Sperm form and function in the absence of sperm competition


  • Gerhard van der Horst,

    Corresponding author
    1. Department of Medical Bioscience, University of the Western Cape, Bellville, South Africa
    • Corresponding author:

      Department of Medical Bioscience, University of the Western Cape, Private Bag X17, Bellville 7535, South Africa. E-mail: gvdhorst@uwc.ac.za

    Search for more papers by this author
  • Liana Maree

    1. Department of Medical Bioscience, University of the Western Cape, Bellville, South Africa
    Search for more papers by this author

  • “If sperm does not need a competitive advantage to fertilize the oocyte, will its quality be maintained or decline?”


Sperm competition is a post-copulatory, sexual selection force that, together with phylogeny and fertilization mode, has been regarded as one of the main factors explaining the diversity in sperm size across species. This universal sperm selection mechanism favors traits that enhance a male's fertilizing ability and paternity success. Surprisingly, however, sperm characteristics and semen quality in monogamous species, with low risk of sperm competition, have barely received any attention. In this review, we consider sperm competition and monogamy as two ends of the selective spectrum, and discuss its effect on sperm structure and function. We address the issue of a lack of sperm competition by comparing sperm traits of essentially monogamous species—their largely degenerative sperm features and high degree of polymorphisms could be norms for monogamous species. Further, the level of sperm competition in humans is discussed by comparing its mating strategy, relative testis size, and sperm traits to other primate species. In terms of sperm concentration, sperm swimming speed, and sperm morphology, humans seem to be closer aligned to the low-risk sperm competition situation in gorillas than to promiscuous chimpanzees. Mol. Reprod. Dev. 2013. © 2013 Wiley Periodicals, Inc.


Gustaf Retzius (1842–1919) must be considered the most important pioneer of sperm-structure analysis. He described basic sperm morphology of almost 400 species in a period of about 5 years (1904–1909), with an accuracy that has only been matched by electron microscopy (Afzelius, 1995). Most importantly, Retzius has shown that despite immense variation among sperm of different animals, related species have a similar sperm structure. This fundamental concept assisted greatly in finding phylogenetic relationships among species, and is still utilized more than a century later (Jamieson, 1991; Breed, 2005; Higginson et al., 2012). Retzius was also the first to note that there is a tendency for larger mammals, such as elephants and whales, to have smaller sperm and for marine mammals to have shorter sperm midpieces (Afzelius, 1995).

Although this early research inherently paved the way for future investigators to try and understand the variation in sperm structure and sperm biology in general, a number of basic questions still remain unanswered or speculated upon. Why is there a 2,000-fold variation in sperm size across species, when they all have the same fundamental task (Gage, 2012)? Does a particular fertilization environment influence the size and shape of sperm, or is it determined by a predominant mating strategy? To what extent do sperm characteristics such as “normal” sperm morphology assist to determine mating strategy?

Parker (1970) can be considered the modern father of sperm competition as he defined this method of sexual selection as the competition between sperm of two or more males to fertilize the oocytes of a single female. The concept of sperm competition has been expanded by a number of theoretical models in the past four decades. It has been described as a raffle principle, where the number of sperm a male produces is directly related to the probability of fertilization if there is a risk of sperm competition (Parker, 1990). Additionally, sperm competition is assumed to affect the evolution of sperm size and involves a direct or indirect trade-off between sperm size and sperm number (Parker, 1982, 1993). High levels of sperm competition have also been associated with increased testes size (for increased sperm production) and thus influencing the volume of the ejaculate (Parker, 1998). The detailed studies of Short (1979) on primates were one of the first to relate testis size to body size and testis-to-body-mass ratio to mating system. Subsequently, most comparative studies have used the gonadosomatic index, or relative testis mass, as an indicator of differences in sperm competition levels.

Much less attention has been given to sperm structure and semen quality in species where there is apparently a low risk of sperm competition. In dioecious species where there is single male to single female mating (monogamy), it can be expected that sperm competition is virtually absent. Although other reproductive strategies involving behavior or attractiveness to the opposite sex may be more important for mate selection in monogamous species (Tilbrook, 1987; Andersson, 1994), extra-pair matings occur more often than once thought (Birkhead and Møller, 1996; Kokko and Morrell, 2005). In polygyny, one male mates with more than one female. Accordingly, if considered from the perspective of each of these females, this would be equivalent to monogamy. If the risk of sperm competition is low, the only post-copulatory or post-spawning sperm selection in the monogamous mating strategy then most probably involves female cryptic choice (if extra-pair copulation is absent) (Reeder, 2003; Wagner et al., 2004). Here the female reproductive system or maybe even the oocyte “selects” the best quality sperm or sperm of a specific male, or alternatively discards the sperm (Eberhard, 1996). Polyandry is defined as a female mating with multiple males, and implies a high risk of sperm competition. There are, of course, many variations on the theme of multiple-male mating; we will pursue the more basic and fundamental aspects.

In this review, we consider sperm competition and monogamy (low risk of sperm competition) as two ends of the selection spectrum in terms of mating strategy. We pursue mating strategy as a key factor influencing sperm form and function by first discussing sperm traits related to high levels of sperm competition, then we focus on sperm traits in monogamous species, and finally turn our attention to primates, including humans. In the event of diminished sperm competition, would the reproductive investment in sperm production be less? If sperm does not need a competitive advantage to fertilize the oocyte, will its quality be maintained or decline?


Although sperm competition strictly refers to the arms race occurring between sperm from more than one male to fertilize a single female's ova (Parker, 1970), the selection of sperm to compete in such an arena already starts in the male during sperm development. The general trend reported across numerous species is that increased sperm competition levels results in increased sperm numbers (increased semen volume and sperm concentration), increased sperm length, and a higher percentage of sperm with normal morphology, progressive motility, and viability. Table 1 indicates examples of specific sperm traits studied in relation to sperm competition in various taxa and species, but are not discussed in detail as it is not the focus of this review. There are, however, a number of on-going debates about the relationship between sperm competition and several of these sperm traits, of which a few are considered below, with an emphasis on vertebrates (see review by Snook (2005) for non-mammalian species).

Table 1. Examples of Sperm Traits Influenced by Sperm Competition Across Taxa
Sperm traitSpecies/taxonEvidenceSource
Sperm numberSeveral taxaIncrease in sperm allocation with increased risk sperm competitiondelBarco-Trillo (2011)
Fish, guppyMales with more and faster sperm, greater share of paternityBoschetto et al. (2011)
Insect, birdsSperm number-sperm size tradeoffImmler et al. (2011)
Malurid passerinesSperm competition positively associated with sperm numberRowe and Pruett-Jones (2011)
Sperm morphology/morphometryDiving beetlesFemale reproductive tract determines complex sperm formHigginson et al. (2012)
Ant and beeLess variable sperm form with more competitionFitzpatrick and Baer (2011)
NematodesSperm volume increases when subject to greater competitionLaMunyon and Ward (2002)
Green sea urchinVariable sperm size between rather than within individualsManier and Palumbi (2008)
SquidNo evidence for large sperm favoredIwata et al. (2011)
Fish, in generalHigher risk sperm competition, shorter sperm lengthStockley et al. (1997)
Fish, African cichlidsHigher risk sperm competition, longer sperm lengthBalshine et al. (2001)
BirdsSperm length variation predicts extra-pair paternityLifjeld et al. (2010)
BullfinchesSimplified/abnormal sperm type—low-risk sperm competitionBirkhead et al. (2007)
Malurid passerinesPost-copulatory selection and sperm phenotypeRowe and Pruett-Jones (2011)
Mammals generalSperm length relates to metabolic rateGomendio et al. (2011)
RodentsNo selection for sperm hook morphologyFirman et al. (2011)
Naked mole-ratAbnormal sperm types increase—absence of sperm competitionvan der Horst et al. (2011)
Muroid rodentsVariable sperm length—lower risk of sperm competitionBreed et al. (2007)
Sperm motilityMuroid rodentsLarger sperm swim faster with higher risk sperm competitionMontoto et al. (2011a)
PrimatesPromiscuous species; swim faster than humanMaree (2011)
PrimatesIncreased midpiece size associated with higher motilityAnderson and Dixson (2002)
Post-copulatory sperm viabilityInsects, scorpionsHigher rates of mating, lower viability of spermEngqvist (2012)
Fish, swordtailSperm viability and velocity trade-offSmith (2012)
Muroid rodentsSperm viability unaffected by sperm competitionMontoto et al. (2011b)

Sperm Number, Sperm Size/Length, and Sperm Velocity

The theoretical model that increased levels of sperm competition will result in increased sperm numbers (Parker, 1990) has been generally accepted. This theory is supported by studies that reported different testis-related characteristics [e.g., increased rate of spermatogenesis (Fujii-Hanamoto et al., 2011; Lüpold et al., 2011; Montoto et al., 2012); higher relative number of seminiferous tubules (Rowe and Pruett-Jones, 2011); and increased testis size (Liao et al., 2011; Vahed and Parker, 2012)] and sperm numbers are positively correlated with higher risk or intensity of sperm competition.

The effect of sperm competition on sperm size is still not resolved (Snook, 2005). According to some theoretical models, vertebrate sperm size and number will be a trade-off due to fixed reproductive reserves (Parker, 1982, 1993). Thus, if sperm competition increases sperm numbers, this has to be compensated for by producing smaller gametes. A contradictory hypothesis was presented by Gomendio and Roldan (1991), stating that in primate and rodent species with promiscuous females, males produce longer sperm that also swim faster, which are probably adaptations to reach the ova first. This positive correlation between sperm competition and sperm length and/or swimming speed has been confirmed in numerous comparative and empirical studies on amphibians (Byrne et al., 2003), fish (Boschetto et al., 2011), birds (Lüpold et al., 2009), and mammals (Gomendio and Roldan, 2008). Nevertheless, Snook (2005) cautioned that different conclusions were reached in some taxonomic groups where this relationship was studied more than once.

Sperm Quality

Apart from sperm quantity, sperm quality also seems to be significantly influenced by sperm competition. When the risk of sperm competition is high, every sperm counts and selection will favor the production of high-quality sperm (Pitnick et al., 2009). In muroid rodents, differences in levels of sperm competition were associated with the proportion of sperm that undergoes capacitation and responds to progesterone (Gomendio et al., 2006). A more detailed study on 18 rodent species suggests that an “overall sperm quality” parameter, obtained by principle component analysis, had a stronger correlation with relative testis mass than any individual sperm quality parameter measured (normal sperm morphology, acrosome integrity, viability, motility) (Montoto et al., 2011b). Such an association between sperm motility, viability, and normal morphology versus sperm competition level was also reported for Australian maluridae (Rowe and Pruett-Jones, 2011), but not in two snake species where sperm quality was instead related to environmental temperature (Tourmente et al., 2011a).

In order to produce sperm with superior quality, various male characteristics have been under evolutionary pressure to assist with sperm selection. Strong evidence exists for sperm competition to be associated with the evolution of a number of reproductive, non-sperm factors that assist with the release and survival of sperm from the male reproductive tract. For instance, Drosophila males exposed to rivals mate for longer and transfer more seminal fluid proteins during this time (Wigby et al., 2009). These males also seem to adaptively tailor the composition of proteins in the ejaculate depending on whether the female has already mated with another male or not (Sirot et al., 2011). Additionally, bank vole (Myodes glareolus) males exposed to high levels of sperm competition developed larger accessory reproductive glands (Lemaître et al., 2011) and increased baculum width (Lemaître et al., 2012).


It is interesting to note that, while the field of sperm competition showed remarkable development during the last forty years, very few studies focused on the opposite side of the spectrum to determine if a low risk of sperm competition also have an influence, albeit in a negative sense, on sperm traits. Here we outline three special cases in vertebrates exhibiting exceptional or modified sperm morphology in predominantly monogamous species or species where the risk of sperm competition is extremely low.

First, Birkhead et al. (2006) found that sperm morphology of the Eurasian bullfinch (Pyrrhula pyrrhula), a passerine bird species, was distinctly different from all other passerine species studied, and this was possibly related to relaxed selection and lack of sperm competition. Additionally, the mean sperm velocity of two Eurasian bullfinch males (21.7 µm/sec) was the second slowest when compared to ten other passerine species (21–43 µm/sec; Birkhead et al., 2006). An ultrastructural study (Birkhead et al., 2007) confirmed that this species' sperm contains several atypical features for passerine birds, including an ellipsoid rather than cylindrical nucleus, dispersed or disorganized chromatin, and a small midpiece with several mitochondria clustered around the nuclear-axonemal junction rather than a single, long and helical mitochondrion. Birkhead et al. (2007) suggested that these “abnormalities” in sperm morphology are the result of suppression of the final stages of spermiogenesis and the retention of an immature morphology (almost a neotenous type of sperm). These authors furthermore speculated that the adaptive significance of such suppression may be an energetic saving response to reduced sperm competition. In a subsequent study on the same species, Durrant et al. (2010) used microsatellite data to rule out the possibility that this bird's unusual sperm morphology was due to a genetic bottleneck and a resulting reduction in its genetic variation, strengthening the hypothesis that relaxed sperm competition is the operational evolutionary force. In a recent study by Lifjeld et al. (2013) on the Azores bullfinch (Phyrrhula murina, sister species to Eurasian bullfinch), similar ultrastructural features and sperm morphology was reported as for the Eurasian bullfinch, as well as high coefficients of variation in total sperm length. The unusual sperm structure thus seems to be an ancestral trait that evolved before these two species diverged. Despite reporting a much higher swimming speed for Azores (157 µm/sec) compared to Eurasian bullfinch (Birkhead et al., 2006) and reduced genetic diversity at neutral loci for the Azores bullfinch due to a genetic bottleneck, Lifjeld et al. (2013) are still in agreement that a lack of sperm competition is operational in these two related bird species.

In a second example, sperm morphology of the greater bandicoot rat (Bandicota indica), described by Thitipramote et al. (2011), shows remarkable similarities to the bullfinch in terms of having predominantly abnormal sperm, a simplified acrosome, and chromatin dispersal reminiscent of fragmentation, but also a vast heterogeneity in sperm form. In previous studies on murine rodents (Breed and Taylor, 2000; Breed et al., 2007), it was found that, apart from the bandicoot rat, species from separate lineages, for example, the Tete veld rat (Aethomysi neptus) from southern Africa and the spinifex hopping mouse (Notomys alexis) from Australia, all had smaller testes mass and ejaculates containing sperm populations with highly divergent and variable head shape compared to closely related species with monomorphic sperm populations. Additionally, these species' sperm also had considerable intra-specific variability in midpiece length and total length of the sperm tail. Breed et al. (2007) stated that when selective pressure is relaxed due to depressed levels of inter-male sperm competition, as suggested by smaller testes mass, poor quality ejaculates appear to occur.

The third case involves the naked mole-rat (Heterocephalus glaber), in which sperm shares all the vast morphological abnormalities with the previous two examples. In the naked mole-rat, there is only about 7% normal sperm, less than 15% of sperm are motile, and those that are possibly have the lowest swimming velocity (curvilinear velocity ± 30 µm/sec) measured for any mammalian species (van der Horst et al., 2011). The sperm of this species show further abnormal features, especially in the sperm tail, and the authors believe that these can only be described as degenerative (Fig. 1). They revitalized the term degenerative orthogenesis, indicating that once a sperm has lost key sperm features, it is on a path of no return and this degeneration can be seen as a selective process in view of the absence of sperm competition. The naked mole rat is a eusocial species, and the “queen” controls the population of workers. Only one male is selected by the female for a lifetime as long as 28 years (essentially monogamous). Consequently, sperm competition would appear to be extremely unlikely in naked mole-rats, and this species exhibits features that could possibly be considered the most extreme case in the absence of sperm competition among mammals. It is important to note that these abnormal features of naked-mole-rat sperm are not considered to be a consequence of inbreeding (van der Horst et al., 2011).

Figure 1.

Bright-field and transmission electron micrographs showing examples of the high number of severely abnormal/degenerative sperm present in a monogamous species such as the naked mole-rat, which exhibit no risk of sperm competition (van der Horst et al., 2011). A: The range of abnormal sperm forms. B: Some common forms, such as a lobed nucleus (left) and a “dag-like” midpiece defect (right bottom). C: Low-magnification transmission electron micrographs showing that sperm in the cauda epididymis exhibit almost 100% fragmentation of nuclear material. D: Enlarged view of a single sperm exhibiting several head defects. Sperm of the monogamous bullfinch and bandicoot rat show remarkable similarities to sperm morphology of naked mole-rat. Scale bars, ∼5 µm.

It is quite remarkable that such major concurrent abnormal or divergent sperm features are observed in three diverse species. All three species have small testes in relation to body mass and low sperm numbers (Birkhead et al., 2006; Thitipramote et al., 2009; van der Horst et al., 2011), which is typically associated with low risk of sperm competition and monogamy. Surprisingly, no compromised male fertility was found in any of these species, and males produced viable offspring (Thitipramote et al., 2011; van der Horst et al., 2011). Since these species' males produce heterogeneous sperm populations and poor-quality ejaculates, it is possible that the female produces a large number of high-quality oocytes that may have specialized mechanisms to select for the best spermatozoa (female cryptic choice; van der Horst et al., 2011). In one of the most recent studies on the greater bandicoot rat, Dorman et al. (2013) reported a possible co-evolution of the gametes in this species. Apart from the unusual sperm features, the oocyte structure also differs from other murine rodents; these alterations probably relate to the acrosome reaction and sperm binding to the zona pellucida (Dorman et al., 2013).

Evidence from the sperm traits of these three species seems to add a new dimension to understanding the relationship between sperm structure and function and a low risk of sperm competition. It was suggested by Parker (1998) that energy investment in producing high-quality sperm is costly and that there will be selection against it if the costs are not equal to or outweighed by the benefits (fertilizing the oocyte). Thus, in the absence of sperm competition, there may be little benefit in investing energy on the quality of sperm produced (Bauer and Breed, 2006).

In summary, we propose that the unusual sperm morphology of the three species discussed be used as a benchmark for the low risk of sperm competition in future studies. It is furthermore suggested that these severe, abnormal sperm characteristics be considered degenerative, and that, from a selection perspective, they represent part of a process of degenerative orthogenesis in view of the virtual absence of sperm competition. It is also clear that sperm quality, in relation to the risk of sperm competition, cannot be assessed using a single, specific sperm trait; it rather is a summation of a range of particular morphometric parameters as well as other traits, such as sperm motility, ability to undergo the acrosome reaction, and sperm DNA fragmentation.


Sexual selection, and more specifically sperm competition, has been studied and debated in great detail in primates. The most important aspect of the debate is trying to relate mating strategy to facets such as sperm competition, testis size, ejaculate volume, sperm concentration, and sperm number in the ejaculate (Gould, 1990; Dixson and Anderson, 2001; Fujii-Hanamoto et al., 2011). An important question resulting from a comparative reproductive study on primates by Dixson and Anderson (2001) is: Do human sperm compete? In order to try and ascertain sperm competition risk, we discuss a number of reproductive facets, such as mating strategy, relative testis size, and sperm quantity and quality in humans.

In a review by Vasey and Forrester (2011) of a recent book by Alan Dixson on sexual selection in primates (2009), they show the elegance by which Dixson provides evidence for the lack of sperm competition in humans. Over a period of about three million years during the evolution of Homo spp., human ancestors apparently lived in small family groups and most probably formed pairs or were polygynous. It is worth noting that the anthropological record indicates that approximately 85% of human societies (not 85% of the world population; possibly a small overall percentage) have permitted men to have more than one wife (polygynous marriage) (White et al., 1988). The date when monogamy evolved in the human lineage is, however, intensely debated, and there are vastly different views in paleoanthropology and from genetic studies. Reno et al. (2003) indicates that there are two prevailing views on the evolutionary history of monogamy in humans: monogamy might have developed early in the human lineage (Reno et al., 2003) versus monogamy evolved very recently, less than 20,000 years ago (Dupanloup et al., 2003). It appears that the genetic evidence for the evolution of monogamy in humans is complex: “While female effective population size (the number of individuals successfully producing offspring thus contributing to the gene pool), as indicated by mitochondrial-DNA evidence, increased around the time of human (not hominid) expansion out of Africa about 80,000–100,000 years ago, male effective population size, as indicated by Y-chromosome evidence, did not increase until the advent of agriculture 18,000 years ago. This means that before 18,000 years ago, many females would be reproducing with the same few males” (Dupanloup et al., 2003).

In a monograph on human sperm competition, Smith (1984) states that, from an evolutionary point of view, there is a potential likelihood of sperm competition in humans on the basis that sperm of several males may be deposited in one female over a full menstrual cycle. Smith (1984) also argues that “males could not have been selected for promiscuity if females had historically always denied them opportunity for expression of the trait. If strict monogamy was the singular human female mating strategy, then only rape would place ejaculates in position to compete, and the potential role of sperm competition as a force in human evolution would thus be substantially diminished.” Although marriage systems are distinct from mating strategies, the role of cultural evolution favoring monogamous marriage should also be noted. Although monogamous marriage suppresses intra-sexual competition and reduces fertility, it also reduces crime rates and infant mortality rate while increasing paternal investment (Henrich et al., 2012).

In terms of relative testis size, Dixson (2009) points out that human testis sizes are consistent with an evolutionary history that involved pair formation as typical, and states that sperm competition pressure would have been low under these conditions. It is indeed true that humans occupy an intermediate position between typically polygamous species, such as monkeys on the one hand and the monandrous or harem-based mating systems of gorillas and orangutans on the other hand, in relation to testes mass and penis size (Dixson and Anderson, 2001). This has often been interpreted that there is a risk of sperm competition in humans, and that they may rather be more closely related to promiscuous chimpanzees in terms of sperm competition than to gorillas. Below we will provide evidence of a contrasting picture since comparative sperm traits in primates indicate that human and gorilla sperm share striking similarities, and that this probably indicates a low risk of sperm competition. Other examples from the literature where relative testis size do not always mirror levels of sperm competition in primates include two lemur species. In the fork-marked lemur, testes size was surprisingly small compared to 23 other lemuroid primates, a result that is in contrast to predictions made by the sperm competition theory (Schülke et al., 2004). In this instance, no factors such as behavioral, morphological, or physiological adaptations to mate guarding nor sampling biases, phylogenetic constraints, or population density effects could explain the absence of large testes in this species, which exhibits high extra-pair paternity in males and extra-pair copulations in females (Schülke et al., 2004). In a population of polygamous Milne-Edwards' sifaka, it was found that the positive relationship between body and testis size found in the non-breeding season dissipates in the breeding season, when the testis of lighter males grow more and match the testis size of heavier males, giving lighter males a higher relative testis mass than heavier males (Pochron and Wright, 2002).

Although the Dixson and Anderson (2001) do not deny the importance of relative testis size being larger in humans than in monogamous gorillas, with the consequent implications, we want to indicate that a more detailed analysis needs to be performed to establish a relationship between testes size and actual sperm traits/sperm quality and what it signifies. Recent literature has largely ignored the classical studies on primate sperm and comparisons of sperm of the big apes to human by Gould and coworkers (Gould et al., 1978, 1985; Gould and Kling, 1982; Gould, 1990). Comparisons of ejaculate volume, sperm concentration, percentage sperm motility, and sperm vitality show similarities and huge overlaps between gorillas and humans, but not in relation to promiscuous chimpanzees and other primates. Furthermore, sperm concentration in chimpanzees (ranging from 910 to 2,000 × 106/ml) appears to be 10- to 20-fold that of gorillas and humans, who have only a sperm concentration of 20–160 × 106/ml (Gould, 1990). Nascimento et al. (2008) reported significant differences between swimming speeds of human and gorilla sperm (lower than human). Yet, only four human sperm samples were compared to two gorilla samples in this study, and the investigators used different media for the two primate species. Despite the fact that the experiment was not controlled for medium and numbers to make any valid comparisons, there was a huge overlap in swimming speeds between humans and gorillas. Sperm kinematic parameters (swimming characteristics) may furthermore be related to mating pattern in general among primates, as sperm of the common marmoset (monogamous) (Hernández-López et al., 2005) show almost similar sperm kinematic characteristics to those of humans.

Apart from the sperm quantity and quality aspects discussed above, it appears that assessment of more detailed sperm parameters, for example, size of sperm midpiece and sperm morphology, could assist in finding relationships between mating strategy and sperm traits. The midpiece is of particular interest since it houses sperm mitochondria and is thus related to energy production for various cellular processes, including sperm motility. Although ATP produced in the midpiece is required throughout the flagellum for the contractile activity of the axoneme, enough ATP is produced by the mitochondria to be transported to the distal end of the flagellum through diffusion (Nevo and Rikmenspoel, 1970) or through ATP shuttles and kinase activity (Ford, 2006). Several studies have also indicated the existence of a direct relationship between midpiece length or volume and flagellar length (Cardullo and Baltz, 1991; Gage, 1998; Montoto et al., 2011a) as well as sperm swimming speed (Anderson and Dixson, 2002; Firman and Simmons, 2010; Montoto et al., 2011a). Mammalian sperm mitochondria can also make use of high levels of lactate (relative to glucose) and oxygen in the female reproductive tract via oxidative phosphorylation (as reviewed by Ruiz-Pesini et al., 2007; Storey, 2008).

The sperm midpiece is significantly larger in primate species whose females mate with multiple partners, and males have large testes in relation to body weight (Dixson and Anderson, 2004). Additionally, Dixson (2009) mentioned that the size of the midpiece of human sperm is smaller than that of the gorilla, and indicated that sperm midpiece volume is positively correlated with larger relative testis size and female promiscuity (also see Anderson and Dixson, 2002). In our own investigations (Maree, 2011), some 50 sperm parameters in three old world monkeys (vervet monkey, rhesus monkey, and chacma baboon—all multiple-male mating species) and humans were compared by studying the most motile sperm as selected by swim-up technique, using the same medium for each species. These parameters included detailed quantitative and objective sperm morphometry of all components, measurement of mitochondrial activity using flow cytometry, and sperm kinematic parameters (Fig. 2). In each instance, the promiscuous old world monkeys had significantly higher values for sperm velocity, mitochondrial activity, as well as much more extensive midpieces (length, volume, and number of gyres) than humans. This further supports the view of Dixson (2009) that humans basically exhibit a low risk of sperm competition when compared to other primates. Although Dixson makes the mistake to also relate midpiece size in gorillas to the ability of these sperm to hyperactivate (correct) and states that human sperm do not exhibit hyperactive motility (false). Hyperactivation does occur in human sperm, and has been extensively studied in this species over the last three decades. One of the best-known studies in this respect is by Burkman (1984), who showed a positive relationship between percentage hyperactivation and fertilization success in humans.

Figure 2.

Detailed sperm parameter measurements of four primate species. A: Human (Homo sapiens). B: Chacma baboon (Papio ursinus). C: Rhesus monkey (Macaca mulatta). D: Vervet monkey (Chlorocebus aethiops). 1: Bright-field sperm morphology showing lengths (µm) of the sperm head (HL), midpiece (MPL), and tail (TL). 2: Transmission electron micrographs of the midpiece showing the number of mitochondrial gyres per 2 µm of midpiece (in red), midpiece volume (MPV, µm2), and mitochondrial height (MH, µm). 3: Spermatozoa stained with MitoTracker Red CM-H2XRos (midpiece) and Hoechst (nucleus) showing staining intensity (INT). 4: Sperm motility tracks captured at 50 frames per second showing rapid progressive (red), rapid (green), medium progressive (blue), and static sperm (yellow crosses). The three kinematic parameters indicated are curvilinear velocity (VCL, µm/sec), linearity (LIN, %), and amplitude of lateral head displacement (ALH, µm). Values labeled with different superscript letters were significantly different (P < 0.05).

In terms of sperm morphology, what are its relationships to testis size in primates, particularly in relation to mating strategy? Dixson and Anderson (2001) commented on the so-called Baker pleomorphism model, which proposes that various morphological forms of human sperm have evolved to fulfill different roles in relation to sperm competition, such as kamikaze and egg-getter sperm (Baker and Bellis, 1995). We agree with Dixson that “although these startling claims achieved notoriety via the popular media, they have received little scientific support.” There is, however, almost no other commentary in sperm competition studies, in general and in primates, on one of the most important parameters when evaluating the quality of sperm, namely the percentage of normal sperm morphology. In human sperm, it has been convincingly shown that many of the pleomorphic or abnormal-looking sperm are indeed defective in some way or another (World Health Organization, 2010). For example, there are sperm that show nuclear fragmentation and have many morphological abnormalities, such as tapered heads, deformed acrosomes, and small or defective midpieces. In this regard, Gould (1990) has furthermore shown that gorillas and humans are the only two primates that share a high incidence of pleomorphisms, and even these pleomorphisms are very similar among the two species. In contrast, promiscuous chimpanzees and other primates have virtually no pleomorphisms and very high percentages of normal sperm (Gould, 1990). Accordingly, the low percentage normal sperm morphology in humans appears to be a consequence of long-term sperm selection under conditions of low levels of sperm competition. As mentioned before, Parker (1998) suggested that if sperm competition is almost absent, there will be less investment in producing “perfect” sperm as this is a costly process and not required under these conditions.

Furthermore, sperm morphology analysis might have been oversimplified in the past due to the absence of quantitative objective sperm morphometry analyses. Valle et al. (2012) assessed common marmoset sperm morphology and found that after cluster analysis, a variable number (from 3 to 7) of sperm morphometric subpopulations were identified with defined sperm dimensions and shapes. There were differences in the distribution of the sperm morphometric subpopulations in all ejaculates among the four donors analyzed. The results from this study by Valle et al. (2012) is not only another example of sperm pleomorphisms occurring in a monogamous primate species, but also indicates how computerized sperm analysis methods combined with principal component analysis clustering were useful to identify, classify, and characterize various morphometric sperm-head subpopulations in non-human primates. It also emphasizes the need for using quantitative measures to define mating strategy in primates.

Thus, if ranked on the basis of sperm traits such as sperm concentration, sperm swimming speed, and sperm morphology, humans will be closer aligned to a monogamous and lower-risk sperm competition situation than to the promiscuous chimpanzee. Relative testes size in monogamous marmosets is probably bigger than in humans, yet their sperm morphometry (heterogeneous) and normal sperm morphology (only 49%) (Cui et al., 1991; Valle et al., 2012) seem to be better indicators of a correlation with mating strategy. The authors also want to reiterate the seemingly many similarities among humans, gorillas, marmosets, bullfinches, bandicoot rats, and naked mole-rats, which all have a very high degree of polymorphism and abnormal sperm and a low risk of sperm competition. The high percentage of abnormality in all these examples is sufficient for producing offspring under conditions of reduced sperm competition, and seems to be a result of the lack or lower risk of sperm competition.

Consequently, when deciding upon mating strategy and its relationship with testes size and sperm traits in humans, it may be more important to look at characteristics such as total number of sperm in the ejaculate and ejaculatory frequency. The point is that if relative testes size in humans is almost as large as in chimpanzees, but sperm output in humans is as small as in gorillas, then relative testis size may not be such a good indicator of sperm competition in humans. Surely the idea of larger testes is primarily associated with having more Sertoli cells and accordingly a much greater capacity for sperm production (Parker, 1998; Fujii-Hanamoto et al., 2011; Lüpold et al., 2011; Montoto et al., 2012). In addition, even under conditions of greater sperm output per testis, the question of sperm quality also needs to be defined. In humans and gorillas, it should be considered that lower sperm quality is evident based on the percentage normal sperm morphology.

Then why do humans occupy an intermediate position between gorillas and chimpanzees in terms of testes size/weight if sperm traits and possibly sperm output aligns them better to gorillas? Two ideas are put forward in this context: First, humans might have experienced greater sperm competition in their distant history than during the last 100,000 years; and/or second, they used a large, almost hairless scrotum containing large testes for display to advertise or attract females or serve as a threat to discourage other males. The above ideas are further elaborated by Smith (1984), who indicates that early hominids such as Australopithecus spp. would have had a male dominance polygyny/female monogamy mating system, and thus little or no sperm competition. Smith (1984) then postulates that cooperative hunting destabilized “monogamy” and set conditions that led to a promiscuous mating system and high levels of sperm competition. It is therefore conceivable that during this phase, an increased testes size would have been favored. Smith (1984) furthermore suggests that “unequal distribution of the product of cooperative hunting probably caused some males to begin consortships that would have reduced sperm competition, improved individual paternity assurance, and begun to destabilize the promiscuous mating system. Improved paternity assurance permitted evolution of longer-term pair bonding and paternal investment” and potentially favor monogamous mating systems.


In this review we aimed to demonstrate how sperm structure and function are determined by mating strategy and different levels of sperm competition, focusing on the two ends of the evolutionary spectrum, namely high and low risk of sperm competition. Although molecular studies on sexual selection is numerous–and it was our intention to include examples of this for different levels of sperm competition–no such studies were found for single-mating systems or in the absence of sperm competition. A better understanding of the influence of sperm competition on sperm traits will probably be gained with advanced sperm-trait assessment and standardized analysis techniques.

Due to the comparative nature of many sperm competition studies, it necessitates the use of standardized experimental procedures and objective measurements. This will ensure that possible relationships reported among sperm parameters are not due to or influenced by biased, subjective, or inaccurate measurements and evaluations (Comhaire et al., 1992; Barratt, 1995). Although manual methods have been used in the evaluation of semen quality for many decades, quantitative and objective techniques–Computer Aided Sperm Analysis (CASA) for example–are available to routinely assess sperm concentration, motility, morphology, DNA fragmentation, and vitality (Agarwal et al., 2003; Mocé and Graham, 2008).

Despite the wealth of quantitative sperm analysis techniques available, sperm-trait analysis in sperm competition studies mainly measures basic parameters such as linear morphometric measurements (total sperm length, midpiece length) and sperm motility or velocity. The authors, for instance, found that much more variation is present in the mammalian sperm midpiece when assessing a larger number of parameters than compared to previous studies that only assessed midpiece length (Maree, 2011). Apart from midpiece volume, which was shown to be related to sperm competition in mammals (Anderson and Dixson, 2002; Anderson et al., 2007), thickness of the mitochondrial sheath, mitochondrial height, and total number of gyres in the midpiece all seem to be selected for in mammalian species (Fig. 2; Cardullo and Baltz, 1991; Maree, 2011).

The combined effect of different parameters (e.g., sperm quality), rather than individual sperm traits, and its correlation with fertilization rate also need to be evaluated (Simmons and García-González, 2008; Barbosa, 2011; Boschetto et al., 2011; Firman and Simmons, 2011; Gasparini and Pilastro, 2011; Rowe and Pruett-Jones, 2011; Montoto et al., 2011b; Higginson et al., 2012). Furthermore, measures of fitness or sperm function should be identified and assessed, such as the ability to initiate Ca2+ signaling pathways for changes in sperm flagellar output associated with chemotaxis (Darszon et al., 2006; Kaupp et al., 2008) or sperm hyperactivation (Suarez and Ho, 2003; Suarez, 2008). Additional factors influencing sperm trait selection (e.g., cryptic female choice, metabolic rate, and phylogeny) should also be considered in conjunction with sperm competition (Iossa et al., 2008; Holt and Fazeli, 2010; Gomendio et al., 2011; Tourmente et al., 2011b; van der Horst and Jarvis, 2012).

Finally, we propose that degenerative sperm traits and a high degree of variation in abnormal sperm forms reported for essentially monogamous species should be used as a reference for future studies to identify or consider the evolutionary consequence of a lack of sperm competition. The generally low percentage of normal sperm morphology and the relatively poor semen characteristic in humans and gorillas, compared to chimpanzees and other promiscuous primates, also seem to be an evolutionary consequence of a low risk of sperm competition.


This study was funded by National Research Foundation (NRF), Thuthuka Program, grant no. 76299.