The Concept of Estrus
One dictionary definition of estrus is “the periodic state of sexual excitement in the female of most mammals, excluding humans, that immediately precedes ovulation and during which the female is most receptive to mating” (American Heritage Dictionary of the English Language). In mammalian reproductive biology, the estrous cycle is equivalent to the ovarian cycle. The estrous phase refers to the phase of high fertility and ovulation in the cycle. Estrus is typically synonymous with estrous phase. Many biologists do not refer to reproductive cycles of female nonhuman Old World primates as estrous cycles. Rather, cycles in these primates are often referred to as menstrual cycles, in reference to the blood flow that occurs at approximately 30-day intervals. As we emphasize later, however, the mid-cycle phase occurring within females of these primates appears to share homologies with the estrous phase of other mammalian species (e.g., Dixson, 1998; Nelson, 1995). Other scholars reserve the term menstrual cycle to refer to the ovarian cycles of human females exclusively, which reflects the widespread assumption—one that, we argue in chapter 9, is clearly wrong—that, of all mammalian species, humans alone lack estrus. (This view is expressed in the dictionary definition we quoted above.) Behavioral estrus is typically defined as a restricted period of proceptivity and receptivity characterized by mammalian females’ behavioral readiness to mate, in addition to attractiveness to males, usually, though not invariably, coinciding with relatively high probability of conception (e.g., Beach, 1976; Nelson, 1995; Symons, 1979). A synonym for behavioral estrus is heat (Nelson, 2000).
Estrus Is Not Restricted to Mammals
The Homology of Estrus
Though estrus has traditionally been applied solely to mammalian females, this convention is arbitrary. The estrous phase and behavioral estrus can be observed in all vertebrate taxa, and, we propose, estrus is homologous within vertebrates. Physiological machinery leads female goldfishes and garter snakes to emit, as incidental by-products, hormones or derivatives associated with egg maturation. These emissions accompany enhanced female sexual motivation and attractivity to males (Mendonca & Crews, 1996; Shine et al., 2003). These mechanisms also possess homology with the physiology of the estrous phase and behavioral estrus in the female house mouse (see later discussion). More generally, the similarity of physiological machinery typically associated with fertility across vertebrate taxa arises, in part at least, from its descent from a common female vertebrate ancestor—one that possessed estrogen-facilitated egg maturation, accompanied by enhanced sexual motivation and attractivity.
Analogies to estrus exist in nonvertebrates. Female moths secrete a hormone (restricted to arthropods and their close relatives) soon after the onset of adulthood. It stimulates ovarian development and female mating behavior (for a review, see Nation, 2002; see also Ringo, 1996). In vertebrates, a quite different hormone with a different phylogenetic origin, estrogen, plays a functionally similar role (Nelson, 2000). In vertebrates, as well as many invertebrates, fertile females emit scents highly attractive to males and are simultaneously behaviorally receptive to mating.
Ichthyologists, herpetologists, and ornithologists rarely describe the reproductive seasonality of female fish, frogs, toads, salamanders, or reptiles (including birds) in terms of estrus (for rare exceptions, see Jones et al., 1983, for a discussion of estrus in Anolis lizards; Aldridge & Duvall, 2002, on pit vipers). Nor do these biologists speak of fertile-phase females in these species that exhibit sexual proceptivity, receptivity, and attractivity (i.e., females characterized by the defining qualities of mammalian behavioral estrus) as estrous females. Although Whittier and Tokarz (1992) described the sexual behavior of female reptiles in terms of these qualities, for instance, they did not go so far as to refer to reptilian females exhibiting these qualities as estrous females.
Again, the reproductive cycles of all female vertebrates are regulated by physiological mechanisms, hormonal and neural, that are homologous in part. Vertebrates share a pattern of hormones that typifies high fertility within female reproductive cycles (reviews in Crews & Silver, 1985; Jones, 1978; Lange, Hartel, & Meyer, 2002; Liley & Stacey, 1983; Lombardi, 1998; Nelson, 2000; Smock, Albeck, & Stark, 1998; Whittier & Tokarz, 1992). For example, in all nonmammalian vertebrate species studied, females’ estrogen levels are above a basal concentration at the time when they mate with males (Crews & Silver, 1985), precisely the pattern observed across diverse taxa of mammals (e.g., Nelson, 2000). As well, the hormones associated with ovulation appear to promote female attractivity in vertebrates in general; typically, the attractivity of fertile females is mediated by effects of estrogen (Nelson, 2000). And ovariectomy suppresses female sexual behavior within all vertebrate taxa (though, as we discuss later, nonovarian hormones may primarily affect sexual behavior in some species; Adkins-Regan, 2005; Nelson, 2000). Although studied less intensively than hormonal homology, similar, apparently homologous neurological structures appear to produce heightened female sexual motivation at peak fertility in the reproductive cycle across vertebrate taxa (e.g., Lombardi, 1998; Smock et al., 1998).
In light of these homologies, the convention of using distinct taxon-specific language to describe the sexuality of vertebrate females at peak fertility in their reproductive cycles makes little sense. It fails to recognize that important aspects of the physiology underlying the sexuality reflect homologies. Worse yet, it hinders that recognition. For this reason, we apply the term estrus to the fertile state of all female vertebrates in their reproductive cycles. We furthermore argue that this usage makes scientific sense, because, we propose, estrus is homologous across all vertebrates. It first appeared 400–450 million years ago in a species of fishlike animal ancestral to all vertebrates. As estrogen-facilitated female sexual motivation at high fertility in the reproductive cycle apparently characterizes all (or virtually all; see later discussion) vertebrates, the principle of parsimony supports our proposal (see chapter 2, this volume).
Phylogeny of Estrus Figure 8.1 depicts vertebrate phylogeny, as generally accepted (e.g., Tree of Life website). We propose that significant events in the phylogeny of estrus occurred at time points A, B, and C (based on data from Thornton, 2001). At 450 million years ago (time point A in the tree), gnathostomes (jawed vertebrates) and lampreys diverged. The common ancestor of these two lineages, evidence suggests, possessed an estrogen receptor (a protein that binds with estrogen) that subsequently evolved into two estrogen receptors within gnathostomes (ERα and ERβ). The presence of an estrogen receptor, in turn, is a signature that estrogen was, in these species, physiologically functional. Estrogen in the modern-day lamprey regulates the reproductive maturation and behavior of both sexes; indeed, blood levels of estrogen are sexually monomorphic. It is reasonable to infer that estrogen levels in the common ancestor of lampreys and gnathosomes were also sexually monomorphic.
Figure 8.1 The phylogeny of the vertebrates indicating the timing of significant evolutionary events pertaining to the phylogeny of estrus. A is dated at 450 million years ago and marks the divergence of lampreys from jawed vertebrates (gnathostomes). Lampreys possess an estrogen receptor that is homologous with and ancestral to the gnathostome estrogen receptors, α and β. B is the ancestral species of all the gnathostomes (vertebrates proper). It had estrogen α and estrogen β receptors. The teleost and tetrapod lineages diverged at 400 million years ago. The ancestral species that gave rise to these two lineages (C) had estrogen α and estrogen β receptors. This phylogeny of estrogen is based on findings in Thornton (2001).
Estrogen-facilitated feminization of reproductive maturation and behavior, then, made its phylogenetic debut in the species of fish-like animal ancestral to gnathostomes (at time point B in the tree). This species had ERα and ERβ. This dimorphism was maintained in the common ancestral species of the teleost fishes and tetrapods, which diverged at time point C in the tree, approximately 400 million years ago. And it was subsequently maintained in all branches of teleosts and tetrapods (Thornton, 2001).
Estrogen appears to play a role in the regulation of reproduction in certain mollusks, branchiostomes, and echinoderms. It is the most ancient form of steroid regulation of reproduction (Thornton, Need, & Crews, 2003). Estrogen regulation of female reproduction, however, is not synonymous with estrus. When estrogen evolved the capacity to specifically regulate female reproductive maturation and simultaneously affected female sexual motivation and sire discrimination (see later discussion), estrus came to be. Again, based on the phylogeny we have sketched out, it is reasonable to assume that the reproductive behavior of both females and males of the ancestral species of the lampreys and the gnathostomes was influenced by estrogen, and possibly equally so. No clearcut evidence of estrus in this species exists. Only with the divergence of lampreys and gnathostomes do we see female-specific estrogen effects. Hence, we propose that estrus originated in the ancestral species of the gnathostomes—at time point B. (Though no clear evidence for an earlier origin exists, it is possible, we note, that estrus first emerged in the species of Bilateria that was ancestral to the protostomes and deuterostomes; see phylogeny of steroid receptors in Thornton et al., 2003. If so, however, estrogen-regulated reproduction was lost secondarily in the ancestral species of the Ecdysozoa [arthropods, nematodes and relatives]; Thornton et al., 2003.)
Estrus in many species involves steroid hormones other than estrogen as well. Progesterone plays an important role in some vertebrates (Nelson, 2000). Progesterone is involved in the regulation of female reproduction across nearly all vertebrate groups (e.g., Nelson, 2000; Rasmussen & Murru, 1992), though its role arose more recently than did that of estrogen. The progesterone receptor originated early in the gnathostome lineage (Thornton, 2001). Androgen also affects female sexual motivation and may facilitate estrus in many vertebrate groups (e.g., Nelson, 2000; Rasmussen & Murru, 1992). The gnathostome androgen receptor evolved very early in vertebrates, though more recently than the progesterone receptor (Thornton, 2001).
In some vertebrates, notably externally fertilizing fishes and amphibians, nonsteroids such as prostaglandins also control female sexual motivation (Adkins-Regan, 2005; Argiolas, 1999; Liley & Stacey, 1983). Liley and Stacey (1983) distinguish two types of hormonal regulation of female sexual behavior in vertebrates. In externally fertilizing species, in which mating occurs at oviposition rather than at ovulation (as in internally fertilizing vertebrates), hormones that regulate oviposition, such as prostaglandins, are important proximate causes of female sexual behavior. In contrast, estrogens emitted by maturing egg-bearing follicles regulate female sexual motivation in internally fertilizing species. Estrous sexuality may also play an important role in externally fertilizing species, however, as estrogen plays a causal role in the process of egg development; in that role, it may affect female sexual psychology in ways that regulate mate choice later, when oviposition occurs. In one anuran studied in detail (the túngara frog), estrogen may affect female sexual behavior at oviposition (Lynch, Crews, Ryan, & Wilczynski, 2006).
In addition to estrogen, other hormones may be involved in regulation of estrous mate choice in specific species. Eliminating activity of a gene responsible for oxytocin in house mice (through a gene knock-out method) prevents expression of females’ preference for unparasitized males (Kavaliers et al., 2005). Hence a combination of hormones (estrogen, progesterone, oxytocin, and perhaps others) plays a causal role in mice.
In summary, major hormonal regulators of estrus—estrogen, progesterone, and androgen—existed in early vertebrates; estrogen regulation arose first. Although all three hormones affect female sexual motivation in many vertebrate taxa, estrogen appears to do so in the most widespread fashion (e.g., Nelson, 2000).
The Case of the Musk Shrew One female mammal long thought to exhibit sexual behavior not facilitated by estrogen is the musk shrew (Soricidae, Insectivora). Female shrews copulate prior to follicular development, a time when blood estrogen levels are very low. The shrew aromatizes testosterone into estrogen in the brain, however, and this estrogen controls female sexual motivation (Rissman, 1991). The musk shrew is not an exception to the general rule that estrogen facilitates female sexual motivation in vertebrates.
Functional Similarity of Estrus Across Vertebrata
We have proposed that estrus is homologous (or, perhaps more precisely, possesses certain fundamental homologous features) across vertebrates. We also propose that estrus shares a basic function across all vertebrates: to obtain sires of superior genetic quality. Indeed, we specifically propose that the phylogenetic conservation of estrus within all vertebrate lineages (even if it has been lost in rare instances; e.g., see chapter 11) has occurred because it has been maintained by selection for this common functional effect—good-genes female choice. Hence estrus in fishes and amphibians has been maintained by selection for the same function that has maintained it in reptiles and mammals.
We note that these claims are not tautological. Traits that possess common functions may be analogues, not homologues. And a trait arising in a common ancestor may be maintained in many lineages without it being an adaptation, let alone an adaptation with a common function across the lineages. Incidental effects can be universally homologous within a phylum. The bones within teleost and tetrapod vertebrates are white or nearly white in color, but this feature has not been maintained by selection for its beneficial effects. Whiteness is an incidental by-product of selection for physiology (largely, densely packed calcium phosphate) conferring structural strength. Later in this chapter, we present evidence that, unlike the whiteness of bones, estrus was retained in vertebrate lineages by direct selection for it, not because estrus covaried with another directly selected trait.
Our proposal that estrus is homologous in vertebrates does not imply that estrus is identical in all vertebrate species. Naturally, many specific vertebrate species have evolved specialized, lineage-specific estrous adaptation, which coexists with the homologous features of estrus universal among vertebrates. Hence the female house sparrow and jungle fowl solicit copulations during the fertile phase of the reproductive cycle, and this phase within the species shares some homologous behavioral, motivational, morphological, and physiological similarities. Estrus in each of the two species is also dissimilar to that of the other in particular ways that function in the particular sexual ecology of each species. Hence, for instance, estrous female house sparrows prefer conspecific males with large, melanin-based breast badges, whereas estrous hens prefer roosters with large combs. Though females of these bird species both solicit copulations by crouching, females of other species are receptive at estrus in different ways: lordosis in the female rat, neck-bending in the female Anolis lizard, posterior body straightening in female snakes, or, as in many species, simply standing still to permit mounting (for descriptions of these estrous behaviors, see Nelson, 2000; Whittier & Tokarz, 1992).
The features that fully characterize estrus within any particular vertebrate species, then, had multiple origins during its descent with modification by selection. Hence, consider estrus in the house mouse (Mus musculus). The first estrous novelty to appear in ancestors of house mice, which was selected and maintained throughout its lineage, arose, we have argued, in the ancestral fish-like species from which all vertebrates are descended, a close phylogenetic relative of hagfish and lampreys. Other novelties may have originated at more recent time points along the lineage: for example, in the species of fish ancestral to sharks and other gnathostomes; in the species of sarcopterygiian fish ancestral to all tetrapod vertebrates; in the amphibian species that gave rise to all the amniote vertebrates (reptiles and mammals); in the reptilian species ancestral to all mammals; in the mammalian species ancestral to all rodents; in the species of rodent ancestral to the genus Mus. Rodent estrus is facilitated by the ratio of progesterone to estrogen (Nelson, 2000), for instance, and this specific feature of estrus in rodents possibly arose only in the species ancestral to rodents but not ancestral to other mammals.
The estrus of human females, too, has had multiple origins during its descent with modification by selection for sire choice. Women’s estrus shares more homologous traits with estrus of other Old World primates (Catarrhini) than with New World primates (Platyrrhini) or the primates comprising the Strepsirhini. As a result of common ancestry, women’s estrus is likely more similar in many ways to the estrus of the house mouse than to the estrus of goldfish or hen.
The fact that estrus within particular taxa is characterized by features very different from those of estrus of other species does not, of course, mean that estrus within the different taxa functions in completely different ways or does not possess common origins. Birds, for instance, are an unusual group of reptiles in that olfaction appears to play little role in their sexual behavior (but see Hagelin, Jones, & Rasmussen, 2003). Naturally, however, this fact does not imply that birds lack estrus. The attractivity of estrus in birds is due to features detected by males largely or solely through visual and acoustic modalities. In other vertebrates, males typically assess female-emitted scent associated with cycle fertility through olfaction and taste (Halpern, 1992; Mason, 1992; Thornhill, 1979). At some time point after birds diverged from other reptiles, novelties in attractivity of estrous female birds arose.
Ecological settings in which adaptive estrous choice for good genes occurs will generally be more similar within than between lineages. Nonetheless, widespread exceptions may exist, reflecting similar selection on estrous females in distantly related lineages. For example, certain estrous adaptations of lekking birds and mammals (e.g, choice for males that hold central territories) may reflect convergence. Similarly, estrous females in some fishes, pinnipeds, and penguins may have independently evolved to assess male genetic quality via their ability to swim well or fast. And, possibly, convergent evolution has rendered certain features of women’s estrus more similar to taxonomically distant groups more similar functionally than comparable features in close relatives. (As we later discuss, for instance, women’s estrus may have been shaped by pair bonding, which may have shaped estrus in many birds as well.)
The Importance of Recognizing Estrus as Taxonomically Widespread
The concept of estrus, we have argued, should be applied to the state of selective sexual motivation and related activities of females in the fertile phase of the reproductive cycle, regardless of the vertebrate taxon to which females belong. But why should it matter? Why is it useful or meaningful to designate the fertile state in all female vertebrates as estrus?
Use of a common term to recognize a common origin and a shared function is not merely semantic. Common usage is embedded within and given meaning by a theoretical framework—a theory about the nature, historical causes, and function of fertile sexuality in all vertebrates. This theoretical framework can promote research and discovery in three ways. First, it encourages analysis of the homologous traits involved in estrus across relatively closely related vertebrate species (e.g., across lizards or birds), as well as across distantly related vertebrates (e.g., fish and mammals). Phylogenetic analyses tend to examine traits within specific orders (e.g., primates) and to be restricted to particular traits that may co-occur with estrus (sexual skins and swellings; see Dixson, 1998; Sillén-Tullburg & Møller, 1993; Strassmann, 1996b). The theoretical framework we propose encourages phylogenetic reconstruction of a much broader set of morphological, physiological, and behavioral features that characterize vertebrate estrus.
Second, recognition that estrus is a general phenomenon promotes the application of the comparative method in the study of functional design and thus in understanding of historically effective selection. The comparative method uses data on divergence in adaptation of closely related taxa and convergence of adaptation in distantly related species to identify function. If woman’s estrous sexuality possesses design to obtain good genes for offspring and homologous estrous sexuality of a female house mouse and a sage grouse possess similar design, a comparative approach yields convincing evidence for a fundamental function of estrus.
Third, this theoretical perspective promotes recognition that, in species in which extended sexuality occurs, there may well exist two functionally separable (even if overlapping) sets of sexual adaptations operating during different phases: one operating during an estrous phase typically associated with ovulation and high probability of conception and a phase of female extended sexuality associated with low or no probability of conception. Sage grouse hens visit potential mates at a lek during a restricted estrus corresponding to the egg-laying period and pick sires that are healthy (resistant to malaria and lice; Boyce, 1990). Female collared flycatchers, by contrast, mate outside of the fertile phase, especially with pair-bond mates (see chapters 3 and 10). These differences, however, should not blind us to seeing that sage grouse hens and female collared flycatchers share common features: They both possess estrus, which functions to obtain good genes. A difference between sage grouse hens and female collared fly-catchers is that the latter also possess extended sexuality. Similar to extended sexuality where it occurs in mammals, female extended sexuality in collared flycatchers and other passerine birds appears to function to obtain material benefits (e.g., see chapter 3). The distinction between estrous sexuality and extended sexuality within vertebrates, broadly considered, can aid comparisons across species of the function(s) of estrus and of extended sexuality—specifically, the nature of the benefits females gain by sexual motivation inside and outside the cycle phase of peak fertility, through which selection shaped forms of female sexuality.
In some species (e.g., some bats and snakes), females mate in one season, store sperm, and then ovulate and produce offspring in a later season (Aldridge & Duvall, 2002; Birkhead & Møller, 1993b; Crews & Moore, 1986). In these species with “dissociated reproduction,” according to our framework, estrus does not co-occur with ovulation or fertilization. Components of estrous adaptation, we argue, likely function to secure genetically superior sires. Hence, female choice that is focused on certain male traits because they connote superior genetic quality is estrous sexuality, whether or not it is associated with ovulation and fertilization. Females in species with dissociated sexuality possess estrus. By contrast, female matings that focus on obtaining nongenetic material benefits from males and have little or no prospect of fertilization exemplify extended sexuality. Females of species with dissociated reproduction may or may not possess extended sexuality. In these species, estrogen may underlie the sexual motivation of estrous females, despite the fact that estrus does not co-occur with ovulation. In at least one bat species, female sexual motivation appears to be independent of ovarian hormones (Mendonca et al., 1996). As in the musk shrew, however, estrous sexuality may involve nonovarian estrogen. If estrogen is in fact not involved in estrous sexuality in this bat, it would be a rare exception to the general pattern observed across mammals.
Adaptive extended sexuality, in our framework, typically requires two circumstances. One is male delivery of nongenetic material resources in exchange for sexual access to females. The second is that females have significant control of paternity of offspring, which, we hypothesize, is accomplished through the functional design of estrus. As male material assistance to females did not arise with the earliest vertebrates, estrus evolved earlier than extended sexuality. Indeed, estrus may have been the sole form of female sexuality for a long period of early vertebrate history. The occurrence and distribution of extended female sexuality in the fishes, amphibians, and nonavian reptiles will remain unknown until studies assess the extent of female sexual behavior outside estrus in these taxa. Such studies require awareness that estrus and extended sexuality are distinct functional forms of female sexuality. Here again, we hope that our framework will promote exploration of female sexuality in various nonmammalian vertebrates along lines not yet widely considered.
The Design of Estrus
Theory
The term estrus (introduced in the late 1800s to describe the female equivalent of male rut; Wikipedia, The Free Encyclopedia) derives from the Greek word for the gadfly, oistros. It refers to the frenzied state of the resisting cow when a bot fly (gadfly) attempts to lay eggs on her. By analogy, the estrous-phase female mammal is purportedly in a frenzied state of madness in her desire to mate. The analogy, though perhaps appropriate in some ways, may unfortunately imply that “sex-crazy” estrous females may be driven to and be satisfied by mating with any male. Many mammalogists and other biologists share this view. Alexander and Noonan (1979), for example, explicitly referred to the “relatively uninhibited receptivity of some estrous female mammals, which may accept essentially any male” (p. 442).
This view typically arises from the belief that estrous sexuality functions to ensure conception. Females are fertile during estrus. Within that time window, and only within it, females can conceive offspring. Hence, females seek sperm to achieve conceptions. That is, according to this view, female estrous behavior functions to solicit sperm from males.
As we discussed in another context (female signaling of fertility, or lack thereof), the view that females pay large costs to solicit sperm is theoretically problematic (see chapter 5). As we detailed there, sexual selection on males typically ensures that males will find and be willing to inseminate fertile females. The problem the typical female is likely to face is not a lack of males ready and willing to copulate and offer sperm, but rather having far too many of them around.
Modern evolutionary thinking, then, leads to a very different conceptualization of estrus. At the time of estrus, females can produce offspring, which represents allocation of parental investment (see chapter 5). Adaptive allocation of parental investment demands that females scrutinize ecological conditions when deciding how and when to expend this investment. A widespread, important ecological condition affecting the adaptive expenditure of female parental investment appears to be the quality of the paternal genes that offspring will receive (see chapter 7). Hence female choices of whom to mate with during estrus represent choices of how to allocate valuable and limited reproductive effort to offspring. Selection should shape female estrous adaptations to make these choices wisely. Some males are better sires for a female’s offspring than are other males. Estrous adaptations should function to lead females to discriminate between males and prefer to mate with males that represent adaptive sire choice.
In many vertebrate species, and mammalian species in particular, males provide very little or no material benefits to females or their offspring. In these instances, adaptive sire choice should substantially be about choice of a mate who can deliver genetic benefits to offspring (as well as minimization of costs of superfluous, nonadaptive, and outright maladaptive mating). In species in which males do deliver material benefits (e.g., including but not exclusively pair-bonding species with biparental care), males provide these material benefits but also still deliver genes to offspring. In these species, then, females should still generally possess estrous adaptations to assess and discriminate males’ ability to deliver genetic benefits to offspring. At the same time, females in these species cannot ignore the implications of sire choice for the expected future flow of material benefits they receive from pair-bond partners. Hence during estrus they must gauge not only the genetic benefits they could secure from sires but also the impact of sire choice for the flow of material benefits. These aspects of estrus clearly apply to humans, and we discuss them in chapters to come (chapters 10 and 12). In this chapter, we emphasize estrous adaptations for choosing sires able to deliver good genes to offspring.
We note that adaptation for seasonal achievement of reproductive condition by females and estrous adaptation are functionally distinct, despite being expressed during the same time period in most species. Adaptation for seasonal reproduction evolves by selection for timing that coincides with the most ecologically suitable time for offspring production for mothers or offspring or both. Estrus involves female sexual adaptation for obtaining sires of superior genetic quality.
Examples of Estrous Female Choice
If, in fact, sire choice meaningfully affects offspring success, we should expect that fertile females have been shaped to possess adaptations that favor choice of some males—those who possess superior genetic quality—to sire offspring over other males, all else being equal. And, indeed, evidence points to such adaptation in a variety of nonhuman mammals. In the Asian elephant, a bull’s degree of dominance (predicted by testosterone level) predicts peak-fertility females’ positive response to his scent, which is stronger than females’ preference for this scent outside of peak fertility (Schulte & Rasmussen, 1999). Estrous-phase American bison cows approach dominant males but run away from subordinate ones. Accordingly, dominant males obtain more matings than do subordinates (Wolff, 1998). Similar behavior characterizes female African elephants and pronghorn antelopes in estrus (Byers, Moodie, & Hall, 1994; Poole, 1989). Indeed, estrous preference of pronghorns demonstrably results in offspring with good genes for survival (Byers & Waits, 2006). Estrous topi antelopes prefer lek males over resource-holding males (Bro-Jorgensen, 2003). Red deer in estrus prefer the roars of larger males over the roars of smaller males (Charlton, Reby, & McComb, 2007). Estrous meadow voles prefer males with good spatial ability over males with poor spatial ability (Spritzer, Meikle, & Solomon, 2005). Guinea pigs in estrus prefer heavier, more vigorously courting males over other males (Hohoff, Franzin, & Sachser, 2003). Similarly, female pademelons (a species of marsupial) prefer to associate with the largest male when presented with males of different sizes, but only during estrus (Radford, Croft, & Moss, 1998). In house mice, only females in estrus show an olfactory mate preference for the scent of males with wild-type t-alleles (over males bearing a t-allele that lowers fertility and survival of offspring; Williams & Lenington, 1993). Estrous house mice also prefer the scent of males carrying MHC dissimilar alleles (Potts, Manning, & Wakeland, 1991) and males with relatively few parasites (Kavaliers & Colwell, 1995a, 1995b). Estrous snow voles prefer the scent of males with relatively high levels of hematocrit (an indicator of good nutrition and health) and developmental stability (low FA; Luque-Larena, López, & Gosálbez, 2003). Preferences by estrous females for male-produced odor stimuli have been documented in a variety of other rodents (see reviews in Gosling & Roberts, 2001; Hurst & Rich, 1999). Fallow deer time their estrus to coincide with the presence of older dominant males (Komers, Birgersson, & Ekvall, 1999). The scent of dominant males, but not that of subordinate males, induces estrus in the house mouse (Novotny, Ma, Zidek, & Daer, 1999), and only female house mice in estrus prefer the scent of dominant males (Rolland, MacDonald, de Fraipont, & Berdoy, 2004). Beach (1970) noted that his studies of sexual behavior in domestic dogs revealed that females in estrus “are not indiscriminately receptive, but accept some males much more readily and enthusiastically than others” (p. 445; see also Le Boeuf, 1967). Female choice during estrus also has been observed in tree squirrels (Koprowski, 2007), 13-lined ground squirrels, and feral domestic cats (see discussions later in this chapter). More generally, biologists increasingly recognize that female choice by estrous females in nonprimate mammals importantly determines male mating success (Ginsberg & Huck, 1989).
And the same appears to be true of nonhuman primates. Estrous pygmy lorises prefer competitive males (Fisher, Swaisgood, & Fitch-Snyder, 2003). In the laboratory, estrous-phase rhesus macaques prefer males whose faces have been experimentally manipulated to reveal exaggerated red coloration, a testosterone-facilitated male sexual ornament (Waitt et al., 2003). The copulation calls of female catarrhines may be estrous female-choice adaptations that function to promote conception with high-quality males and exclude mating with other males (see chapter 5). In female nonhuman primates with extended sexuality, copulation with dominant males often coincides with peak fertility (see chapter 5, this volume). Though the relative impacts of female choice, male-male competition, male sexual coercion, and male choice on male access to females are not always clear, female choice appears to play a major role in many nonhuman primates (Dixson, 1998; Pazol, 2003; Stumpf & Boesch, 2005; Wallen, 2000).
Consider, once again, common chimps (see also chapter 3). Males socially dominate females. Yet females show more selective mating during peak estrus (fertile phase of the estrous cycle of about 3–4 days) than during the low-to-zero fertility days of the cycle surrounding both sides of peak estrus (Stumpf & Boesch, 2005). Female rates of proceptivity (which decrease during estrus) and resistance (which increase at that time) affect male access to mating, with (in the small sample Stumpf and Boesch studied) up-and-coming dominant males having greater access. By contrast, during the extended female sexual phase, female chimps mate more promiscuously with multiple males. Peak estrous sexuality of chimps may well function largely to obtain good genes for offspring, unlike extended sexuality, which arguably secures material benefits (see chapter 3).
Because estrus has not been thought to characterize nonmammalian vertebrates, the estrous behavior of females in these species has been less well studied than that of mammals. As we previously discussed, collared flycatchers preferentially mate with extra-pair males when fertile and prefer as extra-pair males those who possess a purported indicator of good genes (see chapter 3 and also chapter 10, this volume; cf. Brommer et al., 2007). Recent research reveals that, as female túngara frogs approach the time for egg deposition (and hence egg fertilization), their ability to discriminate acoustical signals of conspecific males increases (Lynch, Rand, Ryan, & Wilczynski, 2005). Female midwife toads too possess enhanced ability to discriminate males’ calls during the ovulatory phase of their reproductive cycle (Lea, Halliday, & Dyson, 2000). In an African cichlid fish, gravid females in fertile phase, but not females in infertile phase, prefer territorial males with high levels of behavioral activity (Clement, Grens, & Fernald, 2005).
Finally, in the guppy, estrus lasts a few days. The ova develop just before the birth of a litter. Female guppies are sexually receptive, respond to males, and mate within an interval of several days prior to giving birth. They are sexually unreceptive at other times (i.e., they lack extended sexuality). Males are sexually attracted to a scent that receptive females emit (Houde, 1997). Though fertile-phase guppies often mate with multiple males during a single estrus, they partly control the size of inseminates. As a result of female preference for their semen, more highly ornamented males transfer larger ejaculates (Pilastro, Evans, Sartorelli, & Bisazza, 2002), which may account for these males’ greater paternity in broods with multiple sires (Pitcher, Neff, Rodd, & Rowe, 2003).
In sum, across a wide variety of mammals and other vertebrate taxa, estrous females are choosy, not indiscriminate. Furthermore, in a host of systems, evidence clearly points toward estrous female choosiness for mates of superior genetic quality.
Estrus and Constraints on Female Choice
Conception-Assurance Adaptation
The common belief that estrous females are indiscriminate and interested in mating with most any male, despite the many studies that contradict it, may partly be rooted in misleading observations made in laboratory settings in which females possess little choice. As Nelson (2000) wrote:
Female mammals in estrus have often been portrayed as “out of control” because they appear to be indiscriminate about their mating partners, but part of this portrayal results from the laboratory testing situations used, especially for rodents. Dogs and many other mammals (especially primates) display substantial selectivity when in estrus. (p. 281)
At the same time, the view that estrous sexuality importantly reflects adaptation that functions to secure sires of high genetic quality does not preclude female sexual adaptation to assure conception. When females are conditionally constrained in their mate choices (or pay heavy search costs or costs for resisting available males), their best option may be to relax their standards of mate choice and mate with available males. In these species, however, we hypothesize that estrous adaptations that function to obtain sires with superior genes also exist. That is, we propose that, in some species, females possess conditional conception-assurance adaptation in addition to estrous adaptations to obtain genetic benefits for offspring. We also suspect, but cannot know for certain, that female conception-assurance adaptation is relatively uncommon across species. In most species, we suggest, sexual selection on males will have designed them to reliably find and inseminate fertile females, and females will not have been exposed to the conditions that could favor conditional conception-assurance adaptation.
When costs of assessing male genetic quality are high (e.g., due to predation risks or energy expenditure), estrous females should be naturally selected to reduce them—for instance, by limiting their movement. Some biologists assume that high costs of choice lead females to sample just a few males, thereby restricting female ability to obtain sires with good genes (e.g., Crowley et al., 1991; Pomiankowski, 1987; Real, 1990). In a review of evidence on 11 species, Gibson and Langen (1996) found that searching females sampled only an average of four males (for additional evidence on the pied flycatcher and bowerbird, see also Dale, Rinden, & Slagsvold, 1992; Uy, Patricelli, & Borgia, 2001).
At the same time as females constrain their movement to reduce search costs, however, males should be sexually selected to encounter sedentary estrous females (see chapter 5). Hence females in many species may be presented with and able to choose males of relatively high quality, even when they limit search costs. Of course, we do not argue that female choice is cost-free in terms of energy and time. Rather, we simply suggest that future studies consider the possibility that sexual selection on males typically means that suitable males are available to fertile-phase females. (Cost reduction may also take the form of relatively low-cost choice strategies, such as copying the choices of other females observed to choose mates; Pruett-Jones, 1992; Godin, Herdman, & Dugatkin, 2005.)
We similarly expect that sexual selection on males to find fertile females generally means that seasonal constraints on reproduction do not typically limit the ability of females to have their eggs fertilized by males of high genetic quality. Exceptions may exist during late season if, at that time, many males have died.
Another situation that may have been recurrent at sufficiently high rates to yield selection for conception-assurance adaptation and hence reduced discrimination is that in which females must deposit eggs (to be externally fertilized) within a short period of time following egg maturation, as in anurans and many fishes. For example, túngara frogs must lay their eggs soon after egg maturation (Lynch et al., 2005). (In contrast, female midwife toads can store mature eggs for a considerable period of time; Lea et al., 2000.) Accordingly, selection has produced a female sexual adaptation of reduced discrimination, leading female túngara frogs to mate with any males exhibiting any call type. Reduced discrimination, however, is highly conditional, characteristic only of females that are highly gravid with eggs. Moreover, this conception-assurance adaptation coexists with estrous sexuality for obtaining good genes. Indeed, egg-laden female túngara frogs retain a strong preference for male calls that combine the whine-chuck, which may be a preference for a sire of superior genetic quality (see Lynch et al., 2005, 2006). Only when offered no choice, then, will gravid females mate with lower quality males. In this species, males of high genetic quality may not be present, despite sexual selection on males to find egg-laden females, because of high rates of sex-specific predation or limitations imposed by breeding habitat (e.g., as in other anurans, the absence of a local, suitable body of water for mating aggregations and egg deposition).
As Nelson (2000) notes (see the preceding quotation), artificial laboratory studies may have misled researchers to the view that estrous female mammals mate indiscriminately. If an estrous female is placed in a cage with a single male, mating typically occurs. It need not follow that estrous females possess no preferences. First, in most mammalian species, males compete for access to mates, and hence females may perceive single males with whom they are caged as winners of a competition. (Otherwise, why would they have gained access to a female, with no interference from other males?) Second, female willingness to mate in these situations may reflect the same kind of adaptation that leads female túngara frogs to mate indiscriminately when highly gravid: back-up adaptation to assure conception under severe mate-choice constraints.
By no means, however, do females generally express adaptation to ensure conception when mate choices are limited. Zoos and captive breeding programs, in fact, frequently face the problem that females reject males that zookeepers provide as mates, and not merely because cage conditions are unnatural (Møller & Legendre, 2001). Indeed, the Allee effect can be observed in many species: In conditions of low population density and, accordingly, poorer opportunities for mate choice, females reproduce at lower rates, even when healthy and well nourished (for a review, see Møller & Legendre, 2001). Hence females of many species appear to possess adaptation to reduce immediate expenditure of valuable parental investment when mate choice options are restricted.
Sexually Ardent Males Constrain Female Choice
As we have emphasized, female reproductive success is generally limited by female abilities to optimize the expenditure of their parental investment, where optimization partly depends on ability to place good paternal genes in offspring through appropriate choice of sires. Ability to secure good genes by females, however, is not typically limited by male willingness or ability to deliver them by mating; males are selected to be willing and able to deliver them. Female choice may nonetheless be constrained widely by the outcomes of intense male-male competition for females. Male sexually selected adaptations that function to promote male reproductive success may lead to outcomes that are not in female reproductive interests. Hence males may be selected to manipulate and/or control female reproduction. Sexually antagonistic selection, which gives rise to a “battle of the sexes” (intersexual antagonistic coevolution; Arnqvist & Rowe, 2005), dates in the history of life to the appearance of anisogamy (Gowaty, 1996; Parker et al., 1972), and it continues. Males are relentlessly selected to circumvent female mate choice by control, manipulation, and coercion, and females in turn are selected to control the timing and other events surrounding adaptive parental investment, including sire quality. (See also Gowaty, 1997, on “sexual dialectics theory.”) Earlier, we mentioned female resistance to males and other female traits that apparently evolved in this context (e.g., estrous female guppies’ behaviors of seeking stream habitats that are too costly for low-quality males to occupy and female primate copulatory calls that reduce coercion by unwanted sires; see chapter 5).
From a theoretical perspective, it makes sense that females who possess free choice can attain greater reproductive success than females whose choice is constrained (assuming that females choose males who provide genetic benefits). Empirical studies demonstrate, in dramatic fashion, this phenomenon in Madagascar cockroaches, fruit flies, house mice, and mallard ducks (Bluhm & Gowaty, 2004a, 2004b; Drickamer, Gowaty, & Holmes, 2000; Moore, Gowaty, & Moore, 2003; Partridge, 1980). In these studies, females achieved higher reproductive success when allowed to reproduce with preferred males than when assigned a male partner by researchers or (in studies of cockroaches and ducks) when males were permitted to interfere with female mate choice through manipulation or coercion (see review in Møller & Legendre, 2001).
Estrus, Polyandry, and Multiple Paternity
Multiple Paternity in Relation to Types of Genetic Benefits
In many vertebrate species, females typically mate with more than one male during a single estrus. One might think that, if females seek genetic benefits for offspring during estrus, they should mate only with the single male who, of available mates, displays the greatest potential for good genes. In fact, however, that need not be the case.
As we discussed in chapter 7, females may seek three different kinds of good genes (Jennions & Petrie, 2000; Zeh & Zeh, 2001): intrinsic good genes, compatible or complementary genes, and diverse genes. Females may seek any one, or even all three, within the estrous phase. Indeed, in some species across diverse taxa females may produce offspring with multiple paternities within one litter or clutch (for birds, see chapter 10, this volume; for mammals, see partial review in Zeh & Zeh, 2001; for lizards, Morrison, Keogh, & Scott, 2002; for snakes, McCracken, Burghardt, & Houts, 1999; for turtles, Pearse, Janzen, & Avise, 2002; for fishes, Kelly, Godin, & Wright, 1999). Offspring with different fathers may optimize different good-genes choice. Clearly, if females choose mates for diverse genes, the point of sire choice should be to obtain offspring with different fathers. This strategy appears to characterize the choices of females of the ruff, a lekking bird without male parental care (Lank et al., 2002). But females may also seek multiple fathers if some males offer intrinsic good genes and others offer compatible genes, even in absence of selection for diverse genes. And, as we discussed in chapter 7, females may select compatible genes through postcopulatory choice mechanisms; some forms of genetic compatibility may be most readily assessed when male DNA can be scrutinized within the female reproductive tract (Zeh & Zeh, 1997).
Sexually Ardent Males and Multiple Paternity
Male sexual ardor may also partly explain multiple paternity in some species. Against the interests of females, selection may favor males that can achieve conception despite failure of females to choose these males as sires. In feral domestic cats, multiple paternity is rare, and socially dominant males sire most offspring when the density of males is moderate (Say, Devillard, Natoli, & Pontier, 2001; Say, Pontier, & Natoli, 2002), presumably due to female choice (Ishida, Yahara, Kasuya, & Yamane, 2001). By contrast, when male density is unusually high, as in many urban areas, multiple paternity in single litters is common. When males are densely distributed, single males cannot monopolize estrous females. Similarly, in the common toad, multiple paternity within single egg batches is due to high density and an excess of males, which leads to multiple males amplexing a female at oviposition (Sztatecsny, Jehle, Burke, & Hoedl, 2006).
Multiple paternity within litters in the 13-lined ground squirrel may similarly be the result of ardent males partly circumventing female preference. In these squirrels, estrus lasts a few hours, during which females typically mate with multiple (usually two) males. Males who first find a female sire 75% of her litter. Females may prefer these males because they may produce sons skilled in mate searching or refinding females after interruption of sexual interactions by later-arriving males (as first-finding males are) or may be in better condition (as first-finding males also are; see Schwagmeyer & Parker, 1987, 1990). By contrast, the paternity that late-arriving males achieve may be due largely to their ability to obtain fertilization despite female preference for skilled mate-searchers in good condition.
Perhaps the best illustration of both female choice during estrus and effort by ardent males to inseminate females contributing to multiple paternity in single broods is provided by the guppy (Kelly et al., 1999). In this species, males are attracted to the scent produced by estrous females and court them. Females prefer males with colorful ornamentation and that vigorously courtship display. When rates of predation by larger fish are high, however, males adaptively reduce their courtship effort and, instead, are more likely to attempt forced copulation without courtship. In the absence of courting males, females in turn relax their otherwise high standards for mate choice. Kelly et al. (1999) found that multiple paternity within broods varied considerably across 10 natural populations. In those populations in which level of predation was relatively high, so too was the level of multiple paternity. When the level of predation was low, the level of multiple paternity was low as well. This pattern is explained if, when female choice is adaptively relaxed because males provide less information through courtship, sexually coercive males achieve greater success in conception than when female choice fully determines paternity.
Are Estrous Females Designed to Promote Sperm Competition?
As just discussed, multiple paternity may occur despite female preference when nonpreferred males coerce matings with estrous females. Alternatively, estrous females may seek multiple paternity through polyandrous matings, as when females pursue multiple types of genetic benefits during a single estrus.
One additional hypothesis discussed widely in the literature is that females are adapted to promote sperm competition (e.g., Baker & Bellis, 1995; Birkhead & Møller, 1998; Shackelford & Pound, 2006). According to this idea, by mating with multiple males during a fertile period, females establish sperm competition. If winners come from males with highly competitive ejaculates, sons too may possess competitive ejaculates. If winners come from males more fit in general (and hence able to produce more viable sperm), sons may also be relatively fit.
The sperm competition hypothesis requires that individual females prefer multiple mates during a single estrus. Multiple mating solely due to coercive strategies of males is not consistent with it. Studies of polyandry have not explored in detail whether females prefer multiple mates during a single estrus. Rather, researchers typically record polyandry by individual females without regard to cycle phase or cause (e.g., female preference for multiple mating vs. avoidance of costs of resistance; see, e.g., the many studies of mammals reviewed by Wolff & Macdonald, 2004).
One exception is research on red jungle fowl by Ligon and Zwartjes (1995). Hens in laying condition (i.e., estrus) were offered a choice of two roosters. The roosters were tethered to prevent forced matings with hens. Each hen was run in multiple mating trials across several days with the same pair of roosters. Nearly all hens chose to mate with both males over successive trials, revealing that jungle fowl hens choose to obtain sperm from more than one male during the production of a single clutch of eggs. It should be noted that males in this study were isolated before the trial and prevented from engaging in male-male competition both before and during the trials. This procedure may elevate and equalize roosters’ self-assessment of social dominance and hence their display of dominance to choosing hens. Hence the results show only that estrous hens prefer insemination by multiple socially dominant males. They do not demonstrate that hens prefer multiple mating with males of lower quality. The results are consistent with an explanation that hens run sperm competition races to be inseminated with good sperm. They are also consistent with female postcopulatory choice of compatible genes (e.g., Zeh & Zeh, 1997).
In chapter 10, we discuss the issue of whether women are adapted to promote sperm competition.
Estrus Will (Nearly) Always Accompany Extended Female Sexuality
In many species, females have estrus with no accompanying extended female sexuality. The converse, however, should occur rarely, if ever: Females should not possess extended sexuality without estrus. Female extended sexuality makes little adaptive sense without estrus. Female choice adaptations of estrus provide control over paternity, which allows extended sexuality to reap male nongenetic material benefits provided by an unsuitable sire at low to zero probability of fertilization by him. Extended female sexuality, in other words, will be favorably selected only when there is adaptation that functions to control sire choice in place in the female’s reproductive repertoire.
Because males in species with extended female sexuality imperfectly discriminate female fertile cycle state from infertile states, it behooves them to copulate when opportunity presented by female sexual interest arises, irrespective of the female’s cycle phase. Hence, when females are available for mating across their cycles, extended male sexual competition occurs. Selection then can favor females that choose mates during low to zero fertility cycle phases, but typically females will use different criteria when exercising choice at these times than during estrus. As we discuss in the next chapter, the most detailed support for different female choice criteria during extended sexuality than during estrus, is found in research on human females.
In some bird species with biparental care, females do not appear to prefer mating with males other than primary mates, even when fertile. In these species, adaptive sire choice may simply constitute choice of the social partner, regardless of his relative genetic quality, and female sexuality during the fertile phase may differ minimally from female sexuality during extended sexuality. As we mentioned earlier, implications of choosing a sire other than a pair-bonded partner for the flow of nongenetic material benefits received by that partner cannot be ignored if female choice of a sire during estrus is to be adaptive; in some cases, the implications of losing a partner’s investment may outweigh any potential benefits derived from seeking another male’s genes for offspring. We discuss possible examples in chapters 10 and 12. As we also discuss in these chapters, however, species in which females possess no adaptation to choose sires other than social partners are relatively rare among socially monogamous birds. Furthermore, humans do not qualify as such a species.
The Musk Shrew Probably Has Estrus
As noted earlier, the musk shrew has been widely claimed to be a mammal that lacks estrus. Females mate prior to, during, or after ovulation. Hence, unlike the typical mammalian female, female musk shrews do not limit sexual behavior to the period of ovulation (Schiml, Wersinger, & Rissman, 2000). For this reason, researchers have argued that this species lacks estrus. Female mating motivation across the reproductive cycle does not, however, imply an absence of estrus. (See chapter 9 on women.) A claim that a species lacks estrus despite extended sexuality is a claim that females of the species apply precisely the same choice criteria to mating choice within and outside of fertile states. Little is known about the sexual behavior and mating system of musk shrews in nature, though much laboratory research has been conducted on their reproductive biology (Jameson, 1988). Female musk shrews are energy limited due to a high metabolic rate and a very limited ability to store energy. They have high rates of food consumption when food is available, and food availability determines female reproductive capability (Temple, Schneider, Scott, Korutz, & Rissman, 2002). Females acquire food from territories held by males. Males are sometimes polygynous, and multiple females cohabit the territory held by a single male.
We suggest that female musk shrews do have both extended sexuality and estrus. Extended sexuality in this species functions to allow females to occupy and feed in males’ territories, which, in turn, allows females to achieve an energy threshold sufficient for ovulation. That is, acquisition of material benefits may explain why female shrews routinely mate outside the ovulatory phase of the cycle. At estrus, we propose, female musk shrews exhibit mate preference for male traits associated with high genetic quality. Evidence supporting the existence of estrus in the musk shrew would be comparable to that in other mammals. If, for example, future research reveals that female shrews at the ovulatory phase of high conception probability show a sexual preference for certain males (e.g., socially dominant males) but during other cycle phases are sexually attracted to a less restricted subset of males, the appropriate conclusion is that musk shrews possess estrus. Clearly, female musk shrews have extended sexuality. Again, we propose that it only makes theoretical sense that they also possess estrus in their reproductive cycle, which functions to control the quality of their offsprings’ sire(s).
Estrous Phase Is Temporally Restricted
Estrus in Nonhuman Vertebrates
Estrous behavior typically occurs during a restricted time period, whether it is coupled with conception or seasonally dissociated with ovulation. The duration of estrus is often about 12 hours to 2 days in ungulates (e.g., domestic cow, goat, pig; white-tailed deer; sable antelope; Landaeta-Hernandez et al., 2002; Romano, 1997; Steverink et al., 1999; Thompson & Monfort, 1999; White, Hosack, Warren, & Fayrerhosken, 1995; Young, Nag, & Crews, 1995). Female lemurs are in estrus about one day (Stanger, Coffman, & Izard, 1995; Wrogemann & Zimmermann, 2001). In tree squirrels (Sciuridae), females may attract males for several days before estrus, but females are typically in estrus for less than 1 day and often less than 8 hours (Koprowski, 2007). Based on data for the duration of female sexual receptivity, we surmise that estrous behavior typically lasts several days in lizards, snakes, and turtles (Whittier & Tokarz, 1992; Young et al., 1995; Weiss, 2002). Estrus in female birds appears to coincide with the period just prior to and during egg-laying and thus the reproductive cycle phase of maximum conception probability (see Birkhead & Møller, 1992). Estrus is said to last 6–9 days in African wild dogs, but mating coincides with a brief peak of estrogen levels (Monfort et al., 1997).
Only Estrous Copulation Has Maternal Investment Implications
Biologists have often assumed that copulation in general is costly to females (e.g., Symons, 1979; Trivers, 1972). We suggest that copulation is less costly to female reproductive success (particularly in species with extended female sexuality) than often assumed. The higher cost of copulation to females than to males reflects females’ higher obligate investment, relative to males’, necessary for offspring production. Female sexual behavior need not be costly, however, when females are not fertile. Copulation by female birds or women occurring during extended sexuality has no cost in terms of its implication for expending maternal investment on a resulting conceptus. (We note, however, that costs of infertile sex need not be equal for the two sexes, as other costs of mating may be sexually asymmetric. For instance, the cost of contracting sexually transmitted disease may be greater for females—who, for instance, are more likely to suffer infertility as a result; see, e.g., Nunn, 2003—than for males. Similarly, male seminal fluid may contain toxins or neuroactive substances designed to manipulate female behavior against her interests; see, e.g., Rice, 1996, and references therein.)
The low cost of copulations during extended sexuality to females in terms of their parental investment suggests that the benefits females obtain from extended sexuality need not be as large as assumed under the traditional view that female copulations involve heavy costs in potential parental investment. If, in fact, copulation outside of the fertile phase is not highly costly to females, male-delivered material benefits need not be particularly great for them to exceed those costs, yielding adaptive extended sexuality. Even material benefits that are seemingly minor or subtle can pay for the costs of extended female sexuality—for instance, nuptial feeding of hens by roosters; mate guarding by some male birds, reducing risk of sexual coercion for females; grooming of some nonhuman female primates by males; and males forming temporary social alliances with females that assist females to access status and its associated resources (see chapter 3).
As we discussed in chapter 4, the capacity of men’s material benefits and services to promote women’s reproduction is major and nonsubtle in humans living as hunter-gatherers and probably accounts for the extreme degree of extended sexuality seen in human females.
Human Estrus Was Not Lost or Extended
Our analysis of estrus is contrary to typical views in the literature about women’s estrus. As noted in chapter 1, many authors have concluded that human females lost estrus during their evolutionary history or that they extended estrus through the evolution of permanent sexual ornaments that deceptively signaled high conception probability in the menstrual cycle, thereby disguising cycle-related fertility (e.g., Alexander, 1990; Alexander & Noonan, 1979; Burt, 1992; Morris, 1967; Steklis & Whiteman, 1989; Strassmann, 1981; Symons, 1979; Szalay & Costello, 1991; Turke, 1984; and many others). In general, these authors have assumed erroneously that estrus functions to signal peak conception probability in the cycle and that estrus in a primate is equivalent to the presence of female sexual swellings. Burt (1992), for example, argued that sexual swellings, and hence estrus, was lost in the hominin line, not through direct selection associated with advantages of concealed estrus but because swellings are costly and, in human females, the benefits of signaling ovulation with swellings did not pay for their costs (see also Pawlowski, 1999a).
Estrus, however, is not about signaling to males a female’s fertile state. More generally, then, phylogenetic analyses of the loss of a particular type of sexual ornament of Old World primates—female sexual swellings (Sillén-Tullberg & Møller, 1993; Strassmann, 1996b)—by no means address the loss of estrus among these species. Sexual swellings were independently lost in primates about 6 (Strassmann, 1996b) or 8–11 (Sillén-Tullberg & Møller, 1993) times. We have no reason to believe, however, that estrous sexuality has ever been lost through evolution in primates.
As we have emphasized in this chapter, estrous sexual adaptations partly function to lead females to prefer as sires males that offer superior genes for offspring. In chapter 9, we lay out the evidence that supports the claim that women have not lost estrous adaptations of this sort. If correct, woman’s estrus may have had its first phylogenetic origin in the fish-like ancestor of all the vertebrates, though, similar to other vertebrates, specific features of women’s estrus have had multiple origins (including some features with origins since hominins diverged from their common ancestor with chimpanzees).
Again, because scholars have often linked estrus with signaling of fertility, they focus on the loss of female signaling of fertility in humans—which, we argue, is misguided because ancestral hominins never had adaptations to signal fertility. At the same time, selection on ancestral hominins may have directly favored female traits that disguise the incidental physiological and other correlates of cycle-related high fertility, as well as disguise sexual interest at peak cycle fertility in sires with good genes, except when mating with them. We discuss this evidence in chapter 11. In a very circumscribed sense, then, women may be thought to have lost behavioral estrus (heat), as women’s overt proceptivity, receptivity, and attractivity, we suspect, does not greatly change across the cycle (though see later chapters). This possibility, however, should not be confused with the broader claim that women have lost estrus and estrous adaptations (see chapter 9).
Summary
Although the estrogen receptor arose earlier, estrogen-facilitated discriminative female sexual motivation at the high fertility phase of the reproductive cycle, thus estrus, had its phylogenetic debut in the fish-like ancestor of all the vertebrates. Estrus, then, is homologous across all vertebrates. The maintenance of estrus after its phylogenetic origin involved selection for its effect of good-genes sire choice. Furthermore, lineage-specific selection accounts for the different forms of good-genes preferences exhibited by females in different taxa. In many species, estrus is facilitated by steroids other than estrogen and, in externally fertilizing species, by certain nonsteroids. However, estrogen remains a fundamental proximate cause across vertebrates. The homology of estrus across the vertebrates is seen too in neurological structures affecting female sexual motivation at the fertile phase of the female reproductive cycle. Estrus occurs in the musk shrew and in species with dissociated female sexuality.
Although the term estrus has been applied almost exclusively to female sexuality at the fertile cycle phase in nonhuman mammals, it is important to recognize the homology of estrus across vertebrates. Such recognition promotes phylogenetic research on female sexuality, as well as the separation and functional analysis of the two types of female sexuality in species with dual female sexuality.
Studies of a variety of nonhuman mammals and other vertebrate species reveal that estrous females are choosy, not indiscriminate as often thought, preferring male traits that reflect actual or likely high genetic quality. Estrus does not function to obtain sperm but instead functions to achieve adaptive sire choice. Adaptation to choose sires that possess genes that increase offspring reproductive value is almost always, if not always, a fundamental component of estrus. Conception-assurance adaptation likely is uncommon because sexual selection on males has designed them to find and inseminate fertile females. Estrus is retained in species with conception-assurance adaptation.
Estrus may function in some species to obtain a combination of genetic benefits for offspring (diverse, compatible, and intrinsic good genes), leading to multiple paternity or sperm competition. If females have adaptation to promote sperm competition, they will exhibit a preference for multiple males during a single estrus. Multiple mating by estrous females and multiple paternity may arise commonly as a result of sexually ardent males coercing copulations.
In species with biparental care and pair bonds, adaptive sire choice must consider the impact of sire choice on the future flow of material benefits and their implications for reproductive success. Only rarely, however, do these impacts fully override potential benefits of seeking a sire other than a social partner for good genes, such that females possess no adaptation for seeking good genes during their fertile phase. Humans do not qualify as such a species (see chapters 10 and 12).
Typically in vertebrates estrus is the sole type of female sexuality. Extended female sexuality will not evolve without accompanying estrus because estrus serves to assure sire quality. Only estrous copulations have potential for maternal investment. Extended sexuality matings do not have this cost.
Loss of estrus and concealed estrus are not equivalent. Human estrus was not evolutionarily lost. Nor was it extended in the form of permanent female ornamentation that deceptively signals female cycle fertility. The evolutionary loss of sexual swellings in certain Old World primates is not equivalent to loss of estrus.