Sexual Selection: Perspectives and Models from the Neotropics

By Glauco Machado

Sexual Selection: Perspectives and Models from the Neotropics presents new sexual selection research based upon neotropical species. As neotropical regions are destroyed at an alarming rate, with an estimated 140 species of rainforest plants and animals going extinct every day, it is important to bring neotropical research to the fore now.

Sexual selection occurs when the male or female of a species is attracted by certain characteristics such as form, color or behavior. When those features lead to a greater probability of successful mating, they become more prominent in the species. Although most theoretical concepts concerning sexual selection and reproductive strategies are based upon North American and European fauna, the Neotropical region encompasses much more biodiversity, with as many as 15,000 plant and animal species in a single acre of rain forest.

This book illustrates concepts in sexual selection through themes ranging from female cryptic choice in insects, sexual conflict in fish, interaction between sexual selection and the immune system, nuptial gifts, visual and acoustic sexual signaling, parental investment, to alternative mating strategies, among others. These approaches distinguish Sexual Selection from current publications in sexual selection, mainly because of the latitudinal and taxonomic focus, so that readers will be introduced to systems mostly unknown outside the tropics, several of which bring into question some well-established patterns for temperate regions.

• Synthesizes sexual selection research on species from the Neotropics
• Combines different perspectives and levels of analysis using a broad taxonomic basis, introducing readers to systems mostly unknown outside the tropics and bringing into question well-established patterns for temperate regions
• Includes contributions exploring concepts and theory as well as discussions on a variety of Neotropical vertebrates and invertebrates, such as insects, fish, arthropods and birds

• Language: English
• Publisher: Academic Press
• Release date: Sep 25, 2013
• ISBN: 9780123914569

Book preview

1

Macroecology of Sexual Selection

Large-Scale Influence of Climate on Sexually Selected Traits

Rogelio Macías-Ordóñez,¹, Glauco Machado² and Regina H. Macedo³,    ¹Red de Biología Evolutiva, Instituto de Ecología, A.C., México,    ²Departamento de Ecologia, Instituto de Biociências, Universidade de São Paulo, Brazil,    ³Departamento de Zoologia, Instituto de Biologia, Universidade de Brasília, Brasília, Brazil

Abstract

The exuberant variety of neotropical life forms has lured naturalists for centuries. Darwin himself was amazed by this diversity, and among the first to suggest that many of their traits were not the result of natural (viability) selection but rather of an additional and sometimes opposite selective force: sexual selection. Does this imply that sexual selection is stronger or acts differently in tropical environments? Although sexual selection is arguably the most studied evolutionary mechanism nowadays, a broad geographic perspective is seldom applied to answer this question. Our aim is to provide a theoretical framework to approach the study of sexual selection in a broad geographic scale using the Neotropics as a reference, highlighting the region’s great and frequently overlooked environmental diversity. We define macroecology of sexual selection as the large-scale influence of climatic conditions on sexually selected traits. Broad predictions are postulated concerning the effects of abiotic and abiotic factors that co-vary with latitude on life history, reproductive behavior, and sexually selected traits of arthropods, and of ectothermic and endothermic vertebrates.

Keywords

life history; climate types; sexual selection; neotropics; mating systems

Acknowledgment

This chapter was greatly improved by the suggestions made by Dr Robert Ricklefs, whose ongoing interest in and enthusiasm for the tropics is a source of inspiration to us all.

Introduction

The exuberant variety of life forms in the Neotropics, with their many shapes, sounds, smells, and colors, have lured naturalists for centuries. Darwin himself was amazed by this complexity (Darwin, 1862), and was among the first to suggest that many of these traits were not the result of natural (viability) selection, but rather of an additional and sometimes opposite selective force: sexual selection (Darwin, 1871). Does this imply that sexual selection is stronger in tropical environments? Or at least that sexual selection acts differently in the tropics when compared to more temperate or colder regions? Although sexual selection is arguably the most studied evolutionary mechanism nowadays, a broad geographic perspective is seldom applied to answer these kinds of questions. The aim of this chapter is to provide a theoretical framework to approach the study of sexual selection in a broad geographic scale using the Neotropics as a reference to compare what we know of sexual selection and reproductive strategies in this and other environments of the planet. In this chapter, we will postulate broad predictions concerning the effect of environmental factors on reproductive behavior and sexually selected traits of several animal groups, with the objective of stimulating research along these lines.

Defining the Neotropics

A strictly geographic definition of the Neotropics would exclude any area of the American continent north of the Tropic of Cancer and south of the Tropic of Capricorn. Yet the common perception for this region clearly extends beyond the 23° latitudes. The limits, though, vary widely depending on the criteria used when attempting a formal definition. On the one hand, the Neotropical Floristic Kingdom excludes southern South America and northern Mexico (Good, 1964; Takhtajan, 1969). On the other hand, Udvardy’s (1975) Neotropical Realm includes all of South America and the southern tip of Florida in the southeastern United States, but excludes not only northern Mexico but also the highlands of central and southern Mexico, Guatemala, and eastern Honduras. Biogeographical Realms were later adopted by the World Wildlife Fund (WWF) with modifications as ecoregions (Bailey, 1998). A strictly climatic approach would include only those regions in the Americas with a tropical climate, defined as those areas with an average temperature of the coldest month above 18°C (Peel et al., 2007). This approach would lead to the inclusion of areas from the southern tip of Florida, in the southeastern United States, to Brazil, Paraguay, Bolivia, and Peru, but would exclude Chile, Argentina, and Uruguay altogether. A common concept in the literature is that the Neotropics equates to Latin America, encompassing the areas from Mexico southwards. This interpretation of what constitutes the Neotropics corresponds roughly to one of the eight biogeographic realms defined by Olson et al. (2001), but it excludes southern Florida while including temperate and cold regions of the extreme south of South America. This view was explored under a historical perspective in the Preface of the book.

As we hope will become clear, for the scope of this chapter and the rest of this book we really do not need to choose (and thus restrict ourselves to) one definition for the Neotropics; neither do we need to coin a new one. In fact, we will avoid discussing the limits of the Neotropics and focus on the patterns expected along environmental gradients. When addressing macroevolution at large geographical scales, Darlington (1958) suggested that "we do not need a full or precise definition (of the tropics), but one that will emphasize the significant differences between the tropics and the (north) temperate zone". We are interested in the study of selective forces shaping reproductive behavioral, morphological, and physiological traits, and in order to do so we need to contrast environmental conditions shaping those forces. We have decided to use the Neotropics as a landmark or reference for this comparison for several reasons. This is the area we are most knowledgeable about, both from personal as well as professional perspectives, since it is where we carry out our research, and this is also true for almost all authors of this book. The Neotropics also encompass a region relatively well studied compared to other tropical regions of the world, although still relatively understudied when compared to most parts of the northern hemisphere (see below). Whatever limits we may adopt for the Neotropics, it is clear the region is far from homogeneous in terms of climatic conditions, landscapes, and vegetation types, and thus offers a great opportunity to explore how the action of sexual selection in different environments results in distinctive arrays of traits. Furthermore, we are interested in comparing what we know from the Neotropics with what we know from other regions. This yields the possibility of exploring a yet wider range of environmental variables, encompassing the greater wealth of research on sexual selection available from temperate and colder regions.

Tropical environments, as will be more formally defined later, present very different challenges and opportunities from those in more extreme latitudes (see review in Schemske et al., 2009). However, there is an oversimplified view of the Neotropics (and the tropics overall) that, in our opinion, severely constrains the potential to address macro-evolutionary patterns in broad environmental gradients. Although Darlington (1958) defined the tropics as the zone within which, when other conditions are suitable, warmth and stability of temperature permit development of what we call tropical rain forest with its associated complex fauna, the same author acknowledged that many definitions are possible and no simple one is entirely satisfactory, for the tropics vary in climate (wet to dry), vegetation (rain forest to desert), and animal life. We need an environmental framework with enough variability to accommodate predictions along environmental gradients, yet broad enough to be able to reflect nearly global trends. We have chosen to define this environmental framework using the current climate classification (Peel et al., 2007). Climate types provide an independent set of environmental variables that should cover major selective forces shaping sexual selection at a broad geographical scale.

Climatic Regions

A good place to start is the Köppen–Geiger climate classification (Table 1.1; Fig. 1.1). With few modifications, this climate classification has endured the test of time among climatologists and is now widely accepted as the standard in climatology (Peel et al., 2007). This section introduces this classification to the reader, as it is an important foundation point for the chapter. Climate has long been suggested as a powerful axis to address macro-evolutionary patterns (Darlington, 1958). Although latitude has more recently been used as a simpler proxy of climatic gradients (see, for example, Blanckenhorn et al., 2006; Schemske et al., 2009; Moya-Loraño, 2010), it will become evident in this chapter why, for our purposes, latitude is an oversimplification that does not allow predictions in specific combinations of environmental conditions, especially in such a rich mosaic of climate types as can be found within the Neotropics.

TABLE 1.1

Description of Köppen Climate Symbols and Defining Criteria (modified from Peel et al., 2007)

aMAP, mean annual precipitation; MAT, mean annual temperature; Thot, temperature of the hottest month; Tcold, temperature of the coldest month; Tmon10, number of months in which the temperature is above 10°C; Pdry, precipitation of the driest month; Psdry, precipitation of the driest month in summer; Pwdry, precipitation of the driest month in winter; Pswet, precipitation of the wettest month in summer; Pwwet, precipitation of the wettest month in winter. Pthreshold varies according to the following rules: if 70% of MAP occurs in winter then Pthreshold = 2 × MAT, if 70% of MAP occurs in summer then Pthreshold = 2 × MAT + , otherwise Pthreshold = 2 × MAT + 14. Summer (winter) is defined as the warmer (cooler) 6-month period of Oct–Nov–Dec–Jan–Feb–Mar and Apr–May–Jun–Jul–Aug–Sep.

FIGURE 1.1 The Köppen–Geiger climate classification map (modified from Peel et al., 2007) and frequency of field studies on sexual selection in America and Europe (dotted line).

The area where each climate type occurs is expressed in millions of km². For descriptions and criteria of each climate type, see Table 1.1. See color plate at the back of the book.

Although not entirely independent from biological factors, climate probably offers the most independent array of environmental variables defining plant (and thus biomes; Audesirk and Audesirk, 1996) and animal distributions, and thus key environmental factors of natural (viability) and sexual selection. The Köppen–Geiger climate classification is based on a nested set of climatic regimes defining 29 climate types identified by two- or three-letter combinations (Table 1.1). The first level is the broader climate classification of five climate types: tropical (A), arid (B), temperate (C), cold (D), and polar (E) (Peel et al., 2007; Fig. 1.1). With the exception of B (arid), all major climate types are defined using only temperature criteria.

As stated above, a tropical (A) climate includes those regions with monthly average temperature of the coldest month above 18°C. A second nested classification of tropical climate is defined by precipitation regime. The tropical rainforest climate (Af) is defined by precipitation in the driest month above 60 mm. The western Amazon basin is emblematic of this climate (Fig. 1.1). The tropical monsoon climate (Am) is defined by a somewhat more seasonal but still relatively humid driest month precipitation. The eastern Amazon basin, for instance, is characterized by this climate (Fig. 1.1). The tropical savannah climate (Aw) includes a severe dry season contrasting with a heavy rainy season. Most of southern Mexico and central Brazil have this climate type (Fig. 1.1).

The arid (B) climate is defined using a precipitation criterion and comprises regions with very low annual precipitation. A first division of this climate defines two different environments associated with two rain regimes: steppes (BS) and deserts (BW). Within those environments, a temperature criterion defines cold steppes (BSk) or deserts (BWk), and hot steppes (BSh) or deserts (BWh), depending on whether the mean annual temperature is below or above 18°C, respectively. A mosaic of the four different types of arid climate is characteristic of northern Mexico, the southwestern United States, the western slopes of the Andes in Peru, Bolivia, and northern Chile, and the eastern slopes in Argentina (Fig. 1.1).

Temperate (C) and cold (D) climates are both defined by mean average temperature of the hottest month above 10°C, but a mean average temperature of the coldest month either above (C) or below 0°C (D), respectively. Both climates have the same two additional levels of subdivisions. The first level divides each of these climates based on precipitation regimes. Cf and Df do not have a well-defined dry season. Cw and Dw have a dry season during the winter, but a somewhat humid summer. Cs and Ds also have a fairly dry summer. An additional temperature criterion divides all of these climates in two to four climate types depending on the temperature they reach during the summer: a for hot summer, b for warm summer, c for cold summer, and d for very cold winter.

Temperate climates with dry, hot and warm summers (Csa and Csb), known as Mediterranean climates, are also characteristic of the Pacific Coast of the United States and central Chile (Fig. 1.1). Temperate humid climates with hot summers (Cfa) are characteristic of the southeastern United States, southern Brazil, and northern Argentina (Fig. 1.1). Temperate humid climates with warm summers (Cfb) are characteristic of western Europe (Fig. 1.1). Temperate climates with dry winters and hot or warm summers (Cwa and Cwb) are restricted to the central Mexican plateau, the eastern slope of the central Andes, and southeastern Brazil (Fig. 1.1). Temperate climates with cold summers are extremely rare in the Neotropics, and are restricted to very small regions around mountain peaks. Cold humid climates with hot, warm, or cold summers (Dfa, Dfb, and Dfc) are dominant among other cold climates in northern North America and Europe. They cover immense areas in the northern United States, Canada, and northern and eastern Europe. As might be expected, the summer temperature defining each of these climate types shows a clear latitudinal pattern (Fig. 1.1).

The polar (E) climate characterizes those regions where the average monthly temperature of the hottest month is below 10°C. Another temperature criterion divides this climate into polar tundra (ET) if the average monthly temperature is above 0°C, and polar frost (EF) if it is not. Within Europe and North America, polar frost is restricted to Greenland (Fig. 1.1). The greatest extensions of polar tundra can be found in the northern and southern extremes of the American continent, and also in isolated spots in northern Europe and around mountain peaks in the Andes, central Mexico, and central Europe (Fig. 1.1).

In the Neotropics, an intricate mosaic of climate types is the result of complex orographic systems: the two Sierras in North America that converge in central Mexico, a volcanic belt in Central America, the Andean ridge along the Pacific coast, and the long mountain chains along the Brazilian Atlantic coast in South America. The strong but very different influences of the cold Pacific and the warmer Atlantic oceans result in a fragmented pattern of climates, very distinct from the same latitudes in Africa or Asia (Fig. 1.1). Tropical Africa is mostly covered by a continuous area of arid climate and another of tropical climate, with some temperate regions in the southern end of the continent. A similar pattern describes tropical Asia, Australia, and the Pacific Islands (Fig. 1.1). Thus, the Neotropics forms clearly the most diverse and complex region in terms of climates in the world.

How Much do We Know About the Influence of Sexual Selection in Each Environment?

One of the most frequent claims about the Neotropics is that our knowledge about its diversity is very limited. Indeed, the number of new plant and animal species to be discovered in this region is probably very high when compared with temperate and cold regions (see, for example, Adis, 1990; Fouquet et al., 2007). Richness, however, is only part of this diversity, closely related to taxonomy. Ecologists may also ask how much we know about the behavior and evolutionary mechanisms of neotropical species. The answer is probably very little, but we would like to know how little, particularly if we restrict this fundamental question to a specific field of animal behavior: sexual selection. A recent extensive review of latitudinal patterns of biological interactions (Schemske et al., 2009) describes this topic as one of the less addressed under a broad geographic–environmental perspective (see also Twiss et al., 2007). For this chapter, we looked for an answer by reviewing the past 14 years of the four journals of animal behavior with the highest impact factors: Animal Behaviour, Behavioral Ecology, Behavioral Ecology and Sociobiology, and Ethology. These journals publish high-quality international research in many different fields of animal behavior, and there is no a priori reason to suspect that there is any bias in the acceptance of papers according to taxonomic group or author nationality.

Using the Web of Science database, the first step of our search was to select a set of key words that should appear in the title of the papers: breed∗ or mate or mating or reproduct∗ or sexual. Because we needed the precise location where the studies were conducted, we restricted the search to the period from 1998 to October 2011, for which all four journals have pdf files available. We then filtered the results, selecting papers whose abstracts contained the words field or nature or natural or population or wild, but did not contain the words cage∗ or captiv∗ or laboratory. Our aim was to focus our search only on studies conducted under natural conditions, where study animals were subjected to the effects of environmental variables. We read the abstracts of all the selected papers and removed those that were not related to sexual selection. The remaining papers were downloaded and their content was searched for two basic pieces of information: geographic coordinates of the study site (when this was not available in the text, we obtained them using Google Maps®); and studied taxon, which we lumped into major categories or functional groups – arthropods, other invertebrates, ectothermic and endothermic vertebrates (see below). We only selected papers for studies conducted in the Americas and Europe (delineated by the dotted line in Fig. 1.1). Finally, we plotted the coordinates for study sites on the most recent version of the Köppen–Geiger climate map (Peel et al., 2007) to obtain the number of studies conducted in each climate type. Directly from the authors of this map, we obtained a detailed database containing the total area covered by each climate type in the Americas and Europe, and compared the proportion of studies undertaken in each climate type to the relative area covered by the respective climate to provide a %bias index that represented any research bias relative to specific climate types. This index represents the percentage by which the number of studies should increase (negative values) or decrease (positive values) in order to be proportional to the area covered by each climate type.

We came up with 254 studies conducted in 272 sites in 36 countries. Endothermic vertebrates (birds and mammals) account for nearly 57% of the studied taxa, while ectothermic vertebrates and arthropods account for 19.5% and 22.4%, respectively. This pattern still holds when we individually consider the five major types of climate (A to E). Even though the tropical climate (A) covers around 25% of the combined area of the Americas and Europe, only 13% (n = 35) of the studies were carried out in this climate. The arid climate (B) covers 13% of the area, while 10% (n = 26) of the studies were performed in arid environments. Temperate climates cover 19% of the area, but 34% (n = 92) of the studies were carried out in this environment. Cold climates cover 31% of the area, with 40% (n = 109) of the studies. Finally, polar climates cover 13% of the area, but only 4% (n = 10) of the studies were conducted in this environment (Fig. 1.1). There are many ways to read these data, but overall we can say that studies in tropical and arid environments are somewhat under-represented according to their areas in the Americas and Europe. Studies in temperate and cold environments are somewhat over-represented, and polar environments are greatly under-represented. Despite the fact that one out of four studies on sexual selection in the wild has been carried out in the (climatically defined) Neotropics, these results still provide support for the widespread notion that our knowledge on sexual selection is biased towards species living in temperate regions. This is especially true if one considers the ratio of the number of studies to the potentially very high number of species available in the Neotropics, instead of area. The consequences of this bias relative to our assumptions about general patterns are discussed in detail in Chapter 2, which is devoted to rules and exceptions in sexual selection.

Macroecology of Sexual Traits

Studies at large geographic scales, covering areas such as whole continents, are relatively recent in the ecological literature. The great majority of such studies investigate large-scale variation in species richness, abundances, distributions, and body sizes through space (Ricklefs and Schluter, 1993; Gaston and Blackburn, 2000). Macroecological studies make several assumptions, some of which are closely connected to large-scale variation in physiological responses (Chown and Nicolson, 2004). Two widely known examples are the Rapoport effect, i.e., the increase in the latitudinal range sizes of species towards higher latitudes as a consequence of wider climatic tolerances (Stevens, 1989), and Allen’s rule, which states that endotherms from colder climates have shorter limbs than their relatives from warmer climates as a mechanism of heat conservation (Alho et al., 2011, but see also Nudds and Oswald, 2007). We argue here that combining some of these large-scale ecological patterns may result in emergent trends directly related to reproductive ecology. Blanckenhorn et al. (2006), for instance, explored the topic of latitudinal changes in sexual size dimorphism both in vertebrates and in invertebrates by combining the effect of Bergmann’s rule (organisms are larger at higher altitudes or in colder climates) and Rensch’s rule (male body size varies, or evolutionarily diverges, more than female body size among species within a lineage).

In contrast to large-scale studies concerning the variation of morphological, physiological, or diversity attributes exemplified above, and despite the fact that sexual selection is one of the most intensively studied evolutionary forces, there is no solid theoretical framework directly relating the behavioral ecology of sexual traits to their variation in space, i.e., a macroecology of sexual traits. Large-scale rules are virtually absent, and only very recent empirical studies have drawn attention to how spatial environmental heterogeneity can produce great fluctuations in both the strength and the direction of sexual selection (reviewed in Cornwallis and Uller, 2009). Environmental conditions, including both biotic and abiotic factors, may exert a marked influence on many behavioral, morphological, physiological, and life-history traits (Bradshaw, 2003; Chown and Nicolson, 2004; Fig. 1.2). At least a set of these traits are directly or indirectly related to reproduction and are under sexual selection (Fig. 1.2). For instance, it has recently been suggested that a latitudinal gradient of temperature and humidity could be related to movement rate, resulting in more frequent and more diverse biotic interactions, such as parasitism and predation (Schemske, 2009; Moya-Loraño, 2010; Fig. 1.2). In fact, there is strong empirical evidence that parasitism is more prevalent in tropical climates (reviewed in Schemske et al., 2009), and therefore tropical hosts should invest more heavily in parasite defense than their non-tropical counterparts (Møller, 1998). Given that life-history theory predicts a trade-off between investment in defense against parasites versus investment in other fitness components, intense parasitism may affect both the evolution and the expression of condition-dependent sexually selected traits (Møller, 1990; Lochmiller and Deerenberg, 2000).

FIGURE 1.2 Scheme showing the influence of large-scale variation of environmental conditions (biotic and abiotic factors) on sexually selected