Fifty years of sperm competition: the structure of a scientific revolution

1. Introduction

Kuhn [1] recognized that scientific advances are made through periods of ‘normal science’ and ‘revolutionary science’. During periods of normal science researchers explore their field through a series of experimental tests that challenge the central dogma of an existing paradigm. Occasionally someone might proffer a new way of thinking. Sometimes this will be countered by the findings of ‘normal science’. Rarely, it can result in a scientific revolution in which a new paradigm is established and our understanding of the natural world takes a quantum leap forward. Geoff Parker's [2] insight into the evolutionary consequences of sperm competition 50 years ago was nothing short of a scientific revolution.

For a new paradigm to become established, Kuhn [1] argued that it must preserve a large part of the accumulated knowledge that has already accrued, it must resolve some generally recognized problem that can be met in no other way, and it must have significant promise for future problem-solving. As part of his own scientific revolution, Darwin [3] had made a case for sexual selection being responsible for the evolution of exaggerated secondary sexual traits in males, because of the advantages they gave some individuals over others in competition for access to mating opportunities. Darwin saw sexual selection acting mostly on males through direct competition for access to females, and through female choice of elaborately adorned males. Darwinian sexual selection thereby acted through variation in mating success. Though Darwin may have been aware that the females of many species mate with multiple males with implications for paternity [4], he nevertheless failed to recognize that sexual selection could continue after mating. Prior to 1970, research on poeciliid fishes [5] and Drosophila [6] had revealed how females mate with and store sperm from multiple males, and produced broods of mixed paternity. Seminal fluid peptides had been discovered and their effect on the reproductive behaviour and physiology of females documented [7–9]. Even the term ‘sperm competition’ had been coined in several reports of multiple paternity [10–12], and as far back as 1693 John Ray had asked why males should produce so many sperm [4]. Despite the knowledge accumulated on premating sexual selection, there was a problem; contrary to Darwin, single mating by females could no longer be assumed. As a graduate student in the 1960s, Geoff Parker's analysis of the reproductive behaviour and ecology of the yellow dungfly [13] led him to the realization that selection arising from multiple mating by females could explain the hitherto puzzling array of traits such as ejaculate peptides, mating plugs and extended ‘passive phases’ that occurred after mating, because these traits would promote a male's fertilization success when in competition with the sperm of rivals. In laying down his reasoning in Sperm competition and its evolutionary consequences in the insects [2], Parker initiated a paradigm shift in our understanding of postmating sexual selection that has since developed into a mature research discipline [14].

2. Slow start

A Web of Science search using the keywords sperm competition reveals the structure of Parker's scientific revolution (figure 1). During the first decade, few researchers even noted Parker's paper, perhaps because of its taxonomic focus or perhaps because evolutionary biologists were busy shifting their mindset from the group selectionist thinking that had dominated the first half of the twentieth century, to individual selectionist thinking, a paradigm shift that Parker himself had played a large part in generating through his studies of yellow dungflies [4]. Nonetheless a second landmark paper by Parker et al. published in 1972 showed how sperm competition could drive the evolution of anisogamy, providing an answer to Ray's question of why there are so many tiny sperm [15]. But in 1979 four influential papers appeared from a new cohort of behavioural ecologists. Waage [16] discovered that the damselfly penis is endowed with spines that function in the removal of rival sperm before the male delivers his own ejaculate, while Smith [17] showed how male waterbugs copulate with their mates repeatedly throughout oviposition to ensure they gain paternity of the eggs they brood. Waage and Smith were among a small group of young insect behavioural ecologists that would exchange ideas at annual symposia of the Florida Entomological Society, the proceedings of which were published in the Florida Entomologist. Among them was John Sivinski who argued in 1979 that the rapid and divergent evolutionary patterns of sperm morphology were likely owing to sperm competition [19], and John Lloyd who recognized the importance of postmating sexual selection in the evolution of insect genitalia. Lloyd's argument was, given that ‘females, to one extent or another, subvert male interests by internal manipulation of ejaculates, it is not inconceivable that males will have evolved little openers, snippers, levers and syringes that put sperm in places females have evolved for sperm with priority usage—collectively, a veritable Swiss Army Knife of gadgetry!’ [43, p. 22]. Elsewhere, Short [18] suggested that patterns of testis size among the great apes could be explained by the level of sperm competition expected for their breeding systems, a pattern of variation that has since been demonstrated in formal comparative analyses of taxa ranging from parasitic worms to primates. Indeed, in this themed issue Lüpold et al. [44] provide a meta-analysis of this body of comparative work, finding a general effect size in the region of 0.6 for the relationship between sperm competition and relative testis mass, offering broad support for the role of sperm competition in the evolution of male ejaculate expenditure. They also find a significant effect size, though smaller in magnitude, for the role of sperm competition in the evolution of sperm morphology envisaged by Sivinski [19].

Figure 1.

The rise of Parkerian sexual selection. (1) Parker [2] publishes his review of sperm competition and its evolutionary consequences in the insects. (2) Parker et al. [15] show how sperm competition can drive the evolution of anisogamy. (3) Waage [16] discovers the dual function of the damselfly penis, Smith [17] documents the role of sperm precedence in paternally caring waterbugs, Short [18] proposes that testis size may be indicative of sperm competition and Sivinski [19] argues that sperm competition may be responsible for rapid divergent evolution of sperm form and function. (4) Thornhill [20] formally defines the process of cryptic female choice, and with Alcock publishes The evolution of insect mating systems in which sperm competition and cryptic female choice in insects are reviewed [21]. (5) Smith's edited book provides a taxonomic overview of sperm competition in the evolution of animal mating systems more generally [22]. (6) Eberhard [23] publishes his volume on the evolution of animal genitalia and Pennington [24] documents sperm limitation in broadcast spawners. (7) DNA fingerprinting reveals sperm competition in birds [25,26]. (8) Sperm competition is identified as a postmating pre-zygotic barrier to hybridization [27]. (9) Parker [28,29] develops his sperm competition games. (10) Gage shows how males strategically adjust the number of sperm relative to sperm competition risk [30,31]. (11) Birkhead & Møller's [32] volume on sperm competition in birds is published; the biennial Biology of Sperm meetings begin. (12) Seminal fluid proteins are found to reduce female lifespan in Drosophila [33]. (13) Antagonistic coevolution of male harm and female resistance to harm are demonstrated in Drosophila [34]; Eberhard [35] compiles evidence for mechanisms of cryptic female choice. (14) Birkhead & Møller [36] publish their edited taxonomic overview of sperm competition and sexual selection. (15) First evidence emerges that the evolution of reproductive proteins was driven by sperm competition [37]. (16) Simmons's synthesis of insect sperm competition research is published [38]. (17) Birkhead, Hosken & Pitnick (eds) release Sperm biology: an evolutionary perspective, with collected contributions from the biennial Biology of Sperm meeting [39]; trade-offs between premating weapons of sexual selection and sperm production are demonstrated [40]. (18) First theoretical models of strategic allocation to seminal fluid are published [41]. (19) Parker [42] presents his unifying concept of the sexual cascade.

3. The consolidation of ideas

Three influential books appeared in the early 1980s, which both consolidated the conceptual basis for Parkerian sexual selection and extended its scope across the animal kingdom. Two attendees at the Florida Entomologist symposia were Randy Thornhill and John Alcock. In 1983, they published their volume on The evolution of insect mating systems [21] within which they offered a comprehensive review of both premating and postmating sexual selection. In addition to a chapter devoted to insect sperm competition that updated Parker's original work, in their chapter on female choice they laid out the foundations for the female perspective. Parker had recognized that ‘the female cannot be regarded as an inert environment in and around which this form of adaptation [sperm competition] evolves' [2, p. 559] and he had explicitly modelled the evolutionary consequences of conflicts of interest between males and females over mating and fertilization [45]. But it was Thornhill and Alcock who explicitly proposed a number of mechanisms by which females might manipulate sperm transfer and storage to their own gain. In the same year, Thornhill [20] first coined the term ‘cryptic female choice’ as the female postmating equivalent of Darwinian premating female choice; cryptic because it is manifest in terms of paternity outcome through mechanisms that often occur unseen within the female's reproductive tract.

The second important volume had its origins a few years earlier, in 1980 when Robert Smith organized a symposium on sperm competition and the evolution of animal mating systems in Tucson, inviting Geoff Parker as a guest speaker and with an agenda to review evidence of sperm competition across the animal kingdom. The proceedings of this meeting appeared as an edited volume in 1984 [22] and although more than 50% of the chapters focused on reviews of various insect taxa, there were chapters on poeciliid fish, amphibians, reptiles, birds and mammals, including humans. Many of these chapters were prospective of the potential for sperm competition to operate in diverse taxonomic groups, providing ideas for the future development of research into postmating sexual selection across the animal kingdom. While many of the taxon-focused reviews in Smith were short on empirical support, the contributions to this themed issue now provide overwhelming empirical evidence that postmating sexual selection is responsible for the widespread evolution of behaviour, morphology and physiology among spiders [46], fishes [47], reptiles [48], birds [49] and mammals [50].

The third influential volume of the 1980s appeared in 1985. In his volume, Sexual selection and animal genitalia [23], Eberhard, another attendee of the Florida Entomologist symposia, amassed evidence from the taxonomic literature to argue how female choice during and following copulation offered a general explanation for the observed patterns of rapid and divergent evolution in male genital morphology envisaged by Lloyd [43]. It is now widely accepted that sexual selection is responsible for the evolution of male and female genital morphology, as the contributions on reptiles [48] and mammals [51] in this themed issue discuss, but it still remains difficult to disentangle the precise mechanisms involved: sperm competition, cryptic female choice or sexual conflict [52,53].

In 1985, Pennington demonstrated that for broadcast spawning echinoderms, eggs may remain unfertilized because sperm released into aquatic environments can be lost rapidly from the immediate fertilization environment [24]. In their review of sperm limitation in the sea, Levitan and Peterson [54] presented the alternative view that it may be eggs that compete for a limited supply of sperm, and that sperm limitation rather than sperm competition may have been the fundamental driving force in the evolution of anisogamy. These two views have since been reconciled, with both processes thought to play a role [55]. Nevertheless, studies of broadcast spawning invertebrates offer considerable insight into sperm competition in ‘the ancestral condition’, as Evans and Lymbery discuss in their contribution to this themed issue [56]. Finally, in 1989 Hewitt et al. discovered how postmating sexual selection could function as a barrier to hybridization between races of Podisma grasshoppers [27]. When a female was mated to two males, one from their own population and one from a different population, the authors found a bias in fertilization success toward conpopulation males. Other examples followed, revealing the importance of postcopulatory sexual selection for speciation [57]. Garlovsky et al. [58] address this area of research in their contribution to this issue. Using the fruit fly Drosophila montana, they show that conpopulation sperm precedence is not affected by the strength of sperm competition operating within different populations, suggesting that postmating prezygotic isolation may be driven more by cryptic female choice than sperm competition [58].

4. Rapid growth

It was in the 1990s, two decades after Parker's [2] original insight, that research on sperm competition gathered momentum (figure 1). Observing the outcome of sperm competition in insects had been possible because of the sterile male technique used in insect pest control whereby mutations in the male germline are induced via exposure to radiation, rendering the eggs that irradiated males fertilize inviable [59]. In 1987, the ability to quantify multiple paternity became more widely available with the advent of DNA fingerprinting in birds [25,26]. Extra-pair copulations were well known in birds, but for the most part they had been dismissed as inconsequential. The advent of molecular paternity assignment revealed that for birds, monogamy was the exception rather than the norm. In their contribution to this issue, Birkhead and Montgomerie trace the history of sperm competition research in birds [49]. Molecular paternity testing coupled with the accumulated knowledge and conceptual foundations of postmating sexual selection fuelled a rapid increase in research on this taxon, spurred on by Birkhead and Møller's volume Sperm competition in birds [32] published in 1992. In their contribution to this issue, Carleial et al. [60] provide new data on sperm competition dynamics in red jungle fowl, finding that in free-ranging populations repeated mating and mating last with a given female are the principal factors determining a male's competitive fertilization success, while increased age, female resistance to copulation and her degree of polyandry all decrease a male's success in gaining fertilizations. The effects were found to be time-dependent as they became stronger over the course of experimental trials, indicating cumulative effects of different behavioural mechanisms. These data shed new light on the fine-scale dynamics of sperm competition in birds [60].

It was also in the early 1990s that Parker began his game theory approach for predicting evolutionary responses in ejaculate expenditure to selection from sperm competition [14], an approach foreshadowed 10 years earlier [61]. The first ‘sperm competition games' paper asked how a male should allocate sperm when sperm competition conformed to a ‘fair raffle’, in which each male gains paternity relative to the number of sperm present at the time of fertilization, or a ‘loaded raffle’, where one male's sperm are devalued in some way, for example by being the first or second male to mate. The models predict that in general males should increase their investment in sperm production when there is a risk of sperm competition, but decrease their investment with the degree of unfairness in the raffle [28]. Support for increased male expenditure on sperm in response to sperm competition is now well established from comparative analyses [44]. Subsequent models asked how males should invest when mating opportunistically, such as when in the role of an extra-pair mate or as a ‘sneaker’ in species with alternative mating tactics [29]. Many iterations would follow as the models were refined to specific biological scenarios [14,62], and in developing this theory base Parker provided the framework for future problem-solving that was necessary to firmly establish postmating sexual selection as a new paradigm. Empiricists were quick to begin testing these new models. In 1991, Gage showed how male insects exposed to rivals during copulation increased the number of sperm they transferred to females [30,31] and studies across a diverse array of taxa have since reported similar patterns, indicating that in general males will allocate their sperm strategically in response to the risk of sperm competition [63]. In their contribution to this themed issue, Kustra and Alonzo [64] review the literature on alternative reproductive tactics (ARTs), finding that a consistent pattern across those species that adopt ARTs is that males in the sneaking role have larger testes, and so produce more sperm, than those in the guarding role.

Three contributions in the mid-1990s fuelled the already accelerating research effort into Parkerian sexual selection. In 1995, Chapman et al. [33] discovered that the long-observed longevity cost of mating for Drosophila females was owing to the seminal fluid proteins that males transfer in the ejaculate to suppress future female receptivity. Male adaptations to sperm competition were found to be harmful to females [65]. Using a clever experimental design in which he arrested female evolution, Rice showed in 1996 that males evolved to be more harmful when females were prevented from coevolving to counter males' harm [34]. Parker had, of course, discussed explicitly how sexual conflict could generate arms races between males and females when their reproductive interests did not coincide, both in his studies of yellow dung flies [13] and in his theoretical treatment of sexual conflict published in 1979 [45]. But like sperm competition before it, it took over 10 years before the full significance of sexual conflict was recognized [66]. Likewise, through the 1980s researchers had been testing ideas around ejaculate manipulation as a mechanism of cryptic female choice [20,67–69] but it was not until 1996 when Eberhard published his review of the many ways in which females could potentially control paternity [35] that the significance of cryptic female choice became forefront in the postmating sexual selection paradigm. In consequence, the late 1990s saw a surge of interest in both sexual conflict and the female perspective in postmating sexual selection.

By the end of the decade, it was time to take stock once more with Simmons's Sperm competition and its evolutionary consequences in the insects [38] amassing 30 years of research on insect sperm competition, and Birkhead and Møller's edited volume Sperm competition and sexual selection [36] updating Smith's taxonomic overview of a decade earlier. In the latter volume, Anders Møller argued that for a complete understanding of the action of sexual selection we needed to consider premating and postmating episodes of sexual selection as a continuous process, because different episodes have the potential to act either synergistically to intensify net selection, or antagonistically so giving a false impression of the strength of premating sexual selection. Indeed, subsequent research has shown this to be the case [70,71]. While attractive males can have a greater competitive fertilization success [72], males with greater mating success can also have decreased fertilization success if they become sperm depleted [73,74]. Importantly, Parker has since shown theoretically how polyandry can reduce the strength of sexual selection acting on males because the reproductive return from each mating decreases when paternity is shared [75]. For example, in their contribution to this issue Glavaschi et al. [76] show how in guppies the imminent risk of predation increases the strength of sexual selection acting on males because females become less willing to mate polyandrously and the consequent reduction in postmating sexual selection increases the Bateman gradient.

5. The genomic era

A major advance in the 2000s came with the demonstration that postmating sexual selection was responsible for rapid evolutionary divergence at the molecular level. Evidence of positive selection acting on male reproductive proteins had been accumulating from studies of sperm and seminal fluid proteins in Drosophila, abalone and rodents [77]; reproductive proteins tend to have high rates of non-synonymous nucleotide substitutions. In 2000, Wyckoff et al. [37] showed that the sperm-associated protamines of great apes showed patterns of evolutionary divergence consistent with differences in the strength of selection arising from sperm competition. Subsequent work confirmed these patterns in Drosophila [78] and rodents [79]. Sperm competition research thereby entered the genomic era, asking questions focused on the functional significance of seminal fluid and sperm proteins, and their genomic evolution. The fruit fly Drosophila melanogaster has been a key model in the study of seminal fluid function. In their contribution to this issue, Wigby et al. [80] provide a catalogue of known seminal proteins from D. melanogaster and contrast its seminal proteome with that of humans. They find evidence of similarities in proteins between these flies and humans, suggesting that some seminal fluid proteins, including those involved in postcopulatory sexual selection, have highly conserved functions. They review what we know of the functions of D. melanogaster seminal fluid proteins, identifying considerable scope for future research.

The ejaculate contains both sperm and seminal fluid fractions. Given that sperm production increases with risk of sperm competition among species, and that males allocate their sperm strategically in response to sperm competition risk, it seemed logical to ask whether male expenditure on seminal fluids might show similar patterns of variation. The first models of seminal fluid allocation appeared in 2007. Cameron et al. [41] predicted distinct patterns of investment depending on the function of seminal fluid proteins, whether they served to increase the competitiveness of the sperm or whether they served to promote female fecundity. In his contribution to this themed issue, Ramm reviews theoretical models of the evolution of male investment in seminal fluids, and presents the accumulating empirical support for the idea that sperm competition favours increased male expenditure on seminal fluid production among species, and the strategic allocation of resources to seminal fluid components of the ejaculate in response to sperm competition within species [81]. Much of this work was pioneered using D. melanogaster, wherein males have been found to adjust their allocation to two different seminal proteins, sex peptide and ovulin, in different ways; males reduce their allocation of ovulin, but not sex peptide, to already mated females because mated females will have already been stimulated to produce and lay eggs by their previous mates [80]. This is an area of sperm competition research that is expanding rapidly. Indeed, in their contribution to this issue Bayram et al. now show how in house mice, males of low social status have a higher concentration of seminal fluid proteins in their ejaculates than do dominant males, and that a handful of seminal fluid proteins that influence competitive fertilization success seem to be in greater abundance in subordinate males, perhaps compensating for their disadvantaged role in sperm competition [82].

Parker's sperm competition games were firmly grounded in life-history theory, assuming that males face a trade-off between allocating resources to ejaculates rather than acquiring additional females. While researchers were quick to test the predictions arising from sperm competition games, tests of the underlying assumptions began much later in 2006 when Simmons and Emlen [40] reported a trade-off between male expenditure on testes growth and horn growth among onthophagine dung beetles. Comparative studies now offer broad support for the assumption that species that invest more in the weapons and ornaments of Darwinian sexual selection often invest less in ejaculate expenditure [83], a pattern that is consistent with Parker's unifying theory of the sexual cascade published in 2014 [42]. Parker envisaged that Darwinian sexual selection arose late in evolutionary history as the inevitable consequence of transitions from the ancestral state in which sperm competition, and so gamete production, among externally fertilizing species was at its maximum. The evolution of increased mobility led to the rise of premating sexual selection in which adaptations arose for the monopolization of mates, reducing selection on ejaculate expenditure in favour of increased male allocation to the weapons and ornaments of premating sexual selection.

6. Future prospects

The past decade has seen Parkerian sexual selection enter the phase of ‘normal science’, with the number of studies published each year stabilizing at approximately 250. Yet, there is still much we do not fully understand, and exciting avenues for future research remain. One of Parker's early sperm competition games modelled the effects of haploid versus diploid control over sperm phenotypes [84]. This remains a poorly studied area, in part owing to the difficulties in examining within-ejaculate competition, and in part owing to the dogma that DNA within sperm is supposed to be transcriptionally silent. But as Sutter & Immler [85] highlight in their contribution to this issue, there is increasing evidence of gene expression within sperm and new technologies for studying within-ejaculate sperm competition are opening up exciting new avenues for research.

Male fertility is profoundly important for both male and female reproductive success, but is constantly under threat; sperm performance and production are acutely sensitive to biotic and abiotic factors including diet, temperature and other forms of stress. In humans, oxidative damage to sperm and the integrity of their DNA cargo are strongly linked to infertility. While oxidative stress has been studied extensively in the context of assisted reproduction, more broadly oxidative stress can be an important player in life-history and senescence, and there is evidence to suggest that oxidative stress plays a role in premating sexual selection by maintaining honesty in the expression of sexual ornamentation. Yet, despite calls for research into the role of oxidative stress in postmating sexual selection [86], progress in this area has been slow. In their contribution to this issue, Friesen et al. [87] review evidence of the effects of oxidative stress on sperm, offering many exciting new avenues for future research. Selfish genetic elements (SGEs) can also profoundly affect male fertility, because they can promote their own transmission via negative impacts on sperm that do not carry them. In their contribution to this issue, Verspoor et al. [88] examine the history of research into the mechanisms by which SGEs affect sperm, and how selection can shape the evolution of male and female traits that mitigate against the costs imposed by these selfish passengers. Although predicted to be widespread, we know of only few examples of SGEs and Verspoor et al. [88] point to a number of exciting avenues for future research.

Finally, while male reproductive fluids have become the focus of considerable research attention, the role of female reproductive fluids in postmating sexual selection has not been fully explored. In their contribution to this issue, Gasparini et al. [89] review evidence of how female reproductive fluids can act as chemoattractants for sperm, affecting sperm swimming speeds, behaviour and longevity. In so doing, female reproductive fluids have the potential to exert cryptic female choice at the gametic level. While Parkerian sexual selection may have entered an era of ‘normal science’, the contributions to this themed issue show that there remains considerable potential for future problem solving, and we look forward to the exciting advances that we hope this themed issue will generate over the next 10 years.

Data accessibility

This article has no additional data.

Authors' contributions

L.W.S. drafted the paper, N.W. edited and approved the final version.

Competing interests

We declare we have no competing interests.

Funding

We received no funding for this study.