Keywords

12.1 Introduction

Bottlenose dolphins (Tursiops spp.) live in complex societies with high fission-fusion dynamics including fluid changes in group size and composition (Connor et al. 2000b). Within this fluid structure, preferences for same-sex associates are common (Wells 2014; Ermak et al. 2017; Galezo et al. 2018; Ham et al. 2023, this book; but see Lusseau et al. 2003 and Wiszniewski et al. 2010) and likely reflect sex-specific reproductive strategies. Due to long gestation and lactation periods (Whitehead and Mann 2000), individual female bottlenose dolphins are unavailable to breed for several years at a time. The resulting male-biased operational sex ratio can lead to high variation in male reproductive success and intense male-male competition (Connor et al. 2000b; Karniski et al. 2018; Gerber et al. 2022; Würsig et al. 2023, this book). Male mating strategies are then constrained by the ability of males to monopolize either females or the resources that are valuable to them (Emlen and Oring 1977). Dolphins’ food resources are often patchily distributed and highly mobile, making territorial defense difficult (Connor et al. 2000b). Females, however, are defensible resources, and mate guarding can be effective at ensuring paternity and increasing reproductive success (Wells 1991; Connor et al. 1992). Although some male bottlenose dolphins appear to use a solitary approach to gain mating access, in several populations, males demonstrate a relatively rare mating tactic – cooperative mate guarding within long-term alliances (Table 12.1). Male alliances generally consist of a pair or trio of males that cooperate to sequester a fertile female (Connor et al. 1992; Owen et al. 2002). While mating can be shared, fertilization is an indivisible resource, making intrasexual reproductive cooperation paradoxical and recurring cooperation among the same individuals uncommon among animals (Díaz-Muñoz et al. 2014). Yet, cooperative mate guarding likely increases male reproductive success as it improves the odds of winning contests against other males and of successfully sequestering fertile females (Connor et al. 1992).

Table 12.1 Summary of social structure studies conducted on Tursiops spp. that specifically indicated male social complexity. Populations are listed in order of increasing population density; interbirth intervals (IBIs) are provided for additional context

Despite mate guarding attempts by males, bottlenose dolphins have a polygynandrous mating system; both sexes mate with multiple partners in a given breeding season (Connor et al. 1996; Boness et al. 2002). The benefits of polygynandry vary between the sexes; males likely increase their reproductive success by maximizing the number of mating partners (Bateman 1948), whereas females may reduce infanticide risk (Wolff and Macdonald 2004) and/or increase the genetic quality of offspring by mating with multiple males (Stockley 2003). Populations vary with respect to the seasonality of reproduction. Conceptions and births may occur year-round, but diffuse seasonal peaks corresponding with warm water temperatures are common (e.g., Urian et al. 1996; Mann et al. 2000). Mating can occur in a variety of positions, but males mounting along the side (lateral-ventral) or dorsum (dorsoventral) of the female are more commonly documented than ventrum-ventrum (Tavolga and Essapian 1957; Connor et al. 2000b). Mounting and goosing (rostrum to genital area contact) are the most conspicuous sociosexual behaviors, with intromission more difficult to observe and not necessarily indicative of reproduction (Connor et al. 2000b; Connor and Vollmer 2009; Furuichi et al. 2014). Nonreproductive sociosexual behaviors also occur throughout the year and may facilitate pleasure, learning, and establishing or mediating relationships (Connor et al. 2000b; Brennan et al. 2022; da Silva and Spinelli 2023, this book; Ham et al. 2023, this book).

Both sexes likely use conditional, rather than fixed, strategies with alternative mating tactics to optimize reproductive success. Conditional mating tactics are shaped by a combination of morphological, demographic, environmental, and social variables (Gross 1996), which vary greatly among populations of Tursiops spp. The following sections review population-specific examples of male bottlenose dolphin mating tactics and hypotheses that may explain variation in alliance complexity. The sociosexual behavior and potential countertactics used by females are also reviewed.

12.2 Male Mating Tactics

Male bottlenose dolphins engage in agonistic endurance competitions in which they compete for mating opportunities by roving among females; they either depart soon after mating or follow/herd the female to prevent other males from mating with her (i.e., mate guarding; Wells 1991; Connor et al. 1992). Copulation does not guarantee fertilization nor siring offspring if females have multiple mating partners. The degree of sperm competition in a species is typically correlated with testis size relative to body size and sperm count per ejaculate (Harvey and Harcourt 1984; Connor et al. 2000a). Bottlenose dolphins have relatively small testis mass and a moderate degree of sexual size dimorphism compared to other delphinids (Connor et al. 2000a), suggesting sperm competition may not be important, especially if males mate guard (Perrin and Reilly 1984). Mate guarding duration ranges from a few minutes to several months to competitively exclude rival males from copulation during the female’s estrus (Connor et al. 1992, 1996). The predicted number of receptive females and male competitors may influence the length of time males spend guarding individual females (Magnusson and Kasuya 1997). This mate guarding tactic can be temporally costly to males as ensuring paternity with one female may reduce the time available to mate guard others. However, if males do not guard a female for long enough, the likelihood of paternity may be greatly reduced. Connor et al. (1996) observed that females in Shark Bay, Australia, were guarded (and presumably mated) by up to 13 males in a single breeding season. Mate guarding and male-female associations may also be longer than the typical estrus period and/or begin prior to the breeding season as males may be preemptively mate guarding before a female reaches peak attractiveness (Connor et al. 1996; Owen et al. 2002; Robeck et al. 2005).

When cooperatively mate guarding, allied male dolphins frequently travel abreast behind the female or flank her on either side and slightly behind (i.e., a consortship; Fig. 12.1, Connor et al. 1992). Males pursue a female by angling out on either side of her, a feat more difficult to accomplish alone or in deep waters where a female has depth as an escape route (Connor et al. 2000b). The term herding describes coercively maintained consortships (Connor et al. 1996). Mate coercion is a common component of polygynandrous mating systems without strong or long-term intersexual bonds (Smuts and Smuts 1993). To constrain a female’s movements and prevent extra-pair copulations, males threaten females through posture, vocalizations, and charges or by aggressively biting or colliding into females (Connor et al. 1992, 2000b; Connor and Smolker 1996). Intersexual aggression has also been documented through analysis of conspecific tooth-rake marks. In Shark Bay, cycling females have more new rake marks than non-cycling females (Scott et al. 2005), and younger females may receive more aggression from males than do older females, suggesting male preferences for females with high calving success (Watson 2005; Karniski et al. 2018). There is currently no evidence of forced copulation, as females have been observed rolling away from mounting males; however, males may use intimidation tactics to coerce females into copulating (Connor and Vollmer 2009). In Sarasota, Florida, mate coercion occurs less frequently, and allied and non-allied males increase associations with females in the nonbreeding season compared to the breeding season, suggesting males may attempt to form affiliative relationships to influence future mating success through female choice (Owen et al. 2002).

Fig. 12.1
A photograph of 4 dolphins swimming synchronously in an ocean.

Two adult male first-order allies surfacing synchronously in herding formation behind an adult female and her dependent calf in the St. Johns River, Florida. Photo taken by Q. Gibson under authorization of NOAA Fisheries GA LOC 14157

Intrasexual (male-male) aggression is also evident from tooth-rake marks and opportunistic sightings of violent interactions (Connor et al. 1992, 2000b; Scott et al. 2005; Hamilton et al. 2019). However, the rates and severity of aggression may be underestimated as internal wounds from body slamming may not be externally visible and tooth-rake mark scars typically regain pigmentation within 20 months (Lockyer and Morris 1990; Ross and Wilson 1996). Several studies have found a significant sex difference in the prevalence of conspecific tooth-rake marks on bottlenose dolphins; more males have rake marks than females (Scott et al. 2005; Marley et al. 2013; Lee et al. 2019). This consistent sex-specific pattern suggests that aggression occurs in the context of male-male competition for access to mates. Patterns of rake mark coverage appear to vary among populations. In Sarasota, there was no observed sex difference in rake mark coverage (Tolley et al. 1995), whereas in Scotland, males had greater rake mark coverage than females (Marley et al. 2013). This sex difference may reflect the lack of male-male cooperation (i.e., alliances) in Scotland, resulting in increased competition and aggression (Marley et al. 2013).

12.2.1 Variation in Male Mating Tactics

Significant variation in male cooperation exists as not all populations of bottlenose dolphins exhibit reproductive cooperation (i.e., no alliances; Wilson 1995; Lusseau et al. 2003), males within a population may utilize different tactics (i.e., solitary vs. allied; Owen et al. 2002; Wiszniewski et al. 2012a), and alliances may be multilevel (i.e., first-order vs. second-order; Ermak et al. 2017; Connor et al. 2022; Table 12.1). Quantitative measures used to identify alliances differ among researchers (Table S1), which likely influences some of this variation. Qualitatively, first-order alliances are consistently defined as enduring relationships among males with repeated instances of cooperation within a reproductive context (i.e., jointly sequestering and coercing reproductive females; de Waal and Harcourt 1992). In contrast to more opportunistic coalitionary relationships, alliance associations occur year-round during all behavioral states, can last over seasons or years, and are more stable than other ephemeral relationships within dolphin societies (Wells et al. 1987; Connor et al. 1992, 1996; van Hooff and van Schaik 1994). This complex behavior is distinct in that individuals exhibit mutual tolerance, cooperation, and partner preferences to reduce intrasexual competition (Díaz-Muñoz et al. 2014). To mediate social bonds and potentially reduce tensions during consortships, allied males regularly engage in synchronous surfacing (Fig. 12.2; Connor et al. 2006), with the degree of synchrony increasing between partners with weaker bonds (McCue et al. 2020).

Fig. 12.2
A photograph of 2 dolphins swimming synchronously in an ocean.

Two adult male first-order allies surfacing synchronously in the St. Johns River, Florida. Photo taken by Q. Gibson under authorization of NOAA Fisheries GA LOC 14157

12.2.1.1 Populations Without Confirmed Male Alliances

To our knowledge, there is currently no published evidence of confirmed male alliances in populations at the northern and southern limits of bottlenose dolphins’ range (e.g., Scotland, Wilson 1995; New Zealand, Lusseau et al. 2003). Table 12.1 details populations where male alliances have been noted as absent. In Doubtful Sound, New Zealand, no direct mating competition or mate guarding has been observed; Lusseau (2007) hypothesized that mate guarding may be too costly due to both increased female maneuverability in the fjord’s depths and difficulties excluding rivals in the large group sizes (x̄ = 17.2). Male-male aggression, however, is regularly documented; males with higher intrasexual associations were less likely to suffer from aggression (i.e., headbutting) from other males, and they maintained bonds with potential coalition partners through affiliative behavior (i.e., mirroring; Lusseau 2007). While these coalitions function in a non-mate guarding context, coalitions had heterogenous association rates with receptive females and new mothers, suggesting that the maintenance of intrasexual relationships may still be important in this population (Lusseau 2007).

Solitary male mating tactics may not be as conspicuous as the cooperative mate guarding behavior of allied males, so less is known about the variation in solitary tactics across populations (Connor et al. 2000b). It is currently unknown whether individual males consort or attempt to mate guard females, but it is likely that solitary males employ similar tactics to allied males (e.g., roving, mate following/guarding, aggression, and/or displaying to influence female choice). In Sarasota, Florida, “roving” non-allied males have secured paternities, albeit fewer than allied males (Wells 2000; Duffield and Wells 2002; Owen et al. 2002). Stable associations with females may allow a male to be selected as a preferred mate during the breeding season (i.e., female choice; Owen et al. 2002). Although uncommon across populations, preferred male-female associations are a prominent feature of social structures in Ireland (Baker et al. 2020), the Gulf of Trieste, Slovenia (Genov et al. 2019), and Doubtful Sound (Lusseau et al. 2003), where alliance formation has not been documented. Intersex affiliation may play a strong role in determining reproductive success in small populations where alliances are absent and where strong male-female bonds occur (Lusseau et al. 2003; Augusto et al. 2012; Blasi and Boitani 2014; Louis et al. 2017; Baker et al. 2020).

12.2.1.2 Populations with Probable First-Order Male Alliances

Several study sites have indicated probable alliance occurrence based on strong male-male associations but are pending further behavioral analyses or longer study durations to determine the nature of these male bonds (Cedar Key, Florida: Quintana-Rizzo and Wells 2001; Moreton Bay, Australia: Chilvers and Corkeron 2001; San Luis Pass, Texas: Maze-Foley and Würsig 2002; Normano-Breton Gulf, France: Louis et al. 2015; Golfo Dulce, Costa Rica: Moreno and Acevedo-Gutiérrez 2016; Cres-Lošinj archipelago, Croatia: Rako-Gospić et al. 2017). Researchers in Coffin Bay, Australia, identified interconnected male social clusters ranging in size from two to five males resemblant of second-order alliances (Diaz-Aguirre et al. 2018). These preferred associates likely function as alliances, although neither mate guarding nor coercion was documented and male-male aggression appeared to be absent (Diaz-Aguirre et al. 2018). Similarly in Alvarado, Mexico, male dyads and trios had moderate bonds between them; however, researchers noted that detailed behavioral observations to determine the nature of these associations were limited (Morteo et al. 2014).

12.2.1.3 Populations with Confirmed First-Order Male Alliances

The presence and complexity of male alliances vary considerably within and among populations depending on their socio-ecological environments. Males in several nearshore populations cooperatively mate guard through an alliance to decrease intrasexual competition and increase reproductive success (Wells et al. 1987; Connor et al. 1992; Wiszniewski et al. 2012b). To our knowledge, first-order alliances have been reported in Florida (Owen et al. 2002; Bouveroux and Mallet 2010; Ermak et al. 2017; Brightwell et al. 2020), the Bahamas (Parsons et al. 2003; Elliser and Herzing 2011), Ecuador (Félix et al. 2019), Japan (Nishita et al. 2017), and Australia (Smolker et al. 1992; Möller et al. 2001; Chabanne et al. 2022). Table 12.1 provides a list of bottlenose dolphin populations with confirmed alliances.

The size and stability of first-order alliances vary. Across populations, pairs are the most commonly documented alliance size (Owen et al. 2002; Parsons et al. 2003; Elliser and Herzing 2011; Nishita et al. 2017; Félix et al. 2019; Brightwell et al. 2020). In Shark Bay, Australia, trio formation is the preferred alliance size (Connor et al. 1999), but the number of partners participating in a consortship is influenced by the habitat’s ecological variation (Connor et al. 2017). Greater intrapopulation variation has been observed in Port Stephens, Australia, and the St. Johns River, Florida, where alliances ranged from pairs to quads, with pairs most common (Wiszniewski et al. 2012a; Ermak et al. 2017; Fig. 12.3). Wiszniewski et al. (2012a) documented considerable variation among Port Stephens alliances that encapsulates the continuum of alliance tactics across populations: males in strong highly stable alliances, males in weaker and more labile alliances, and males that were allied for a short duration. At the longest running behavioral study sites, Sarasota, Florida, and Shark Bay, Australia, researchers have documented alliances ranging in duration from labile (e.g., changing each season or consortship) to stable partnerships lasting decades (Wells 1991; Connor et al. 1999, 2001; Connor and Krützen 2015). Disappearances can cause partner changes on shortened timescales, and males may form new alliances with unallied males whose partners may have also disappeared (Connor et al. 2000b). However, partner switches also occur when a previous alliance partner remains present in the same geographic area, indicating changes in association preferences (Wiszniewski et al. 2012a, Karle 2016; Brightwell et al. 2020).

Fig. 12.3
A social network depicts the pacts that range from dyadic and triadic alliances. The different colors indicate the variations.

Social network of 23 dyadic and 2 triadic alliances in bottlenose dolphins (Tursiops truncatus) in the St. Johns River, Jacksonville, Florida, from April 2011 to March 2018. Edge weights correspond to association strength calculated using the simple ratio index (SRI). Associations less than twice the nonzero male mean (SRI = 0.114) were removed. Node colors denote first-order alliance membership with second-order alliances sharing similar colors: yellows are the 6 dyads and 1 trio that form only first-order alliances; pinks, oranges, and reds are the 6 dyads that only form 1 second-order alliance each; and purples, blues, and greens are the 11 dyads and 1 trio that are part of larger second-order complexes wherein some (but not all) of the first-order alliances form multiple second-order alliances. SRIs were calculated in SOCPROG 2.9 (Whitehead 2009) and nodes arranged using the Force Atlas 2 algorithm in GEPHI (Bastian et al. 2009)

In populations where alliances have been documented, solitary (unallied) males are also present. The relative percentage of allied vs. unallied males varies; in some populations, solitary males are as prevalent, or more so, than allied males (≥50% unallied males in Little Bahama Bank, Bahamas, Elliser and Herzing 2011; St. Johns River, Ermak et al. 2017; Indian River Lagoon, Florida, Brightwell et al. 2020). In other populations, most males form alliances (<30% unallied males in Shark Bay, Smolker et al. 1992; Port Stephens, Möller et al. 2001; Sarasota, Owen et al. 2002). It is possible that solitary males are successfully using an alternative mating tactic. For example, Krützen et al. (2004) found that unallied juvenile males sired offspring.

12.2.1.4 Populations with Documented Multilevel Male Alliances

Second-order alliances consist of multiple first-order alliances that cooperate in contests over females (e.g., attempted thefts or defense of females from rival males; Connor et al. 1992). Quantitatively, the social bonds among members of second-order alliances are more moderate in strength than those among first-order alliance partners (Connor et al. 1992, 1999; Ermak et al. 2017; Table S1). This level of male social complexity is extremely rare; multilevel bottlenose dolphin alliances have been documented in only two populations to date: Shark Bay, Australia, and the St. Johns River, Florida. The majority of Shark Bay males are members of second-order alliances ranging in size from 4 to 14 members, with alliance size potentially related to the members’ foraging tactics (Connor and Krützen 2015; Bizzozzero et al. 2019; O’Brien et al. 2020). Second-order alliances are believed to be the core male social unit in the Shark Bay population (Connor and Krützen 2015), as males choose their first-order (herding) partner(s) from within their second-order alliances (Connor et al. 2011; King et al. 2021). While the identity of some first-order pairs and trios is stable (i.e., high partner fidelity), many second-order alliances demonstrate much greater flexibility in the formation of pairs and trios (Connor and Krützen 2015). This frequent partner switching is believed to maintain cooperative relationships within a larger group (Connor et al. 1999). Second-order alliances can endure for 20 years, ending due to gradual attrition more often than relationship changes (Connor and Krützen 2015). Surviving members of second-order alliances that have dissolved to the size of a first-order alliance (“lone trios”) still form relationships with other alliances, but at the association level of third-order alliances (Connor et al. 2011).

In contrast, second-order alliances do not appear to be the core male social unit in the St. Johns River, as males in this community exhibit a variety of mating tactics. As shown in Fig. 12.3, males may be unallied, form only a first-order alliance, form only one second-order alliance, or form multiple second-order alliances (Ermak et al. 2017). Among allied males, partner fidelity is high with most alliances dissolving due to a partner’s death or disappearance (Brightwell and Gibson, unpublished data). However, some alliances have reduced associations despite partners remaining in the area (Karle 2016). Switching herding partners between consortships, as observed in Shark Bay, has not been documented in the St. Johns River. Second-order alliance duration also appears to be more variable within the St. Johns River than in Shark Bay.

A third level of alliance formation, cooperation among multiple second-order alliances, has also been reported (Connor and Krützen 2015). Although the functions of first- and second-order alliances differ (i.e., consortships vs. female theft/defense), Connor and Krützen (2015) proposed that second- and third-order alliances are functionally similar. Third-order alliances in Shark Bay increased consortship duration by increasing the likelihood that allies were nearby (Connor et al. 2022). While there have been observations of groups containing multiple second-order alliances in the St. Johns River (Fig. 12.3), it is not yet clear if third-order alliances are present.

12.2.2 Cooperation Benefits

Populations with bisexual philopatry (i.e., both sexes remain in the same area postweaning) allow for association and affiliative bonding with kin postweaning (van Hooff and van Schaik 1994; Tsai and Mann 2013; Wells 2014; Wallen et al. 2017). As fertilizations cannot be shared, cooperation among individuals can provide direct (i.e., increased reproductive success) and/or indirect (e.g., kin selection) fitness benefits (Hamilton 1964; Würsig et al. 2023, this book). Although relatedness is not yet documented for many populations, where it has been studied, there does not appear to be a clear pattern across populations. Mean genetic relatedness was higher within than between alliances off Abaco, Bahamas (Parsons et al. 2003). Similarly, in Coffin Bay, Australia, preferential associates were more likely to be related than by chance (i.e., probable alliances; Diaz-Aguirre et al. 2018). In contrast, alliance members in Port Stephens, Australia (Möller et al. 2001), and Sarasota (Duffield and Wells 2002), were primarily unrelated, despite the presence of male relatives in the population. Findings from Shark Bay are mixed; early reports indicated that males in small, stable first- and second-order alliances were more related than those in a large second-order alliance with more labile herding partners (i.e., first-orders within a “super-alliance,” Krützen et al. 2003). Recent Shark Bay analyses evaluated individual male relatedness, as opposed to average group relatedness, and found that while kinship explained adolescent associations, similar ages between males were a better predictor of adult associations (Gerber et al. 2021), similar to the patterns observed among allied pairs in Sarasota (Wells 2014).

Alliance partner preferences for close relatives may not be a successful tactic given differences in sexual and social maturity among siblings due to demographic constraints (i.e., single births, extended interbirth intervals), and joint skill may be a more important driver than relatedness in partner selection (Möller 2012; Diaz-Aguirre et al. 2018; Gerber et al. 2021). However, depending on the population, partner choice is likely influenced to varying degrees by a mixture of kin selection and a form of reciprocity or by-product mutualism based on the availability of similar sexually and socially mature individuals (Trivers 1971; West-Eberhard 1975; Díaz-Muñoz et al. 2014). The adaptive benefits of reproductive cooperation, in the shape of increased reproductive success, likely offsets any incurred costs due to sharing copulations with unrelated allies.

Alliance membership is believed to be advantageous to reproductive success in populations that exhibit this male mating tactic. In Sarasota, both solitary and allied males sired offspring, with allied males siring disproportionately more calves despite appearing to associate equally with females (Wells 2000; Duffield and Wells 2002; Owen et al. 2002). In Shark Bay, non-allied males sired few, if any, offspring (Krützen et al. 2004), as males with more homogenous social bonds with second-order partners obtained the most paternities (Gerber et al. 2022). In Port Stephens, paternities were positively correlated with the number of males in an alliance and evenly distributed among members (Wiszniewski et al. 2012b), yet alliance social bond strengths were not predictors of success. Wiszniewski et al. (2012b) hypothesized that the variance in male reproductive success was attributed to mate guarding within a diffuse breeding season.

While alliances function in a reproductive context, they may also provide additional advantages through protection (e.g., reduced predation risk; Wells 1991; Hill and Lee 1998). Allied males in Sarasota, Florida, had larger home ranges, and although they acquired more shark bite scars, they lived longer than solitary males (Wells 1991). This pattern suggests that alliance partners may provide increased predator detection or enable cooperative defense (Wells 1991). Predation risk can be approximated through documentation of shark bite scars in field observations or through postmortem reports, as relatively few predation attempts have been directly observed by researchers (e.g., Gibson 2006). However, predation risk is likely underestimated in all areas; typically only survivors of predation attempts are observed by researchers in the field, and carcass recovery may not be feasible. Males may also guard a partner during recovery from an injury (Wells 1991). This hypothesis was supported by observations that alliances remained stable after anthropogenic injuries were incurred, with the exception of a male that died post-injury (Greenfield et al. 2021). In contrast, two Gulf of Guayaquil, Ecuador, alliances dissolved during a partner’s entanglement in fishing gear, and did not resume alliance status with their original partner after disentanglement (Félix 2021).

12.3 Hypotheses on Differences

We examine several socio-ecological factors that may help explain the variation in male bottlenose dolphin alliance complexity among populations. Encounter rates, in concert with the operational sex ratio and sexual size dimorphism, likely affect a male’s choice of mating tactic. In populations with a high rival encounter rate, limited availability of breeding partners, and minimal intersexual size differences, alliance formation should be favored if it leads to increased mating opportunities for allied males that can outcompete lone males or smaller alliances (Whitehead and Connor 2005). In contrast, in populations with low encounter rates, with stable availability of receptive females, and where males are large enough to effectively monopolize a female, alliance formation may not confer any significant reproductive benefits (Connor et al. 2000b; Möller 2012).

12.3.1 Sexual Size Dimorphism

Without a large degree of sexual size dimorphism (SSD), it may be difficult for a lone male to sequester and monopolize a female in a three-dimensional environment. Alliances are likely beneficial in that males can coordinate their spatial positions to effectively restrict female movements, while more robust males may be able to intimidate females on their own and not need assistance in mate guarding (Connor et al. 2000a). Bottlenose dolphin SSD is constrained, particularly with respect to body shape and size variations, possibly due to the energetic costs associated with increasing drag (Connor et al. 2000a). If SSD is present, the most pronounced differences are in robustness and modes of propulsion (males 11–47% heavier than females; Tolley et al. 1995; McFee et al. 2010); however, differences in mass (kg) are less often reported. Population differences in the degree of SSD and alliance formation tend to follow this predicted pattern (Connor et al. 2000a). There is minimal SSD in Shark Bay, Australia (van Aswegen et al. 2019), where multilevel alliances are present; slight-to-moderate SSD in Florida (Tolley et al. 1995; McFee et al. 2010), which has first-order alliances; and more moderate SSD in Scotland (Cheney et al. 2018a) and Brazil (Fruet et al. 2012) where alliances are absent. Although this comparison may be confounded by species-specific (T. truncatus vs. T. aduncus) differences in morphology, second-order alliances have been documented in both T. aduncus (Shark Bay; Connor et al. 1992) and T. truncatus (St. Johns River, Florida; Ermak et al. 2017). In the Bahamas, bottlenose dolphins are much larger than the sympatric spotted dolphins (Stenella frontalis), and bottlenose dolphin males attempt interspecific matings without the assistance of alliance partners (Elliser and Herzing 2016).

12.3.2 Operational Sex Ratio

Alliance formation would be expected in populations with a strongly male-biased operational sex ratio (OSR) as a tactic to reduce male-male competition (Daly and Wilson 1983; Whitehead and Connor 2005). Although the ratio of reproductively available males to females can be difficult to assess directly, the average interbirth interval (IBI) of females in a population can serve as a proxy for the OSR. Due to long gestation and lactation periods (Mann et al. 2000; Henderson et al. 2014), individual female bottlenose dolphins are unavailable to breed for several years at a time which can influence the degree of male-male competition. Few studies have reported a mean IBl of <3 yr for surviving calves, with most documented IBIs ranging between 3 and 4 years (Table 12.1). Among the populations with mean IBIs >4 yr, which suggests high levels of male-male competition, the full continuum of male social complexity (from no alliances to multilevel alliances) is observed. Thus, a male-biased OSR (and longer IBIs) may be a contributing factor for alliance formation, but it is unlikely to be the primary driver. However, calf mortality rates should also be considered due to their impacts on IBIs and the OSR (Mann et al. 2000; Karniski et al. 2018).

12.3.3 Encounter Rates

The encounter rate with rival males, which is often estimated using population density (dolphins/km2; Connor et al. 2000b), likely impacts alliance formation; however, density can vary among and within study sites as it may be influenced by demographics, predation pressure, resource availability, and habitat (Heithaus and Dill 2002; Wiszniewski et al. 2012a; Connor et al. 2017). Theoretically, given a set population density within a community, an increase in daily travel distance could increase the male-male encounter rate with adjacent communities, and a more open habitat would increase the detectability of rivals through better sound propagation (reviewed in Connor et al. 2000b). When the likelihood of encountering potential rivals is high, males may reduce competition and increase reproductive success via cooperative mating tactics (i.e., alliance formation; Connor and Whitehead 2005). The costs of sharing mating opportunities would be lower than the accrued benefits of gaining and maintaining access to fertile females. As population density and thus competition increase, cooperation benefits and alliance sizes should increase as well. However, the spatiotemporal distribution of male-male competition varies; clusters of increased competition may lead to the formation of clusters in the distribution of alliance sizes (e.g., pairs and trios; Whitehead and Connor 2005). An alternative explanation for this potential correlation between population density and alliance formation is that social complexity is easier for researchers to document in populations with high density. Table 12.1 summarizes male alliance complexity with respect to population density across populations.

The two locations with the greatest alliance complexity (i.e., multilevel alliances and multiple/shifting tactics [intrapopulation variation]) also have some of the highest reported population densities (Shark Bay, Australia, Bejder et al. 2006; St. Johns River, Florida, Ermak et al. 2017 and Mazzoil et al. 2020). In Shark Bay, alliance range overlap increases during the breeding season, and consortship size (male pairs vs. trios) and aggression (new tooth-rake marks) increase at the study site’s transition from shallow banks to open habitat, suggesting alliance size is being driven by both encounter rate and rival detection (Whitehead and Connor 2005; Connor et al. 2017; Hamilton et al. 2019). St. Johns River dolphins also demonstrate seasonal shifts in habitat use during the breeding season, which coincides with a large influx of transients and seasonal residents whose core areas are concentrated near the mouth of the river (Mazzoil et al. 2020; Szott et al. 2022). Upriver range expansion is limited due to low salinity levels which compacts dolphin density within the river despite seasonal habitat shifts (Ermak et al. 2017; Mazzoil et al. 2020).

Encounter rates in Sarasota, Florida, may be affected by home ranges in a shallow and fragmented habitat, where rival detectability can be restricted. Allied males have larger ranges compared to unallied males (Wells et al. 1987; Owen et al. 2002), and males occasionally leave the study area for months at a time, thereby increasing their encounters with males in adjacent communities (Wells 1991; Urian et al. 2009). Further, range overlap among communities can increase competition in those areas (Wells et al. 1987). In Port Stephens, Australia, males with more labile alliance partners, larger group sizes, and larger social networks relative to the population’s averages concentrated their spatial use close to the entrance of the embayment where they would encounter males from the coastal population (Wiszniewski et al. 2010, 2012a). As with the fluid first-order alliances within Shark Bay’s larger second-order alliances, Port Stephens males with a large social network likely have reduced costs of partner switching due to maintenance of social bonds among potential partners (Connor et al. 1999; Whitehead and Connor 2005; Wiszniewski et al. 2012a). In the Gulf of Guayaquil, Ecuador, the two communities with the most survey effort demonstrate a male-biased OSR (3:1 and 2:1; Félix and Burneo 2020) with sightings concentrated at channel mouths (Félix et al. 2017). Alliances with low male-female associations had wider home ranges than the alliance with stronger male-female bonds (Félix et al. 2019), suggesting these males may be forced to rove between communities for mating opportunities.

Bottlenose dolphin alliance formation on the Little Bahama Bank, Bahamas, is likely influenced by both intra- and interspecific encounter rates with the sympatric spotted dolphin (Stenella frontalis) population. Cross-species mating and suspected hybridization have been reported in the Bahamas (Herzing et al. 2003; Herzing and Elliser 2013), effectively doubling the population density and increasing male-male competition in this area (average 100 individuals of each species per season; Volker and Herzing 2021). Bottlenose dolphin alliance members are often observed alone during mixed-species encounters (Elliser and Herzing 2016), as their larger size allows them to outcompete the small spotted dolphin males for mating opportunities. Herzing and Johnson (1997) found that it takes six spotted males to chase away one bottlenose dolphin.

In populations with relatively stable population density and low encounter rates, alliance formation is less commonly reported (Table 12.1; Connor and Whitehead 2005). Doubtful Sound, New Zealand, is a small, closed population wherein density remains relatively stable and there is no need to restrict resident females from accessing males from other communities (Lusseau et al. 2003). Moray Firth, Scotland, is also composed of a small population that has increased in abundance from approximately 100 to 200 individuals since the 1980s. Yet dolphins have also expanded their range along the east coast of Scotland (Wilson et al. 2004), keeping encounter rates low.

12.4 Female Mating Tactics

The mating tactics of female bottlenose dolphins have received relatively little research attention compared to those of males; in many cases, female mating tactics can be masked by male-male competition and sexual coercion (Clutton-Brock and McAuliffe 2009), and there are likely more female tactics than are currently reported. The cost of poor mate choice is higher for females than males given the discrepancy in parental investment; female bottlenose dolphins have a yearlong gestation period, produce a single offspring per reproductive event, and exhibit extended interbirth intervals due to long lactation periods (Table 12.1; Whitehead and Mann 2000). Firstborn offspring of young adult females tend to have low survival rates, potentially due to inexperience in parenting, mate choice, or toxic offload (Wells 2000; Schwacke et al. 2002). Calf survival also decreases with maternal age due to reproductive senescence (Karniski et al. 2018). Although physiological factors play a strong role in female reproductive success, social factors such as associations with kin and other females in the same reproductive state can influence fitness as well (Mann et al. 2000; Möller and Harcourt 2008). Mate guarding by males can be costly to females by altering their foraging patterns and energetic budgets due to range and habitat shifts during consortships (Wallen et al. 2016), and it likely also limits their ability to select a preferred mate, at least outside those consorting her. Non-mutually exclusive female countertactics to mate guarding involve polygynandrous mating, preferential association with potential mates, and male avoidance.

Paired with polygynandry, repeated estrus cycles can counteract conception monopolization and reduce harassment, obscure paternity, and improve the genetic quality of offspring (Robeck et al. 2005; Watson 2005; Furuichi et al. 2014). Mate fidelity is uncommon (Duffield and Wells 2002; Wiszniewski et al. 2012b), and the risk of rejecting males can increase harassment, aggression, and injury during herding (Scott et al. 2005; Watson 2005). Mothers with calves may also attempt to avoid adult and juvenile males to reduce the threat of infanticide or aggression; in Shark Bay, sexual segregation is driven by female avoidance of aggressive males (Gibson and Mann 2008; Galezo et al. 2018). Calf-directed aggression and infanticide are favored in species with seasonal breeding, where lactation duration exceeds gestation duration, and year-round intersexual association occurs (Connor et al. 2000a). Males may be less likely to commit infanticide when there is a possibility that calves may be their offspring; thus, it is in a female’s best interest to mate with multiple males and not exhibit mate or alliance fidelity (van Schaik and Kappeler 1997; Wiszniewski et al. 2012b; Chap. 11) and in a male’s best interest to exert paternity control through mate guarding.

Multiple estrus cycles may enable females to mate with non-preferred males during one cycle and a preferred male during the next (Connor et al. 1996; Robeck et al. 2005). Males may have imperfect fertility detection, as suggested by the finding that male habitat use and ranging patterns shift during consortships, regardless of the female’s cycling status (Wallen et al. 2016). Consortships occur year-round, even though there can be seasonal peaks in reproduction (Connor et al. 1996; Mann et al. 2000; Karle 2016). However, both sexes may use these opportunities to strengthen bonds and utilize countertactics. Males may consort non-cycling females to strengthen male-male bonds and provide consortship practice prior to the mating season; females may be attempting to confuse paternity and/or evaluate males’ fitness (Connor et al. 1996; Furuichi et al. 2014). Connor et al. (1996) proposed that females may attempt escapes during consortships to test a male’s physical fitness.

Preferentially seeking out or associating with preferred males can facilitate female mate choice (Watson 2005). Females’ associations with males were high during breeding seasons in which the female was cycling (Wells et al. 1987; Smolker et al. 1992), and both non-agonistic and preferred female-male associations have been observed (Connor et al. 1996; Owen et al. 2002; Lusseau 2007; Wiszniewski et al. 2010). Sarasota alliances begin associating with females in the middle of the nonbreeding season, potentially to create affiliative relationships to influence female choice (Owen et al. 2002); this is further supported by the relatively low observations of mate coercion in this population (Tolley et al. 1995). Synchronous surfacing and displays by males facilitate social bonding among males (Fig. 12.4) but may also indicate mate quality to females (Connor et al. 2006; Sakai et al. 2010). Australian alliances (i.e., Shark Bay and Swan Canning Riverpark) have been observed conducting displays near female consorts, suggesting females may utilize these displays as a choice criterion among alliance members (Connor et al. 1992, 2000b; Chabanne et al. 2022).

Fig. 12.4
A photograph of three adult dolphins swimming synchronously in an ocean.

Three adult dolphins performing a synchronous display in the St. Johns River, Florida; the placement of the pectoral fin on the dorsal body of another dolphin is an indication of bonding. Photo taken by Q. Gibson under authorization of NOAA Fisheries GA LOC 23796

In bisexually philopatric populations, evasion of related males can reduce the cost of inbreeding. Inbreeding can reduce fitness through lower calving success and extended weaning age (Frère et al. 2010). Mating with unrelated males can increase the chances of better genetic compatibility by obtaining genetically diverse sperm (Jennions and Petrie 2000). Shark Bay females almost never associate with their sons while cycling but do preferentially associate with sons compared to non-sons during anestrous periods, suggesting they may be mitigating for inbreeding risk during estrous (Wallen et al. 2017).

When sexual coercion occurs, modified genitalia may provide females with a mechanism for cryptic female choice and the ability to evade fertilization (Eberhard 1996). While ventrum-ventrum mating has been observed (Tavolga and Essapian 1957), males attempting mating alongside females at the surface or by lateral-ventral or dorsoventral mounting may have better fertilization success (Connor et al. 2000b). Optimal copulatory fit corresponds to a dorsoventral positioning, and penile penetration may be curtailed by a vaginal fold (Orbach et al. 2016, 2017); females can subtly shift position and may obstruct a male’s fertilization success by redirecting sperm or the penis from non-preferred partners into vaginal recesses.

12.5 Conclusions and Future Directions

Within the polygynandrous mating systems of bottlenose dolphins, each sex exhibits conditional mating tactics to optimize sex-specific reproductive success. In estuarine or nearshore coastal populations, males rove between receptive females, solitarily or cooperatively mate guard females, or form preferential intersex bonds (Wilson 1995; Lusseau et al. 2003; Connor and Krützen 2015). Females counteract mate guarding through multi-male mating, evasive behaviors, and preferential intersex bonds (Boness et al. 2002; Galezo et al. 2018; Baker et al. 2020). Less is known about the sexual strategies of offshore dolphins. Offshore studies provide logistical challenges, and larger group sizes can make it difficult to maintain proximity to specific individuals; it is hypothesized that larger groups and deeper depths make it difficult for males to sequester and monopolize a female (Gowans et al. 2008).

Further research on contiguous study sites using similar methodological approaches to each other would be beneficial for modeling predictive parameters of alliance formation, while filling in data gaps on morphological, genetic, demographic, and socio-environmental differences (e.g., SSD, OSR/IBI, population density) would enhance a global comparison. Technological advances can reduce some of these data gaps. For example, laser photogrammetry can be used to assess sexual size dimorphism, and drones can provide greater context during behavioral studies (Cheney et al. 2018a; King et al. 2021). Where possible, focal follows should be conducted on individuals of both sexes to provide a more thorough understanding of the context in which alliances form and insight into both solitary male and female mating tactics.