and Elizaga flakes as prototools, including projectiles, in a foraging con- 2008), include unfashioned stick clubs and projectiles used text. Also, because large brains and problem solving are phylo- by free-living white-faced capuchins (Cebus capucinus)against genetically correlated among primates and carnivores (Reader a terciopelo pit viper (Bothrops asper; Boinski 1988) and by and Laland 2002; Benson-Amram et al. 2016), similar under- semifree tufted capuchins (Sapajus sp.) against a leopard tor- lying cognitive innovations might be more broadly homoplas- toise (Stigmochelys pardalis; Hamilton and Fragaszy 2014). tic among mammals, including for percussive tool use (e.g., Spear making has arisen repeatedly in primate history, for sea otters; Fujii et al. 2015). yielding lances, javelins, and hafted projectiles within Homo; lances in savanna-woodland chimps; and, arguably, protolances in bearded capuchins. In all three instances, those weapons increase the distance between primates and prey or Lances and Javelins otherwise dangerous adversaries, diminishing the risk of fatal Probing with sticks (e.g., for termites) and aiming projectiles retaliation and fostering emotional distance (Pickering 2013). (e.g., stones) are widespread practices among anthropoids, Like human hunters, savanna-woodland chimps appear in- but fashioned weapons (modified to inflict bodily injury) were tently composed while trying to skewer their victims, whereas long believed restricted to our lineage, as documented by a rich forest chimps are highly agitated while hand-capturing mon- archaeological and historical record (e.g., Pickering 2013; keys—a difference used to argue that hunting and aggression Emlen 2014). Spears are long-shafted, penetrating weapons, were decoupled early in hominin evolution (Pickering 2013). of which the most ancient known for hominins are ∼500,000- Furthermore, also consistent with the savanna-woodland hy- year-old hafted, stone-tipped lances (thrusting spears; Wilkins pothesis of human ancestry (e.g., Pickering and Domínguez- et al. 2012); a 400,000-year-old wooden spear tip (delivery Rodrigo 2010; Fleagle 2013; Pickering 2014), chimps in that mode unknown; Schoch et al. 2015); 300,000-year-old wooden habitat have converged with early Homo by taking shelter lances and javelins (light throwing spears; fig. 3A; Thieme 1997; in caves, behaving as if familiar with fire, and at least to my Schoch et al. 2015); and 280,000-year-old hafted, stone- eye, being more sparsely haired than forest-dwelling individ- tipped throwing spears (Sahle et al. 2013). The ancient jav- uals of their species (Pruetz and LaDuke 2010; see images in elins are remarkable for having been made with stone tools Roach 2008). from trunks of small spruce (Picea sp.) and pine (Pinus sylvestris) trees, with their shaft and tips carved offset from Private Experiences the softer central axis of the wood and then polished (Schoch et al. 2015). Nonetheless, savanna-woodland chimpanzees Internal activities are increasingly subject to phylogenetic in Senegal break off selected saplings, roughly sharpen an perspectives, involving considerations of, among other be- end with their teeth, and use those lances to puncture their haviors, consciousness (e.g., Cartmill 2000), cognition (e.g.,

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:50 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). A

Figure 3: A, Wooden spear and horse bones from Germany, dated at ∼300,000 years ago; note pointed tip (arrow, upper right) and thicker area (arrow, center) typical of javelins. B, Field sample of lances fashioned by chimpanzees for hunting bushbabies in Senegal. Photo credits: P. Pfarr, State Agency for the Heritage of Lower Saxony (A); J. Pruetz (B).

Figure 4: Bearded capuchin jabbing with protolance (upper arrow) at plausibly dangerous pit viper mimic (lower arrow) in a rock crevice (from video). Photo credit: T. Falótico.

MacLean et al. 2012), emotion (e.g., Clark 2013), and musi- Are? (de Waal 2016a) mentions homology; that word and cality (Honig and Ploeger 2012) in diverse animals; of social homoplasy are lacking from the indices and, as far as I can intelligence among primates and carnivores (Reader and Laland tell, otherwise unmentioned in Animal Wise: How We Know 2002; Holekamp et al. 2007; Benson-Amram et al. 2016); of Animals Think and Feel (Morrell 2013), How Animals Grieve fairness, empathy, and other putative precursors of moral- (King 2013), and Beyond Words: What Animals Think and ity among anthropoids (Brosnan and de Waal 2003; de Waal Feel (Saﬁna 2015). Whether thinking, grieving, and feeling 2016a,2016b); and of psychological attributes more gener- might not be the same behaviors across species as phyloge- ally (Ereshefsky 2007). Although such studies typically ad- netically disparate as whales, elephants, and primates—and dress private experiences with reference to homology and ho- thus, perhaps, not have shared a common evolutionary der- moplasy, central to understanding behavioral evolution, that ivation—is variously explicit (e.g., King 2013, pp. 7–8) or not perspective is not yet true of popular writing on the subject. in the other volumes. And although those authors are moti- Among four recent widely and favorably reviewed books, vated by conservation and animal welfare concerns (as am I), only Are We Smart Enough to Know How Smart Animals they do not grapple in detail with how whether private expe-

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:50 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S80 The American Naturalist riences in other species are homologous or homoplastic with is better or worse as a model for thinking about ourselves? our own might affect ethical debates on those topics (see, e.g., (For a critical overview favoring chimps, see Pickering and Cartmill 2000). Domínguez-Rodrigo 2010.) In fact, robustly inferring ances- Beyond matters of definition and historical explanation, tral states for hominins could require observations of addi- robust evolutionary scenarios for private experiences will re- tional taxa, including extinct species, and for some attributes quire denser and broader sampling than currently available. might never be possible (Wood and Harrison 2011). Mini- In terms of denser representation among relatively basal pri- mizing political problems surely will follow, at least in part, mates, given logistical and ethical restrictions on captive re- from transparency in scenario presentation with regard to search, field experiments might more likely yield new in- shaky underpinnings, as well as from admissions of bias by sights (e.g., for strepsirrhines: Huebner and Fichtel 2015; for critics and proponents of diverse viewpoints (for extended tarsiers: Gursky 2005; for a cercopithecoid: Isbell and Etting discussions, see Borofsky 2004; Prindle 2009; Dreger 2011). 2017). As for phylogenetic breadth, many of the species dis- As even Eldredge (1979) and Gould (1980) acknowledged, cussed by King (2013), Morrell (2013), Safina (2015), and and as exemplified by the above examples, scenarios can play de Waal (2016a) are mammals, and with more distant taxa positive roles in anthropology, biology, and society more it becomes more difficult to imagine their inner worlds, let generally. By highlighting homologies and homoplasies in alone experimentally address them. However, if we could ac- the diversification of clades, we focus on the controversial cess the private experiences of eyelid-less, limbless creatures extent to which adaptation, constraint, stasis, and repeated like snakes—“in which we detect no joy and no emotion” evolution of sameness have characterized life’s history (e.g., (Skutch 1980, p. 257)—that effort might inspire new ways Brooks 1996; Kappeler and Heymann 1996; Wake 1996; to illuminate more familiar and accessible species (Burghardt Losos 2011; Wake et al. 2011; Edwards and Donoghue 2013; 1977, 1991). Bridgham 2016; Agrawal 2017). Moreover, by laying out detailed case studies of Darwin’s descent with modification, scenarios can inspire public understanding of evolution and Discussion: Now What? our long-term roles in the current extinction crisis, as well Almost four decades after Eldredge (1979) conceptualized as inform management strategies and appreciation for bio- evolutionary scenarios, I judge that their shortcomings have diversity (e.g., Brooks and McLennan 2010; Greene 2013, been substantially overstated. With respect to the deep his- pp. 183–187; Werdelin 2013; Cavender-Bares 2016; Sullivan tory of primates, including of our own species, we have an et al. 2017)—we might well regard island macaques, bearded ever-growing empirical basis for studying the evolution of capuchins, and savanna-woodland chimps as especially wor- morphology, physiology, and behavior in the context of ecol- thy of conservation, given their homoplastic resemblances to ogy and archaeology—but then again, back in the 1970s and hominins, the possibility of deeper homologies shared among 1980s, who would have imagined material culture for non- us, and thus the contributions of those other primates to illu- human primates dating back more than four millennia or minating our own past. Likewise, evolutionary scenarios for that savanna-woodland chimps fashion crude spears? Now, a sense of fairness, grief, and other private experiences, by as a generalization and granting further refinements of ana- clarifying the nature and origins of those phenomena, could lytic methods, scenario construction seems unproblematic— enrich our deliberations about ethical treatment of other limited primarily by lack of phylogenies and natural history species. details for many taxa, and, for better or worse, by restrictions With that last point in mind and a nod to de Waal’s (1999) on invasive neurobiological studies of primates (Preuss 2012). notion of anthropodenial—“blindness to the human-like char- Nevertheless, concerns that evolutionary scenarios can be acteristics of animals, or the animal-like characteristics of used to support ideological and political agendas sometimes ourselves” (p. 258)—I will close with two koan-like tensions. are justified (e.g., Klopfer 1999; Marks 2015). As narratives, First, some anthropologists and biologists, myself included, scenarios indeed are vulnerable to distortions and self-serving strive to confirm private experiences similar to our own in rhetoric, including by critics; they also are susceptible to in- other species, all the while yearning to justify treating them terpretive excesses and lapses (for verbatim examples, see better (e.g., Greene et al. 2002, pp. 198–199; King 2013; Mor- Prindle 2009, pp. 135–138), as well as faulty interpretation rell 2013; Safina 2015; de Waal 2016a). Other scholars (and because of obscure terminology and concepts. Christina Cau- perhaps some of the same ones) insist on human exception- terucci’s (2016) blog post asks, for example, “Why do we idol- alism and minimize evolution’s relevance for knowing our- ize chimps when we could be imitating female bonobos?”— selves (e.g., Lewontin 1998; Klopfer 1999; Marks 2015). It but even setting aside why we might aspire to a close relative’s is as if we could be inspired by two metrics, yet ignore their social system (e.g., Cartmill 2002), how many lay readers will fundamental irreconcilability—one whereby nonhuman ani- understand that we are equally related to bonobos and chimps, mals resemble us ever more closely with each new discovery, such that as a generalization and all else equal, neither species the other with which, no matter those revelations, we always

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:50 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Scenarios and Natural History S81 must be profoundly unlike even bonobos and chimps. Sec- Allmon, W. D. 2009. The structure of Gould: happenstance, human- ond, having spent a career advocating for snakes, I long have ism, history, and the unity of his view of life. Pages 3–68 in W. D. regarded their persecution as ignorant, gratuitous, and shame- Allmon, P. H. Kelley, and R. M. Ross, eds. Stephen Jay Gould: reflections on his view of life. Oxford University Press, Oxford. ful—especially by the likes of Albert Schweitzer (who fa- “ ” Allmon, W. D., P. J. Morris, and L. C. Ivany. 2009. A tree grows mously revered all life ) and Alexander Skutch (an acclaimed in Queens: Stephen Jay Gould and ecology. Pages 147–170 in ornithologist), both of whom routinely killed those reptiles W. D. Allmon, P. H. Kelley, and R. M. Ross, eds. Stephen Jay Gould: (Greene 2013, pp. 125, 181). Now, though, I ponder such neg- reflections on his view of life. Oxford University Press, Oxford. ative biases, my own motivations, and prospects for con- Atz, J. 1970. The application of the idea of homology to behavior. serving serpents in the light of some 75 million years of mor- Pages 53–74 in L. R. Aronson, E. Tobach, D. S. Lehrman, and J. S. tal conflict between our two lineages (Isbell 2009; Headland Rosenblatt, eds. Development and evolution of behavior. Freeman, San Francisco. and Greene 2011). Well-corroborated scenarios of primate Autumn, K., M. J. Ryan, and D. B. Wake. 2002. Integrating historical evolution, however uncomfortable to contemplate, will likely and mechanistic biology enhances the study of adaptation. Quarterly play important roles in clarifying these and other contempo- Review of Biology 77:383–408. rary dilemmas. Bateman, R. M. 1996. Nonfloral homoplasy and evolutionary scenarios in living and fossil land plants. Pages 91–130 in M. J. Sanderson and Note Added in Proof L. Hufford, eds. Homoplasy: the recurrence of similarity in evolution. Academic Press, San Diego, CA. While this article was in press, I encountered Tattersall and Baum, D. A., and A. Larson. 1991. Adaptation reviewed: a phylogenetic Eldredge’s (1977) earlier formulation of scenario in a phylo- methodology for studying character macroevolution. Systematic Zo- – genetic context; although not cited by Eldredge (1979), my ology 40:1 18. fl Benson-Amram, S., B. Dantzer, G. Stricker, E. M. Swanson, and K. E. quotes herein from the latter reference faithfully re ect views Holekamp. 2016. Brain size predicts problem-solving abilities in advanced in the former. mammalian carnivores. Proceedings of the National Academy of Sciences of the USA 113:2532–2537. Acknowledgments Bezanson, M. 2012. The ontogeny of prehensile-tail use in Cebus capucinus and Alouatta palliata. American Journal of Primatology The American Naturalist was the journal I most eagerly read 74:770–782. as a graduate student and most aspired to publish in as a sole Boinski, S. 1988. Use of a club by a white-faced capuchin (Cebus ca- author, and I am grateful to A. Agrawal for organizing the pucinus) to attack a venomous snake (Bothrops asper). American – symposium that finally led to realizing that goal. I also thank Journal of Primatology 14:177 179. Borgerhoff Mulder, M., and B. A. Beheim. 2011. Understanding the G. M. Burghardt, my PhD advisor, for more than 40 years of nature of wealth and its effects on human fitness. Philosophical mentorship; D. B. Wake and M. H. Wake for their friend- Transactions of the Royal Society B 366:344–356. ship, scholarly examples, and support at Berkeley; W. D. Borofsky, R., ed. 2004. Yanomami: the fierce controversy. University Allmon, W. Bemis, M. Bezanson, C. J. Campbell, D. Cundall, of California Press, Berkeley. K. de Queiroz, F. B. de Waal, H. Drummond, T. Falótico, Bridgham, J. T. 2016. Predicting the basis of convergent evolution. T. N. Headland, K. R. Hill, D. M. Hillis, K. E. Holekamp, Science 354:289. L. A. Isbell, R. E. Keen, B. J. King, J. B. Losos, K. C. Brooks, D. R. 1985. Historical ecology: a new approach to studying the evolution of ecological associations. Annals of the Missouri Botan- MacKinnon, J. Marks, G. Mayer, A. R. McCune, G. B. Pauly, ical Garden 72:660–680. T. R. Pickering, J. Pruetz, A. Quinn, J. P. Sparks, P. Weldon, ———. 1996. Explanations of homoplasy at different levels of bio- A. D. Yoder, and K. R. Zamudio for feedback and/or inspira- logical organization. Pages 3–36 in M. J. Sanderson and L. Hufford, tion; E. Behrens, R. Bish, T. Falótico, M. Haslam, G. Legant, eds. Homoplasy: the recurrence of similarity in evolution. Academic N. A. Mason, A. Meyer, R. C. Nali, P. Oxford, J. Pruetz, and Press, San Diego, CA. T. Terberger for assistance with figures; an anonymous re- Brooks, D. R., and D. A. McLennan. 1991. Phylogeny, ecology, and viewer for cogent criticisms; and the National Science Foun- behavior: a research program in comparative biology. University of Chicago Press, Chicago. dation for support from Opportunities for Promoting Un- ———. 2010. The biodiversity crisis: lessons from phylogenetic sagas. derstanding through Synthesis grant 1354156. Evolution Education and Outreach 3:558–562. Brosnan, S. F., and F. B. M. deWaal. 2003. Monkeys reject unequal Literature Cited pay. Nature 425:297–299. Brown, J. L. 1982. The adaptationist program. Science 217:884, 886. Agrawal, A. A. 2017. Toward a predictive framework for convergent Burghardt, G. M. 1967. Chemical-cue prey preferences of inexperi- evolution: integrating natural history, genetic mechanisms, and enced snakes: comparative aspects. Science 157:718–721. consequences for the diversity of life. American Naturalist 190 ———. 1977. Of iguanas and dinosaurs: social behavior and com- (suppl.):S1–S12. munication in neonate reptiles. American Zoologist 17:177–190. Alcock, J. 1998. Unpunctuated equilibrium in the natural history essays ———. 1991. Cognitive ethology and critical anthropomorphism: a of Stephen Jay Gould. Evolution and Human Behavior 19:321–336. snake with two heads and hognose snakes that play dead. Pages 53–

90 in C. A. Ristau, ed. Cognitive ethology: the minds of other ani- eds. Cultural transmission and material culture: breaking down bound- mals. Lawrence Erlbaum, Hillsdale, NJ. aries. University of Arizona Press, Tucson. ———. 1997. Amending Tinbergen: a fifth aim for ethology. Pages Collard, M., S. J. Shennan, B. Buchanan, and R. A. Bentley. 2007. Evo- 254–276 in R. W. Mitchell, N. S. Thompson, and H. L. Miles, eds. lutionary biological methods and cultural data. Pages 203–223 in Anthropomorphism, anecdotes, and animals. State University of R. A. Bentley, H. D. G. Maschner, and C. Chippindale, eds. Hand- New York Press, Albany. book of archaeological theories. Altamira, Lanham, MD. ———. 2004. Play: how evolution can explain the most mysterious Cooper, N., G. H. Thomas, and R. G. FitzJohn. 2016. Shedding light on behavior of all. Pages 231–246 in A. Moya and E. Font, eds. Evo- the “dark side” of phylogenetic comparative methods. Methods in lution: from molecules to ecosystems. Oxford University Press, Ox- Ecology and Evolution 7:693–699. ford. Crabb, P. B., and A. Elizaga. 2008. The adaptive value of tool-aided de- ———. 2005. The genesis of animal play: testing the limits. MIT Press, fense against wild animal attacks. Aggressive Behavior 34:633–638. Cambridge, MA. Cracraft, J. 1981. The use of functional and adaptive criteria in phy- ———. 2016. Mediating claims through critical anthropomor- logenetic systematics. American Zoologist 21:21–36. phism: commentary on Key on fish pain. Animal Sentience 2016.024: Daanje, 1950. On locomotor movements in birds and the intention 1–4. movements and behavior derived from them. Behaviour 3:49–98. Burghardt, G. M., and R. I. Bowers. 2017. From instinct to behavior sys- de Queiroz, A., and P. H. Wimberger. 1993. The usefulness of behav- tems: an integrated approach to ethological psychology. Pages 333– ior for phylogeny estimation: levels of homoplasy in behavioral and 364 in J. Call, ed. APA handbook of comparative psychology. Vol 1. morphological characters. Evolution 47:46–60. Basic concepts, methods, neural substrate, and behavior. American de Queiroz, K. 1985. The ontogenetic method for determining char- Psychological Association, Washington, DC. acter polarity and its relevance to phylogenetic systematics. System- Burkhardt, R. W., Jr. 2005. Patterns of behavior: Konrad Lorenz, Niko atic Zoology 34:280–299. Tinbergen, and the founding of ethology. University of Chicago ———. 1988. Systematics and the Darwinian revolution. Philosophy Press, Chicago. of Science 55:238–259. Caplan, A. L., ed. 1978. The sociobiology debate: readings on the eth- Desutter-Grandcolas, L. 1995. Toward the knowledge of the evolution- ical and scientific issues concerning sociobiology. Harper and Row, ary biology of phalangopsid crickets (Orthoptera: Grylloidea: Phal- New York. angopsidae): data, questions, and evolutionary scenarios. Journal of Cardoso, R. M., and E. B. Ottoni. 2016. The effects of tradition on prob- Orthopteran Research 1995(4):163–175. lem solving by two wild populations of bearded capuchin monkeys in de Waal, F. 2016a. Are we smart enough to know how smart animals a probing task. Biology Letters 12:20160604. are? W. W. Norton, New York. Carpenter, A. 1887. Monkey opening oyster. Nature (May 19):53. ———.2016b. Apes know what others believe. Science 254:39–40. Cartmill, M. 2000. Animal consciousness: some philosophical, method- de Waal, F. B. M. 1999. Anthropomorphism and anthropodenial: con- ological, and evolutionary problems. American Zoologist 40:835– sistency in our thinking about humans and other animals. Philosoph- 846. ical Topics 27:255–280. ———. 2002. Paleoanthropology: science or mythological charter. Di Fiore, A., and D. Rendall. 1994. Evolution of social organization: a Journal of Anthropological Research 58:183–201. reappraisal for primates by using phylogenetic methods. Proceedings Cartmill, M., and F. H. Smith. 2011. The human lineage. Wiley Black- of the National Academy of Sciences of the USA 91:9941–9945. well, New York. Domínguez-Rodrigo, M. 2014. Is the “savannah hypothesis” adead Cauterucci, C. 2016. Why do we idolize chimps when we could be im- concept for explaining the emergence of the earliest hominins? Cur- itating female bonobos? Slate: XX Factor: What women really think rent Anthropology 55:59–81. (Sept. 14). http://www.slate.com/blogs/xx_factor/2016/09/14/why Dominy, N. J. 2015. Ferment in the family tree. Proceedings of the _do_we_idolize_chimps_when_we_could_be_imitating_feminist National Academy of Sciences of the USA 112:308–309. _bonobos.html. Accessed October 16, 2016. Donoghue, M. J. 1989. Phylogenies and the analysis of evolutionary se- Cavender-Bares, J. 2016. Diversity, distribution and ecosystem services quences, with examples from seed plants. Evolution 43:1137–1156. of the North American oaks. International Oaks 27:37–48. Dreger, A. 2011. Darkness’s descent on the American Anthropological Cerling, T. E., F. K. Manthi, E. N Mbjua, L. N. Leakey, M. G. Leakey, Association: a cautionary tale. Human Nature 22:225–246. R. E. Leakey, F. H. Brown, et al. 2013. Stable isotope-based diet Drummond, H. 1981. The nature and descriptions of behavior patterns. reconstructions of Turkana Basin hominins. Proceedings of the Na- Pages 1–33 in P. P. G. Bateson and P. Klopfer, eds. Perspectives in tional Academy of Sciences of the USA 110:10501–10506. ethology. Vol. 4. Plenum, New York. Clark, J. A. 2013. Intersections between development and evolution Dugatkin, L. A., and H. K. Reeve, eds. 2000. Game theory and animal in the classification of emotions. Development and Psychobiology behavior. Oxford University Press, Oxford. 55:67–75. Edwards, E. J., and M. J. Donoghue. 2013. Is it easy to move and easy Coddington, J. A 1988. Cladistic tests of adaptational hypotheses. to evolve? evolutionary accessibility and adaptation. Journal of Ex- Cladistics 4:3–22. perimental Botany 64:4047–4052. Collard, M., B. Buchanan, A. Ruttle, and M. J. O’Brien. 2011. Niche con- Eldredge, N. 1979. Cladism and common sense. Pages 165–198 in struction and the toolkits of hunter-gatherers and food gatherers. J. Cracraft and N. Eldredge, eds. Phylogenetic analysis and paleon- Biological Theory 6:251–259. tology. Columbia University Press, New York. Collard, M., and S. J. Shennan. 2008. Patterns, process, and parsimony: Eldredge, N., and J. Cracraft. 1980. Phylogenetic patterns and the evo- studying cultural evolution with analytic techniques from evolution- lutionary process: method and theory in comparative biology. Co- ary biology. Pages 17–33 in M. T. Stark, B. J. Bower, and L. Horne, lumbia University Press, New York.

Eldredge, N., and S. J. Gould. 1972. Punctuated equilibria: an alter- ———. 1994. Homology and behavioral repertoires. Pages 369–391 native to phyletic gradualism. Pages 193–223 in T. J. M. Schopf, in B. K. Hall, ed. Homology: the hierarchical basis of comparative ed. Models in paleobiology. Freeman Cooper, San Francisco. biology. Academic Press, San Diego, CA. Emlen, D. J. 2014. Animal weapons: the evolution of battle. H. Holt, ———. 1999. Natural history and behavioural homology. Pages 173– New York. 188 in G. R. Bock and G. Cardew, eds. Homology (Novartis Founda- Ereshefsky, M. 2007. Psychological categories as homologies: lessons tion Symposium 222). J. Wiley and Sons, Chichester. from ethology. Biology and Philosophy 22:659–674. ———. 2005. Organisms in nature as a central focus for biology. ———. 2012. Homology thinking. Biology and Philosophy 27:381– Trends in Ecology and Evolution 20:23–27. 400. ———. 2013. Tracks and shadows: field biology as art. University of Falótico, T., and E. B. Ottoni. 2014. Sexual bias in probe tool man- California Press, Berkeley. ufacture and use by wild bearded capuchin monkeys. Behavioural Greene, H. W., and G. M. Burghardt. 1978. Behavior and phylog- Processes 108:117–122. eny: constriction in ancient and modern snakes. Science 200:74–77. ———. 2016. The manifold use of pounding stone tools by wild ca- Greene, H. W., P. May, D. L. Hardy Sr., J. Sciturro, and T. Farrell. puchin monkeys of Serra da Capivara National Park, Brazil. Behav- 2002. Parental behavior by vipers. Pages 179–205 in G. W. Schuett, iour 153:421–442. M. Höggren, M. E. Douglas, and H. W. Greene, eds. Biology of the Fitzhugh, K. 2016. Phylogenetic hypotheses: neither testable nor fal- vipers. Eagle Mountain, Eagle Mountain, UT. sifiable. Evolutionary Biology 43:257–266. Griffiths, O. W., D. G. Blackburn, M. C. Brandley, J. U. Van Dyke, Fleagle, J. G. 2013. Primate adaptation and evolution. 3rd ed. Academic C. M. Whittington, and M. B. Thompson. 2015. Ancestral state re- Press, San Diego, CA. constructions require biological evidence to test evolutionary hypoth- Foster, S. A. 2013. Evolution of behavioural phenotypes: influences eses: a case study examining the evolution of reproductive mode in of ancestry and expression. Animal Behaviour 85:1061–1075. squamate reptiles. Journal of Experimental Zoology (Molecular and Fragaszy, D., P. Izar, E. Visaberghi, E. B. Ottoni, and M. Gomes de Developmental Evolution) 324B:493–503. Oliveira. 2004. Wild capuchin monkeys (Cebus libidinosus) use an- Gumert, M. D., M. Kluck, and S. Malaivigitnond. 2009. The physical vils and stone pounding tools. American Journal of Primatology characteristics and usage patterns of stone ax and pounding ham- 64:359–366. mers used by long-tailed macaques in the Andaman Sea region of Fujii, J. A., K. Ralls, and M. T. Tinker. 2015. Ecological drivers of var- Thailand. American Journal of Primatology 71:594–608. iation in tool-use frequency across sea otter populations. Behavioral Gursky, S. 2005. Predator mobbing in Tarsius spectrum. International Ecology 26:519–526. Journal of Primatology 26:207–221. Gómez, J. M., M. Verdú, A. Gonzáles-Megías, and M. Méndez. 2016. Guthrie, R. D. 2005. The nature of Paleolithic art. University of Chicago The phylogenetic roots of human lethal violence. Nature 538:233– Press, Chicago. 237. Hall, B. K., ed. 1994. Homology: the hierarchical basis of compara- Goodall, J. 1964. Tool-using and aimed throwing in a community of tive biology. Academic Press, San Diego, CA. free-living chimpanzees. Nature 201:1264–1266. ———. 2013. Homology, homoplasy, novelty, and behavior. Devel- Gould, S. J. 1980. Introduction. Pages ix-xviii in B. Kurtén. Dance of opmental Psychobiology 55:4–12. the tiger: a novel of the Ice Age. Random House, New York. Hamilton, C., and D. M. Fragaszy. 2014. Observation of weapon use ———. 2002. The structure of evolutionary theory. Harvard Univer- in a group of semi-free tufted capuchins (Sapajus spp.). Neotrop- sity Press, Cambridge, MA. ical Primates 21:199–201. Gould, S. J., and R. C. Lewontin. 1979. The spandrels of San Marco Harmand, S., J. E. Lewis, C. S. Feibel, C. J. Lepre, S. Prat, A. Lenoble, and the Panglossian paradigm: a critique of the adaptationist pro- X. Boes, et al. 2015. 3.3-million-year-old stone tools from Lomekwi 3, gramme. Proceedings of the Royal Society B 205:581–598. West Turkana, Kenya. Nature 521:310–315. Gould, S. J., and E. Vrba. 1982. Exaptation—a missing term in the Harrison, T., ed. 2011. Paleontology and geology of Laetoli: human science of form. Paleobiology 8:4–15. evolution in context. Springer, New York. Gowaty, P. A. 1981. Sexual terms in sociobiology: emotionally evoca- Haslam, M., and T. Falótico. 2015. Nasal probe and toothpick use by tive and, paradoxically, jargon. Animal Behaviour 30:630–631. a wild female bearded capuchin (Sapajus libidinosus). Primates 56: Grandcolas, P., R. Nattier, F. Legendre, and R. Pellens. 2011. Map- 211–214. ping extrinsic traits such as extinction risk or modelled [sic] biocli- Haslam, M., A. Hernandez-Aguilar, V. Ling, S. Carvalho, I. de la Torre, matic niches on a phylogeny: does it make sense at all? Cladistics 27: A. DeStefano, A. Du, et al. 2009. Primate archaeology. Nature 460: 181–185. 339–344. Gray, R. D., S. J. Greenhill, and R. M. Ross. 2007. The pleasures and Haslam, M., L. Luncz, A. Pascual-Garrido, T. Falótico, S. Malaivijit- perils of Darwinizing culture (with phylogenies). Biological Theory nond, and M. Gumert. 2016a. Archaeological excavation of wild 2:360–375. macaque tone tools. Journal of Human Evolution 96:134–138. Greene, H. W. 1983. Dietary correlates of the origin and radiation of Haslam, M., L. V. Luncz, R. A. Staff, F. Bradshaw, E. B. Ottoni, and snakes. American Zoologist 23:431–441. T. Falótico. 2016b. Pre-Colombian monkey tools. Current Biology ———. 1986. Diet and arboreality in the emerald monitor, Varanus 26:R515–R522. prasinus, with comments on the study of adaptation. Fieldiana, Headland, T. N., and H. W. Greene. 2011. Hunter-gatherers and Zoology (New Series) 31:1–12. other primates as prey, predators, and competitors of snakes. Pro- ———. 1992. The behavioral and ecological context for pitviper evo- ceedings of the National Academy of Sciences of the USA 108:20865– lution. Pages 107–117 in J. A. Campbell and E. D. Brodie Jr., eds. 20866, E1470–E1474. Biology of the pitvipers. Selva, Tyler, TX. Hoevers, E. 2015. Tools go back in time. Nature 521:294–295.

Holekamp, K., S. T. Sakai, and B. L. Lundrigan. 2007. Social intelligence Lorenz, K. 1941. Vergleichende Bewegungsstudien bei Anatiden. in the spotted hyena (Crocuta crocuta). Philosophical Transactions Journal für Ornithologie 89:194–294. of the Royal Society B 362:523–538. Losos, J. B. 2009. Lizards in an evolutionary tree: ecology and adap- Honing, H., and A. Ploeger. 2012. Cognition and the evolution of music: tive radiation of anoles. University of California Press, Berkeley. pitfalls and prospects. Topics in Cognitive Science 4:513–524. ———. 2011. Convergence, adaptation, and constraint. Evolution 65: Hrdy, S. B. 2011. Mothers and others: the evolutionary origins of 1827–1840. mutual understanding. Belknap, Cambridge, MA. Luncz, L. V., T. Falatico A. Pascual-Garrido, A. Corat, H. Mosley, and Huebner, F., and C. Fichtel. 2015. Innovation and behavioral flexibility M. Haslam. 2016. Wild capuchin monkeys adjust stone tools ac- in wild red-fronted lemurs (Eulemur rufifrons). Animal Cognition 18: cording to changing nut properties. Scientific Reports 6:33089. 777–787. Lycett, S. J., M. Collard, and W. C. McGrew. 2009. Cladistic analyses of Huffman, M. A., J.-B. Leca, and C. A. D. Nahallage. 2010. Cultured behavioural variation in wild Pan troglodytes: exploring the chim- Japanese macaques: a multidisciplinary approach to stone handling panzee cultural hypothesis. Journal of Human Evolution 57: 337–349. behavior and its implications for the evolution of behavioral tra- Lynch Alfaro, J. W., J. de Sousa e Silva Jr., and A. B. Rylands. 2012. How dition in non-human primates. Pages 185–213 in N. Nakagawa, different are robust and gracile capuchin monkeys? an argument for M. Nakamichi, and H. Sugiura, eds. The Japanese macaques. Springer, the use of Sapajus and Cebus. American Journal of Primatology 74: Berlin. 273–286. Hull, D. L. 1988. Science as a process: an evolutionary account of the MacLean, E. L., L. J. Matthews, B. A. Hare, C. L. Nunn, R. C. Anderson, social and conceptual development of science. University of Chicago F. Aureli, E. M. Brannon, et al. 2012. How does cognition evolve? phy- Press, Chicago. logenetic comparative psychology. Animal Cognition 15:223–238. Isbell, L. A. 2009. The fruit, the tree, and the serpent: why we see so Marks, J. 2015. Tales of the ex-apes: how we think about human evo- well. Harvard University Press, Cambridge, MA. lution. University of California Press, Berkeley. Isbell, L. A., and S. F. Etting. 2017. Scales drive visual detection, atten- Mayr, E. 1983. How to carry out the adaptationist program. Amer- tion, and memory of snakes in wild vervet monkeys (Chlorocebus ican Naturalist 121:324–334. pygerythrus). Primates 58:121–129. McCune, A. R., and J. C. Schimenti. 2012. Using genetic networks Kappeler, P. M., and E. W. Heymann. 1996. Nonconvergence in the and homology to understand the evolution of phenotypic traits. evolution of primate life history and socioecology. Biological Jour- Current Genomics 13:74–84. nal of the Linnean Society 59:297–326. McGrew, W. C. 1992. Chimpanzee material culture: implications for Katoh, S., Y. Beyene, T. Itaya, H. Hyodo, M. Hyodo, K. Yagi, C. Gouzu, human evolution. Cambridge University Press, Cambridge. et al. 2016. New geological and palaeontological age constraint for McGrew, W. C., and C. E. G. Tutin. 1973. Chimpanzee tool use in den- the gorilla-human lineage split. Nature 530:215–218. tal grooming. Nature 241:477–478. King, B. J. 2013. How animals grieve. University of Chicago Press, Mercador, J., H. Barton, J. Gillespie, J. Harris, S. Kuhn, R. Tyler, and Chicago. C. Boesch. 2007. 4,300 year-old chimpanzee sites and the origins Klopfer, P. H. 1973. Does behavior evolve? Annals of the New York of percussive stone technology. Proceedings of the National Acad- Academy of Sciences 223:113–119. emy of Sciences of the USA 104:3043–3048. ———. 1999. Politics and people in ethology: personal reflections on Meulman, E. J. M., C. M. Sanz, E. Visalberghi, and C. P. van Schaik. the study of animal behavior. Bucknell University Press, Lewisburg, 2012. The role of terrestriality in promoting primate technology. PA. Evolutionary Anthropology 21:58–68. Kurtén, B. 1980. Dance of the tiger: a novel of the Ice Age. Random Morrell, V. 2013. Animal wise: how we know animals think and feel. House, New York. Crown, New York. Landau, M. 1984. Human evolution as narrative: have hero myths Norell, M. A., J. M. Clark, L. M. Chaippe, and D. Dashzeveg. 1995. A and folktales influenced our interpretations of the evolutionary nesting dinosaur. Nature 378:774–776. past? American Scientist 72:262–268. O’Brien, M. J., B. Buchanan, M. Collard, and M. T. Boulanger. 2012. Langin, K. M., T. S. Sillett, W. C. Funk, S. A. Morrison, M. A. Cultural cladistics and the early prehistory of North America. Desrosier, and C. K. Ghalambor. 2015. Islands within an island: Pages 23–42 in P. Pontarotti, ed. Evolutionary biology: mecha- repeated adaptive divergence in a single population. Evolution 69: nisms and trends. Springer, Berlin. 653–665. O’Brien, M. J., M. Collard, B. Buchanan, and M. T. Boulanger. 2013. Larson, A., and J. B. Losos. 1996. Phylogenetic systematics of adapta- Trees, thickets, or something in between? recent theoretical and tion. Pages 187–220 in M. R. Rose and G. V. Lauder, eds. Adaptation. empirical work in cultural phylogeny. Israeli Journal of Ecology Academic Press, San Diego, CA. and Evolution 59:45–61. Leakey, M. D. 1981. Discoveries at Laetoli in northern Tanzania. Pro- O’Hara, R. 1988. Homage to Clio: or, toward an historical philoso- ceedings of the Geological Association 92:81–86. phy for evolutionary biology. Systematic Zoology 37:142–155. Lewontin, R. C. 1978. Adaptation. Scientific American 293:212–228. ———. 1992. Telling the tree: narrative representation and the ———. 1998. The evolution of cognition: questions we will never an- study of evolutionary history. Biology and Philosophy 7:135–160. swer. Pages 107–132 in D. Scarborough and S. Sternberg, eds. An in- Ostrom, J. 1972. Were some dinosaurs gregarious? Palaeogeography, vitation to cognitive science. Vol. 4. Methods, models, and conceptual Palaeoclimatology, Palaeoecology 11:287–301. issues. MIT Press, Cambridge, MA. Pickering, T. R. 2013. Rough and tumble: aggression, hunting, and Liao, W. B., S. L. Lou, Y. Zeng, and A. Kotrscal. 2016. Large brains, small human evolution. University of California Press, Berkeley. guts: the expensive tissue hypothesis supported within anurans. Amer- ———. 2014. Comments on Domínguez-Rodrigo (2014). Current ican Naturalist 188:693–700. Anthropology 55:73–74.

Pickering, T. R., and M. Domínguez-Rodrigo. 2010. Chimpanzee Schoch, W. H., G. Bigga, U. Böhner, P. Richter, and T. Terberger. referents and the emergence of human hunting. Open Anthropol- 2015. New insights on the wooden weapons from the Paleolithic ogy Journal 3:94–100. site of Schöenigen. Journal of Human Evolution 89:214–225. Pinker, S. 2002. The blank slate: the modern denial of human nature. Scholes, E. 2008. Courtship ethology of Wahnes’ parotia Parotia Penguin, New York. wahnesi (Aves: Paradisaeidae). Journal of Ethology 26:79–91. Preuss, T. M. 2012. Human brain evolution: from gene discovery to Segerstråle, U. 2000. Defenders of the truth: the battle for science in the phenotype discovery. Proceedings of the National Academy of sociobiology debate and beyond. Oxford University Press, Oxford. Sciences of the USA 109:10709–10716. Seilacher, A. 2007. Trace fossil analysis. Springer, Berlin. Prindle, D. F. 2009. Stephen Jay Gould and the politics of evolution. Sessions, S. K., L. Bizjak Mali, D. M. Green, V. Trifona, and M. Ferguson- Prometheus, Amherst, NY. Smith. 2016. Evidence for sex chromosome turnover in proteid Proctor, H. C. 1996. Behavioral characters and homoplasy: percep- salamanders. Cytogenetics and Genome Research 148:305–313. tion versus practice. Pages 131–149 in M. J. Sanderson and L. Shettleworth, S. J. 2010. Clever animals and killjoy explanations in Hufford, eds. Homoplasy: the recurrence of similarity in evolu- comparative psychology. Trends in Cognitive Sciences 14:477–481. tion. Academic Press, San Diego, CA. ———. 2012. Do animals have insight, and what is insight anyway? Proffitt, T., L. V. Luncz, T. Falótico, E. B. Ottoni, I. de la Torre, and Canadian Journal of Experimental Psychology 66:217–226. M. Haslam. 2016. Wild monkeys flake stone tools. Nature 539:85–88. Shultz, S., C. Opie, and Q. D. Atkinson. 2011. Stepwise evolution of Pruetz, J., and P. Bertolani. 2007. Savanna chimpanzees, Pan troglodytes stable sociality in primates. Nature 479:219–222. verus, hunt with tools. Current Biology 17:412–417. Skutch, A. F. 1980. A naturalist on a tropical farm. University of Pruetz, J. D., P. Bertolani, K. Boyer Ontl, S. Lindshield, M. Shelley, and California Press, Berkeley. E. G. Wessling. 2015. New evidence on the tool-assisted hunting Sullivan, A. P., D. W. Bird, and G. H. Perry. 2017. Human behavior as a exhibited by chimpanzees (Pan troglodytes verus) in a savannah hab- long-term ecological driver of non-human evolution. Nature Ecology itat at Fongoli, Senegal. Royal Society Open Science 2:140507. and Evolution, http://dx.doi.org/10.1038/s41559-016-0065. Pruetz, J. D., and T. C. Laduke. 2010. Reaction to fire by savanna Tattersall, I., and N. Eldredge. 1977. Fact, theory, and fantasy in hu- chimpanzees (Pan troglodytes verus) at Fongoli, Senegal: concep- man paleontology: controversy in the study of human evolution tualization of “fire behavior” and the case for a chimpanzee model. reflects inadequacies in the formulation of hypotheses more than American Journal of Physical Anthropology 141:646–650. it does the supposed inadequacies of the fossil record. American Quinn, A. 2016a. William Whewell’s philosophy of architecture and Scientist 65:204–211. the historicization of biology. Studies in History and Philosophy Tëmkin, I., and N. Eldredge. 2007. Phylogenetics and material culture. of Biological and Biomedical Sciences 59:11–19. Current Anthropology 48:146–154. ———.2016b. Charles Girard: relationships and representation in Thieme, H. 1997. Lower Paleolithic hunting spears from northern nineteenth century systematics. Journal of the History of Biology, Germany. Nature 385:807–810. http://dx.doi.org/10.1007/s10739-016-9447-6. Tinbergen, N. 1963. On aims and methods of ethology. Zeitschrift Radinsky, L. B. 1968. Evolution of somatosensory specialization in für Tierpsychologie 20:410–429. otter brains. Journal of Comparative Neurology 134:495–505. Treves, A., and C. Bonancic. 2016. Humanity’s dual response to dogs Reader, S. M., and K. N. Laland. 2002. Social intelligence, innova- and wolves. Trends in Ecology and Evolution 31:489–491. tion, and enhanced brain size in primates. Proceedings of the Na- Visalberghi, E., G. Sirianni, D. Fragaszy, and C. Boesch. 2015. Per- tional Academy of Sciences of the USA 99:4436–4441. cussive tool use by Taï Western chimpanzees and Fazenda Boa Richerson, P. J., and R. Boyd. 2005. Not by genes alone: how culture Vista bearded capuchin monkeys. Philosophical Transactions of transformed human evolution. University of Chicago Press, Chicago. the Royal Society B 370(1682):20140351. Rivas, J. A., and G. M. Burghardt. 2002. Crotalomorphism: a meta- Wake, D. B. 1996. Introduction. Pages xvii–xxv in M. J. Sanderson phor for understanding anthropomorphism by omission. Pages 9– and L. Hufford, eds. Homoplasy: the recurrence of similarity in 17 in M. Bekoff, C. Allen, and G. M. Burghardt, eds. The cognitive evolution. Academic Press, San Diego, CA. animal: empirical and theoretical perspectives on animal cogni- Wake, D. B., M. H. Wake, and C. D. Specht. 2011. Homoplasy: from tion. MIT Press, Cambridge, MA. detecting pattern to determining process and mechanism of evo- Roach, M. 2008. Almost human. National Geographic 213:127–144. lution. Science 331:1032–1035. Roffman, I., S. Savage-Rumbaugh, E. Rupert-Pugh, A. Stadler, A. Wake, M. H. 1992. Evolutionary scenarios, homology and conver- Ronen, and E. Nevo. 2015. Preparation and use of varied natural gence of structural specializations for vertebrate viviparity. Amer- tools for extractive foraging by bonobos (Pan paniscus). American ican Zoologist 32:256–263. Journal of Physical Anthropology 158:78–91. Wanntorp, H.-E. 1983. Historical constraints in adaptation theory: Safina, C. 2015. Beyond words: what animals think and feel. H. Holt, traits and non-traits. Oikos 41:157–160. New York. Wanntorp, H.-E., D. R. Brooks, T. Nilsson, S. Nylin, F. Qonquist, S. C. Sahle, Y., W. K. Hutchings, D. R. Braun, J. C. Sealy, L. E. Morgan, Stearns, and N. Wedell. 1990. Phylogenetic approaches in ecology. A. Negash, and B. Atnafu. 2013. Earliest stone-tipped projectiles from Oikos 57:119–132. the Ethiopian Rift date to 1279,000 years ago. PLoS ONE 8:e78092. Watanabe, K., N. Urasopon, and S. Malaivijitnond. 2007. Long-tailed Sanderson, M. J., and L. Hufford, eds. 1996. Homoplasy: the recur- macaques use human hair as dental floss. American Journal of Pri- rence of similarity in evolution. Academic Press, San Diego, CA. matology 69:940–944. Sarringhaus, L. A., L. M. MacLatchy, and J. C. Mitani. 2014. Locomo- Wenzel, J. W. 1992. Behavioral homology and phylogeny. Annual tor and postural development of wild chimpanzees. Journal of Hu- Reviews of Ecology and Systematics 23:361–381. man Evolution 66:29–38. Werdelin, L. 2013. King of beasts. Scientific American (Nov.):36–39.

Werdelin, L., and M. E. Lewis. 2013. Temporal change in functional Wood, B., and T. Harrison. 2011. The evolutionary context of the richness and evenness in the East African Plio-Pleistocene carnivo- ﬁrst hominins. Nature 470:347–352. ran guild. PLoS ONE 8:e57944. Wrangham, R. 2009. Catching ﬁre: how cooking made us human. White, T. D., B. Asfaw, Y. Beyene, Y. Haile-Selasse, C. O. Lovejoy, Basic, New York. G. Suwa, and G. WoldeGabriel. 2009. Ardipithecus ramidus and Yamaguchi, N., A. Cooper, L. Werdelin, and D. W. Macdonald. 2004. the paleobiology of early hominids. Science 326:75–86. Evolution of the mane and group-living in the lion (Panthera leo): White, T. D., and G. Suwa. 1987. Hominid footprints at Laetoli: facts a review. Journal of Zoology 263:329–342. and interpretations. American Journal of Physical Anthropology 72: Yeaves, J. D., N. J. Dominy, P. L. Koch, and M. Mangel. 2013. Func- 485–514. tional morphology, stable isotopes, and human evolution: a model Wilkins, J., B. J. Schoville, K. S. Brown, and M. Chazan. 2012. Evi- of consilience. Evolution 68:190–203. dence for early hafted hunting technology. Science 338:942–946. Zamudio, K. R., R. C. Bell, R. C. Nali, C. F. B. Haddad, and C. P. A. Wilson, D. E., and D. M. Reeder. 2005. Mammal species of the world: a Prado. 2016. Polyandry, predation, and the evolution of frog re- taxonomic and geographic reference. 3rd ed. Johns Hopkins Uni- productive modes. American Naturalist 188(suppl.):S41–S61. versity Press, Baltimore. Wilson, E. O. 1975. Sociobiology: the new synthesis. Harvard Uni- versity Press, Cambridge, MA. Symposium Editor: Anurag A. Agrawal

“Then follows a general ‘popular’ account of the forms of apes, their geographical distribution, dwelling-places, food, motions, social life, language, reproduction, education, rearing of young, diseases, life in confinement, and of the apes figured on the Egyptian temples. . . . [Figured:] the head of a five-year-old tschego in the Dresden Zoological Garden, which was brought from the Loango coast.” From the review of Brehm’s Animal Life (The American Naturalist, 1877, 11:557–559).

Symposium Convergence, Consilience, and the Evolution of Temperate Deciduous Forests*

Erika J. Edwards,1,† David S. Chatelet,1,‡ Bo-Chang Chen,2 Jin Yao Ong,3 Shuichiro Tagane,4 Hironobu Kanemitsu,5 Kazuki Tagawa,5 Kentaro Teramoto,5 Brian Park,6 Kuo-Fang Chung,2,7 Jer-Ming Hu,3 Tetsukazu Yahara,4 and Michael J. Donoghue6

1. Department of Ecology and Evolutionary Biology, Brown University, Providence, Rhode Island 02912; 2. School of Forestry and Resource Conservation, National Taiwan University, Taipei 10617, Taiwan; 3. Institute of Ecology and Evolutionary Biology, National Taiwan University, Taipei 10617, Taiwan; 4. Department of Biology, Kyushu University, 744 Motooka, Fukuoka 819-0395, Japan; 5. Graduate School of Systems Life Sciences, Kyushu University, 744 Motooka, Fukuoka 819-0395, Japan; 6. Department of Ecology and Evolutionary Biology, Yale University, New Haven, Connecticut 06520; 7. Biodiversity Research Center, Academia Sinica, Taipei 11529, Taiwan Dryad data: http://dx.doi.org/10.5061/dryad.k505s. abstract: The deciduous habit of northern temperate trees and Introduction shrubs provides one of the most obvious examples of convergent evo- Biologists have long been attracted to convergent evolution, lution, but how did it evolve? Hypotheses based on the fossil record and for good reason, as it provides powerful evidence of posit that deciduousness evolved first in response to drought or dark- natural selection on organismal performance. Many con- ness and preadapted certain lineages as cold climates spread. An alter- fl native is that evergreens first established in freezing environments and vergent traits re ect organismal interactions, both mutual- later evolved the deciduous habit. We monitored phenological pat- istic and antagonistic (Fenster et al. 2004; Agrawal and terns of 20 species of Viburnum spanning tropical, lucidophyllous Fishbein 2006; Wilson et al. 2007), but past climate change (subtropical montane and warm temperate), and cool temperate Asian has undoubtedly also driven convergence on a global scale. forests. In lucidophyllous forests, all viburnums were evergreen plants For example, the latter half of the Cenozoic witnessed the fl that exhibited coordinated leaf ushes with the onset of the rainy sea- formation and spread of deserts and grasslands, resulting son but varied greatly in the timing of leaf senescence. In contrast, deciduous species exhibited tight coordination of both flushing and se- in the repeated evolution of succulent life-forms and C4 nescence, and we found a perfect correlation between the deciduous photosynthesis (Edwards et al. 2010; Arakaki et al. 2011). habit and prolonged annual freezing. In contrast to previous stepwise Convergence also provides insights into the relative evo- hypotheses, a consilience of independent lines of evidence supports a lutionary accessibility of certain phenotypes and how struc- lockstep model in which deciduousness evolved in situ, in parallel, and tural features of organisms may influence their evolutionary concurrent with a gradual cooling climate. A pervasive selective force response (Sanderson and Hufford 1996; Donoghue and Ree combined with the elevated evolutionary accessibility of a particular 2000; Christin et al. 2013). The degree to which a particular response may explain the massive convergence of adaptive strategies that characterizes the world’s biomes. character repeatedly emerges must be a function of both the pervasiveness of the selection pressure(s) and the rela- Keywords: Viburnum, biome assembly, freezing tolerance, leaf habit, tive ease of its evolution. As climate change is experienced climate change, phylogeny. by virtually all organisms in a region concurrently, it makes sense that it would be one of the most powerful agents of massive convergent evolution. However, whether such massive convergence is the outcome depends on whether a given adaptive response is also the most evolutionarily accessible * This issue originated as the 2016 Vice Presidential Symposium presented at (e.g., Weinreich 2006; Meyer et al. 2012; Agrawal 2017) in the annual meetings of the American Society of Naturalists. multiple lineages. The combination of a ubiquitous selec- † Corresponding author; e-mail: [email protected]. tive force with the evolutionary accessibility of a particular ‡ Present address: Biomedical Imaging Unit, University of Southampton, South- adaptation virtually guarantees rampant convergence. ampton SO16 6YD, United Kingdom. On the surface of it, massive convergence would appear ORCIDs: Kanemitsu, http://orcid.org/0000-0002-4332-2152; Hu, http://orcid .org/0000-0003-2739-9077. to lend itself perfectly to comparative phylogenetic analy- – q ses. It is only through a phylogenetic lens that convergence Am. Nat. 2017. Vol. 190, pp. S87 S104. 2017 by The University of Chicago. fi 0003-0147/2017/190S1-57320$15.00. All rights reserved. can be identi ed, after all, and a multitude of comparative DOI: 10.1086/692627 methods have been developed to measure convergence and

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:54 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S88 The American Naturalist the factors that are correlated with it, perhaps causally. In egories, our fieldwork revealed a natural decomposition of characters with moderate levels of convergence, phyloge- phenological behavior into two distinct and potentially in- netic approaches can work well, but for extremely labile dependently evolving elements: the flushing of new leaves traits, uncertainties in phylogenetic inference can become and the senescing of old ones. It also exposed the inadequacy limiting (Schultz et al. 1996; Schluter et al. 1997). This sit- of the standard binary breakdown of biome or habitat type uation is complicated further by the observation that con- into tropical versus temperate forests. Specifically, we found it vergence itself is often distinctly concentrated in particular important to also recognize the existence of subtropical mon- regions of the tree of life (Edwards and Donoghue 2013). tane and warm temperate lucidophyllous forests (Tang 2010;

For example, C4 photosynthesis has evolved over 60 times Tang et al. 2013). These Asian forests are characterized by in plants, but fully two-thirds of these origins occur in just the dominance of broad-leaved angiosperm trees (typically two flowering plant lineages, the grasses and the Caryophyl- Lauraceae and Fagaceae) in a climate that is distinctly mon- lales (Sage et al. 2011), and even within grasses, C4 origins soonal in comparison to tropical rainforests but where freez- are clustered yet again in a clade that includes roughly half ing is infrequent and winter temperatures are mild compared of all grass species (Christin et al. 2013). Furthermore, if to cold temperate forests. The Viburnum species that we evolutionary responses are rapid enough, it becomes im- studied in these forests show patterns of leaf flushing and se- possible to identify the evolutionary sequences and transi- nescence that provide important clues to the transition be- tional conditions that might provide us with clues about tween the archetypal evergreen and deciduous conditions. cause and effect, especially about evolutionary events that Considered in isolation, our phylogenetic and field stud- happened long ago. Faced with these difficulties, convinc- ies each provide insight into the evolution of deciduousness ing answers may not be achieved using phylogenies alone but not overwhelming support for a particular evolutionary (Christin et al. 2010; Hancock and Edwards 2014). Prog- pathway. Considered together, however, the consilience of ress, then, depends on the integration of phylogenetic stud- evidence leads us to propose a model in which the decidu- ies with other lines of evidence (Weber and Agrawal 2012; ous habit evolves quite rapidly, in lockstep with a gradual in Olson and Arroyo-Santos 2015) and, as we highlight, per- situ transition to a routinely freezing climate. We hypothe- haps by further deconstruction of the traits and potential size that the novel behavior that emerged with the decidu- selective factors of interest. ous habit was the tight coordination of leaf senescence, per- Here we focus on the evolution of the deciduous leaf haps as a means of ensuring the resorption of nutrients habit, one of the most obvious and highly convergent adap- prior to leaf death by freezing (Feild et al. 2001; Niinemets tations of woody plants to temperate environments that ex- and Tamm 2005). To our surprise, this lockstep model ap- perience an annual period of prolonged freezing temper- pears to be novel, as prior theories have envisioned a dis- atures. Deciduous plants shed their leaves with the onset of tinctly stepwise process. As we review below, these entail ei- the cold period and remain leafless until they flush a new ther the evolution of the deciduous habit, first in response set of leaves in the spring. This habit dominates northern to a period of drought or darkness (with establishment in hardwood forests and has evolved independently many times freezing climates emerging only later), or the establishment in distantly related plant groups (e.g., maples and oaks). It is of evergreen plants in freezing climates, followed later by important to note that deciduousness has also evolved many the evolution of deciduousness. We suggest that the step- times in settings that we will not consider here, particularly wise perspective that has explicitly or implicitly oriented in dry tropical forests where there is also a season unfavor- many phylogenetic studies of adaptation may be inappro- able for growth—in this case, determined by drought, not priate or even misleading when studying past evolutionary cold (Murphy and Lugo 1986). We do not know of any at- responses to the gradual climate changes that have shaped tempts to infer the number of independent origins of the de- the assembly of the world’s biomes. ciduous leaf habit in angiosperms, but considering the extremely broad phylogenetic distribution of deciduous plants and its demonstrated lability within smaller lineages (e.g., Background Information Schmerler et al. 2012), we guess that it has evolved more fre- Evolution of the Deciduous Habit in Temperate Forests quently than other well-known convergent plant traits (e.g., 1 1 62 origins of C4 photosynthesis [Sage et al. 2011]; 130 or- In general, the deciduous leaf habit in woody plants is un- igins of bilaterally symmetrical flowers [Reyes et al. 2016]). derstood to be a response to long periods of time that are As a case study, we analyze the evolution of deciduous- unfavorable for growth, which can result from episodes of ness in the woody plant clade Viburnum, combining stan- freezing or drought (Chabot and Hicks 1982). By definition, dard phylogenetic analyses with a phenological field study annually deciduous plants must have leaf life spans of less along a latitudinal transect in eastern Asia. Independent than 12 months, with many cold temperate woody plants of the traditional evergreen versus deciduous leaf habit cat- averaging no more than 6 or 7 months (Reich et al. 1992;

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:54 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Consilience, Convergence, and Phylogeny S89 van Ommen Kloeke et al. 2012). Deciduous plants also tend which experience strong seasonality in both rainfall and to fall on the fast or resource-acquisitive end of the leaf eco- temperature but rarely freeze. Axelrod hypothesized that nomic spectrum (Wright et al. 2004), and their leaves are the deciduous leaf habit first appeared in such forests and typically thinner, with higher nitrogen contents and higher that this later enabled these plants to expand into the freez- photosynthetic rates than evergreen species. Soil fertility ing zone. has thus been identified as an important secondary variable: Wolfe and Upchurch (1986) and Wolfe (1987) explained nutrient-poor, cold sites tend to have a higher representa- the origin of deciduous forests in another way. They pro- tion of broad-leaved evergreens, presumably because leaf posed that deciduousness evolved in situ at high latitudes, nitrogen contents are too low to maintain the higher pho- when even areas at 807N did not experience significant tosynthetic rates required to balance investment costs in a freezing. However, winters at such high latitudes were as short-lived leaf. These arguments, revolving around trade- dark as they are today, and they hypothesized that the de- offs in leaf carbon and nutrient economy, are well developed ciduous habit evolved there as a response to many months elsewhere (Chabot and Hicks 1982; Kikuzawa 1991; Givnish with little or no sunlight. At the Cretaceous-Paleogene (K-Pg) 2002), and our aim here is not to elaborate on the adaptive boundary, the brief impact winter was credited with reorganiz- significance of the deciduous leaf habit. Instead, we simply ing plant communities at lower latitudes, and Wolfe and col- assume that deciduousness is adaptive for all of the reasons leagues suggested that it was during this time that polar de- above and consider a related question that has received far ciduous lineages migrated south and became established as less attention, namely, how the deciduous habit evolved from elements of midlatitude floras. Only during the late Mio- an evergreen state. In particular, we analyze leaf flushing and cene cooling did they outcompete warm-adapted evergreen leaf senescence as two potentially independent behaviors lineages to establish a new deciduous biome. that became coordinated with the emergence of the decid- It is noteworthy that both the Axelrod and Wolfe hy- uous habit. potheses rest on the idea that deciduousness was a preadaptation to freezing—that is, it evolved first as a response to some other stress but then allowed lineages to tolerate freez- Climate Change and the Origin of ing climates when they were eventually exposed to them. Temperate Deciduous Forests Neither author explained why they preferred such an order In spite of its importance as a modern biome, there are rel- of events as opposed to the perhaps simpler hypothesis that atively few hypotheses about the origin of broad-leaved de- deciduousness evolved directly in response to freezing. More ciduous forests. Inferences based on fossil leaves indicate recently, Zanne et al. (2014), without reference to these older that deciduous angiosperm species have existed for a long hypotheses, addressedthe samequestion using a megaphylog- time, appearing as elements of fossil floras from the Late eny for angiosperms and concluded that most lineages first Cretaceous through the Cenozoic (Wolfe 1987). There is became established in the freezing zone as evergreen plants evidence for the dominance of the deciduous habit in the and only later evolved the deciduous habit. Elsewhere we highest northern latitudes (polar deciduous forests) during reanalyzed the data of Zanne et al. and raised doubts about the Late Cretaceous, when global temperatures were signif- their conclusions (Edwards et al. 2015), but we highlight icantly warmer than today. These polar forests persisted this study here because it focused directly on freezing as a through the Paleocene and Eocene periods, with broad- driver. Like Axelrod and Wolfe, Zanne et al. envisioned a leaved evergreen forests occupying significant areas in mid- stepwise evolutionary process but with a reversed order of latitude and tropical regions (Axelrod 1966; Wolfe and events. In the terminology of Zanne et al. (2014), Axelrod Upchurch 1986; Wolfe 1987). Post-Eocene global cooling and Wolfe favored a trait-first hypothesis, whereas they fa- shifted many of these midlatitude zones toward the tropics, vored a climate-first hypothesis. but it was not until the late Miocene that the modern distribution of cool temperate deciduous forests was established The Viburnum Study System (Tiffney and Manchester 2001; Utescher and Mosbrugger 2007; Pound et al. 2011). Viburnum (Adoxaceae, Dipsacales, Campanulidae) is an Axelrod (1966) proposed that deciduousness first evolved angiosperm clade of ∼165 species of shrubs and small trees in a warm, subtropical climate zone that experienced mild that is broadly distributed (and widely cultivated) around winter drought. He hypothesized that such forests were the Northern Hemisphere. Viburnum species all occupy me- widespread throughout the midlatitudes of the Northern sic forests, but they have adapted to a range of climatic con- Hemisphere in the Late Cretaceous and early Cenozoic but ditions and forest environments (fig. 1). At one end of the became restricted during the Miocene. As modern analogues, spectrum, multiple (and distantly related) species in tropical he specifically cited the lucidophyllous forests of Asia (par- forests are never subjected to freezing temperatures. At the ticularly Taiwan) and the cloud forests of southern Mexico, other end, upwards of 10 Viburnum lineages appear to have

clemensiae cassinoides nudum lentago elatum rufidulum prunifolium punctatum lepidotulum macrocephalum utile carlesii schensianum cotinifolium rhytidophyllum buddleifolium veitchii mongolicum burejaeticum lantana nervosum sympodiale lantanoides furcatum grandiflorum foetens sieboldii odoratissimum farreri suspensum taitoense henryi brachybotryum chingii erubescens subalpinum oliganthum plicatum lutescens colebrookeanum amplificatum taiwanianum urceolatum tinus rigidum atrocyaneum propinquum cinnamomifolium davidii mullaha hupehense brachyandrum foetidum flavescens ichangense lobophyllum dilatatum wrightii japonicum integrifolium parvifolium tashiroi erosum phlebotrichum luzonicum betulifolium lancifolium melanocarpum sempervirens acerifolium orientale kansuense cylindricum coriaceum beccarii vernicosum hispidulum inopinatum sambucinum evergreen edule koreanum trilobum deciduous opulus sargentii ellipticum australe bracteatum molle restricted to New World rafinesquianum tropical dentatum scabrellum recognitum lucidophyllous sulcatum acutifolium venustum cold temperate discolor blandum microphyllum cloud forest microcarpum hartwegii costaricanum uncertain stellato-tomentosum villosum disjunctum obtusatum lautum jucundum hallii ayavacense pichinchense triphyllum toronis -22 °C 0 °C 20.6 °C jamesonii stipitatum anabaptista seemenii undulatum minimum temperature of the coldest week tinoides lasiophyllum obtectum

Figure 1: Phylogeny, climatic envelope, forest habitat, and leaf habit of Viburnum. A 120-taxon phylogeny, pruned from Spriggs et al. (2015). Branch colors represent inferred ancestral minimum weekly temperatures (BIO6). Pie charts at internal nodes are ancestral state estimates of leaf habit, inferred using the threshold model (Revell 2012). Species at tips are coded by leaf habit in the left column (evergreen vs. deciduous) and habitat in the right column (tropical, lucidophyllous, cold temperate, or cloud forests). Species with names in boldface were included in the phenological monitoring study. evolved independently into cool temperate forests that expe- ceptions, in cool temperate forests they are seasonally de- rience consecutive months of below freezing temperatures. ciduous. Of the 165 species of Viburnum that we currently In Asia, members of several clades occupy lucidophyllous recognize, we consider 84 to be evergreen and 81 to be de- forests, and members of the Neotropical Oreinotinus clade ciduous. Eight of the well-supported major clades within occupy montane cloud forests, with reduced temperature Viburnum include both evergreen and deciduous species, seasonality and only rare and intermittent freezing temper- implying a minimum of eight evolutionary shifts. However, atures. we infer no fewer than 20 transitions in leaf habit using any Evolutionary shifts between these different environments of our recently published phylogenetic trees (Clement et al. have been accompanied by transitions in leaﬁng habit (ﬁg. 1). 2014; Spriggs et al. 2015; Eaton et al. 2017), including some In tropical, lucidophyllous, and cloud forests, Viburnum clear-cut evolutionary shifts in both directions. For exam- plants are most often evergreen, while, with very few ex- ple, the deciduous species Viburnum plicatum appears to

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:54 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Consilience, Convergence, and Phylogeny S91 have been derived from evergreen ancestors within the Lu- ods described in Edwards et al. (2014). At the start of the tescentia clade. Likewise, the evergreen species Viburnum study, each branch usually had only a few apical meristems, rhytidophyllum and Viburnum utile appear to have origi- but due to the growth architecture of most Viburnum spe- nated from deciduous ancestors within the Euviburnum cies (stems typically terminate in an inflorescence, with clade, as did Viburnum sempervirens and its several ever- growth recommencing from two subtending buds; Don- green relatives within Succotinus. oghue 1981; Edwards et al. 2014), by the end of the study As we have shown previously (Schmerler et al. 2012), each branch consisted of multiple independent leafy shoots, these repeated evolutionary shifts in forest environment sometimes bearing over 100 leaves. and leafing behavior have been accompanied by specific Viburnum-wide climate data, leaf habit, and forest type. changes in leaf form. Adaptation to cool temperate forests We downloaded georeferenced locality data for Viburnum at more northerly latitudes has entailed coordinated shifts from the following databases: Global Biodiversity Informa- to rounder leaf shapes with more marginal teeth or lobes. tion Facility (http://www.gbif.org/), Chinese Virtual Her- Marked seasonal heteroblasty in the temperate species (the barium (http://www.cvh.org.cn/), Interactive Agricultural development of preformed leaves with more derived leaf Ecological Atlas of Russia and Neighboring Countries (http:// shapes and neoformed leaves with more tropical shapes) sug- www.agroatlas.ru/), Plant DNA Bank in Korea (http:// gested to us that the evolution of temperate leaf forms may pdbk.korea.ac.kr/), and Virtual Viburnum (http://viburnum relate to the development of leaf primordia within the resting .peabody.yale.edu/). We also georeferenced occurrence rec- buds of deciduous species (Edwards et al. 2016). In any case, ords from Hara’ s (1983) treatment of the Viburnum species Viburnum provides clear evidence, directly in line with global of Japan. The raw data were filtered to remove spurious trends (Bailey and Sinnott 1916; Wolfe 1995; Royer and Wilf records. Specifically, we matched taxon names with those 2006; Peppe et al. 2011), of repeated evolutionary changes in in Spriggs et al. (2015) and removed records with coordi- both leaf habit and leaf form in connection with shifts be- nates at 07,07 (and those from botanical gardens and her- tween tropical and temperate forests. baria) and records of the same species with identical coor- dinates. We also plotted the records for each species using the R package ggmap (Kahle and Wickham 2013) and re- Material and Methods moved points outside of the known geographic ranges of Phenology.TomonitorViburnum leafing behavior across a these species based on floristic treatments and herbarium wide range of environments, we established a latitudinal specimens examined in preparation for our worldwide mono- transect spanning 277 latitude, from montane tropical for- graph of Viburnum. ests in Kinabalu National Park, Sabah, Malaysia, at 67Nto We used the CliMond database (Kriticos et al. 2012) at a temperate forests in the mountains surrounding Fukuoka, 10-min resolution (∼18 km2 at the equator) to extract es- Japan, on Kyushu Island at 337N(fig. 2). Sites were chosen timates of climate for each locality in the filtered data set us- to represent different forest types (tropical rain forest in ing the R package raster (Hijmans and van Etten 2012). Borneo, both lucidophyllous and cool temperate forests in This spatial grain best matches the georeferencing precision Taiwan and Japan) and also to maximize the number of of our data set. The species locality data were then spatially co-occurring Viburnum species in each location. From May rarefied so that there was no more than one occurrence per to June 2013, we selected a total of 19 populations represent- climate grid cell using SDMtoolbox (Brown 2014), which ing 18 species in Taiwan and Japan, and in February 2014 we increased the evenness of coverage across the species’ entire added two species in Borneo. For logistical reasons, not all range. Species with fewer than three grid cell occurrences populations were monitored for the same length of time. were removed from the data set. We calculated the mean The Borneo populations were monitored for 13 consecutive minimum temperature of the coldest week (BIO6) and sea- months, the Japanese populations for 10–19 consecutive sonality of precipitation (BIO15) for each species. We note months, and most of the Taiwanese populations for 25 con- that although CliMond estimates BIO6 as a weekly mea- secutive months. We achieved our best sampling in Taiwan, sure, their variable is interpolated from monthly tempera- where we were also able to monitor two populations of Vibur- ture data and is strongly correlated with the more commonly num luzonicum that spanned its elevational range (120 m, used BIO6 variable, the mean minimum temperature of the 2,100 m). coldest month (R2 p 0:995). In all, our data set consists of In each population of each of our focal species, we tagged 7,718 records for 120 Viburnum species. We also used the three to six individuals (mostly four individuals per species) CliMond database to extract mean monthly temperature for monitoring. We marked four branches per individual and precipitation data for the sites of our monitored popu- and made biweekly to monthly census trips to record the lations. persistence of each leaf on these branches, the emergence We scored each species in our data set as (1) belong- of new leaves, and all events of leaf senescence using meth- ing in the traditional evergreen or deciduous category and

Figure 2: Locations of the phenological monitoring transect in eastern Asia. Insets provide greater resolution of our different population locations. Top inset, Sites situated around Fukuoka, Japan, on the island of Kyushu. Middle inset, Location of our low-elevation population of Viburnum luzonicum in the suburbs of Taipei, Taiwan. Bottom inset, The majority of our Taiwan monitoring sites scattered around the Hehuanshan Mountain district of the Central Mountain Range. In both Japan and Taiwan, sites occupied both freezing and nonfreezing climates.

(2) occurring in tropical, lucidophyllous, cool temperate, or dividual and averaged across individuals within a popula- cloud forests. These scores were based on monographic and tion. Only a small subset of leaves were observed for their floristic treatments of Viburnum (e.g., Rehder 1908; Killip entire life span in the evergreen species in our sample; the and Smith 1931; Morton 1933; Kern 1951; Hara 1983; Yang majority of leaves were either present when we initiated and Malecot 2011). For the majority of these species we the study and senesced partway through the monitoring pe- have first-hand knowledge based on our own field and her- riod or emerged during the study and were still present at barium studies, but for Japanese species we also relied on the end. Population-level patterns in flushing and senes- Hara (1983) and for Chinese species we followed Yang and cence were visualized using the ribbon function in ggplots2 Malecot (2011). We were uncertain about assignment to for- (Wickham 2016) in R. Most species appeared to exhibit est type for seven species and scored these as unknown. strong seasonality in leaf flushing and sometimes also in se- Analysis of phenological data. We recorded leaf gain and nescence. To evaluate the statistical significance of these loss using hand-drawn sketches of branches, and dates of patterns, for each species we compared our observed data emergence and disappearance of each leaf were transcribed with a null distribution of phenological events. Because the into a spreadsheet. The total number of leaves gained or lost timing of our population visits was not perfectly spaced at each census was summed across branches within an in- throughout the year, we could not assume a flat null distri-

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:54 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Consilience, Convergence, and Phylogeny S93 bution; just by chance we would expect to see more events where we adjusted these assignments. We ran Pagel tests in spring and fall seasons, when our visitation rates were using both the curated and the uncurated binary variable. slightly higher. For each species, we pooled all visitation All data and scripts for analyses are available in a GitHub dates (as Julian days) and generated 1,000 data sets each repository (https://github.com/ejedwards/amnat_2017/final of new leaf and senesced leaf observations, sampled with re- _analyses/phylogeny). All data used in all analyses are avail- placement from the pool of Julian day visits, with the total able in the Dryad Digital Repository: http://dx.doi.org/10.5061 number of events equal to that in the observed data. We /dryad.k505s (Edwards et al. 2017). then tested whether observed distributions were statistically different from the null distribution using a Mann-Whitney test. All data and scripts for analyses are publicly available Results in a GitHub repository (https://github.com/ejedwards/amnat _2017/phenology/null_analyses). Phenology. The two tropical Viburnum species that we fol- Analysis of Viburnum-wide climate data, leaf habit, and lowed in Borneo (Viburnum clemensiae, Viburnum verni- forest type. We conducted comparative analyses using the cosum) are evergreen and exhibited little leaf turnover dur- maximum-clade credibility phylogeny from Spriggs et al. ing the course of our study (fig. 3). This was especially true (2015). We pruned the phylogeny to the 120 species (of for leaf senescence, with some individuals losing no more the 165 species of Viburnum) represented in our data set us- than a single leaf over the course of 13 months. Viburnum ing the drop.tip function in ape (Paradis et al. 2004). These clemensiae produced leaves from October to April and no species represent all of the named clades in the comprehen- leaves at all during the months of May–September. In con- sive phylogenetic classification of Viburnum developed by trast, V. vernicosum produced leaves throughout the year Clement et al. (2014). We conducted several analyses to test despite the fact that these two species occupy the same mon- the relationship between leaf habit and climate. First, we es- tane tropical forest, with individuals less than 1 km apart. timated the significance of the relationship between these The very low rates of leaf turnover in both species neces- traits using phylogenetic independent contrasts (Felsenstein sitates a much longer period of data collection to make ro- 1985) as implemented for use with a discrete and continu- bust observations of seasonality. ous character in the function crunch in the R package caper In both Taiwan and Japan, we included evergreen and (Orme 2013), treating leaf habit as the dependent variable deciduous species in our sample, though the ratios were and either BIO6 or BIO15 as the independent, continuous inverted: in Taiwan we monitored three deciduous species variable. Second, we inferred evolutionary shifts in leaf habit and six evergreen, while in Japan we monitored six decid- using the threshold model first proposed by Wright (1934) uous and three evergreen. In each location, one of the ever- and developed as an explicitly phylogenetic method by Fel- green species is perhaps better described as a leaf exchanger; senstein (2012). The threshold model assumes that discrete Viburnum luzonicum (Taiwan) and Viburnum erosum (Ja- character changes are governed by shifts in an underlying, pan) plants shed almost all of their leaves as they produced unobserved continuous variable called the liability. As the a new flush. value of the liability reaches a particular threshold, the bi- Across all species and all sites in Taiwan and Japan, nary character changes. This is an appealing approach to ex- there was a striking coordination of the timing of new leaf amine possible evolutionary relationships between observ- flushes; all species, whether deciduous or evergreen, pro- able continuous and discrete characters by evaluating the duced a significant flush of new leaves in the months of relationship between the continuous character and the esti- March, April, and May (fig. 3; table A1). In sharp contrast, mated liability parameter. We estimated the liability values senescence patterns were more variable across species and underlying shifts in leaf habit using the function thresh- across the seasons. In deciduous species, as expected, leaf BAYES in phytools (Revell 2012), running Markov chain senescence was highly seasonal and varied little between Monte Carlo for 106 generations, sampling every 1,000, and species or locations; all deciduous species demonstrated discarding the first 20% of steps as burn-in. We regressed coordinated leaf shedding as winter approached, typically estimated liability values to inferred ancestral estimates of in November and December. In evergreen species, senes- BIO6 and BIO15. Third, we utilized standard methods for cence periods were typically much longer, occurring in assessing the correlated evolution of two binary characters the months preceding, during, and after spring leaf flushes with likelihood ratio tests (Pagel 1994). In this case, we con- (fig. 3; e.g., Viburnum propinquum, Viburnum taitoense). verted BIO6 into a binary character, binning species into The tightest coordination of flushing and senescence ap- whether their mean value of BIO6 was greater or less than peared in V. luzonicum and V. erosum, but as noted, these 07C. Because six species with mean values just above or just do not fit neatly into the strictly evergreen or deciduous below 07C were assigned to categories incongruent with our category. Indeed, they may have a short leafless period field knowledge, we created an additional binary variable in one year but not the next (e.g., V. luzonicum was evergreen

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 20 10 80 V. dilatatum 60 0 40 V. parvifolium 40 30 V. integrifolium 80 20 V. furcatum

This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:54AM 20 40 10

60 0 V. sympodiale 80 V. phlebotrichum 40 30 V. luzonicum* 40 20 20 80 10 V. sieboldii 40 Jul 2013 Jan 2014 Jul 2014 Jan 2015 Jul 20150 V. propinquum 30 80 Borneo – evergreen V. urceolatum 20 40

6 10 V. clemensiae Jul 2013 Jan 2014 Jul 2014 Jan 2015 0 Japan – evergreen 4 V. taitoense 30 30 V. awabuki 2 20 20 10 10 0 0 0 6 V. vernicosum 30 V. erosum* V. taiwanianum 20 30 10 4 20 0

2 10 30 V. japonicum 20 0 10 0 Jul 2013 Jan 2014 Jul 2014 Jan 2015 Jul 2015 0 Apr 2014 Jul 2014 Oct 2014 Jan 2015 Jul 2013 Jan 2014 Jul 2014

Figure 3: Flushing and senescence patterns of 20 Viburnum species living along a tropical-temperate gradient. Lines represent the mean number of leaves gained or lost per observational period across three to six individuals per species; shading represents 51 SE. Green lines indicate new leaf flushes, and yellow lines represent leaf senescence. Species are grouped by geographical area (Borneo, Taiwan, Japan) and leaf habit (evergreen, deciduous). Asterisks indicate species that may be better described as leaf exchangers rather than strictly evergreen. Consilience, Convergence, and Phylogeny S95 in 2014 but deciduous in 2015; V. erosum was deciduous arate the influence of temperature from the influence of in 2014 but evergreen in 2015). drought, as our evergreen and deciduous species in these The significance of these visual patterns was confirmed areas all experienced similar periods of relative drought by comparison to our null distributions of flushing and se- from November to February (fig. 4). nescence (table A1). All deciduous species exhibited highly The clear association between the deciduous leaf habit significant clustering of both flushing and senescence. Flush- and regular exposure to freezing temperatures but not ing patterns were all significant in our evergreen species; changes in precipitation seasonality was corroborated in in contrast, in most cases senescence was indistinguishable our Viburnum-wide analyses. Figure 1 maps the mean min- from our null distributions (table A1; see additional files imum temperature of the coldest week (BIO6) on a Vibur- and analyses at https://github.com/ejedwards/amnat_2017 num phylogeny derived from Spriggs et al. (2015), together /final_analyses/phenology/null_analyses). The three exceptions with ancestral reconstruction of leaf habit inferred from the were Viburnum japonicum in Japan and V. propinquum threshold analyses. This implies that the first viburnums and Viburnum integrifolium in Taiwan. These species all lived in warm climates without significant freezing and that exhibit a similar senescence pattern, with a period of se- there were multiple independent transitions into colder (and nescence directly preceding and then directly following the then back again into warmer) environments. Scanning the spring leaf flush (fig. 3). We note that both V. japonicum distribution of the leaf habit and forest type characters gives and V. integrifolium belong to the Succotinus clade along the impression of a strong correlation: species living in with V. erosum and V. luzonicum and likely evolved an ev- warmer climates tend to be evergreen and to occupy trop- ergreen habit from a deciduous ancestral state. ical or lucidophyllous forests, while species living in colder Correlation of phenology with climate.Wechosesites climates are generally deciduous and occupy cool temperate in both Taiwan and Japan that spanned freezing and non- forests. This is confirmed by our independent contrast anal- freezing climates, reflecting differences in elevation. In both yses, which showed a significant relationship between leaf countries, there is a pronounced monsoonal climate, with habit and BIO6 (R2 p 0:607, P p :00001) but not BIO15 the heaviest rains falling during the warm summer months. (R2 p 0:037, P p :3785). Likewise, our threshBAYES anal- As precipitation is indistinguishable across the freezing/ ysis yielded a strong positive relationship between leaf habit nonfreezing boundary, we were able to compare phenology and BIO6 (R2 p 0:868, P p 2:2e216) but not BIO15 (R2 p in contrasting temperature regimes while holding rainfall 1:6e25, P p :9647). Similarly, our binary tests strongly fa- patterns constant (fig. 4). vored a model of correlated evolution of leaf habit and hab- As noted above, leaf flushes were coordinated across all itat (uncurated BIO6 binary: DAIC p 58:56, P p 1:22e213; sites and species in both freezing and nonfreezing zones. curated BIO6 binary: DAIC p 73:2, P p 9:63e217). These flushes occurred in the spring as temperatures warmed and monsoonal rains returned. Due to the correlation between temperature and rainfall, it is not possible at this time Discussion to discern which climatic factor the flush is primarily re- sponding to. Ancillary evidence that rainfall may be the pri- Flushing, senescence, and emergence of the deciduous habit. mary driver is provided by our two populations of V. lu- The deciduous habit so characteristic of Northern Hemi- zonicum (fig. 5). Our high-elevation site showed a strong sphere temperate forests involves two distinct behaviors: monsoonal rainfall pattern and, correspondingly, a well- in the spring, there is a rapid and dramatic flush of new defined spring flush; our low-elevation site experienced leaves, and in the fall, there is an equally dramatic, coor- pronounced temperature seasonality but with high rainfall dinated, and often colorful display of leaf senescence. By year-round, and its leaf flushing patterns were considerably monitoring the flushing and senescing of leaves as indepen- more erratic (fig. 5). dent behaviors in both evergreen and deciduous Viburnum In contrast to flushing, we found evidence for a single en- species, we can begin to understand how these two aspects vironmental driver of coordinated senescence and decidu- of phenology relate to different climatic factors and how ousness: in both Taiwan and Japan, our deciduous species their coordination may have evolved along an evergreen- all lived in areas experiencing routinely freezing temper- to-deciduous evolutionary trajectory. atures during the coldest months of January and February The preliminary data on Viburnum phenology presented (fig. 4). In contrast, all of our evergreen species lived in areas here clearly document several patterns. Unfortunately, the where minimum monthly mean temperatures remained very low leaf turnover in the essentially aseasonal tropical above freezing year-round. It is important to note that this site (Borneo) limits interpretation of that data set for now. does not mean that they never experienced occasional frosts, We note only that our single year of data collection provides only that the mean monthly minimum temperatures never some evidence of seasonality to new leaf production (fig. 3) fell below 07C. Fortunately, our sampling allows us to sep- and that periods of leaf flushing are not uncommon in tropical

30 20 20 10 10 All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 0 0 0.015

0.10 0.010 This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:54AM

0.05 0.005

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Japan – deciduous Taiwan – deciduous 600 600 S96 400 400 200 200

20 10 10 5 0 0

0.04 0.02 0.03

0.02 0.01 0.01

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Figure 4: Seasonal drought, freezing, and the coordination of ﬂushing and senescence. Box plots represent mean monthly precipitation (in mm), and line plots represent mean monthly maximum temperature (red) and minimum temperature (blue) in 7C. Density curves below the climate plots were derived from pooling all new (green) and senesced (orange) observation events from all species within that category (e.g., evergreen Japan includes Viburnum japonicum, Viburnum awabuki, and Viburnum erosum). Climate data are from one representative ﬁeld site per category. In cases where we had multiple locations of evergreen or deciduous populations within a country, their climate diagrams were virtually indistinguishable. A 30 40 20

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 10 30 0

20 600 This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:54AM 400 10 200

0 0 J F M A M J J A S O N D B 30

S97 40 20 leaves flushed or senesced 10 30 0

20 600

400 10 200

0 0 J F M A M J J A S O N D Jul 2013 Jan 2014 Jul 2014 Jan 2015 Jul 2015

Figure 5: Variation in phenological behavior in Viburnum luzonicum. Lines on left-hand graphs represent the mean number of leaves gained or lost per observational period across four to six individuals per species; shading represents 51 SE. Green lines represent new leaf flushes, and yellow lines represent leaf senescence. Climate diagrams on the right are similar to figure 4. A, V. luzonicum population in the Danfeng region just outside of Taipei, Taiwan, at an elevation of 120 m. B, V. luzonicum population in the Hehuanshan area of the Central Mountains of Taiwan at an elevation of 2,124 m. S98 The American Naturalist plants more generally, perhaps in response to herbivore pres- evergreen habit. This makes sense given that a coordinated sure (Coley and Barone 1996). We focus instead on our obser- senescence period had already evolved in deciduous species; vations in Taiwan and Japan, where all of our sites are mesic in the absence of freezing, senescence may have become de- but with strong rainfall seasonality. Here it is clear that a pro- layed until it effectively co-occurred with new leaf produc- nounced and coordinated leaf flush is present in all species. tion. A single peak of senescence might have been disrupted Senescence patterns, on the other hand, are quite varied, and in shifting to a more fully evergreen condition, as evidenced leaf drop appears to be less concentrated in time. Although by V. integrifolium and V. japonicum. our evergreen species varied considerably, most exhibited pat- A close comparison of the climates occupied by our ever- terns that were indistinguishable from a random null distribu- green versus deciduous Viburnum species highlights the tion (table A1). The senescence period could be quite extended climatic factor that most likely drove the initial evolution in these species, and the intensity (i.e., the absolute number of coordinated senescence. All of our sites in Taiwan and Ja- of leaves gained or lost per event) of senescence was generally pan experience an annual dry season beginning in October– reduced compared with leaf flushing. Our results are remark- November, but in the evergreen species this is not associated ably similar to an earlier study of phenology in lucidophyllous with coordinated leaf loss; instead, coordinated senescence Japanese forests, where most species produced a single spring appears only when plants are subjected to prolonged freez- flush of leaves but presented at least four different senescence ing. All of our deciduous species and only our deciduous patterns (Nitta and Ohsawa 1997). In marked contrast, our species occur in locations where minimum monthly tem- deciduous species all had short and coordinated senescence peratures routinely dip below freezing in winter (January– periods that commonly matched their spring flush in inten- February; fig. 4). This perfect correlation in our 20-species sity. Senescence also occurred concurrently, in the late fall, data set was corroborated by our Viburnum-wide phyloge- in all of these species. Owing to the high level of homoplasy, netic analyses, which strongly supported minimum tempera- the directionality of character change in Viburnum is still dif- ture of the coldest week (BIO6) as a significant correlate of ficult to infer in some areas, but our sample of nine deciduous shifts in leaf habit across the clade and strongly rejected any species likely captures a minimum of five independent origins influence of precipitation seasonality (BIO15). Taken together, of the deciduous habit (i.e., within the Euviburnum, Pseudo- our data indicate that periods of leaf flushing are universal in tinus, Urceolata, Solenotinus,andSuccotinus clades; Clement Viburnum, perhaps even in the tropical forest species (where et al. 2014; Spriggs et al. 2015). Therefore, this provides a clear more data are clearly needed). A more concentrated burst of case of the convergent evolution of coordinated leaf senes- leaf flushing may have evolved first in lucidophyllous forests cence. that experience monsoonal rainfall patterns with a period of Several species stand out as notable exceptions. Two, Vi- drought in the cool season. Specifically, we suggest that the burnum luzonicum and Viburnum erosum, are what we de- initiation of flushing might be triggered by the onset of spring scribe as leaf exchangers, as their senescence events directly rains (generally in April–May), though given the obvious cor- preceded or were concurrent with their flushes. The range relation, it is difficult to disentangle the relative effects of rain- of V. luzonicum extends to warm temperate forests in south- fall versus temperature or even rainfall versus lengthening ern China and south to the Philippines. In contrast, we sam- photoperiod at that time of year. In the case of leaf senescence, pled V. erosum in the warmest part of its range in Japan; it on the other hand, the environmental trigger seems quite extends northward into colder forests, where it is certainly clear. Our data suggest that a regular and prolonged period deciduous (Hara 1983). Two other exceptions are Vibur- of freezing temperatures was the primary driver of highly co- num integrifolium and Viburnum japonicum, two evergreen ordinated leaf senescence and consequently the emergence of species that exhibit senescence patterns that appear to be a fully deciduous leaf habit. slightly concentrated in two bouts, one preceding and one How do our results square with expectations from the following their annual leaf flush. Interestingly, all of these Wolfe, Axelrod, and Zanne et al. hypotheses outlined above? species are nested within the large eastern Asian Succotinus Although we cannot rule out Wolfe’s hypothesis that dark- clade and have close relatives in cool temperate forests ness at high latitudes favored the evolution of deciduousness (Clement et al. 2014; Spriggs et al. 2015). We infer that these in the Late Cretaceous in some plant lineages (Wolfe and species have extended their ranges back into warmer climates Upchurch 1986; Wolfe 1987), we doubt that this explanation and have shifted (at least in the southern parts of their geo- applies to Viburnum. Our phylogenetic analyses, the Vibur- graphic ranges) in the direction of an evergreen habit. With- num fossil record, and the absolute dates that we have in- out a phylogenetic context, one might consider a leaf ex- ferred (Spriggs et al. 2015) are not consistent with Cretaceous changing phenology to be an intermediate phenotype in the origins of deciduousness in Viburnum. Instead, most origins transition from an evergreen to a deciduous habit. However, probably occurred in the late Miocene, when temperate veg- in the case of Viburnum it seems more likely to be an inter- etation was becoming more common at lower latitudes mediate phenotype in the transition from a deciduous to an (Utescher and Mosbrugger 2007; Pound et al. 2011).

Axelrod’s hypothesis that deciduousness evolved in warm ever, do not support this order of events. One expectation of but monsoonal forests with seasonal drought (Axelrod 1966) such a stepwise model is the existence of one or more ances- does not apply neatly to Viburnum either, considering both trally evergreen clades living in freezing climates, with one the phylogenetic and phenological analyses. Breaking decid- or more deciduous species nested within it. However, there uousness into two components—leaf flushing and leaf senes- is no unambiguous pattern of this type in Viburnum phy- cence—allows us to consider the possibility that these may logeny. Furthermore, in the species that we monitored, ever- have had separate evolutionary causes. Temporally restricted greens never occur in forests that experience predictable and coordinated leaf flushing might have evolved in response freezing. On these grounds we judge the Zanne et al. (2014) to the strong rainfall seasonality of monsoonal climates, with explanation to be unlikely in this case, although the hypoth- the onset of leafing possibly controlled by the onset of rains esis could be salvaged by imagining the systematic extinc- or increasing temperatures in the spring. Restricted and co- tion of evergreen species in cold climates. ordinated leaf senescence, on the other hand, might have An alternative model of evolution. As we have empha- been driven by exposure to freezing temperatures. Drought sized, the previously proposed hypotheses envision a se- is potentially also a factor in senescence, as Axelrod sup- quence of steps ending in the association of cold climates posed, but as we have shown, our evergreen species also ex- with a deciduous habit. A slightly different view seems much perience similar periods of significant drought in the fall simpler in the face of the evidence that we have presented for without dropping their leaves. Freezing, not drought, seems Viburnum. We might consider the possibility that the decid- to be most directly related to coordinated senescence in our uous habit and in particular the coordinated senescence of deciduous species. leaves in the fall evolved in some plant lineages directly in re- We concur with a simple physiological explanation for a sponse to freezing. By “directly” we mean that the senescence direct link between freezing and deciduousness: a coordi- component of the deciduous habit evolved in situ as popula- nated period of senescence allows plants to remobilize key tions were subjected to a gradually cooling climate and the nutrients prior to the inevitable death of leaves by predict- establishment of a prolonged and predictable period of freez- able freezing (Feild et al. 2001; Keskitalo et al. 2005; Niine- ing during the winter. We envision that this change in cli- mets and Tamm 2005). In all of the varied senescence pat- mate took place in fits and starts over the course of thousands terns observed in our evergreen species, it is worth noting to millions of years (Zachos 2001; Herbert et al. 2016), and that no peaks in senescence coincide with the onset of the this is a timescale during which the final steps in the evolu- dry season; instead, they are scattered throughout the rainy tion of a fully deciduous habit could certainly have evolved. season. It appears that viburnums can weather dry periods Under this hypothesis the deciduous habit is not a preadap- by either tolerating lower water potentials or maintaining tation that first originated as the solution to some other prob- low stomatal conductance. A modest drought tolerance com- lem, nor did it evolve in lineages that got there first and bined with interannual variation in the strength of drought adapted later. Instead, we are suggesting that climate change periods may render drought a weaker selection pressure for and evolutionary adaptation occurred together, effectively in coordinated senescence, as many leaves may be able to persist lockstep with one another. This scenario is especially realistic and continue to function during the following rainy season. It for an adaptation related to leafing phenology, which we is important to note, of course, that there are evergreen woody have shown to be quite sensitive to climate even within a sin- plants in the northern temperate zone (e.g., Rhododenron, gle species across its range (fig. 5). Ilex) and that these species instead produce freezing-tolerant The distinction we are making between stepwise and leaves with longer life spans. With very few exceptions (fig. 1), lockstep may seem like a subtle one, but we think it is cru- Viburnum has not evolved this alternative strategy, sug- cially important in considering biome assembly and the gesting that deciduousness was the more evolutionarily acces- causes of concerted convergence. At the very least it war- sible adaptation in this lineage. Investigating the very few rants greater care in our use of phrases like “moved into” instances of evergreen viburnums that experience freezing or “shifted into” in relation to biome transitions. Likewise, (e.g., Viburnum rhytidophyllum, which is nested within a cold the Zanne et al. (2014) terms “climate first” and “trait first” temperate deciduous clade) might provide insights into the quite explicitly describe a stepwise process—in other words, relative advantages of these two distinct ecological strategies that one thing happened first and the other thing happened to the cold temperate zone. later. In the case of the emergence of the temperate decidu- In proposing that deciduousness evolved directly as an ous biome, it appears that a widespread and floristically rich adaptation to predictable and prolonged freezing, we might warm evergreen forest (the boreotropical flora sensu Wolfe appear to be more aligned with the hypothesis of Zanne 1975; Tiffney 1985a,1985b) was subjected to late Miocene et al. (2014). These authors concluded that many lineages cooling, eventually resulting in a climate with annual periods first became established in the freezing zone as evergreen of freezing temperatures. The many existing lineages of woody plants and that deciduousness evolved later. Our data, how- plants in these forests would have experienced this in situ di-

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:54 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S100 The American Naturalist rectional trend. Some of them probably fared poorly, either experienced by all individuals simultaneously, and routine going extinct or shifting their ranges accordingly, but many exposure to prolonged cold could have tightly coordinated others simply kept up with the changing times and stayed the senescence period to a more condensed time period across put. the community. What factors determine the winners and losers as large Consilience and integration in phylogenetic biology.How geographical regions gradually become reconfigured and do we justify our lockstep model? If we approached this new biomes emerge? In our case, the deciduous habit might with only our phylogenetic tools, we would not be entirely have been more accessible to species that already produced satisfied. The high lability of both the deciduous habit and resting buds and that flushed their leaves annually in re- climate makes it difficult to infer with any confidence many sponse to variation in precipitation or temperature. The of the deeper and potentially most relevant transitions key step in such lineages would be the consolidation of leaf (fig. 1). We do know that deciduousness and freezing cli- senescence, and it is this, we argue, that likely was a direct mate are very tightly correlated across our tree, so much response to freezing, not drought. Indeed, such predispo- so that changes in both of them very often co-occur along sitions likely explain the repeated evolution of the decidu- the same branches, making it difficult to establish cause ous habit within particular lineages and in general the clus- and effect (Baum and Donoghue 2001). If there had in fact tering of the origin of certain traits in particular regions of been a repeated order of events, history has largely erased its the tree of life (Edwards and Donoghue 2013; Donoghue signal. We might also be frustrated by our inability to ana- and Edwards 2014). Viburnum provides an excellent exam- lyze both characters as continuous variables, which would ple, with possibly as many as 10 separate origins of decidu- have provided a more direct test of a simultaneous, lockstep ousness, but we suspect that other familiar temperate evolutionary model. Transforming habitat into a continu- lineages that were present in boreotropical forests (e.g., ous climate variable was possible by using the CliMond da- Acer, Carya, Nyssa, Hamamelis, Sassafras, Cornus; Tiffney tabase, but assigning continuous values to species along an 1985a), when analyzed carefully, will show similar patterns evergreen-deciduous continuum is logistically and concep- of in situ convergence. tually more difficult and would require detailed phenolog- It is certainly the case that biome assembly also reflects ical studies of every species, not just the 20 that we have other processes, including preadaptation, habitat filtering presented here. (cf. Ackerly 2004), and even occasional long-distance dis- But what if we consider these conclusions together with persal events (Pennington and Dick 2004; Crisp et al. 2009). the results from the phenological study? Here we also see We also do not doubt that all of these pathways have con- evidence of a very tight association between freezing in par- tributed to the assembly of temperate deciduous forests. ticular and the deciduous habit (fig. 4) and of the extreme However, we believe that these other phenomena may have lability of phenological patterns, even among populations had a relatively minor influence as compared to the in situ within a single species (fig. 5). These observations are com- adaptation of multiple lineages to gradual climate change. pletely consistent with the strong correlation across the Vi- We suggest that such concurrent responses may largely un- burnum phylogeny, and together these lines of evidence derpin the assembly of new biomes and that this process suggesttousthatthe“simultaneous” changes that we see may be responsible for the extremely high number of ori- in phylogenetic analyses may often be real simultaneous gins of certain traits. changes rather than artifacts of extinction. If individual spe- Although we have concentrated here on direct shifts be- cies can be evergreen in one part of their range and decid- tween mesic tropical and temperate forests, we fully appre- uous in another (as in V. erosum and V. luzonicum), grad- ciate that there have likely been other biome pathways and ual and concurrent emergence of the deciduous habit as a in particular that some members of cool temperate forests freezing climate becomes established begins to seem highly may have been derived within lineages that occupied sea- likely. sonally dry tropical forests (e.g., possibly Celtis). Such a The study of phylogenetic comparative methods appears pathway would be far more consistent with a stepwise model, to be shifting into a more introspective phase, with renewed where deciduousness evolved first in response to one vari- scrutiny of the limitations of our inferences (Edwards et al. able (drought) and served later as a preadaptation for an- 2015; Maddison and FitzJohn 2015; Pennell et al. 2015; other (cold). Nevertheless, in both pathways (i.e., with an- Wright et al. 2015). Critiques typically end with a call for cestors in either mesic forests or dry forests) the final steps better models, and we agree completely that there is much in the process may have been similar. In both cases, senes- room for improvement. However, at the end of the day, we cence may have been relatively uncoordinated at first, as in need to remind ourselves of the extreme difficulty of con- the case of dry forests, where leaves tend to senesce at differ- vincingly inferring evolutionary events in the distant past ent times depending largely on individual plant water status with data usually only from extant organisms, especially when (e.g., Reich and Borchert 1984). However, freezing events are convergence is rampant and densely clustered in particular

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:54 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Consilience, Convergence, and Phylogeny S101 clades. Under these circumstances, phylogenetic insights lenge of inferring evolutionary history. Immersion in the will be most powerful when interpreted alongside other, messy details of particular lineages, far from being a distrac- quite independent lines of evidence until (hopefully) con- tion, may be the most fertile ground for achieving consilience— silience emerges (Weber and Agrawal 2012; Olson and the moment when multiple, sometimes weakly supported Arroyo-Santos 2015). findings all point to the same conclusion. From this perspective, we believe that research conducted at the scale of Viburnum will continue to be highly productive, precisely because the results of multiple inde- Acknowledgments pendent analyses can most effectively be brought into alignment in such model lineages. For example, the data pre- We thank A. Agrawal for his invitation to participate in the sented here are relevant to questions about the deciduous 2016 American Society of Naturalists Vice Presidential leaf habit, but they are also directly connected to our studies Symposium and for his patience in handling our manu- of leaf form and function (Chatelet et al. 2013; Scoffoni et al. script. We appreciated insightful comments by D. Ackerly, 2016), growth architecture and leaf life span (Edwards et al. T. Pennington, and two other reviewers, which greatly im- 2014), and even flowering phenology (L. Garrison, M. Don- proved our manuscript. We are also grateful to J. Pereira oghue, and E. Edwards, unpublished manuscript) and wood and the staff of Kinabalu National Park for their help with anatomy (D. Chatelet, M. Donoghue, and E. Edwards, un- phenological monitoring, E. Johnson for help with data published manuscript). Consequently, we have the poten- cleansing and management, and J. de Vos, J. Gauthier, M. tial to establish connections between what might at first ap- Olson, and members of the Edwards and Donoghue labs pear to be unrelated phenomena. It is this form of synthesis for many thoughtful discussions. This work was supported that we find most inspiring and most likely to provide new by National Science Foundation grants IOS-1257262 to conceptual breakthroughs as we face the formidable chal- E.J.E. and IOS-1256706 to M.J.D.

APPENDIX

Supplementary Table Table A1: Significance of leaf flushing and senescence patterns in 20 species of Viburnum along a latitudinal gradient in Asia Total Mean Julian day Mean Julian day Species Location Leaf habit Event leaves (observed) (null) P V. awabuki Japan Evergreen Leaf flush 41 89 194 2.60E-12 V. awabuki Japan Evergreen Leaf drop 48 176 195 .239 V. betulifolium Taiwan Deciduous Leaf flush 211 118 187 3.20E-25 V. betulifolium Taiwan Deciduous Leaf drop 195 336 187 1.90E-88 V. bitchiuense Japan Deciduous Leaf flush 353 102 207 4.00E-102 V. bitchiuense Japan Deciduous Leaf drop 523 305 207 2.60E-115 V. clemensiae Borneo Evergreen Leaf flush 68 110 169 4.20E-06 V. clemensiae Borneo Evergreen Leaf drop 15 183 170 4.80E-01 V. dilatatum Japan Deciduous Leaf flush 353 106 189 6.57E-39 V. dilatatum Japan Deciduous Leaf drop 352 311 189 1.39E-96 V. erosum Japan Leaf exchanger Leaf flush 240 91 195 3.54E-65 V. erosum Japan Leaf exchanger Leaf drop 205 276 195 2.67E-34 V. furcatum Japan Deciduous Leaf flush 229 134 205 1.41E-32 V. furcatum Japan Deciduous Leaf drop 341 311 205 6.47E-106 V. integrifolium Taiwan Evergreen Leaf flush 154 126 187 4.65E-12 V. integrifolium* Taiwan Evergreen Leaf drop 109 170 188 2.00E-02 V. japonicum Japan Evergreen Leaf flush 134 81 195 3.01E-47 V. japonicum* Japan Evergreen Leaf drop 95 156 195 3.28E-04 V. luzonicum Taiwan Leaf exchanger Leaf flush 205 89 187 4.31E-56 V. luzonicum Taiwan Leaf exchanger Leaf drop 133 154 187 3.37E-07 V. odoratissimum Taiwan Evergreen Leaf flush 220 151 187 1.35E-08 V. odoratissimum Taiwan Evergreen Leaf drop 91 180 186 4.26E-01

Table A1 (Continued) Total Mean Julian day Mean Julian day Species Location Leaf habit Event leaves (observed) (null) P V. parvifolium Taiwan Deciduous Leaf flush 405 99 187 2.37E-86 V. parvifolium Taiwan Deciduous Leaf drop 396 332 187 1.95E-174 V. phlebotrichum Japan Deciduous Leaf flush 198 97 202 4.50E-62 V. phlebotrichum Japan Deciduous Leaf drop 196 306 202 1.14E-57 V. propinquum Taiwan Evergreen Leaf flush 295 122 187 8.12E-33 V. propinquum* Taiwan Evergreen Leaf drop 125 165 187 4.10E-02 V. sieboldii Japan Deciduous Leaf flush 333 100 215 3.70E-105 V. sieboldii Japan Deciduous Leaf drop 332 277 215 5.33E-27 V. sympodiale Taiwan Deciduous Leaf flush 508 121 187 3.73E-53 V sympodiale Taiwan Deciduous Leaf drop 399 328 187 1.31E-156 V. taitoense Taiwan Evergreen Leaf flush 273 105 187 9.60E-47 V. taitoense Taiwan Evergreen Leaf drop 109 180 187 7.10E-01 V. taiwanianum Taiwan Evergreen Leaf flush 326 145 187 1.77E-11 V. taiwanianum Taiwan Evergreen Leaf drop 245 208 187 4.00E-01 V. urceolatum Japan Deciduous Leaf flush 96 111 203 1.62E-19 V. urceolatum Japan Deciduous Leaf drop 96 271 204 1.95E-09 V. vernicosum Borneo Evergreen Leaf flush 126 117 169 3.25E-07 V. vernicosum Borneo Evergreen Leaf drop 17 199 170 3.60E-01 Note: Species in boldface exhibited leaf senescence patterns statistically indistinguishable from our generated null distributions. Evergreen species marked with an asterisk exhibited senescence patterns significantly different from the null at P ! .05. All species exhibited highly significant leaf flushes, occurring earlier in the year than the null expectation.

Literature Cited Christin, P. A., R. P. Freckleton, and C. P. Osborne. 2010. Can phylogenetics identify C4 origins and reversals? Trends in Ecology Ackerly, D. D. 2004. Adaptation, niche conservatism, and conver- and Evolution 25:403–409. gence: comparative studies of leaf evolution in the California chap- Christin, P. A., C. P. Osborne, D. S. Chatelet, J. T. Columbus, G. arral. American Naturalist 163:654–671. Besnard, T. R. Hodkinson, L. M. Garrison, M. S. Vorontsova, and

Agrawal, A. A. 2017. Toward a predictive framework for convergent E. J. Edwards. 2013. Anatomical enablers and the evolution of C4 evolution: integrating natural history, genetic mechanisms, and photosynthesis in grasses. Proceedings of the National Academy consequences for the diversity of life. American Naturalist 190 of Sciences of the USA 110:1381–1386. (suppl.):S1–S12. Clement, W. L., M. Arakaki, P. W. Sweeney, E. J. Edwards, and M. J. Agrawal, A. A., and M. Fishbein. 2006. Plant defense syndromes. Donoghue. 2014. A chloroplast tree for Viburnum (Adoxaceae) Ecology 87:S132–S149. and its implications for phylogenetic classiﬁcation and character Arakaki, M., P. A. Christin, R. Nyffeler, A. Lendel, U. Eggli, R. M. evolution. American Journal of Botany 101:1029–1049. Ogburn, E. Spriggs, M. J. Moore, and E. J. Edwards. 2011. Contem- Coley, P. D., and J. A. Barone. 1996. Herbivory and plant defenses in poraneous and recent radiations of the world’s major succulent tropical forests. Annual Review of Ecology and Systematics 27: plant lineages. Proceedings of the National Academy of Sciences 305–335. of the USA 108:8379–8384. Crisp, M. D., M. T. K. Arroyo, L. G. Cook, M. A. Gandolfo, G. J. Axelrod, D. I. 1966. Origin of deciduous and evergreen habits in tem- Jordan, M. S. McGlone, P. H. Weston, M. Westoby, P. Wilf, and perate forests. Evolution 20:1–15. H. P. Linder. 2009. Phylogenetic biome conservatism on a global Bailey, I. W., and E. W. Sinnott. 1916. The climatic distribution of cer- scale. Nature 458:754–756. tain types of angiosperm leaves. American Journal of Botany 3:24–39. Donoghue, M. J. 1981. Growth patterns in woody plants with exam- Baum, D. A., and M. J. Donoghue. 2001. A likelihood framework for ples from the genus Viburnum. Arnoldia 41:2–23. the phylogenetic analysis of adaptation. Pages 22–44 in S. H. Or- Donoghue, M. J., and E. J. Edwards. 2014. Biome shifts and niche evo- zack and E. Sober, eds. Adaptation and optimality. Cambridge Uni- lution in plants. Annual Review of Ecology, Evolution, and System- versity Press, Cambridge. atics 45:547–572. Brown, J. L. 2014. SDMtoolbox: a python-based GIS toolkit for land- Donoghue, M. J., and R. H. Ree. 2000. Homoplasy and developmen- scape genetic, biogeographic and species distribution model analy- tal constraint: a model and an example from plants. American Zo- ses. Methods in Ecology and Evolution 5:694–700. ologist 40:759–769. Chabot, B., and D. Hicks. 1982. The ecology of leaf life spans. An- Eaton, D. A. R., E. L. Spriggs, B. Park, and M. J. Donoghue. 2017. nual Review of Ecology and Systematics 13:229–259. Misconceptions on missing data in RAD-seq phylogenetics with a Chatelet, D. S., W. L. Clement, L. Sack, M. J. Donoghue, and E. J. deep-scale example from ﬂowering plants. Systematic Biology 66: Edwards. 2013. The evolution of photosynthetic anatomy in Vi- 399–412. burnum (Adoxaceae). International Journal of Plant Sciences 174: Edwards, E. J., D. S. Chatelet, B. Chen, J. Y. Ong, S. Tagane, J. Kane- 1277–1291. mitsu, K. Tagawa, et al. 2017. Data from: Convergence, consil-

ience, and the evolution of temperate deciduous forests. American Maddison, W. P., and R. G. FitzJohn. 2015. The unsolved challenge Naturalist, Dryad Digital Respository, http://dx.doi.org/10.5061 to phylogenetic correlation tests for categorical characters. Sys- /dryad.k505s. tematic Biology 64:127–136. Edwards, E. J., D. S. Chatelet, L. Sack, and M. J. Donoghue. 2014. Meyer, J. R., D. T. Dobias, J. S. Weitz, J. E. Barrick, R. T. Quick, and Leaf life span and the leaf economic spectrum in the context of R. E. Lenski. 2012. Repeatability and contingency in the evolution whole plant architecture. Journal of Ecology 102:328–336. of a key innovation in phage lambda. Science 335:428–432. Edwards, E. J., J. M. de Vos, and M. J. Donoghue. 2015. Doubtful Morton, C. V. 1933. The Mexican and Central American species of pathways to cold tolerance in plants. Nature 521:E5–E6. Viburnum. Contributions of the US National Herbarium 26:339– Edwards, E. J., and M. J. Donoghue. 2013. Is it easy to move and easy 366. to evolve? evolutionary accessibility and adaptation. Journal of Ex- Murphy, P. G., and A. E. Lugo. 1986. Ecology of tropical dry forest. perimental Botany 64:4047–4052. Annual Review of Ecology and Systematics 17:67–88.

Edwards, E. J, C. Osborne, C. Stromberg, S. Smith, C4 Grasses Con- Niinemets, Ü., and Ü. Tamm. 2005. Species differences in timing of fi sortium. 2010. The origins of C4 grasslands: integrating evolution- leaf fall and foliage chemistry modify nutrient resorption ef ciency ary and ecosystem science. Science 328:587–591. in deciduous temperate forest stands. Tree Physiology 25:1001–1014. Edwards, E. J., E. L. Spriggs, D. S. Chatelet, and M. J. Donoghue. Nitta, I., and M. Ohsawa. 1997. Leaf dynamics and shoot phenology 2016. Unpacking a century-old mystery: winter buds and the lat- of eleven warm-temperate evergreen broad-leaved trees near their itudinal gradient in leaf form. American Journal of Botany 103: northern limit in central Japan. Plant Ecology 130:71–88. 975–978. Olson, M. E., and A. Arroyo-Santos. 2015. How to study adaptation Feild, T. S., D. W. Lee, and N. M. Holbrook. 2001. Why leaves turn (and why to do it that way). Quarterly Review of Biology 90:167– red in autumn: the role of anthocyanins in senescing leaves of red- 191. osier dogwood. Plant Physiology 127:566–574. Orme, D. 2013. caper: comparative analysis of phylogenetics and Felsenstein, J. 1985. Phylogenies and the comparative method. evolution in R. R package version 0.5.2. http://cran.r-project.org American Naturalist 125:1–15. /packagepcaper. ———. 2012. A comparative method for both discrete and contin- Pagel, M. 1994. Detecting correlated evolution on phylogenies: a gen- uous characters using the threshold model. American Naturalist eral method for the comparative analysis of discrete characters. 179:145–156. Proceedings of the Royal Society B 255:37–45. Fenster, C. B., W. S. Armbruster, P. Wilson, M. R. Dudash, and J. D. Paradis, E., J. Claude, and K. Strimmer. 2004. APE: analyses of phy- Thomson. 2004. Pollination syndromes and floral specialization. logenetics and evolution in R language. Bioinformatics 20:289–290. Annual Review of Ecology, Evolution, and Systematics 35:375– Pennell, M. W., R. G. FitzJohn, W. K. Cornwell, and L. J. Harmon. 403. 2015. Model adequacy and the macroevolution of angiosperm Givnish, T. 2002. Adaptive significance of evergreen vs. deciduous functional traits. American Naturalist 186:E33–E50. leaves: solving the triple paradox. Silva Fennica 36:703–743. Pennington, R. T., and C. W. Dick. 2004. The role of immigrants in Hancock, L., and E. J. Edwards. 2014. Phylogeny and the inference of the assembly of the South American rainforest tree flora. Philo- evolutionary trajectories. Journal of Experimental Botany 65:3491– sophical Transactions of the Royal Society B 359:1611–1622. 3498. Peppe, D., D. Royer, B. Cariglino, S. Oliver, S. Newman, E. Leight, R. Hara, H. 1983. A revision of the Caprifoliaceae of Japan with refer- Enikolopov, et al. 2011. Sensitivity of leaf size and shape to cli- ence to allied plants in other districts and the Adoxaceae. Acade- mate: global patterns and paleoclimatic applications. New Phytol- mia Scientific, Tokyo. ogist 190:724–739. Herbert, T. D., K. T. Lawrence, A. Tzanova, L. C. Peterson, R. Caballero- Pound, M. J., A. M. Haywood, U. Salzmann, J. B. Riding, D. J. Lunt, Gill, and C. S. Kelly. 2016. Late Miocene global cooling and the rise of and S. J. Hunter. 2011. A Tortonian (Late Miocene, 11.61–7.25 modern ecosystems. Nature Geosciences 9:843–847. Ma) global vegetation reconstruction. Palaeogeography, Palaeocli- Hijmans, R. J., and J. van Etten. 2012. raster: geographic analysis and matology, Palaeoecology 300:29–45. modeling with raster data. R package version 2.0-12. http://cran.r Rehder, A. 1908. The viburnums of eastern Asia. Pages 105–116 in -project.org/packagepraster. C. S. Sargent, ed. Trees and shrubs. II. Vol. 2. Houghton Mifflin, Kahle, D., and H. Wickham. 2013. ggmap: spatial visualization with Boston. ggplot2. R Journal 5:144–161. Reich, P. B., and R. Borchert. 1984. Water stress and tree phenology Kern, J. H. 1951. The genus Viburnum (Caprifoliaceae) in Malaysia. in a tropical dry forest in the lowlands of Costa Rica. Journal of Reinwardtia 1:107–170. Ecology 72:61–74. Keskitalo, J., G. Bergquist, P. Gardeström, and S. Jansson. 2005. A Reich, P. B., M. B. Walters, and D. S. Ellsworth. 1992. Leaf life-span cellular timetable of autumn senescence. Plant Physiology 139: in relation to leaf, plant, and stand characteristics among diverse 1635–1648. ecosystems. Ecological Monographs 62:365–392. Kikuzawa, K. 1991. A cost-benefit analysis of leaf habit and leaf lon- Revell, L. J. 2012. phytools: an R package for phylogenetic compar- gevity of trees and their geographical pattern. American Naturalist ative biology (and other things). Methods in Ecology and Evolu- 138:1250–1263. tion 3:217–223. Killip, E. P., and A. C. Smith. 1931. The South American species of Reyes, E., H. Sauquet, and S. Nadot. 2016. Perianth symmetry changed Viburnum. Bulletin of the Torrey Botanical Club 57:245–258. at least 199 times in angiosperm evolution. Taxon 65:945–964. Kriticos, D. J., B. L. Webber, A. Leriche, N. Ota, J. Bathols, I. Mac- Royer, D. L., and P. Wilf. 2006. Why do toothed leaves correlate with adam, and J. K. Scott. 2012. CliMond: global high resolution his- cold climates? gas exchange at leaf margins provides new insights torical and future scenario climate surfaces for bioclimatic model- into a classic paleotemperature proxy. International Journal of Plant ling. Methods in Ecology and Evolution 3:53–64. Sciences 167:11–18.

Sage, R. F., P. A. Christin, and E. J. Edwards. 2011. The C4 plant Weber, M. G., and A. A. Agrawal. 2012. Phylogeny, ecology, and the lineages of planet Earth. Journal of Experimental Botany 62: coupling of comparative and experimental approaches. Trends in 3155–3169. Ecology and Evolution 27:394–403. Sanderson, M. J., and L. Hufford, eds. 1996. Homoplasy: the recur- Weinreich, D. M., N. F. Delaney, M. A. DePristo, and D. L. Hartl. rence of similarity in evolution. Academic Press, San Diego, CA. 2006. Darwinian evolution can follow only very few mutational Schluter, D., T. Price, A. O. Mooers, and D. Ludwig. 1997. Likeli- paths to fitter proteins. Science 312:111–114. hood of ancestor states in adaptive radiation. Evolution 51:1699. Wickham, H. 2016. Getting started with ggplot2. Pages 11–31 in Schmerler, S., W. Clement, D. Chatelet, J. Beaulieu, L. Sack, M. J. ggplot2. Springer, New York. Donoghue, and E. J. Edwards. 2012. Evolution of leaf form cor- Wilson, P., A. D. Wolfe, W. S. Armbruster, and J. D. Thomson. 2007. relates with tropical-temperate transitions in Viburnum (Adoxaceae). Constrained lability in floral evolution: counting convergent origins Proceedings of the Royal Society B 279:3905–3913. of hummingbird pollination in Penstemon and Keckiella. New Phy- Schultz, T. R., R. B. Cocroft, and G. A. Churchill. 1996. The recon- tologist 176:883–890. struction of ancestral character states. Evolution 50:504–511. Wolfe, J. 1995. Paleoclimatic estimates from Tertiary leaf assem- Scoffoni, C., D. S. Chatelet, J. Pasquet-Kok, M. Rawls, M. J. Donoghue, blages. Annual Review of Earth and Planetary Sciences 23:119– E. J. Edwards, and L. Sack. 2016. Hydraulic basis for the evolution of 142. photosynthetic productivity. Nature Plants 2:16072. Wolfe, J. A. 1975. Some aspects of plant geography of the Northern Spriggs, E. L., W. L. Clement, P. W. Sweeney, S. Madriñán, E. J. Hemisphere during the Late Cretaceous and Tertiary. Annals of Edwards, and M. J. Donoghue. 2015. Temperate radiations and the Missouri Botanical Garden 62:264–279. dying embers of a tropical past: the diversification of Viburnum. ———. 1987. Late Cretaceous-Cenozoic history of deciduousness New Phytologist 207:340–354. and the terminal Cretaceous event. Paleobiology 13:215–226. Tang, C. Q. 2010. Subtropical montane evergreen broad-leaved forests Wolfe, J. A., and G. R. Upchurch. 1986. Vegetation, climatic and flo- of Yunnan, China: diversity, succession dynamics, human influ- ral changes at the Cretaceous-Tertiary boundary. Nature 324:148– ence. Frontiers in Earth Science China 4:22–32. 152. Tang, C. Q., C.-R. Chiou, C.-T. Lin, J.-R. Lin, C.-F. Hsieh, J.-W. Tang, Wright, A. M., K. M. Lyons, and M. C. Brandley. 2015. Which came W.-H. Su, and X. Hou. 2013. Plant diversity patterns in subtropical first: the lizard or the egg? robustness in phylogenetic reconstruc- evergreen broad-leaved forests of Yunnan and Taiwan. Ecological tion of ancestral states. Journal of Experimental Zoology 324B:504– Research 28:81–92. 516. Tiffney, B. H. 1985a. Perspectives on the origin of the floristic sim- Wright, I. J., P. B. Reich, M. Westoby, D. D. Ackerly, Z. Baruch, F. ilarity between eastern Asia and eastern North America. Journal Bongers, J. Cavender-Bares, et al. 2004. The worldwide leaf eco- of the Arnold Arboretum 66:73–94. nomics spectrum. Nature 428:821–827. ———. 1985b. The Eocene North Atlantic land bridge: its impor- Wright, S. 1934. An analysis of variability in number of digits in an tance in Tertiary and modern phytogeography of the Northern inbred strain of guinea pigs. Genetics 19:506–536. Hemisphere. Journal of the Arnold Arboretum 66:243–273. Yang, Q.-E., and V. Malécot. 2011. Viburnum. Pages 570–611 in Z.-Y. Tiffney, B. H., and S. R. Manchester. 2001. The use of geological and Wu, P. H. Raven, and D. Y. Hong, eds. Flora of China. Vol. 19. paleontological evidence in evaluating plant phylogeographic hy- Lentibulariaceae through Dipsacaceae. Missouri Botanical Gar- potheses in the Northern Hemisphere Tertiary. International den, St. Louis. Journal of Plant Sciences 162(suppl.):S3–S17. Zachos, J., M. Pagani, L. Sloan, E. Thomas, and K. Billups. 2001. Utescher, T., and V. Mosbrugger. 2007. Eocene vegetation patterns Trends, rhythms, and aberrations in global climate 65 Ma to pres- reconstructed from plant diversity—a global perspective. Palaeo- ent. Science 292:686–693. geography, Palaeoclimatology, Palaeoecology 247:243–271. Zanne, A. E., D. C. Tank, W. K. Cornwell, J. M. Eastman, S. A. Smith, van Ommen Kloeke, A. E. E., J. C. Douma, J. C. Ordoñez, P. B. Reich, R. G. FitzJohn, D. J. McGlinn, et al. 2014. Three keys to the radia- and P. M. van Bodegom. 2012. Global quantification of contrasting tion of angiosperms into freezing environments. Nature 506:89–92. leaf life span strategies for deciduous and evergreen species in response to environmental conditions. Global Ecology and Biogeog- raphy 21:224–235. Symposium Editor: Anurag A. Agrawal

Symposium Geographical Variation in Community Divergence: Insights from Tropical Forest Monodominance by Ectomycorrhizal Trees*

Tadashi Fukami,1,† Mifuyu Nakajima,1 Claire Fortunel,2 Paul V. A. Fine,3 Christopher Baraloto,4 Sabrina E. Russo,5 and Kabir G. Peay1

1. Department of Biology, Stanford University, Stanford, California 94305; 2. Department of Ecology and Evolutionary Biology, University of California, Los Angeles, California 90095; 3. Department of Integrative Biology and University and Jepson Herbaria, University of California, Berkeley, California 94720; 4. L’Institut National de la Recherche Agronomique, Unité Mixte de Recherche Ecologie des Forêts de Guyane, Kourou, French Guiana; and Department of Biological Sciences and International Center for Tropical Botany, Florida International University, Miami, Florida 33199; 5. School of Biological Sciences, University of Nebraska, Lincoln, Nebraska 68588 Online enhancements: supplementary material. Dryad data: http://dx.doi.org/10.5061/dryad.c0kr7. abstract: Convergence occurs in both species traits and commu- Introduction nity structure, but how convergence at the two scales inﬂuences each At the species level, convergence is deﬁned by distantly re- other remains unclear. To address this question, we focus on tropical lated species sharing similar traits, but convergence is also forest monodominance, in which a single, often ectomycorrhizal (EM) tree species occasionally dominates forest stands within a landscape possible at the community level, in which distantly located otherwise characterized by diverse communities of arbuscular mycor- communities develop to have similar species abundance dis- rhizal (AM) trees. Such monodominance is a striking potential ex- tribution, functional group composition, and other aspects ample of community divergence resulting in alternative stable states. of community structure (Fukami 2009). These forms of However, it is observed only in some tropical regions. A diverse suite convergence are thought to result from the predictable re- of AM and EM trees locally codominate forest stands elsewhere. We sponse of independently evolved species (e.g., Reich et al. develop a hypothesis to explain this geographical difference using a 1997; Conte et al. 2012) and separately assembled commu- simulation model of plant community assembly. Simulation results suggest that in a region with a few EM species (e.g., South America), nities (e.g., Samuels and Drake 1997; Li et al. 2016) to simi- EM trees experience strong selection for convergent traits that match lar environmental conditions. Identifying factors that pro- the abiotic conditions of the environment. Consequently, EM species mote or prevent convergence is therefore fundamental to successfully compete against other species to form monodominant the understanding of predictability at both the species and stands via positive plant-soil feedbacks. By contrast, in a region with community levels of biological organization (Stern 2013). many EM species (e.g., Southeast Asia), species maintain divergent Despite this duality of convergence, the link between trait traits because of complex plant-soil feedbacks, with no species having convergence and community convergence remains poorly traits that enable monodominance. An analysis of plant trait data from Borneo and Peruvian Amazon was inconclusive. Overall, this work investigated (Cavender-Bares et al. 2009). It may seem ob- highlights the utility of geographical comparison in understanding the vious that trait convergence automatically translates into relationship between trait convergence and community convergence. community convergence (Melville et al. 2006). As we argue in this article, however, the link may not be so straightfor- Keywords: community assembly, mycorrhizae, plant-soil feedback, ward, particularly when multiple lineages of species con- plant traits, priority effects, species pools. verge in traits among themselves but diverge from members of other lineages, which can promote divergence—rather than convergence—of community structure. How does community convergence and divergence take * This issue originated as the 2016 Vice Presidential Symposium presented at place? Insights on this question can be found as early as the annual meetings of the American Society of Naturalists. † Clements, who developed the climax concept of plant suc- Corresponding author; e-mail: [email protected]. cession (e.g., Clements 1936). According to this well-known ORCIDs: Fukami, http://orcid.org/0000-0001-5654-4785; Nakajima, http:// orcid.org/0000-0001-7778-2029. concept, communities converge to a predictable species com- Am. Nat. 2017. Vol. 190, pp. S105–S122. q 2017 by The University of position determined by the abiotic environment (but see Chicago. 0003-0147/2017/190S1-57233$15.00. All rights reserved. Gleason 1927). More recently, increasing evidence suggests DOI: 10.1086/692439 that community divergence—in the form of alternative sta-

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:57 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S106 The American Naturalist ble states—may be more common than previously thought to several thousand hectares (Peh et al. 2011). The term (Schröder et al. 2005), but understanding when alternative “monodominance” has been variously defined in the trop- stable states emerge remains a major challenge for commu- ical forest literature, but here we consider those cases in nity ecologists (Petraitis 2013). Local communities are con- which one species contributes more than 60% of total basal sidered to be in alternative stable states when they diverge area (Hart et al. 1989) or more than 25% of total stem den- in species composition, even though the communities share sity of a forest stand that is surrounded by mixed forest. the same environmental conditions and the same species There are a number of hypothesized mechanisms by pool (Fukami 2015). This divergence is driven by priority ef- which monodominance can emerge (Torti et al. 2001; Fred- fects, in which the order or initial abundance in which spe- erickson et al. 2005; McGuire et al. 2008; Peh et al. 2011), cies arrive influences the effects that species have on one an- but one factor that is frequently associated with mono- other in local communities (Palmgren 1926; Sutherland dominance is the mycorrhizal status of the plants (Hart 1974; Drake 1991). As factors determining when alternative et al. 1989). Tree species that host any type of mycorrhiza stable states occur, more attention has been paid to the abi- can form monodominant stands. For example, multiple spe- otic conditions of local habitats than to the trait values of po- cies from primarily AM plant families (Brundrett 2009)— tential colonists in the regional species pool (but see, e.g., such as the Apocynaceae, Burseraceae, Euphorbiaceae, Faba- Fox 1987; Fargione et al. 2003). Consequently, the connec- ceae, Lauraceae, Moraceae (reviewed in Peh et al. 2011), and tion between convergence in species traits and in community Rubiaceae (Frederickson et al. 2005)—form monodominance. structure remains poorly understood. In this article, we ex- However, monodominance-forming species are dispropor- plore this connection, using tropical trees and their associa- tionately from EM lineages within the Dipterocarpaceae, Fa- tion with mycorrhizal fungi as an illustrative example. baceae, Fagaceae, and Juglandaceae (reviewed in Peh et al. 2011; Smith et al. 2013; Corrales et al. 2016). One defining feature of monodominant stands of EM host tree species in Mycorrhizal Association and Forest Monodominance Neotropical rainforests is that they are embedded in forest A classic case of convergent evolution in plants is their as- with a more even abundance distribution of primarily AM sociation with mycorrhizal fungi (Brundrett 2002). There host trees. EM host monodominance in these forests is par- are at least 19 evolutionarily independent origins of ecto- ticularly dramatic, given that far fewer tree species form EM mycorrhizal (EM) symbiosis in plants (Koele et al. 2012) compared with AM associations in typical mixed forest stands and more than 60 in fungi (Tedersoo et al. 2010). In addi- and that EM host species are often found at low abundances tion, plants associated with EM fungi may potentially expe- outside of the monodominant patches (Henkel 2003; Cor- rience common selective pressure and, because of it, undergo rales et al. 2016). further convergent evolution in other traits, particularly those The mechanisms that cause mycorrhizal status to gener- relatedtoresourceeconomy(e.g.,Read1991;Cornelissen etal. ate monodominance are not fully understood but likely in- 2001; Phillips et al. 2013; but see Koele et al. 2012). volve positive plant-soil feedback as one form of priority ef- As a result, EM host plant species may form a species al- fects. EM monodominant stands are thought to result in liance (sensu Van Nes and Scheffer 2004). That is, EM host part from positive feedback caused by their EM associa- plants may modify the local environment to make it more tion, in which local soil conditions are made more favor- favorable to the members of their own alliance (EM plant able to EM than to AM host trees (Dickie et al. 2014). These species) than to those of other plants, which are often asso- changes in local soil conditions may involve the species com- ciated with arbuscular mycorrhizal (AM) fungi. Thus, if by position of mycorrhizal fungi and other soil microbes or chance a locality were initially dominated by EM plants, it plant-induced nutrient depletion (Corrales et al. 2016; Peay might persist as such owing to positive feedbacks resulting 2016). Priority effects driven by such self-enforcing habitat from this alliance. Consequently, it is possible that local modification—or niche construction (sensu Odling-Smee plant communities diverge as alternative stable states, each et al. 2003)—can lead to alternative stable states, in which developing as either EM or AM dominated as a result of pri- EM trees may form monodominant stands when they es- ority effects, as we explain in more detail below. tablish earlier than other species after local disturbance but A particularly striking example of such alternative stable are otherwise excluded by AM trees. states is seen in the phenomenon of tree species monodom- Intriguingly, however, these potential instances of alterna- inance in tropical forests, in which forests otherwise char- tive stable states with either high-diversity AM assemblages acterized by highly diverse plant communities are dotted or low-diversity EM monodominant stands are observed in by occasional stands in which one tree species is far more only some tropical regions (e.g., South America and Africa) dominant than any other (Richards 1952; Janzen 1974; Con- but not in others (e.g. Southeast Asia), where a diverse suite nell and Lowman 1989; Hart et al. 1989; Torti et al. 2001). In of AM and EM plant species codominate. To our knowledge, tropical forests, such stands can range in area from one no study has directly addressed why this difference exists,

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:57 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Tropical Forest Monodominance S107 yet this geographical contrast presents itself as an opportu- diated by plant-soil feedbacks. We also examine whether nity to gain a better understanding of how and when mono- empirical leaf functional trait data are consistent with the hy- dominance may arise as an alternative stable state. pothesis developed with our model. Variation in leaf functional traits reflects selection for ecological strategies of resource acquisition and conservation (Donovan et al. 2011; Natural History of Mycorrhizal Association Reich 2014) and is correlated with mycorrhizal status (Read as a Motivation for This Article 1991; Phillips et al. 2013). The leaf data we use for this pur- The purpose of this article is to develop a biogeographical pose are from two typical mixed forests in different geo- hypothesis to explain why EM monodominance arises in graphic regions. One is from a Neotropical forest in the Pe- some tropical forest regions and not in others. More gener- ruvian Amazon, where EM plants are rare and the EM plant ally, our goal is to use monodominance as a case in point to species pool is small. The other is from a Paleotropical forest discuss how community convergence can be influenced by in Borneo, where EM plants are common and the EM plant trait convergence and how the extent of this effect can be species pool is large. mediated by the properties of the species pool. Our motivation for this goal comes from the idea that Simulation Methods EM monodominance may be associated with convergent Overview values of plant traits related to resource economy, such as leaf chemistry and decomposability (Torti et al. 2001; Peh The simulation model we used is a modified version of Fu- et al. 2011). Traits of EM host plant species (such as specific kami and Nakajima’s (2013) individual-based, spatially im- leaf area, leaf C and N content, and C∶Nratio)havebeen plicit model of plant community assembly through stochastic found to differ significantly from those of AM host plant spe- sequential immigration of species from an external species cies (Read 1991; Phillips et al. 2013), suggesting that several pool. In this model, which is built on traditional plant com- lineages of EM plant species may have converged and, in the petition models (Chesson 1985; Pacala and Tilman 1994; process, as a functional group diverged from AM plant spe- Mouquet et al. 2002; Fukami and Nakajima 2011), plant in- cies in these traits. Even so, there can still be substantial trait dividuals compete for local resources during the individ- variation across plant species within the EM group (Koele ual establishment stage. They also affect one another’scom- et al. 2012), which may affect the strength and direction of petitiveness via plant-soil feedback. We assume that plants plant-soil feedbacks. For example, interspecific variation in can take on one of two mycorrhizal associations, EM or leaf C∶N ratios may increase the variation in the strength AM. There are other potential mycorrhizal states (e.g., non- and direction of plant-soil feedbacks among EM species (Ke mycorrhizal or ericoid mycorrhizal), but EM and AM asso- et al. 2015). ciations represent most tree species and individuals in trop- In this article, we seek to link the possibility that inter- ical forests (McGuire et al. 2008; Brundrett 2009). We also specific trait variation is associated with variation in plant- assume that plants that host both EM and AM fungi can be soil feedbacks with one aspect of forest communities that is considered primarily EM associated, because the root sys- highly variable among tropical regions: the size of the local tems of these plants tend to become dominated by EM fungi pool of tree species that host EM fungi (Peay 2016). This (Egerton-Warburton and Allen 2001). size ranges from only a few species in South America (e.g., We vary the size of the EM host species pool (the number Baraloto et al. 2011) to hundreds in Southeast Asia (e.g., of EM host species) while holding constant the number of Brearley 2012). We focus on this difference among regions AM species in the pool, and we examine how the size of the in species pool size and the associated potential variation EM species pool influences the way local plant communities in the strength of species interactions mediated by plant-soil assemble, particularly with respect to the emergence of the feedbacks. We hypothesize that a diverse EM species pool alternative stable state of EM monodominance. By replicat- and the resulting complexity in local interactions among ing community assembly in many local forest patches with EM species makes trait convergence in these species difficult the same species pool, we examine how species abundance and that this obstacle to trait convergence in turn prevents distributions may vary among patches that have the same the strong divergence of local communities that is necessary set of local abiotic conditions within them but differ in the for the emergence of monodominant patches. history of stochastic species arrival. In particular, we assess To develop this hypothesis, we use results from computer the conditions under which the local abundance of EM hosts simulation of plant community assembly. The simulation shows strong bimodality, in which a single EM host species model is not intended to replicate actual community assem- becomes monodominant in some local patches but remains bly to quantitatively predict community patterns, but instead rare or absent in all other patches. to explore possible qualitative outcomes arising from a small As in Fukami and Nakajima’s (2013) model, plant reset of assumptions that characterize species interactions me- cruitment occurs in sites that are arranged in a patch. Plants

are recruited from both seeds produced by established indi- by a set of values (Sij) that defined the strength of plant-soil viduals within the patch and immigration from an external feedback, in which the presence of a plant individual be- species pool. For immigration, species of plants were ran- longing to species j at a recruitment site during a given time domly chosen for immigration each time step from the spe- step changed the competitive ability of a plant belonging to cies pool. The chosen species immigrated as a small number species i at that recruitment site during the following time of propagules to a local forest patch consisting of 1,000 re- step (see details below). cruitment sites. One thousand recruitment sites roughly correspond to the sapling abundance we may expect in a hect- Local Patches are, the scale at which small monodominant stands are observed. All recruitment sites were initially empty. Subse- Local forest patches each consisted of 1,000 recruitment quently, only one plant individual could establish and pro- sites. Each recruitment site could accommodate only one duce propagules in each recruitment site, even when multi- adult tree and was characterized by a value that represents ple propagules arrived from the species pool or from local abiotic conditions (Hk), chosen randomly between 0 and 1 dispersal within the patch. Propagules were distributed ran- from a beta distribution. For the beta distribution for Hk,we p p — — domly to recruitment sites within the patch each time step. set a b 2, which as with the case of Ri produces a Of the propagules that arrived at a recruitment site, the one that belonged to the species that best fit the environmental A 1 condition of the recruitment site could establish and pro- Arbuscular Ecto- duce propagules (determination of environmental fit is de- mycorrhizal mycorrhizal scribed in more detail below). All plants within a patch pro- trees 1 trees duced propagules once each time step (e.g., each year or each mast event) until they died. All plants died with a fixed probability, and when they did, recruitment sites became 1.1 if i = j 1.1 if i = j empty and available for a new plant individual to establish. [1, 1.1] if i ≠ j [1, 1.1] if i ≠ j Real tree species vary in dispersal ability, fecundity, and mor- tality rates, but we kept these constant across species to focus on the mechanisms of monodominance in relation to our hypothesis. This process of immigration, arrival, establish- B v = 0.0001 ment, reproduction, and death was repeated for 400 time 4 steps. All simulations were carried out using Mathematica v = 0.01 8.0 (Wolfram Research, Champaign, IL). Code and data are 3 available as supplementary material, available online.1 v = 0.1 2 v = 1

Species Pools Frequency 1 Regional species pools each contained 100 AM plant spe- 0 cies and one to 20 EM plant species. Each species i was de- 0.2 0.4 0.6 0.8 1.0 fined by a value R , which can be thought of as a multitrait i Species value, R (solid lines) or phenotype that determines how well species perform in lo- i abiotic condition, H (dashed line) cal abiotic conditions during the recruitment stage (see de- k tails below). Values of Ri were chosen randomly between 0 and 1 from a beta distribution, where the probability den- Figure 1: Model description. A, Schematic summary of the strength a21 2 b21 sity for value x was proportional to x (1 x) (Mouquet of plant-soil feedback (Sij) assumed in the simulation model. The val- et al. 2002). We set a p b, which causes the probability den- ues indicated were used for the simulation shown in figures 2–4but changed as specified in figure 5 to evaluate the effect of feedback sity for Ri to have a peak at 0.5, but the value of a was varied strength on community assembly. An Sij of 1 means no feedback. An in order to examine the effect that the amount of variation 1 fi Sij value that is 1 means positive feedback. All intraspeci c feedbacks p in Ri among species had on community assembly. We re- (i.e., i j) were assumed to have Sij of 1.1, and Sij values for interspe- fer to 1=a as interspecific phenotypic variability v. A larger cific feedbacks (i.e., i ( j) were assigned a random value between 1 and 1.1, so that they are all positive, but their strengths vary with species v means that Ri values are more variable among species fi identity and are never greater than intraspecific feedbacks. B, Probabil- ( g. 1B). In addition to Ri, species were also characterized ity distribution for abiotic condition of recruitment sites (Hk; dashed

line) and for species phenotypic value (Ri) among ectomycorrhizal 1. Code that appears in the American Naturalist is provided as a convenience species in the species pool under different degrees of species variabil- =b to the readers. It has not necessarily been tested as part of the peer review. ity (v, which is 1 for the ß distribution for Ri).

peak of Hk values at 0.5 (fig. 1B). For all simulations in a Teste et al. 2017). Similarly, in a large analysis of forest plots run, the Hk values remained the same. in North America and New Zealand, Dickie et al. (2014) found that most tree species exhibited positive density dependence. We therefore assumed in our model that positive Competitive Ability effects outweigh negative ones to result in net positive plant- fi p : soil feedback for conspeci cs. Accordingly, we set Sij 1 1 Values of Hk, Ri,andSij together determined competitive when i p j. ability (Cijk) of species i at recruitment site k when the re- We also assumed that EM tree species shared similar ni- cruitment site was occupied by species j at the previous trogen economy (Corrales et al. 2016) as well as similar as- time step: sociations with EM fungi. As a result, EM plant species can benefit each other to some extent, but the strength of in- p 2 j 2 j : Cijk (1 Hk Ri )Sij terspecific plant-soil feedback is variable among EM plant species pairs and is never greater than that of intraspecific feedback. We believe that these assumptions are plausible p Thus, assuming that Sij 1 (i.e., no effect of species j and because evidence suggests that EM fungi tend to be gener- species i through plant-soil feedback, as explained below), alists with high overlap between co-occurring species (Ken- a species would have a high value of Cijk if it had a value of nedy et al. 2003; Peay et al. 2015; but see Bennett et al. 2017) ’ Ri that is close to that of Hk, as in Mouquet et al. s (2002) and that shared EM fungi can facilitate interspecific recruit- model. In our simulation, a close match between Hk and Ri ment (Horton et al. 1999; Nara 2006b). In addition, there is ’ is most likely when a species Ri value is 0.5, since Hk also evidence that different combinations of EM fungi and plants has a peak at 0.5 (fig. 1B). can have different effects on plant growth, even among closely Plant-soil feedback moderates competitive ability, with related species (Nara 2006a; Fransson et al. 2015). Conse- fi ( the value of Sij de ning the direction and strength of the quently, when i j, we assigned Sij values by taking a uni- fi fi feedback. Speci cally, Sij de nes the effect of species j occu- form randomly drawn value between 1 and 1.1 for each Sij. pying a given recruitment site during a given time step on For simplicity, we assume that this same pattern of intra- the competitive ability of species i in that microsite during and interspecific feedback strengths also holds for AM plant the following time step. Thus, Sij affects the competitive species. In other words, EM and AM host species have sim- ability of species i independently of both recruitment site ilar capacities for positive plant-soil feedback in our model. fi condition (as de ned by Hk) and species phenotype (as de- Finally, we also assumed that EM and AM plant species did fi ned by Ri) and represents the amount by which the differ- not facilitate each other through plant-soil feedback. Thus, ’ p ence between the abiotic environment and the species match Sij 1 between all pairs of an EM species and an AM species. 1 ! to the environment is improved (Sij 1), worsened (Sij 1), The direction and strength of plant-soil feedback in trop- p or not affected (Sij 1) via plant-soil feedback (Fukami and ical forests have not been thoroughly characterized, and it is p fi Nakajima 2013). That is, if Sij 1, there is no net effect of uncertain whether the assumptions speci ed above are al- 1 ! plant-soil feedback, whereas Sij 1 and Sij 1 represent pos- ways realistic. To examine whether model results are sensi- itive and negative plant-soil feedback, respectively. tive to these assumptions, we ran two sets of additional sim-

We set the values of Sij to attempt to represent the inter- ulation. In the first, we assumed that among EM host plants, actions within and between AM and EM plants that are ob- Sij took a uniform randomly drawn value between 1 and 1.1, served in nature (fig. 1A). There are many ways in which even when i p j, in order to relax the assumption that all plant-soil feedback could be structured, but here we assume intraspecific feedbacks were more or equally positive com- that plants facilitated the growth of conspecific individuals pared with any interspecific feedbacks (fig. S1B; figs. S1– via local accumulation of specific mycorrhizal fungi that are S5 are available online). In a second set, we used a combination beneficial to them, via modification of soil nutrient com- of alternative assumptions. Specifically, we assumed (1) var- position to the species’ own benefit via their specific mycor- iable positive intra- and interspecific feedbacks for EM host rhizal association (Corrales et al. 2016) or via better protec- plants, (2) variable intra- and interspecific feedbacks (both tion from pathogens through added physical or chemical positive and negative) for AM host plants, and (3) negative defenses in mycorrhizal roots (Duchesne et al. 1989; News- effects of EM host plants on AM host plants. For assump- ham et al. 1995; Bennett et al. 2017). Plants can also nega- tion 1, we tried two variants. In one (variant 1), we used tively affect conspecifics by, for example, accumulating soil the original assumption as above, that is, that the strength pathogens (Kulmatiski et al. 2008). However, studies that of interspecific plant-soil feedback is variable among EM examine net effects of pathogens plus mycorrhizae often find plant species pairs and is never greater than that of intra- overall positive effects from soil biota (e.g., fig. 1B in Mangan specificfeedback(fig. S1C). The other (variant 2) is that in- et al. 2010; see also Cortois et al. 2016; Bennett et al. 2017; terspecific and intraspecific feedbacks are all positive but

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:57 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). S110 The American Naturalist equally variable, such that some intraspecific feedbacks are After plant establishment was completed for all recruitment less positive than some interspecific feedbacks (fig. S1D). sites, plants occupying a recruitment site died with probability m, which was 0.4 for all species. Thus, competitive ability in our model is not related to dispersal, fecundity, or baseline survival; it dictates only a species’ ability to win a recruitment Community Assembly site. We assembled communities for 400 time steps. From vi- At each time step, each species in the species pool immigrated sual inspection of results, 400 time steps seemed long enough to the local patch with probability I. For each of the 100 AM for most communities to reach an equilibrium state (fig. 2). species in the pool, we used I p 0:05. For each EM species, we used I p 1=(number of EM species in the pool). For Manipulating Species Pool Size and Species Variability example, when we had two or 20 EM species in the pool, I was 0.5 or 0.05 for each EM species, respectively. This We manipulated two factors—the number of EM species in way, we standardized for the total immigration frequency the species pool and the amount of variation in Ri among for all EM plants across the gradient of the size of the EM species pool. We also did additional simulation in which I p 400 0:05 for each species, regardless of the size of the EM species A pool used, in order to decouple the effects of regional species 300 richness per se from those of regional relative abundance. At each recruitment site in the local patch, spe- 2 2 1 = cies i arrived with probability 1 exp[ (Pi FNi) 200 (total number of recruitment sites, i.e., 1,000)] at each time step. Here, Pi is the number of propagules of species i that 100 immigrate from the species pool (20 propagules for species chosen for that time step for immigration from the species pool, and 0 seed for all other species), F is fecundity (50 for all species), and Ni is the number of plants belonging to spe- 200 cies i in the local patch (0 for all species in the first time step, B

p Local abundance i.e., at t 1, which is analogous to modeling recruitment af- 150 ter a stand-replacing disturbance). The speciﬁcations were chosen so as to have values that likely fall within a plausible 100 range. When the number of recruitment sites that were as- 1 signed to receive a propagule of species i exceeded Pi FNi 1 (which rarely happens), Pi FNi recruitment sites were 50 randomly selected from these recruitment sites, and a propagule of the species were assigned only to the selected recruitment sites. 100 200 300 400 Given this probability, there were three possibilities re- Time step garding plant establishment and seed production in each recruitment site. First, if the recruitment site were already occupied by a plant, that plant remained there. In other words, Figure 2: Examples of simulated population dynamics for two forest patches, showing ectomycorrhizal (EM) monodominance (A)and we assumed that seeds could not displace established adults. arbuscular mycorrhizal (AM) dominance (B) as alternative stable Second, if the recruitment site were empty and no species states. Both A and B show results for one instance of local commu- arrived at that recruitment site, it remained empty. Third, if nity assembly under the same species spool, and each line indicates a the recruitment site were empty and one or more species ar- species, with warm colors denoting EM tree species and cold colors rived at that recruitment site, of those species that arrived, denoting AM tree species. In these examples, there were 100 AM tree species and 10 EM tree species in the species pool, and species variabil- the one with the greatest value of Cijk (independent of the ity among EM species (v) was set to be small (v p 0:0001), with all EM ﬁ number of propagules of each species) was assumed to occupy species having an Ri value very close to the optimal, 0.5 (see g. 1B). the recruitment site and produce propagules starting the fol- Depending on assembly history, local communities develop either as lowing time step. The seed-to-adult recruitment processes EM monodominance (A), in which a single EM species becomes dom- are important in many of the dynamics modeled here, but inant (orange line) and all others are AM species, or as an AM-only community (B), in which a few EM species were able to colonize the processes operating during those stages per se were not the forest patch initially but became locally extinct because they were focus of our model, which was intended to identify the min- not common initially and were excluded competitively by AM species imal processes required to produce monodominance. through positive plant-soil feedback among AM species.

EM host species (i.e., species variability v,asdefined in only when all intraspecific feedbacks were more or equally “Species Pools”)—to examine their effects on EM abun- positive compared with interspecific feedbacks (fig. S1C). dance in assembled local communities. To this end, under Otherwise, multiple EM plant species often coexisted as a re- each of the species pools we used (i.e., 16 pools in fig. 3), sult of mutual facilitation among them, with only a small we ran the simulation 100 times, with each run considered number of monodominance patches developing (fig. S1B, one instance of community assembly in a local forest patch. S1D). Therefore, one condition for monodominance forma- Under each species pool, the replicated 100 patches had the tion in our model is stronger positive intraspecific relative same set of recruitment sites, which allowed us to evaluate to interspecific feedback among EM trees, which is consis- whether plant communities that develop in different patches tent with some recent empirical data (Bennett et al. 2017). that share the same set of Hk values in them diverged in spe- Strong intraspecific feedback is possible if, for example, a cies composition as a result of random variation in the his- tree species cultivates an assemblage of highly beneficial EM tory of species arrival from the species pool. If they did di- fungi. The additional simulation in which immigration rate verge, that would be evidence for alternative stable states. I was 0.05, regardless of the size of the EM species pool, in- We were particularly interested in assessing the conditions dicated that the number of monodominant patches is posi- under which EM monodominance arose. We defined EM tively correlated with the immigration rate per EM species monodominance as the case in which only one dominant (fig. S2A,S2C). EM species occupied one-fourth (25%) or more of the avail- Together, these results suggest that there is no effect of able recruitment sites in a patch, with the other sites being species pool size per se on whether EM monodominance occupied by AM species. will emerge or on how strong the bimodal pattern reflecting monodominance formation will be. Rather, it is the amount of phenotypic variability among EM species that shapes the Simulation Results and Discussion pattern of EM abundance in local patches. Specifically, re- Simulation results verified that communities with (fig. 2A) duced variability tends to cause greater frequency of mono- and without (fig. 2B) monodominant EM species could de- dominance (fig. 3). In order to directly test for the relation- velop as alternative stable states in our model. Under the ship between species pool size and phenotypic variability of same species pool, communities were either dominated by EM species, we ran additional simulations to examine how AM species without any EM species present or had at least species pool size influences the relationship between spe-

25% EM individuals, representing two alternative stable cies phenotypic value (Ri) and the abundance of EM spe- states, similar to those seen in some Neotropical rainforests. cies when EM species variability in the species pool is large By varying species pool size and phenotypic variability (v p 1). We found that when species pool size was small,

(fig. 3), we found that EM monodominance arose as an al- there was a relatively good correspondence between Ri value ternative state whenever trait variability among EM species and EM species abundance (fig. 4A). We call this corre- was low. In these cases, all EM species in the species pool spondence trait-environment matching, meaning that the had an Ri value that was nearly optimal for the abiotic con- closer species are in their phenotype Ri to the optimal (i.e., ditions available in the forest patch. Under this condition, a 0.5 in our simulation), the more abundant they tend to be- single dominant EM species was competitive enough to form come. In this case, the abundances of species follow what is monodominantstandswhenitarrivedearlyduringlocalcom- expected on the basis of their match to the environment j 2 j munity assembly but not when it arrived late, as indicated by (i.e., Ri Hk ). the strongly bimodal pattern in the top row in figure 3. One As species pool size was increased, this correspondence of the two modes had no EM species, and the other mode had became weaker (fig. 4), such that species with an optimal only one EM species, even when the species pool contained Ri value (closer to 0.5) were not necessarily more abundant multiple EM species. than those with a suboptimal value (away from 0.5). When

Regardless of species pool size, as Ri values became more the species pool was diverse, two additional abundance peaks variable among EM species in the local species pool (i.e., appeared (at Ri of around 0.2 and 0.8), so that species that did ﬁ movingfromtop to bottom rows in g.3),theEM totalabun- not have an Ri close to 0.5 could nevertheless be as abundant dance became more variable among forest patches, and the as those with an optimal Ri, as most clearly seen in ﬁgure 4D. bimodal pattern became increasingly obscure. The frequency In short, for EM species in our model, it is adaptive to have of monodominance declined with greater phenotypic vari- traits that best match the underlying abiotic environmental ability when there was more than one EM species in the pool conditions if there are not many EM species in the species because, although the number of EM trees in a patch might be pool. However, if the species pool has many EM species, our high, these patches consisted of multiple EM species. results suggest that there can be disruptive natural selection The additional simulation indicated that these results gen- that keeps some species away from having the abiotically erally held under the alternative assumptions examined, but optimal Ri value if the species pool has many species. As a

1.0 44 30 29 44 0.0001 40 50 50 40 40 40 30 1.0 1.0 30 30 30 20

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). 20 20 20 10 10 1.0 10 1.0 1.5 10 1.0 1.0 1.0

200 400 600 800 1000 200 400 600 800 1000 200 400 600 800 1000 200 400 600 800 1000 ectomycorrhizal tree speciesinthepool

1.1 70 0 0 Variability, 1.0 50 0 50 This content downloadedfrom 132.236.027.111 onJuly22,2017 05:14:57AM 50 0 60 52 45 1 2.8 3 40 40 50 40 0.01 30 30 40 30

30 20 v 20 20 2.5 in speciesvalues, , 20 1.8 3.2 10 10 10 2.6 10 2.0 1.0 1.0 1.5 1.0 200 400 600 800 1000 200 400 600 800 1000 200 400 600 800 1000 200 400 600 800 1000 hes (number out of 100) 60 1.0 60 0 50 0 0 50 50 56 50 1 40 20 4 0 40 1.6 40 40 30 0.1 30 30 2.0 R

30 2.0 i 20 , among 20 20 1.9 20 2.3 2.0 10 10 10 1.0 2.8 10 3.0 4.3 1.0 1.0 1.0 2.0 2.03.0 3.0 3.0 3.5 5.0

Number of patc local 200 400 600 800 1000 200 400 600 800 1000 200 400 600 800 1000 200 400 600 800 1000 0 0 0 0 50 30 60 13 60 7 60 3 40 50 50 50

1.0 1 30 40 40 40 30 30 1.7 30 20 1.0 20 20 20 1.0 1.0 2.0 1.7 2.8 10 10 10 1.0 3.3 10 1.4 4.0 1.0 1.0 2.0 2.0 1.8 3.0 1.3 3.5 200 400 600 800 1000 200 400 600 800 1000 200 400 600 800 1000 200 400 600 800 1000

Number of ectomycorrhizal trees in a local patch (number out of 1000)

Figure 3: Histograms showing ectomycorrhizal tree abundance in a local forest patch across a gradient of the number of ectomycorrhizal tree species in the species pool and the species phenotypic variability (v) among ectomycorrhizal tree species in the species spool (see fig. 1B). The Sij values used are as specified in figure 1A. The number above each bar indicates the mean number of ecotomycorrhizal tree species that were present in the patch, averaged across all patches belonging to the bar. The number in the top right corner of each histogram indicates the percentage of the patches that showed monodominance by our criteria (only one ectomycorrhizal tree species in the patch and more than 25% of the stems belong to that species). Each histogram shows results for one species pool, but we replicated species pools for each histogram and found that results were consistent across replicated pools. Tropical Forest Monodominance S113

Number of ectomycorrhizal tree species in the species pool 1 21020 7000 A 3500 B 700 C 350 D 5000 2500 500 250 3000 1500 300 150 1000 500 100 50 0.2 0.4 0.6 0.8 1.00.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0

6000 8000 2500 2500 E 5000 F G 2000 H 6000 2000 4000 1500 4000 3000 1500 across 40 local patches 2000 1000 1000 2000 500 500

Species abundance, summed 1000

0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0

Species value, Ri

Figure 4: Relationship between species value (Ri) and the abundance of ectomycorrhizal tree species across a gradient of the number of ectomycorrhizal tree species in the species pool. A–D, Results for a total of 2,400 community assembly outcomes comprising 100 replicated species pools, under each of which 40 replicated local communities were assembled. The Y-axis is the mean abundance of a species in a local community, averaged across the 40 replicated local communities, which were then averaged across species having similar Ri values (with error bars showing standard errors). Solid lines are the simulation results, and dotted lines show the relationship that would be expected if abundance were determined solely by availability of recruitment sites (i.e., the frequency of recruitment sites having speciﬁc values of – – Hk, abiotic condition of microsite). E H, Results based on the same data as for A D, but instead of showing mean abundances and standard errors, the abundance of each species is indicated (each circle represents a species).

result, convergent trait evolution that results in low interspe- 5D). These results support our hypothesis that it is the var- cific trait variability and tight trait-environment matching iation in interspecific feedback strengths among EM species in the species pool may be more likely when the EM species that makes trait-environment matching poor and mono- pool is small, whereas divergent trait evolution may arise dominance unlikely when the species pool is large. when the EM species pool is large. In summary, our results suggest that two interdependent Why do EM species with suboptimal phenotypes obtain processes influence the likelihood of trait and community comparable densities to those with optimal phenotypes convergence. First, trait-environment matching that leads when the species pool is large (fig. 4D)? We suspected that to trait convergence among EM species may be less likely unit was because the abundance of each EM species is affected der a larger EM species pool because of local plant-soil feed- so strongly by the plant-soil feedback—which is made more backs that vary in strength. Second, weak trait convergence complex by the larger number of EM species that each have may in turn impede the development of EM monodominance, different strengths of plant-soil feedback—that the under- which represents an extreme case of community divergence. lying influence of abiotic conditions is overwhelmed and obscured. To test this idea, we reran the simulation, but this time with the strength of plant-soil feedback among EM Empirical Trait Data Analysis and/or AM species (S ) all set to 1.1 instead of being drawn ij Study Systems and Methods uniform-randomly from [1, 1.1]. Setting all Sij values to 1.1 makes interspecific feedback strengths all the same and To begin to assess whether empirical data are consistent equal to intraspecific feedback strength, eliminating the with the hypothesis developed with the model above, we an- complexity of feedback strength. In this additional simula- alyzed leaf trait data from trees in two geographic regions: tion, we found good trait-environment matching—such as Southeast Asia (Borneo), where EM trees are common in fi — the one seen under small species pools ( g. 4A) if all Sij most lowland forests, and South America (Peruvian Ama- values among EM species were set to 1.1 (fig. 5B). Poor zon), where EM trees are generally rare but can become trait-environment matching was not, however, eliminated monodominant. Both regions contain large areas of low- fi when Sij values among AM species were set to 1.1 ( g. 5C, land tropical rainforests with extraordinary levels of tree di-

Strengths of inter-specific feedback in EM trees ]1.1 ,1[ ]1.1 llA 1.1 Strengths of inter-specif 400 A 500 B 300 400 [1, 1.1] 300 200 200 100 100

0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0 ic feedback in AM trees 120 C 200 D 100 150 1.1 All 80 across 40 local patches 60 100 Species abundance, summed 40 50 20

0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0

Species value, Ri

fi Figure 5: Effect of changing Sij values on the relationship between species value (Ri) and species abundance. Symbols are as in gure 4, except that data for each panel are based on 60 (not 100) replicate species pools. EM, ectomycorrhizal; AM, arbuscular mycorrhizal. versity. The Peruvian Amazon has greater regional tree di- classified at the genus or family level on the basis of Brun- versity (approximately 5,000 species; Pennington et al. 2004) drett (2009) and Peay et al. (2013). Data are deposited in the than Borneo (approximately 3,000 species) probably be- Dryad Digital Repository: http://dx.doi.org/10.5061/dryad cause it is connected to the rest of the Amazon basin, the .c0kr7 (Fukami et al. 2017). world’s largest contiguous area covered by lowland rainfor- For each region, we analyzed trait data from lower fertil- est. Thus, at the scale of a local patch of forest (e.g., Lambir ity (sand) and higher fertility (clay) soil habitats separately Hills in Borneo or Allpahuayo-Mishana in Peru), we expect because these habitats differ in species pool composition the regional tree species pool to be reasonably similar, with within each region. The trait data set from Peru included Peru being slightly larger. However, the pool of EM trees is 365 AM and 13 EM species from clay soils and 140 AM much larger in Borneo than in Peru. and 3 EM species from sandy soils. The trait data set from For our analysis, trait data were taken from two separate Borneo included 98 AM and 17 EM species from clay soils studies of plant functional ecology in Peruvian Amazon (For- and 111 AM and 35 EM species from sandy soils. We took tunel et al. 2012, 2014) and Borneo (Russo et al. 2005, 2013). three approaches to look for differences in EM traits be- Detailed methods on data collection can be found from tween the two species pools (Peru and Borneo) within soil Fortunel et al. (2012, 2014) and in the appendix, respectively. types. First, we compared mean trait values for EM versus Data in Peru were collected from multiple 0.5-ha Gentry AM host species within each species pool, using a two- plots, whereas those in Borneo were collected from within sample t-test, to evaluate our prediction that EM species a single 52-ha forest dynamics plot in Lambir Hills National are functionally distinct from AM species. Second, we com- Park. We used data on four leaf traits that are involved in the pared the variance of trait values for EM versus AM species fast-slow plant economics spectrum (Reich 2014) and that within each species pool, using a one-sided variance ratio are thought to be strongly filtered by environmental condi- F-test, to evaluate our prediction that EM species have larger tions (Fortunel et al. 2014): specific leaf area (SLA), leaf car- trait variability than AM species. Third, we compared the bon and nitrogen concentrations by mass, and leaf C∶N ra- variance in trait values between EM species pools in Borneo tio. Leaf C∶N ratio is considered a particularly important and Peru, using a one-sided variance ratio F-test, to evalu- trait in determining the strength of plant-soil feedbacks (De ate our prediction that there should be greater variation in Deyn et al. 2008; Ke et al. 2015). Data from individual plants trait values among EM host species in Borneo than in Peru, were averaged within species. Species’ mycorrhizal status was where EM monodominance is found. We do not approach

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:57 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Tropical Forest Monodominance S115 this question in an explicit phylogenetic framework, but other nutrients (Baillie et al. 2006). Low soil fertility is our test is conservative because we predict larger trait varia- strongly associated with overall low foliar nutrient concen- tion in the narrower phylogenetic pool, that is, in Borneo, trations and higher C∶N ratios (Chapin et al. 1993), which which had only one EM family (Dipterocarpaceae). could give less latitude for differences between AM and EM host species in Borneo compared with more fertile sites. Trait differences between AM and EM host species in Bor- Trait Analysis Results and Discussion neo may be more tightly linked to leaf structure and physical We found significant differences in some mean foliar trait defense, as suggested by the significant differences between values between AM and EM species in both Peru and Bor- mycorrhizal groups in foliar C, which is likely to affect leaf neo that were consistent with our model prediction (table 1; litter decomposition and, in turn, the strength of plant-soil fig. 6). In Peru, foliar C∶Nratiowassignificantly different feedbacks and the extent of trait-environment matching. between AM and EM species on both soil habitats, and fo- Thereweresignificant differences in the variance of leaf liar nitrogen concentration was different only on clay soil. traits between EM and AM host species in Peru and Bor- In Borneo, foliar carbon concentration differed significantly neo. However, in contrast to our expectations, variance in between AM and EM species on both habitats, but no other traits was often larger for AM than EM host species in Bor- significant differences were observed. The difference found neo (three of four traits), whereas no significant differences in Peru between AM and EM host species in foliar C∶N was were observed in trait variances between AM and EM host particularly large, in contrast to virtually identical means for species in Peru (table 2). Furthermore, results for trait var- this trait in Borneo (fig. 6). iance of EM species pools did not support our model pre- One possible explanation for these inconsistencies lies in diction either. The variance in species traits was not larger the relative differences in the sizes of EM and AM host spe- among EM host species in Borneo than in Peru for any of p : p : cies pools in each forest. To the extent that greater richness the traits measured for clay (SLA F16, 12 0 56, P 86; p : p : of the species pool causes more complex plant-soil feed- leaf carbon concentration [LCC] F16, 12 0 52, P 89; leaf p : p ∶ backs within each mycorrhizal group, in Peru (with fewer nitrogen concentration [LNC] F16, 12 0 05, P 1; C N p : p : p : p EM host species), their traits may track environmental con- F16, 12 0 87, P 61) or sandy (SLA F34, 2 2 32, P : p : p : p : p ditions closely, causing a large divergence from those of AM 35; LCC F34, 2 5 13, P 18; LNC F34, 2 0 03, P 1; ∶ p : p : host species. Since the pool of AM species is larger, they may C N F34, 2 0 49, P 85) habitats. be influenced strongly by the complex plant-soil feedbacks, How could these apparent discrepancies between model reducing their match to underlying abiotic conditions. In predictions and empirical patterns be reconciled? Our fo- contrast, in Borneo, where both mycorrhizal groups are spe- cus in this article has been on contrasting alternative stable cies rich, complex plant-soil interactions might have pre- states between diverse AM and monodominant EM patches, vented species traits from tracking the environment in both but as a family, the Dipterocarpaceae dominate this and EM and AM host species. many other forests in Southeast Asia in terms of basal area. Another explanation lies in the difference in the relative One potential explanation for their trait convergence may soil fertility of these regions. Averaging over soil types and then be that phylogenetic relatedness of dipterocarp species fi p mycorrhizal status, foliar N was signi cantly lower (F1, 781 have caused them to act as if they belonged to one species, : ! : ∶ fi p 133 7, P 001) and C N signi cantly higher (F1, 781 thereby allowing for trait-environment matching in the 109:8, P ! :001) in Borneo than in Peru, and both soils at absence of complex interspecific plant-soil feedback. As for the Bornean site are particularly low in soil N, P, K, and the apparent lack of trait convergence among EM relative

Table 1: Difference in species’ mean values of leaf traits between arbuscular mycorrhizal (AM) and ectomycorrhizal (EM) tree species across two edaphic habitats in Peru and Borneo No. species SLA LCC LNC C∶N Region and habitat AM EM tPtPtPtP Peru: Clay 362 13 .06 .95 1.59 .14 24.18 .001* 6.15 !.001* Sand 140 3 .97 .41 2.15 .89 22.87 .10 4.68 .03* Borneo: Clay 98 17 .98 .34 24.43 !.001* 1.06 .29 21.34 .19 Sand 111 35 1.29 .20 24.55 !.001* .58 .56 2.56 .58 Note: SLA, speciﬁc leaf area; LCC, leaf carbon concentration; LNC, leaf nitrogen concentration; C∶N, leaf C∶N ratio. * P ! :05.

Clay soil in Peru Sandy soil in Peru

AB Density 0.02 0.04 0.06 0.08 0.10 0.00 0.02 0.04 0.06 0.08 0.10 0.00 0 20406080 0 20406080

Clay soil in Borneo Sandy soil in Borneo

C D Density 0.02 0.04 0.06 0.08 0.10 0.00 0.00 0.02 0.04 0.06 0.08 0.10 0 20406080 0 20406080 Foliar C:N Foliar C:N

Figure 6: Probability density plots showing the frequency of tree species having particular foliar C∶N ratios under two soil conditions (less fertile sand or more fertile clay) in Borneo and Peru. Dark gray indicates ectomycorrhizal (EM) host species. Light gray indicates all other species, most of which are arbuscular mycorrhizal (AM) hosts. Dashed lines indicate means. to AM host species in Peru, this result may reflect low statis- species pool has a small number of EM species. Specifically, tical power due to small sample size. The number of EM spe- results summarized in figure 3 indicate that monodomi- cies was much smaller in Peru (13 and three species) than in nance can emerge only when EM species phenotypes (Ri) Borneo (17 and 35 species). closely match the environmental conditions available in lo- fi cal forest patches (Hk). Results presented in gure 4 indicate in turn that such optimal trait-environment matching Discussion is more likely under small species pools because EM spe- Taken together, our simulation results suggest one poten- cies can then more easily track the environment. As a result, tial reason why EM monodominance is favored when the trait evolution of EM species converges toward abiotically

Table 2: Difference in leaf trait variance between arbuscular mycorrhizal (AM) and ectomycorrhizal (EM) tree species across two edaphic habitats in Peru and Borneo No. species SLA LCC LNC C∶N Region and habitat AM EM F ratio PFratio PFratio PFratio P Peru: Clay 362 13 1.11 .45 1.03 .52 .34 1.00 1.65 .16 Sand 140 3 5.71 .16 4.24 .21 .79 .71 3.95 .22 Borneo: Clay 98 17 1.15 .40 3.17 .006* 4.00 .001* 2.55 .02* Sand 111 35 1.42 .12 4.07 !.001* 7.05 !.001* 5.44 !.001* Note: SLA, specific leaf area; LCC, leaf carbon concentration; LNC, leaf nitrogen concentration; C∶N, leaf C∶N ratio. * P ! :05. optimal conditions. Finally, results in figure 5 show that the may influence feedback strength, such as phylogenetic relat- reason why species do not track the environment under di- edness (Gilbert and Webb 2007; Liu et al. 2012), local den- verse species pools is because the complex plant-soil feedback sity (Comita et al. 2010; Mangan et al. 2010; Liu et al. 2012), among the many EM species obscures the effect of environ- and mycorrhizal type (Johnson et al. 2012; Dickie et al. mental heterogeneity on local species abundances. In other 2014; Bennett et al. 2017). words, environmental filtering (Kraft et al. 2015a) is weaker In this study, we attempted to keep the model focused than priority effects. on differences in species pool size by keeping the nature of Although we have focused on mycorrhizal associations, plant-soil feedback identical between AM and EM species. monodominance may arise through other pathways (Peh Through additional simulation (figs. S1, S2), we have made et al. 2011). For example, a defensive ant-plant mutualism an initial effort to examine the robustness of our model pre- generates monodominant stands of Duroia hirstuta known dictions to assumptions regarding plant-soil feedback. Fu- as devil’s gardens in the Amazon (Frederickson et al. 2005). ture iterations of this model could explore more complex Still, many tree species that form monodominant stands in plant-soil feedback structure, such as more negative (or less the tropics are EM. Studies have uncovered evidence sup- positive) intraspecific feedbacks and stronger positive feed- porting a range of ways in which EM associations may gen- backs in EM compared with AM plants. It is also possible erate positive plant-soil feedbacks that lead to monodomi- that EM plants are better able to make the environment less nance, including common mycorrhizal networks (McGuire suitable for AM plants, owing to the abilities of EM fungi 2007), alteration of local nutrient economies (Corrales et al. to access organic forms of nutrients and preempt nutrient 2016), variation in leaf chemistry (Torti et al. 2001), and EM uptake by AM fungi. This scenario can be evaluated further mining of organic nitrogen (Orwin et al. 2011). Our model with our model. Finally, we assigned Ri and Sij values inde- does not distinguish between these different mechanisms. pendently, but in nature, species with similar traits—as re- — Rather, our work is complementary to these studies in that presented by more similar values of Ri might have the ten- we ask why geographic differences in species pools may lead dency to share fungal symbionts and therefore have more to some tropical regions containing EM monodominant positive Sij. Phylogenetic relationships of plant species, which stands and other regions showing local codominance of mul- we did not consider in this article, may prove useful in some tiple tree species of both mycorrhizal types. cases as a proxy for estimating Ri and Sij values. We chose one simple way to represent plant-soil feed- It is well established in plant-soil feedback theory that pos- backs. Namely, we assumed that plant-soil feedbacks were itive feedback causes alternative stable states, whereas nega- net positive and most beneficial to conspecifics. Choosing tive feedback facilitates species coexistence (Bever et al. 2012). appropriate values for plant-soil feedbacks is not trivial, be- Our simulation results are consistent with this prior theory. cause in nature these feedbacks are hard to estimate. Em- However, most previous plant-soil feedback theories have pirical measurement of their strength should take into ac- considered interactions between only two plant species. Here, count multiple contributing factors, such as abiotic effects we have studied interactions among many plant species (see (Waring et al. 2015) and the combined effects of pathogens also Fukami and Nakajima 2011, 2013), which is the only and mutualists(Klironomos 2002) andnatural enemies (Bag- way to directly address the effect of the number of species in chi et al. 2014). In addition, whether net feedbacks are pos- theregionalspeciespoolonlocalcommunityassembly driven itive or negative also depends on the choice of other het- by plant-soil feedback. erospecific comparisons and sterilization methods used to One benefit of our model is that it makes specific pre- generate the reference point. A number of other factors dictions about trait variation of plant species and about

mycorrhiza-driven plant-soil feedbacks under different spe- timates of the Cijk in our model to test the effects of AM and cies pool sizes, specifically that divergence in the values of EM trees on each other’s growth and survival. Third, we com- traits related to local environmental fit should be greater pared just two regions here, but it will be more informative in more diverse species pools. Our trait analysis did not yield to analyze data from more regions—including Africa, where strong support for this prediction, perhaps because these are monodominance also arises—to assess the potential appli- difficult effects to measure empirically, since many factors cability of our hypothesis across geographical regions with not included in our model influence trait variation in natural different evolutionary histories. settings. Moreover, uncertainties regarding the mycorrhizal If species pool properties are key determinants of com- status (e.g., EM or AM) of plant species in different geograph- munity assembly (Ricklefs and Schluter 1993; Zobel 2016), ical locations may also have caused the poor correspondence as we have considered in this article, a fundamental ques- of model predictions and empirical patterns (Forrestel et al. tion is what causes differences in species pools among trop- 2017). In addition, detecting patterns in traits across a gra- ical regions in the first place. For example, what might ex- dient of species pool size is made difficult by the inherent plain the unusually high prevalence of EM lineages in the correlation between species pool size and statistical power Asian tropics? Any understanding of why the EM pool is so for any analysis that uses species as a unit of replication. large in Asia must reckon with the radiation of the Diptero- Our trait analysis is only a first step to begin to evaluate carpoideae (Dipterocarpaceae), which numbers more than the empirical relevance of the hypothesis we have devel- 470 species (Ashton 2002). More generally, many hypothe- oped here through the simulation modeling. Research on ses have been put forward to explain the disparities in tree the links between plant traits, mycorrhizal associations, and species diversity among different tropical regions. However, the strength of plant-soil feedback is still at an early stage of there is no consensus explanation (Couvreur 2014). Some development (Laughlin et al. 2015; Cortois et al. 2016; Ben- authors emphasize the relative importance of extrinsic fac- nett et al. 2017), and the dearth of relevant data may in part tors, such as climate or climatic stability (Couvreur 2014), explain thepoor correspondence of our empirical results with whereas others highlight intrinsic factors, such as history the model predictions. We list three future directions of of diversification of different lineages in different regions, empirical research. First, with respect to the determinants of which have resulted in, for example, a diverse understory tree monodominance, we need to better understand which traits flora only in South America (Gentry 1993; Terborgh et al. are important for tree species’ fit to the local environment 2016). Another potentially important intrinsic factor is EM (Kraft et al. 2008; Fortunel et al. 2014; Laughlin et al. 2015; association and its influence on the origin and maintenance Forrestel et al. 2017) and how correlated different traits are of tropical tree diversity, as outlined here. Clearly, more re- with competitive abilities and the strength of plant-soil feed- search is needed to understand EM influences on patterns backs (Uriarte et al. 2010; Fortunel et al. 2016). For example, of tree monodominance and how this relates to the commu- some traits may be more closely related to competitive ability nity assembly and diversity of regional tree floras. and others to niche differences (Mayfield and Levine 2010; Kraft et al. 2015b; Kunstler et al. 2016). Likewise, the pat- Conclusions terns observed in the relationship between trait variability and monodominance may depend on which traits are in- To our knowledge, this is the first study to develop a hypoth- volved in which kinds of interactions in the system. This pos- esis to explain why alternative stable states of either local EM sibility is supported by the fact that in our analyses of trait monodominance or relative rarity arise in some tropical re- variation, some traits showed patterns consistent with our gions and not others. Specifically, our hypothesis is that an predictions, whereas others did not. To wit, plant-soil feed- increase in the size of the EM tree species pool results in an backs involving EM species in Borneo may be mediated more increase in the complexity of biotic interactions, which in strongly by the effects of trait variation in leaf structure and turn prevents monodominance by a single EM species. We physical defense on litter decomposability than by the effects have proposed that under a diverse EM host species pool, of foliar N because of the generally low fertility of soils in this tree species maintain divergence in traits because of the com- Bornean forest. Thus, whether patterns of variation consis- plex plant-soil feedbacks among the many different tree spe- tent with trait-environment matching are observed may not cies, with no species achieving a monodominant status. In only depend on how correlated the traits examined are with contrast, under a depauperate EM host species pool, tree spe- plant-soil feedbacks but also be constrained by the local envi- cies experience strong selection for evolution toward the ronment. Second, to complement the indirect inference of optimal abiotic conditions, allowing them to become com- species performance and interactions through trait analysis, petitive enough to form monodominant stands via positive more direct measurements of the Cijk values could be helpful. plant-soil feedback that causes priority effects. Overall, this One approach may be leveraging trait values in neighbor- study highlights the importance of geographical variation in hood models (Uriarte et al. 2010; Fortunel et al. 2016) as es- species pools in understanding the conditions under which

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:57 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c). Tropical Forest Monodominance S119 community divergence results in vastly different alternative raw humus, to clays that are shale derived, less nutrient de- stable states. pleted, and less well drained, with little raw humus. Tree com- Returning to the general topic of convergence, our goal in munity composition and demography vary across soil types, this article has been to point out the utility of studying the with most species exhibiting soil habitat specialization and link between trait convergence and community convergence. tree demography varying across soil types (Davies et al. 2005; To this end, we have explored how trait convergence may Russo et al. 2005). Lambir is the site of a 52-ha research plot affect community convergence, using the case of tropical that was established in 1991 as part of the Center for Tropical monodominanceasan illustrative example.Althoughwehave Forest Science Forest Global Earth Observatory plot network developed only one hypothesis, one thing that is clear from (Anderson-Teixeira et al. 2015) to monitor woody plants. All our work is that the natural historical knowledge of how con- trees ≥ 1 cm in diameter at breast height are tagged, mapped, vergent traits influence species interactions is essential to un- and dentified, and their diameters are measured to the near- derstanding community convergence. est 1 mm. Tree species were sampled for the quantification of leaf Acknowledgments functional traits from within and near the Lambir plot (Russo et al. 2013). The species identity of individuals sampled out- We thank A. A. Agrawal for the invitation to contribute this side of the plot was verified using an on-site herbarium and article to the special issue arising from the Vice Presidential by consultation with local botanists (S. Tan). Species were Symposium of the American Society of Naturalists held in selected to encompass a wide range of families—targeting Austin, Texas, in June 2016. Comments from A. A. Ag- taxa contributing substantially to forest basal area in the Lam- rawal, P.-J. Ke, and three anonymous reviewers improved bir plot—to target species-rich genera, such as Shorea (Dip- the article. T.F. thanks the National Science Foundation terocarpaceae) and Diospyros (Ebenaceae), and to include (NSF; DEB 1555786) and the Terman Fellowship at Stan- species with a range of shade tolerance niches. For each spe- fi ford University for nancial support as well as the Center cies, juvenile (1–5 cm in diameter) and adult (110 cm in di- for Macroecology, Evolution, and Climate and the Section ameter) trees were sampled as much as possible, with one to of Microbiology at the University of Copenhagen for sab- 22 individuals sampled per species. batical support. K.G.P. was supported in part by the NSF (DEB 1249342 and RAPID 1361171) and a Department of Energy Early Career Grant (DE-SC0016097). S.E.R. was sup- Quantification of Leaf Functional Traits ported in part by the NSF (RAPID 1361171). Author partici- fi pation: T.F. and K.G.P. designed the study; M.N. conducted From each tree, three to ve mature, sunlit, minimally and analyzed simulation in collaboration with T.F.; C.B., damaged leaves were harvested. The petiole was cut from P.V.A.F., and C.F. collected data from Peru; S.E.R. collected the lamina, which was gently cleaned of debris. Fresh leaf data from Borneo; K.G.P. and S.E.R. analyzed data from Peru laminas were scanned (Canon LiDE 110), and the images and Borneo; T.F. wrote the first draft of the manuscript; and were analyzed with ImageJ software (Schneider et al. 2012) 7 all authors contributed to writing the manuscript. to estimate the area of each. After oven drying at 60 Cfor 3 days, the dry weight of each lamina was recorded. The SLA (cm2/g) was calculated as fresh area divided by dry APPENDIX weight, and SLA was averaged across leaves of each individual. After drying, the midvein was removed from each Methods for Quantifying Leaf Functional Traits lamina. Laminas from each individual were ground to- in Bornean Rain Forest gether to a fine, uniform powder and analyzed by elemen- Study System and Species Selection tal combustion for carbon and nitrogen content (Costech Analytical Elemental Combustion System 4010). For each Lambir Hills National Park (Lambir) is located in the north- individual, percent carbon (C) and nitrogen (N) were cal- western part of Borneo in the Malaysian state of Sarawak 0 0 culated as the mass of C (or N) in the sample/dry mass of (4720 N, 113750 E). Lambir receives approximately 3,000 mm # ∶ 1 the sample 100, and C N ratio was calculated as per- of rainfall annually, with all months averaging 100 mm cent C/percent N. Trait values were averaged across indi- (Watson 1985). The region has the highest tree species rich- viduals to obtain mean values for each tree species. ness recorded in the Paleotropics (Ashton 2005), with species in the Dipterocarpaceae dominating the forest (Lee et al. 2002). The soils and geomorphology of Lambir have been previously described (Baillie et al. 2006; Tan et al. 2009). The Literature Cited soils range from coarse loams that are sandstone derived, Ashton, P. S. 2002. Dipterocarpaceae. Pages 182–197 in K. Kubitzki, leached, nutrient depleted, and well drained with substantial ed. The families and genera of vascular plants. V. Flowering plants:

Dicotyledons. Malvales, Capparales and non-betalain Caryophyl- Couvreur, T. L. P. 2014. Odd man out: why are there fewer plant lales. Springer, Berlin. species in African rain forests? Plant Systematics and Evolution Bagchi, R., R. E. Gallery, S. Gripenberg, S. J. Gurr, L. Narayan, C. E. 301:1299–1313. Addis, R. P. Freckleton, and O. T. Lewis. 2014. Pathogens and in- De Deyn, G. B., J. H. C. Cornelissen, and R. D. Bardgett. 2008. Plant sect herbivores drive rainforest plant diversity and composition. functional traits and soil carbon sequestration in contrasting bi- Nature 506:85–88. omes. Ecology Letters 11:516–531. Baillie, I. C., P. S. Ashton, S. P. Chin, S. J. Davies, P. A. Palmiotto, S. E. Dickie, I. A., N. Koele, J. D. Blum, J. D. Gleason, and M. S. McGlone. Russo, and S. Tan. 2006. Spatial associations of humus, nutrients 2014. Mycorrhizas in changing ecosystems. Botany 92:149–160. and soils in mixed dipterocarp forest at Lambir, Sarawak, Ma- Donovan, L. A., H. Maherali, C. M. Caruso, H. Huber, and H. de laysian Borneo. Journal of Tropical Ecology 22:543–553. Kroon. 2011. The evolution of the worldwide leaf economics spec- Baraloto, C., S. Rabaud, Q. Molto, L. Blanc, C. Fortunel, B. Hérault, trum. Trends in Ecology and Evolution 26:88–95. N. Dávila, I. Mesones, M. Rios, E. Valderrama, and P. V. A. Fine. Drake, J. A. 1991. Community-assembly mechanics and the structure 2011. Disentangling stand and environmental correlates of above- of an experimental species ensemble. American Naturalist 137:1–26. ground biomass in Amazonian forests. Global Change Biology Duchesne, L. C., B. E. Ellis, and R. L. Peterson. 1989. Disease sup- 17:2677–2688. pression by the ectomycorrhizal fungus Paxillus involutus:contri- Bennett, J. A., H. Maherali, K. O. Reinhart, Y. Lekberg, M. M. Hart, bution of oxalic acid. Canadian Journal of Botany 67:2726–2730. and J. Klironomos. 2017. Plant-soil feedbacks and mycorrhizal type Egerton-Warburton, L., and M. F. Allen. 2001. Endo- and ectomy- influence temperate forest population dynamics. Science 355:181– corrhizas in Quercus agrifolia Nee. (Fagaceae): patterns of root col- 184. onization and effects on seedling growth. Mycorrhiza 11:283–290. Bever, J. D., T. G. Platt, and E. R. Morton. 2012. Microbial popula- Fargione, J., C. S. Brown, and D. Tilman. 2003. Community assem- tion and community dynamics on plant roots and their feedbacks bly and invasion: an experimental test of neutral versus niche pro- on plant communities. Annual Review of Microbiology 66:265–283. cesses. Proceedings of the National Academy of Sciences of the Brearley, F. 2012. Ectomycorrhizal associations of the Dipterocarpa- USA 100:8916–8920. ceae. Biotropica 44:637–648. Forrestel, E. J., M. J. Donoghue, E. J. Edwards, W. Jetz, J. C. O. du Brundrett, M. C. 2002. Coevolution of roots and mycorrhizas of land Toit, and M. D. Smith. 2017. Different clades and traits yield sim- plants. New Phytologist 154:275–304. ilar grassland functional responses. Proceedings of the National ———. 2009. Mycorrhizal associations and other means of nutri- Academy of Sciences of the USA 114:705–710. tion of vascular plants: understanding the global diversity of host Fortunel, C., P. V. A. Fine, and C. Baraloto. 2012. Leaf, stem and root plants by resolving conflicting information and developing reli- tissue strategies across 758 Neotropical tree species. Functional able means of diagnosis. Plant and Soil 320:37–77. Ecology 26:1153–1161. Cavender-Bares, J., K. H. Kozak, P. V. A. Fine, and S. W. Kembel. Fortunel, C., C. E. T. Paine, P. V. A. Fine, N. J. B. Kraft, and C. 2009. The merging of community ecology and phylogenetic biol- Baraloto. 2014. Environmental factors predict community func- ogy. Ecology Letters 12:693–715. tional composition in Amazonian forests. Journal of Ecology Chapin, F. S., III, K. Autumn, and F. Pugnaire. 1993. Evolution of 102:145–155. suites of traits in response to environmental stress. American Nat- Fortunel, C., R. Valencia, S. J. Wright, N. C. Garwood, and N. J. B. uralist 142:78–92. Kraft. 2016. Functional trait differences influence neighbourhood Chesson, P. L. 1985. Coexistence of competitors in spatially and tem- interactions in a hyperdiverse Amazonian forest. Ecology Letters porally varying environments: a look at the combined effects of 19:1062–1070. different sorts of variability. Theoretical Population Biology 28: Fox, B. J. 1987. Species assembly and the evolution of community 263–287. structure. Evolutionary Ecology 1:201–213. Clements, F. E. 1936. Nature and structure of the climax. Journal of Fransson, P. M. A., Y. K. Toljander, C. Baum, and M. Weih. 2015. Ecology 24:252–284. Host plant-ectomycorrhizal fungus combination drives resource al- Comita, L. S., H. C. Muller-Landau, S. Aguilar, and S. P. Hubbell. location in willow: evidence for complex species interaction from a 2010. Asymmetric density dependence shapes species abundances simple experiment. Ecoscience 20:112–121. in a tropical tree community. Science 329:330–332. Frederickson, M. E., M. J. Greene, and D. M. Gordon. 2005. ‘Devil’s Connell, J. H., and M. D. Lowman. 1989. Low-diversity tropical rain gardens’ bedevilled by ants. Nature 437:495–496. forests: some possible mechanisms for their existence. American Fukami, T. 2009. Convergence. Pages 188–191 in R. G. Gillespie and Naturalist 134:88–119. D. A. Clague, eds. Encyclopedia of islands. University of Cali- Conte, G. L., M. E. Arnegard, C. L. Peichel, and D. Schluter. 2012. fornia Press, Berkeley. The probability of genetic parallelism and convergence in natural ———. 2015. Historical contingency in community assembly: inte- populations. Proceedings of the Royal Society B 279:5039–5047. grating niches, species pools, and priority effects. Annual Review Cornelissen, J., R. Aerts, B. Cerabolini, M. Werger, and M. van der of Ecology, Evolution, and Systematics 46:1–23. Heijden. 2001. Carbon cycling traits of plant species are linked Fukami, T., and M. Nakajima. 2011. Community assembly: alterna- with mycorrhizal strategy. Oecologia (Berlin) 129:611–619. tive stable states or alternative transient states? Ecology Letters Corrales, A., S. A. Mangan, B. L. Turner, and J. W. Dalling. 2016. An 14:973–984. ectomycorrhizal nitrogen economy facilitates monodominance in ———. 2013. Complex plant-soil interactions enhance plant species a Neotropical forest. Ecology Letters 19:383–392. diversity by delaying community convergence. Journal of Ecology Cortois, R., G. B. De Deyn, T. Schröder-Georgi, and W. H. van der 101:316–324. Putten. 2016. Plant-soil feedbacks: role of plant functional group Fukami, T., M. Nakajima, C. Fortunel, P. V. A. Fine, C. Baraloto, S. E. and plant traits. Journal of Ecology 104:1608–1617. Russo, and K. G. Peay. 2017. Data from: Geographical variation in

community divergence: insights from tropical forest monodominance Liu, X., M. Ligan, R. S. Etienne, Y. Wang, C. Staehlin, and S. Yu. by ectomycorrhizal trees. American Naturalist, Dryad Digital Repos- 2012. Experimental evidence for a phylogenetic Janzen-Connell itory, http://datadryad.org/resource/doi:10.5061/dryad.c0kr7. effect in a subtropical forest. Ecology Letters 15:111–118. Gentry, A. H. 1993. Diversity and floristic composition of lowland Mangan, S. A., S. A. Schnitzer, E. A. Herre, K. M. L. Mack, M. C. tropical forests in Africa and South America. Pages 500–547 in Valencia, E. I. Sanchez, and J. D. Bever. 2010. Negative plant-soil P. Goldblatt, ed. Biological relationships between Africa and South feedback predicts tree-species relative abundance in a tropical for- America. Missouri Botanical Garden Press, St. Louis. est. Nature 466:752–755. Gilbert, G. S., and C. O. Webb. 2007. Phylogenetic signal in plant Mayfield, M. M., and J. M. Levine. 2010. Opposing effects of compet- pathogen-host range. Proceedings of the National Academy of Sci- itive exclusion on the phylogenetic structure of communities. Ecol- ences of the USA 104:4979–4983. ogy Letters 13:1085–1093. Gleason, H. A. 1927. Further views on the succession concept. Ecol- McGuire, K. L. 2007. Common ectomycorrhizal networks may main- ogy 8:299–326. tain monodominance in a tropical rain forest. Ecology 88:567– Hart, T. B., J. A. Hart, and P. G. Murphy. 1989. Monodominant and 574. species-rich forests of the humid tropics: causes for their co- McGuire, K. L., T. W. Henkel, I. G. de la Cerda, G. Villa, F. Edmund, occurrence. American Naturalist 133:613–633. and C. Andrew. 2008. Dual mycorrhizal colonization of forest- Henkel, T. W. 2003. Monodominance in the ectomycorrhizal Dicymbe dominating tropical trees and the mycorrhizal status of non- corymbosa (Caesalpiniaceae) from Guyana. Journal of Tropical Ecol- dominant tree and liana species. Mycorrhiza 18:217–222. ogy 19:417–437. Melville, J., L. J. Harmon, and J. B. Losos. 2006. Intercontinental Horton, T. R., T. D. Bruns, and V. T. Parker. 1999. Ectomycorrhizal community convergence of ecology and morphology in desert fungi associated with Arctostaphylos contribute to Pseudotsuga lizards. Proceedings of the Royal Society B 273:557–563. menziesii establishment. Canadian Journal of Botany 77:93–102. Mouquet, N., J. L. Moore, and M. Loreau. 2002. Plant species rich- Janzen, D. H. 1974. Tropical blackwater rivers, animals, and mast ness and community productivity: why the mechanism that pro- fruiting by the Dipterocarpaceae. Biotropica 6:69–103. motes coexistence matters. Ecology Letters 5:56–65. Johnson, D. J., W. T. Beaulieu, J. D. Bever, and K. Clay. 2012. Con- Nara, K. 2006a. Ectomycorrhizal networks and seedling establishment specific negative density dependence and forest diversity. Science during early primary succession. New Phytologist 169:169–178. 336:904–907. ———. 2006b. Pioneer dwarf willow may facilitate tree succession Ke, P.-J., T. Miki, and T.-S. Ding. 2015. The soil microbial commu- by providing late colonizers with compatible ectomycorrhizal fungi nity predicts the importance of plant traits in plant-soil feedback. in a primary successional volcanic desert. New Phytologist 171: New Phytologist 206:329–341. 187–198. Kennedy, P. G., A. D. Izzo, and T. D. Bruns. 2003. There is high po- Newsham, K. K., A. H. Fitter, and A. R. Watkinson. 1995. Ar- tential for the formation of common mycorrhizal networks be- buscular mycorrhiza protect an annual grass from root pathogenic tween understorey and canopy trees in a mixed evergreen forest. fungi in the field. Journal of Ecology 83:991–1000. Journal of Ecology 91:1071–1080. Odling-Smee, J. F., K. N. Laland, and M. W. Feldman. 2003. Niche Klironomos, J. N. 2002. Feedback with soil biota contributes to plant construction: the neglected process in evolution. Princeton Uni- rarity and invasiveness in communities. Nature 417:67–70. versity Press, Princeton, NJ. Koele, N., I. A. Dickie, J. Oleksyn, S. J. Richardson, and P. B. Reich. Orwin, K. H., M. U. Kirschbaum, M. G. St John, and I. A. Dickie. 2012. No globally consistent effect of ectomycorrhizal status on 2011. Organic nutrient uptake by mycorrhizal fungi enhances eco- foliar traits. New Phytologist 196:845–852. system carbon storage: a model-based assessment. Ecology Letters Kraft, N. J. B., P. B. Adler, O. Godoy, E. C. James, S. Fuller, and J. M. 14:493–502. Levine. 2015a. Community assembly, coexistence and the envi- Pacala, S. W., and D. Tilman. 1994. Limiting similarity in mechanis- ronmental filtering metaphor. Functional Ecology 29:592–599. tic and spatial models of plant competition in heterogeneous en- Kraft, N. J. B., O. Godoy, and J. M. Levine. 2015b. Plant functional traits vironments. American Naturalist 143:222–257. and the multidimensional nature of species coexistence. Proceed- Palmgren, A. 1926. Chance as an element in plant geography. Pages 591– ings of the National Academy of Sciences of the USA 112:797–802. 602 in B. M. Duggar, ed. Proceedings of the International Congress of Kraft, N. J. B., R. Valencia, and D. D. Ackerly. 2008. Functional traits Plant Sciences, Ithaca, NY. and niche-based tree community assembly in an Amazonian for- Peay, K. G. 2016. The mutualistic niche: mycorrhizal symbiosis and est. Science 322:580–582. community dynamics. Annual Review of Ecology, Evolution, and Kulmatiski, A., K. H. Beard, J. R. Stevens, and S. M. Cobbold. 2008. Plant- Systematics 47:143–164. soil feedbacks: a meta-analytical review. Ecology Letters 11:980–992. Peay, K. G., C. Baraloto, P. V. A. Fine. 2013. Strong coupling of plant Kunstler, G., D. Falster, D. A. Coomes, F. Hui, R. M. Kooyman, D. C. and fungal community structure across western Amazonian rain- Laughlin, L. Poorter, et al. 2016. Plant functional traits have glob- forests. ISME Journal 7:1852–1861. ally consistent effects on competition. Nature 529:204–207. Peay, K. G., S. E. Russo, K. L. McGuire, Z. Lim, J. P. Chan, S. Tan, Laughlin, D. C., S. J. Richardson, E. Wright, and P. J. Bellingham. and S. J. Davies. 2015. Lack of host specificity leads to indepen- 2015. Environmental filtering and positive plant litter feedback si- dent assortment of dipterocarps and ectomycorrhizal fungi across multaneously explain correlations between leaf traits and soil fer- a soil fertility gradient. Ecology Letters 18:807–816. tility. Ecosystems 18:1269–. 1280 Peh, K. S. H., S. L. Lewis, and J. Lloyd. 2011. Mechanisms of mono- Li, S. P., M. W. Cadotte, S. J. Meiners, Z. Pu, T. Fukami, and L. Jiang. dominance in diverse tropical tree-dominated systems. Journal of 2016. Convergence and divergence in a long-term old-field succes- Ecology 99:891–898. sion: the importance of spatial scale and species abundance. Ecol- Pennington, T. D., C. Reynel, and A. Daza. 2004. Illustrated guide to ogy Letters 19:1101–1109. the trees of Peru. Hunt, Sherborne.

Petraitis, P. 2013. Multiple stable states in natural ecosystems. Ox- Torti, S. D., P. D. Coley, and T. A. Kursar. 2001. Causes and conse- ford University Press, Oxford. quences of monodominance in tropical lowland forests. American Phillips, R. P., E. Brzostek, and M. G. Midgley. 2013. The mycorrhizal- Naturalist 2:141–153. associated nutrient economy: a new framework for predicting Uriarte, M., N. G. Swenson, R. L. Chazdon, L. S. Comita, W. J. Kress, carbon-nutrient couplings in temperate forests. New Phytologist D. Erickson, J. Forero-Montana, J. K. Zimmerman, and J. Thomp- 199:41–51. son. 2010. Trait similarity, shared ancestry and the structure of Read, D. J. 1991. Mycorrhizas in ecosystems. Experientia 47:376–391. neighbourhood interactions in a subtropical wet forest: implica- Reich, P. B. 2014. The world-wide ‘fast-slow’ plant economics spec- tions for community assembly. Ecology Letters 13:1503–1514. trum: a traits manifesto. Journal of Ecology 102:275–301. Van Nes, E. H., and M. Scheffer. 2004. Large species shifts triggered Reich, P. B., M. B. Walters, and D. S. Ellsworth. 1997. From tropics by small forces. American Naturalist 164:255–266. to tundra: global convergence in plant functioning. Proceedings of Waring, B. G., L. Alvarez-Cansino, K. E. Barry, K. K. Becklund, S. the National Academy of Sciences of the USA 94:13730–13734. Dale, M. G. Gei, A. B. Keller, et al. 2015. Pervasive and strong ef- Richards, P. W. 1952. The tropical rain forest, an ecological study. fects of plants on soil chemistry: a meta-analysis of individual Cambridge University Press, Cambridge. plant ‘Zinke’ effects. Proceedings of the Royal Society B 282: Ricklefs, R. E., and D. Schluter, eds. 1993. Species diversity in ecolog- 20151001. ical communities. University of Chicago Press, Chicago. Zobel, M. 2016. The species pool concept as a framework for study- Russo, S. E., S. J. Davies, D. A. King, and S. Tan. 2005. Soil-related ing patterns of plant diversity. Journal of Vegetation Science 27:8– performance variation and distributions of tree species in a Bor- 18. nean rain forest. Journal of Ecology 93:879–889. Russo, S. E., A. Kochsiek, J. Olney, L. Thompson, A. E. Miller, and S. Tan. 2013. Nitrogen uptake strategies of edaphically specialized References Cited Only in the Online Appendixes – Bornean tree species. Plant Ecology 214:1405 1416. Anderson-Teixeira, K. J., S. J. Davies, A. C. Bennett, E. B. Gonzalez- Samuels, C. L., and J. A. Drake. 1997. Divergent perspectives on com- Akre, H. C. Muller-Landau, S. J. Wright, K. Abu Salim, et al. 2015. – munity convergence. Trends in Ecology and Evolution 12:427 432. CTFS-ForestGEO: a worldwide network monitoring forests in an Schneider, C. A., W. S. Rasband, and K. W. Eliceiri. 2012. NIH Im- era of global change. Global Change Biology 21:528–549. – age to ImageJ: 25 years of image analysis. Nature Methods 9:671 Ashton, P. S. 2005. Lambir’s forest: the world’s most diverse known 675. tree assemblage? Pages 191–216 in D. W. Roubik, S. Sakai, and A. A. Schröder, A., L. Persson, and A. M. de Roos. 2005. Direct experimen- Hamid, eds. Pollination ecology and rain forest diversity, Sarawak – tal evidence for alternative stable states: a review. Oikos 110:3 19. Studies. Springer, New York. Smith, M. E., T. W. Henkel, J. K. Uehling, A. K. Fremier, H. D. Davies, S. J., S. Tan, J. V. LaFrankie, and M. D. Potts. 2005. Soil- Clarke, and R. Vilgalys. 2013. The ectomycorrhizal fungal com- related ﬂoristic variation in the hyperdiverse dipterocarp forest in munity in a Neotropical forest dominated by the endemic diptero- Lambir Hills, Sarawak. Pages 22–34 in D. W. Roubik, S. Sakai, carp Pakaraimaea dipterocarpacea. PLoS ONE 8:e55160. and A. A. Hamid, eds. Pollination ecology and rain forest diversity, Stern, D. L. 2013. The genetic causes of convergent evolution. Nature Sarawak Studies. Springer, New York. – Reviews Genetics 14:751 764. Lee, H. S., S. J. Davies, J. V. LaFrankie, S. Tan, T. Yamakura, A. Sutherland, J. P. 1974. Multiple stable points in natural communities. Itoh, T. Ohkubo, and P. J. Ashton. 2002. Floristic and structural – American Naturalist 108:859 873. diversity of mixed dipterocarp forests in Lambir Hills National Tedersoo, L., T. W. May, and M. E. Smith. 2010. Ectomycorrhizal life- Park, Sarawak, Malaysia. Journal of Tropical Forest Science 14: style in fungi: global diversity, distribution, and evolution of phylo- 379–400. – genetic lineages. Mycorrhiza 20:217 263. Tan, S., T. Yamakura, M. Tani, P. Palmiotto, J. D. Mamit, C. S. Pin, Terborgh, J., L. C. Davenport, R. Niangadouma, E. Dimoto, J. C. S. Davies, P. Ashton, and I. Baillie. 2009. Review of soils on the Mouandza, O. Schultz, and M. R. Jaen. 2016. The African rain- 52-ha long term ecological research plot in mixed dipterocarp for- forest: odd man out or megafaunal landscape? African and Ama- est at Lambir, Sarawak, Malaysian Borneo. Tropics 18:61–86. – zonian forests compared. Ecography 39:187 193. Watson, H. 1985. Lambir Hills National Park: resource inventory Teste, F. P., P. Kardol, B. L. Turner, D. A. Wardle, G. Zemunik, M. with management recommendations. National Parks and Wildlife Renton, and E. Laliberté. 2017. Plant-soil feedback and the main- Ofﬁce, Kuching. tenance of diversity in Mediterranean-climate shrublands. Science 355:173–176. Symposium Editor: Anurag A. Agrawal

This content downloaded from 132.236.027.111 on July 22, 2017 05:14:57 AM All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).