Read about some of the findings from our research...

Female Reproduction
Male Alliances
Calf Development
Foraging Strategies
Sharks and Dolphins


Bottlenose dolphins are very social animals who live in complicated social systems known as fission-fusion societies. Much like humans and chimpanzees, who also live in fission-fusion societies, all individuals in the dolphin community do not stay together; the number of individuals in groups changes over time; and each animal has certain individuals he or she prefers to associate with. The complexity of fission-fusion systems lies in the fact that information about who is associating with whom is not always available. Since relearning social standing every time dolphins encounter each other would presumably waste time and energy, dolphins would benefit from remembering individuals and interactions between individuals not seen on a regular basis. The fact that dolphins live in the water where it is relatively easy to move from place to place means that individuals can interact with larger numbers of other dolphins on a less regular basis than animals that live on land. The average group size in Shark Bay is 4 – 5 dolphins, however this number may depend on the individuals present in the group and what they are doing at the time (Smolker et al. 1992).

Overall, adult male and female dolphins lead fairly different social lives. Male bottlenose dolphins in Shark Bay form hierarchical alliances that cooperate to obtain and sequester females for mating (see section on male alliances). This nested nature of male bottlenose dolphin alliances is more complex than any non-human mammal (Connor et al. 1992). While adult male Shark Bay dolphins are almost always with their alliance partners, female bottlenose dolphins do not form alliances and vary widely in their degree of sociality. These females form loose social networks with their number of associates ranging anywhere from 0-139 (Smolker et al. 1992; Gibson & Mann 2008b). While females do appear to have certain individuals they prefer more than others, they typically spend less than 30% of their time with these top associates (Smolker et al. 1992). In a recent examination of the possible reason for female bottlenose dolphin groups in Shark Bay, researchers found that mothers with calves in their first year of life tend to form larger groups, suggesting that the groups provide these vulnerable calves with some form of protection. The researchers also found evidence that female grouping allows calves, particularly males, to develop social skills before a lack of social savvy negatively influences reproductive opportunities (Gibson & Mann 2008a). Bottlenose dolphins show bisexual philopatry, meaning that as adults, both sexes stay in the general area of their birth. Therefore, social relationships have the potential to form early in life and last into adulthood, however the first in-depth investigation into bottlenose dolphin calf social development was only recently published (Gibson & Mann 2008a). Not surprisingly, calves increased the time spent separated from their mothers as they approached weaning. Additionally, male calves increased the time they spent in groups during separations from their mothers, while female calves decreased their time spent in groups. Researchers also found that the number of associates a mother had was reflected in the number of associates her calf had (Gibson & Mann 2008a).

Few species inhabit social systems as complex as the fission-fusion system of bottlenose dolphins and it is presumed that this complexity persists because it increases survival and reproduction. Research in Shark Bay is addressing this assumption by investigating bottlenose dolphin sociality using a method known as social network analysis.

Click image to open in new windowClick image to open in new window
Figure 1: A visual representation of the complexity of social networks in the Monkey Mia dolphin population.

A social network is defined as a set of actors linked by relationships. Facebook is the most common example of a social network in which people are linked by bonds of “facebook friendship.” Social networks provide more realistic representations of dolphin society and allow us to look at multiple levels of the society by making available information on individuals, subgroups, and the network as a whole. We can look at how connected a dolphin is to others in the network and how tightly groups of dolphins are connected to each other. These different social network measurements offer different information about the same entity and provide us with data not accessible using traditional methods. We are currently using social network analysis to investigate topics including calf social development, potential fitness consequences of sociality, and the transmission of foraging strategies.

Return to top

Female reproduction

Results and figures taken from:

Mann, J., Connor, R.C., Barre, L.M. & Heithaus, M.R. 2000. Female reproductive success in wild bottlenose dolphins (Tursiops sp.): Life history, habitat, provisioning, and group size effects. Behavioral Ecology, 11:210-219.

Our long term study has revealed:

  • The months of October - December are the peak birth months, occuring just before the peak in surface water temperature. (Figure 1)
  • Females give birth to their first calf at ages 12-15. However, a few females, such as a dolphin named "Peglet", have given birth to their first calves at the age of 11.
  • Interbirth intervals for surviving calves range between 3- 8 years. Interbirth intervals of 4 years were most typical. (Figure 5)
  • One third of all females studied have not calved successfully in a 10 year period. Reproductive success was defined as the number of calves surviving to 3 years old within a ten year period. (Figure 6)
Figure 1: Number of births and average temperature by month (SD ranges for temperature= 0.9-1.6). If a birth could not be assigned to one month, it was divided equally among the possible birth months. (N = 74 calves, 47 mothers).

Figure 5: Interbirth intervals when the first calf survived to weaning (N=33). The graph demonstrates that average birth interval is 4.55 years (SD=1.00, median= 4.07). Shaded bar represents a possible gap if she gave birth >1 year after weaning for first calf, then she could have lost a fetus in the interum.

Click images to open in new windowClick images to open in new windowFigure 6: Graph demonstrates 30% of the females had no surviving calves, 37.5% had one surviving calf, 25% had two surviving calves, and 7.5% had three (N=40).

Return to top

Male Alliances

Since 1987 Dr. Richard Connor has focused his research on male alliances. In Shark Bay, alliances of 2-3 males cooperate to herd individual females for periods of up to over a month. Some alliance partners are nearly always found together and their bonds may last for years.

Richard and his colleagues have been following one such pair of males, Real Notch and Hii, for fifteen years (Hii is pictured with close associate Bottomhook on right). They think that by cooperating in an alliance, the males can prevent their female consort from mating with other males and other males from mating with her.

Unfortunately for the males, females appear to have countered this male strategy with one of their own: they come into estrus several times the year they conceive making it very hard for any group of males to insure that one of their members fathers her offspring.

Incredibly, they found that teams of alliances cooperate to attack other alliances to take their females. Thus, there are two levels of male alliances in our dolphin society, the alliances that herd females and teams of alliances that steal them.

This was a very exciting discovery!, because such nested levels of male alliances are common in human society but very rare even in non-human primates. The key is that both levels of alliance occur within a single dolphin society, not between two different societies.

Many birds and mammals form groups to defend their turf, but these are all 'us against them' interactions--nothing complicated about that. But in social groups, dolphin alliances can take on the dimensions of a soap opera, with individuals using friendly behavior to compete for favored allies in a strategic fashion to enhance their social position. Such alliances within groups are common in primates but rare elsewhere-and dolphin males have two levels of such alliances!

But the story does not end there. In the 1990s they documented a 'super-alliance' of 14 males that, not surprisingly, handily defeats other alliances in their area. Members of the super-alliance still get together in groups of 3 for the purpose of herding a female, but to their surprise, they found that after a trio of males finished herding one female, the males often joined a different trio to herd another one (but only with other members of the super-alliance). This came as quite a surprise after years of watching the stable alliances that always stuck together.

Richard and his colleagues suspect that the males have to cooperate with a larger number of super-alliance members to maintain a degree of cohesion in the group. His future research will focus on finding other large alliances to see if alliance stability correlates with group size and on discovering the ecological and genetic bases for alliance formation.

Return to top

Calf Development

A brand new calf is always an exciting event for the dolphins- and the researchers. This is also a difficult time for both mother and calf. The newborn must learn to coordinate his or her movements with the mother, breathe at the surface, dive, nurse, and avoid danger such as tiger sharks. Much of the popular and scientific literature has promoted the view that females “babysit” for each other, referred to as allomaternal care. Births to the provisioned dolphins in the early 1990s allowed us to systematically observe several wild newborns nearly every day from birth. During these and offshore observations, it became clear that during the first week of life, female dolphins that are either inexperienced or who have not successfully calved, attempt to steal the newborn from the mother by swimming rapidly past it. The newborn, with an innate following response, simply swims after the faster moving dolphin. We have called this “natal attraction” since the females seem to be drawn towards the infants. Mothers sometimes attacked females who stole their calves, and they typically had to steal their own calf back using the same method. Adult females are rarely aggressive in any other context. Oddly enough, mothers tolerate those same females swimming away with their calves during the second week of life. We’ve proposed that the first week is an imprinting period and mothers do not tolerate separations from their calves until they’ve learned to recognize the mother (Mann & Smuts 1998).

When calves are born, they will have lumps or folds in their skin, referred to as fetal folds. These folds last for approximately one week. Fetal lines, white lines resulting from the folds, develop and will remain until approximately the third month.

Mothers and calves use distinctive whistles to mediate reunions after separations of up to several hundred meters (Smolker et al 1993). While about two-thirds of calves studied are weaned before their fourth birthday, some calves nurse for up to eight years. Current studies are investigating the reasons behind this high variation in maternal care.

In our population, 44% of calves do not survive to age three, with the highest mortality rates occuring in the first year of life. There are several variables that can contribute to calf mortality as shown in the diagram below (adapted from Mann & Watson-Capps 2005). The majority of calves show signs of poor health prior to their death, and have also been shown to seek out more contact with their mothers when compared to surviving calves.

While calf condition is likely the primary cause of mortality, predation also has significant effects on maternal behavior, with mothers avoiding dangerous deep water habitats during shark season. When it was observed that calves born to provisioned females were experiencing elevated rates of mortality (56% in the first year), our research aided in implementing a management regime that has significantly improved the survival of these calves.

Click image to open in new windowClick image to open in new window
Figure 4: A schematic representing a possible network of factors leading to calf mortality in bottlenose dolphins (in boxes), the direction of their effects (arrowheads), and the strength of their relationships (indicated by the width of the line).

Calves begin catching their own fish at around 3-4 months of age. They usually begin with a technique called "snacking, where they chase small fish belly up, possibly to silhouette their prey against the surface so they can easily snap it up. As calves get older they develop a wide variety of foraging techniques, which they seem to learn from their mothers through observation (Mann et al 2007).

Return to top

Foraging Strategies

Bottlenose dolphins feed on a wide variety of fish, cephalopods (squid, octopodes), crustaceans (e.g., shrimp, prawns) and occasionally stingrays, sharks, eels, and mollusks. Approximately 20 species of prey have been identified for Shark Bay, but we know little about the bulk of dolphin diets. Our observations of fish caught are biased towards large fish because the long time they spend carrying and breaking up these fish (up to an hour in some cases) allows for more accurate identifaction. Examination of stomach contents from Indian Ocean bottlenose dolphins caught in shark nets off Natal, South Africa found about 48 species of fish and cephalopods. Our latest research involves both using fatty acid signatures to match up dolphins with their suspected prey, along with increased underwater filming to capture images of the more benthic species they may be feeding on.

A number of foraging strategies have been described for bottlenose dolphins, and Shark Bay dolphins have been observed at least at least 13 different techniques. They use echolocation or sonar to track fish, but vision may be quite important too. Dolphins might also listen to sounds that fish produce, and use these to locate their prey.The variety and complexity of these techniques, as well as the manner in which they are passed down (primarly from mother to daughter) presents evidence for social learning and may even qualify as a type of animal "culture."

Wedges & the Golden Trevally

In May of 1999, Dr. Janet Mann was surveying a group of dolphins and noted a very tiny newborn calf. The mother, Wedges, stayed close by her side, but left twice to chase a large dusky shark that was pestering the group of females and calves. The calf was clearly born out of season, and hence named "Whoops". Newborn calves are hard to come by, so Janet enrolled Wedges in her long-term study of mothers and calves. During that first follow, Wedges began these tremendous high leaps in rapid pursuit of fish. She caught her prize, a large golden trevally, nearly one meter in length.
In 12 years of watching adult females, Janet had never seen a catch so impressive. Whoops seemed less impressed. She darted back and forth, whistling constantly as her mother b-lined for shallow water and broke the fish up in shallow sand flats. It took Wedges nearly an hour to consume the fish. Poor Whoops didn’t get to nurse the entire time. But there were other things to worry about. Twice, large tiger sharks cruised close by, seemingly attracted by the broken bits of fish. Each time Wedges and Whoops sped to even shallower water to continue feeding. At the time, Janet thought this event was merely a “fluke” and wouldn’t happen again. About two hours after the first trevally catch, Wedges caught another, although this one was slightly smaller. Her team has now followed Wedges for ten years and she catches a trevally every few hours. During one follow in 2002, Wedges finished off a large trevally and then lay still in 4 ft. of water for over four hours. Her slight arching suggested that she may have had an awful stomach ache. No surprise, given the size of the trevally bones. The team anchored and waited with her until sunset. By the next day, she was out hunting again. No other dolphin has been observed catching a fish this large, although Puck was seen carrying trevally twice, but both times Wedges was close behind her. Researchers suspect that Puck may have stolen her fish, a rare occurrence among dolphins. Whoops, who is nearly five and going through the weaning process, is too small to catch such spectacular fish, but they expect that one day, she will.


One of the most exciting foraging strategies, known as "sponging," is the only recorded instance of cetaceans using tools in the wild. Over 70 dolphins have been documented wearing sponges on their rostra (beaks) in Shark Bay. They use these sponges to ferret prey from the seafloor. This is generally a solitary activity, but sometimes more than one sponger will be tens of meters apart. Spongers generally stick to sponging and don't use other foraging techniques, but there are some exceptions. This appears to be a "tradition" of sorts. The daughters of sponge-carriers clearly grow up to be sponge-carriers, but we aren't sure about the sons. New research from the genetics team at the University of New South Wales will shed light on the relatedness of spongers.

Sponging was discovered when Rachel Smolker, one of the founders of the Shark Bay Dolphin Research Project, was told in the mid-80s about a dolphin north of Monkey Mia with a horrendous growth on its nose and half a tail fluke. Not believing this fisher’s tale, she thought little of it. But, a year later, in the location the fisher described, she too, saw the dolphin with the growth and half a tail fluke. As she watched, she noticed that the dolphin managed to change its growth, and at closer inspection, she discovered that this was in fact, a marine sponge. She named the dolphin Half-fluke and soon found that others were carrying sponges too. About one year later, Half-fluke had a calf, Demi. Although Half-fluke and Demi were seen regularly, only Half-fluke carried a sponge. In 1989, when Demi was nearly 3 years old, researcher Janet Mann observed her wearing a tiny sponge on her beak, conical shaped like her mother’s, only much much smaller. She dived like her mother (flukes out) and seemed to be foraging like her. When she went back to nursing position (infant position), she dropped her sponge. Half-fluke had two more calves after Demi, but both died. Demi continued to sponge-carry after weaning and often associated with her aging mother. A year after Half-fluke’s death, Demi had her first calf, Dodger, before her 13th birthday. Janet and her graduate student Brooke Sargeant were very interested in Dodger and would observe her several times each year to see when she would pick up her first sponge. Soon after Dodger’s third birthday, she did. It was such an exciting event that a reporter-writer on our boat fell down and nearly knocked the rest of us off the boat. Dodger was soon joined by a sister, Ashton, who has also begun sponging. The tradition continues…

Other Strategies

Shark Bay dolphins also chase fish "belly-up" near the surface, a behavior we call "snacking." Calves engage in this type of foraging most often. At Point Peron, northwest from our main study area, a small group of dolphins appear to strand-feed, trapping fish in very shallow water. This behavior can be viewed on the National Geographic film, Dolphins: The Wild Side.

In the shallow seagrass beds out east, dolphins will arch their tails high before driving them forcefully into the water, creating a several meter splash and a resounding 'kerplunk' sound. We think dolphins learn the location of fish hiding in the seagrass when they are startled by 'Kerplunks.'

One might also argue that the provisioned females in Monkey Mia, Shark Bay, have developed a unique foraging strategy of begging for fish from boats and tourists. Such "traditions" have continued across at least three generations. This may be true for other foraging tactics as well.

At other sites, dolphins can be seen corkscrewing into the sand after fish (Bahamas), strand-feeding on mud-banks in Portugal, Georgia and South Carolina, or stunning or killing fish with a tail-hit. Found worldwide in warm coastal waters, bottlenose dolphins have also learned to take advantage of human activity. For example, bottlenose dolphins have learned to feed on fish drawn to garbage barges, follow shrimp trawlers as they stir up the bottom, or steal bait from lines or crab pots. In Laguna, Brazil, fishermen and dolphins appear to cooperatively net mullet, with the dolphins herding the fish into the nets and feeding easily off the remains. Historical accounts of Australian aboriginal cooperative fishing with dolphins have also been reported.

Return to top


Genetic analysis has been critical for understanding population structure, gene flow, relatedness, paternity, and more broadly, conservation. Genetic data have been collected since 1997, when shark bay researchers developed a novel biopsy system for small cetaceans, whereby adult dolphins were darted in order to obtain a small skin sample for genetic analysis. The research team had already identified several highly polymorphic regions of interest (Krützen et al. 2001), and these samples allowed for microsatellite and d-loop analysis in 95.8% and for genetic sexing in 99% cases. This was a novel method for reliable and safe tissue sampling of free-ranging cetaceans, with wounds healing completely in about three weeks (Krützen et al. 2002).

Being able to obtain genetic data from our dolphins allowed us to test several interesting questions, such as whether the formation of male alliances was based on relatedness. It turned out that males in stable first-order alliances tend to be highly related, while males in super-alliances with labile first-order alliances aren't any more related than one would expect by chance alone (see male alliances). This suggested that within one population and one sex there are multiple ways to form groups (Krützen et al. 2003).

Genetic data has also given support for natal philopatry, with males retaining their natal home ranges within slightly larger adult ones, which makes sense if they want to obtain inclusive fitness benefits from alliance formation. For females the data showed that adjacent localities were genetically similar for mtDNA, with a gradual shift in haplotype frequencies with increasing distance. Data which supported the existence of a network in which all females were connected to each other, and where dispersal in female is more restricted than that of male dolphins (Krützen et al. 2004a).

Examination of female association patterns in conjunction with genetic data has also shown support for the recognition of biparental kin. While females showed a preference for forming close bonds with matrilineal kin, they also showed a preference for having casual relationships with biparental kin outside their matriline, independent of the level of home range overlap. (Frère et al. 2010a).

Genetic data also allows us to answer questions about the possibility of dolphin culture. Using mitochondrial DNA analyses, our research has shown that the use of marine sponges as foraging tools demonstrates an almost exclusive vertical social transmission within a single matriline from mother to female offspring. Also significant genetic relatedness at the nuclear level suggests that all modern spongers have close coancestry, which may be the result of descendance from a single female, a "Sponging Eve." However, despite their close relatedness, sponging is unlikely to be the result of a genetic propensity. It’s unlikely to be explained by any single-locus mode of inheritance, with any sex limitation or other special expression pattern, and multilocus inheritance is equally unlikely because they would have to be so tightly linked that they behaved as one gene or there would have to be strong assortative mating. Likewise assortative mating, assortative mating is unlikely because since adult males virtually never sponge, any assortment would have to be based on some other (unknown) correlated trait, and sponging females have been shown to conceive from nonsponging males and almost all offspring of spongers are sired by nonsponging males, and there was no observed heterozygosity deficit among all 13 spongers that would be predicted by such assortative mating (Krützen et al. 2005). Further research tested the hypothesis that the propensity for sponging may be a result of enhanced respiratory and diving abilities, but it was found that the regions of mitochondrial DNA that predicted sponging were non-coding and thus wouldn't result in any phenotypic differences between spongers and non-spongers, which again supported the view that genetic transmission was not responsible for this form of tool use (Bacher et al. 2010).

Genetic data allowed for paternity assessment and therefore for us to examine male reproductive success. Such assessments showed that while alliance membership increased a male's chance of fathering offspring, some juvenile males were able to obtain paternities without being a member of an alliance. Reproductive success was skewed within some stable first order alliances, suggesting that the alliances were hierarchically arranged with more dominant males obtaining more matings than others (Krützen et al. 2004b). The data also demonstrated that at least one mating was incestuous, leading our team to later investigate the occurrence and effects of inbreeding within the population.

Levels of inbreeding in Shark Bay dolphins are indeed higher than expected. However, it is not without negative effects. Inbred females have been shown to have lower calving success (Figure 1) and inbred calves take longer to be weaned (Figure 2). While inbreeding has deleterious effects on female reproductive success, the lack of male dispersal and high level of sexual coercion by male alliances may be allowing inbreeding to occur regardless (Frère et al. 2010b).

Figure 1: Significant relationship between mothers' calving success (Cs) and their internal relatedness (m-IR) + the mean internal relatedness of their calves (c-IR).

Click image to view in new windowClick image to view in new window
Figure 2: Significant relationship between calves' weaning age in years and their internal relatedness (c-IR).

One of the most exciting new developments in the field of cetacean genetics is a new technique known as blow sampling. This method allows DNA to be collected from the epithelial cells present in the dolphins' exhalations. It's an exciting and much less invasive alternative to skin biopsying, and will be able to be more ethically applied to young animals and members of more vulnerable populations (Frère et al. 2010c).

Return to top

Sharks and Dolphins

True to its namesake, Shark Bay has a large number and diversity of sharks. From the small black-tipped sharks to the 5m tiger shark, Shark Bay is home to more than a dozen shark species.

The tiger shark, Galeocerdo cuvier, is the most common shark in the bay and has been the subject of intensive study by Mike Heithaus (Florida International University). They have found, using the "crittercam" developed by Greg Marshall at National Geographic, that tiger sharks hunt in deep and shallow habitats, but most of their prey is in shallow water (4-5m).

Studies of their stomach contents by Australian scientist Colin Simpfendorfer, indicate that seasnakes, turtles, and dugongs are the main prey of tiger sharks, but dolphins have been found in tiger shark stomachs at other sites. Over 74% of Shark Bay dolphins bear shark bite scars of varying size, which is much higher than reports for other populations, but this does not mean that sharks are the main cause of dolphin mortality. Both dolphins and sharks like to forage in the shallow seagrass beds where prey density is the highest, but during the warm months when shark density peaks, dolphins switch to resting in deep water and venturing into shallow habitats in larger groups, suggesting that they are seeking to minimize the threat of shark predation.

Researchers commonly see sharks and dolphins in the same area, and the reactions of sharks to dolphins and dolphins to sharks seems to depend on the size and number of sharks, the size and number of dolphins, and probably some element of surprise. We know that sharks sometimes eat dolphins but dolphins occasionally turn the tables. Researchers have seen sharks chase dolphins and dolphins chase or even mob sharks. Dolphin mothers sometimes chase small (1m) sharks from their young calves. Clearly the relationship between sharks and dolphins is complex and deserving of further study.

Return to top


Research demonstrated that calves born to provisioned females had twice the mortality than calves born to non-provisioned females. This was most likely due to over feeding and unregulated feeding. The Department of Parks and Wildlife (DPaW) responded to our research by changing their feeding policies. Calves were no longer fed and since the changes were implemented in 1995, calf survivorship has dramatically improved (Mann et al. 2000; Mann & Kemps 2003).

In addition, it has been shown that aquaculture may have an impact on dolphin ranging and behaviour. These data helped deter expansion of aquaculture in the Shark Bay area, benefiting dolphins and several other important marine species. These data are influencing aquaculture management in other parts of Australia and in New Zealand (Watson-Capps & Mann 2005).

Researcher Lars Bejder has been investigating the long-term impacts of vessel activity on the Shark Bay dolphins. His work has shown that when multiple tour vessels are present in a study area, the dolphin population subsequently declines in abundance in the same area. While such effects may not be catastrophic to the Shark Bay population due to its size and genetic diversity, they are still important to monitor, and can serve as crucial guidelines for the management of more vulnerable populations of cetaceans worldwide (Bejder et al. 2006).

Return to top