Opinion The (mis)concept of species recognition 1 2 Tamra C. Mendelson and Kerry L. Shaw 1 Department of Biological Sciences, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250, USA 2 Department of Neurobiology & Behavior, Cornell University, Seeley G. Mudd Hall, Tower Road, Ithaca, NY 14853, USA To many, the concept of ‘species recognition’ is integral compatibility and mate quality, and the commonly held to the origin and maintenance of species. However, the model of species recognition in which animals determine heuristic value of species recognition is hampered by its taxonomic status before mate status. We conclude by out- reliance on the problematic concept of species. In this lining two main pathways by which diverging lineages paper, we first discuss assumptions associated with become behaviorally isolated and then direct our attention prevailing use of the term, including the typological to the phenotypes involved. We suggest that a shift in our implications of the concept, the false dichotomy of approach to this subject, away from the conceptual frame- compatibility and mate quality, and the commonly held work of ‘species’ toward one based on fitness consequences model of species recognition in which animals deter- and evolving phenotypes, would dramatically improve our mine taxonomic status before mate status. Subsequent- prospects for understanding the role of communication in ly, we propose research directions aimed to improve animal speciation. our understanding of the role of courtship behavior in speciation. We propose two complementary research What is species recognition? approaches, one addressing the processes that drive Despite many papers addressing the topic in some form, an the evolution of mate recognition systems and the other explicit definition of species recognition is difficult to pin- addressing the phenotypic architecture of behavioral point in any major text or influential publication. Anders- isolation. Our approach emphasizes the fitness conse- son [8] considers species recognition to be a subset of mate quences and multidimensional nature of mate choice. recognition. Ryan and Rand [6] define mate recognition ‘‘. .as a behavioral response indicating that one individual Species recognition and the ‘species’ problem considers another an appropriate mate, even if mistaken- As research into the origin of species comes into greater ly,’’ but fail to define species recognition. focus [1,2], clarifying the conceptual foundation of the The question of whether and why animals most often process becomes increasingly imperative. The concept of mate with conspecifics requires rigorous scientific investi- ‘species recognition’ has long been considered a stable part gation based on clear, operational definitions. If we were to of this foundation, emerging from our effort to understand attempt to define species recognition by borrowing from the role of sexual communication in the origin and main- definitions of individual recognition [9] and kin recognition tenance of animal diversity. Since the emergence of mod- [10], species recognition might be stated to be ‘‘a measur- ern evolutionary biology [3–5], special attention has able difference in behavioral response toward conspecifics focused on differences in courtship behavior among closely as compared to heterospecifics’’. However, we suggest that related species, the importance of these differences in a constructive definition of species recognition should be restricting gene flow, and consequently their role in behav- free of associated hypotheses or mechanisms to explain it, ioral isolation and speciation. Differences in mate recogni- and for that reason, even the simplified definition above tion systems that result in behavioral isolation are imposes both conceptual and practical limitations, due to commonly, although not universally, thought to evolve the problematic and relative term ‘species’. due to sexual selection within species and now function in the capacity of species recognition [6,7], apparently The ‘species problem’ forming an ontological bridge between microevolutionary On a conceptual level, our difficulty in finding a working process and macroevolutionary pattern. definition of species recognition is perhaps not surprising, We argue, however, that the term ‘species recognition’ is because that first would require identifying a working poorly defined, and that its current usage obfuscates our definition of species [11,12]. In practice, the task of naming understanding of the role of behavior in speciation by encouraging assumptions that are both unnecessary and Glossary misleading. Here, we examine limitations of the term ‘species’ in this context and highlight assumptions associ- Compatible mate: a mate with whom a focal organism can produce viable, fertile offspring. ated with the prevailing use of the term, including the Mate quality: the effect of a mate on the direct and indirect fitness of its typological nature of the concept, the false dichotomy of partner. Necessary mate recognition signals: sexual signals that are a necessary prerequisite for attracting a compatible mate. Corresponding author: Mendelson, T.C. ([email protected]). 0169-5347/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tree.2012.04.001 Trends in Ecology and Evolution, August 2012, Vol. 27, No. 8 421 Opinion Trends in Ecology and Evolution August 2012, Vol. 27, No. 8 species – hypothesizing boundaries between taxa that another is itself a phenotype that evolves and not a static might continue to share trait variation – has challenged property of species. taxonomists since before Darwin. Theoretical treatments Nevertheless, the appeal of species recognition under of the species problem are no less controversial, resulting the BSC has now become so ingrained that mating with a in dozens of species concepts and no universal consensus on heterospecific typically is considered a ‘mistake’, a ‘recog- the nature of ‘species’ [11], probably due to the varied nition error’, or an ‘incorrect mating decision’ without any means by which species differentiate [13]. Because knowledge of the fitness consequences of that decision. all species designations are model-based hypotheses in Although in many cases data validate the assumption that practice, the term ‘species recognition’ can never be hy- mating with heterospecifics results in relatively low-fitness pothesis-free, and will forever beg the question: what hybrids, this assumption has been disproven in numerous exactly do organisms ‘recognize’ when they express ‘species systems, as shown by documented evidence of adaptive recognition’? hybridization [16–19] (Figure 1). On a more practical level, the expression of species We suggest a fundamental disconnect between theory recognition almost certainly depends on the identity of and practice is to blame for the common assumption that the referenced heterospecific. Species recognition exhib- all ‘good’ species are incompatible with others. Despite the ited by a focal organism in the face of one heterospecific powerful theoretical utility of the BSC [12], species descrip- might disappear when presented with another, more close- tions in practice are rarely based on quantitative measures ly related species, or one of more similar phenotype. Thus, of reproductive isolation, but rather rely on statistical species recognition under any definition would be opera- analyses of genetic and/or phenotypic character variation. tional only when the heterospecific from which conspecifics Although valuable taxonomic characters often include re- are distinguished is explicitly identified. productive phenotypes, species taxonomies should be con- Another problem with both conceptual and practical sidered hypotheses that are later tested through empirical elements is that the term ‘species recognition’ is most often analysis. applied when an organism is distinguishing among indi- viduals of a single sex; for example, when choosing a mate Distinguishing compatibility and quality [6] or responding to a potential competitor [14]. Yet, species Another common practice in the literature on species of sexual organisms comprise two sexes. Whether animals recognition is to distinguish ‘compatibility’ from ‘quality’ recognize both sexes of their own species as a unified when referring to the attributes of potential mates. A category distinct from heterospecifics is a different ques- compatible mate in this context might be defined as ‘a tion from whether they can distinguish between species mate with whom a focal organism can produce a viable, within a single sex. What defines an acceptable mate for a fertile offspring’, and people often draw on this concept to female is likely to be different than for a male, even though distinguish conspecifics from heterospecifics (see [20]). both categories belong to the same ‘species’. Quality, by contrast, is defined as ‘the effect of a mate For these reasons, a logical and consistent use of the on the direct and indirect fitness of its partner’, such that a concept of species recognition will be difficult to achieve. Yet, high quality mate is one that provides greater fitness even were it possible to solve these species problems, pre- benefits to the receiver. Because ‘compatible’ is often used vailing use of the term ‘species recognition’ is further bur- synonymously with ‘species’ (see above), compatibility dened by various additional assumptions, outlined below. becomes a categorical variable, with a potential mate being either