Skip to content

The many forms of cooperation on coral reefs


At a time when the political world seems overfilled with evidence of human greed, selfishness, destructive competitiveness, and utter immorality, I found myself reading Tim Flannery’s 2011 book, Here on Earth: A natural history of the planet.  It’s not perfect, but it is a good read.  It lifted my spirits.

In it, Flannery contrasts the Medea and the Gaia hypotheses.  The Medea hypothesis, developed by Peter Ward, holds that natural selection drives a perpetual arms race among species that ultimately leads to system collapse.  The Gaia hypothesis, coined by James Lovelock, argues that the Earth system continually evolves to become more and more strongly interconnected as a self-organizing complex system with considerable resilience.  Flannery comes down in favor of Gaia, contending that natural selection is just as powerful in shaping affiliative, cooperative interactions and relationships within and among species, as it is in shaping competitive, predatory and other destructive interactions.  Let’s think of these as the win-win (Gaia) and the win-lose (Medea) interactions that characterize all interactions among living beings.  While I remain unconvinced that the planet is in the process of becoming some sort of living super-organism as James Lovelock proposed, I am persuaded by Flannery that we and all other species naturally have the capacity to develop increasingly strong affiliative responses, just as we naturally are capable of developing ever more savage destructive tendencies.

Two hypotheses about our relationship to Earth, Gaia and Medea, as visualized by Sarah Howell.  Image © Sarah Howell/

Because humanity changes through a combination of natural selection and cultural selection, our rates of change, in both positive and negative directions, are blindingly fast compared to those of our pre-civilized forebears, or those of other creatures.  This speed carries great risk because we have become a potent force for planetary-scale change, and change in bad directions (towards less well-integrated, less resilient, less effectively connected ecologies) can have substantial consequences before we are even aware of what is happening.  Flannery likens us to newborn infants still learning to function in our world; an apt metaphor that helps me understand why so many of us still do not appreciate the immense impacts of humanity on the planet over the last century or so, nor the urgency of the need to change course if we want to get to a ‘good future’ without going through hell to get there.

Evolution of cooperation

The easiest affiliative responses to understand are those within a social group of (usually) closely related individuals.  Quite simple genetics can explain the evolution of parental behavior, and the seemingly altruistic behavior directed towards one’s parents or siblings.  In these cases, actions that favor the survival of the relative can be selected for even when they lead to some risk or disadvantage to the actor.  Parents do risk their own lives to save their children, and siblings also behave altruistically to save one another; while we consider such self-sacrificing acts as noble, brave, loyal, they are also readily explained as a consequence of natural selection – taking a risk to help an immediate relative helps ensure the survival of copies of a substantial proportion of your own genes, and evolution is all about getting copies of your own genes to survive.

Extending such altruism to more distantly related members of your social group, or to random strangers, requires more complex genetic reasoning.  One argument made is that groups which contain at least some altruistic members will enjoy enhanced survivorship overall, relative to other groups comprising more completely selfish members.  Extending affiliative responses to other species requires even more genetic gymnastics, but the gymnastics are plausible and cross-species affiliative behavior definitely exists on our planet.  There are wonderful examples on coral reefs, and these are my topic today.  They are one more example of how coral reefs are magnificent, and why we should care more about their continued presence on this planet than we currently do.

Cooperation on coral reefs

Coral reefs are species-rich ecosystems.  Individuals live in close proximity to many others, of their own and other species, so opportunities for interaction are frequent.  Some of these associations among neighbors are symbiotic (literally ‘living together’), as mutualisms, commensalisms, or parasitisms.  Of these, only the parasitic ones are examples of win-lose interactions.  Symbiosis is a common occurrence on reefs, and many people have claimed the high frequency of symbiosis is a unique attribute of reef ecosystems, making them different to other marine systems.  I doubt there is anything ‘special’ about coral reefs in this regard.  So far as I know, nobody has yet demonstrated that symbiotic relationships on reefs are more common than they are in other ecosystems, once one adjusts for the very large number of species typically present there.  To put it bluntly, there are few examples of symbiosis on the tundra because that is an ecosystem comprised of few species, and the reverse is true for coral reefs.

Never mind symbioses for a moment.  A coral reef consists of a number of neighborhoods, each occupied by numerous individuals of many species.  Most of these individuals are quite circumscribed in their movements from day to day; they are resident in their neighborhoods.  Setting aside the sessile creatures such as corals, gorgonians and sponges, it does not take many visits, by snorkel or scuba, to recognize that each neighborhood has its own resident fish, crabs, sea urchins and so on that are reliably present day after day.  The fishes have neighbors that they recognize as individuals whether of their own or of other species.  I suspect the crustaceans are fully as aware; the worms, molluscs, echinoderms likely far less so.  Affiliative interactions do not require such awareness, but its presence enhances the possibilities.

A coral reef, such as this one in the Red Sea, is made up of neighborhoods, each with its own residents of various species.  Plenty of opportunities for interactions, affiliative or not.
© Vladimir Levantovsky.

Cooperative damsels

The thousands of damselfishes of the world can be roughly divided into the small, colorful, planktivorous ones (that help explain why the group was called damselfishes in the first place), and the larger, stockier, more drably colored and downright belligerent herbivores.  People whose experience is primarily in the Caribbean think of damselfishes as these belligerent beasts because they comprise over half the species present.  Those whose experience is primarily Indo-Pacific think of damselfishes as those delicate, colorful damsels hovering in vast shoals over every available coral-covered spur or slope, picking plankton one by one as the water streams past.  For them, the relatively far fewer belligerent ones are atypical of what damselfishes really are.

Lemon damsels (Pomacentrus moluccensis) and humbugs (Dascyllus aruanus) feeding on plankton above branching Acropora.  Not wandering, they are at home.
Photo © Luciano Napolitano.

When we see a shoal of the bright yellow damsel, Pomacentrus moluccensis, hundreds of individuals hovering 50cm to a meter or so above branching Acropora, we tend to assume they are rather boring little creatures with zero individuality.  Some of us simply classify them as ‘minnows’ or ‘fish food’.  Such creatures have not received the detailed attention from behavioral ecologists they might deserve, and I think we’d be surprised if some such attention was paid.  I say this because a long time ago a graduate student of mine, Bruce Mapstone, did pay attention to P. moluccensis, and while his focus was primarily demographic ecology, he tagged a few with subcutaneous latex paint marks, and discovered that the same individuals took up position over the same patch of coral day after day, that they hovered to the left and the right, above and below the same other tagged fish, and despite being ‘silly little fish’ could live a decade or more.  Now Bruce, in his wisdom, wandered away without ever publishing his results, got into more ‘important’ reef ecology, and eventually wandered clear out of the tropics to a career in Tasmania.  But think about the fish.  Day after day, the same fishes, in the same spatial relationship to one another, hovering over branching coral, feeding on plankton.  I’d never have believed that, and I would not be surprised if somebody one day discovers that their social behavior is a bit richer than just foraging beside their buddies!

The belligerent herbivorous damselfishes have received far more attention from behavioral scientists.  What is quickly obvious is that while individuals each tirelessly defend their own small territories from nearly every creature that comes by, they live in groups with contiguous territories, and spend lots of time arguing with each other across the shared borders.  In some cases, the local groups may exist because the habitat is patchy, and they have filled a patch of suitable habitat.  But in other cases, the habitat is not obviously patchy in this way, and these belligerent little fish still persist in living side by side.  Some of my earliest research once I arrived in Australia involved tracking real estate transactions among groups of territorial damselfishes occupying small patches of rubble habitat on the reef slope.  As new young juveniles arrived, as individuals grew, expanding the size of their territories, and as individuals disappeared, presumably because something ate them, space in the rubble patch would get reassigned, borders would be redrawn, and I would see that some individuals were gaining whilst others lost.  What made this particularly interesting to me was that there were three different species of damselfish in my rubble patches.  Their competition did not seem to be leading to clear winner and loser species over time, despite the fact that individuals were arriving and departing and a simple ecologist might be forgiven for expecting that over time the ‘superior’ species would come to hold all the territory.  But that is a whole other story.

I raise territorial damsels here because they do live in social groups, even groups comprised of more than one species.  Why do they do this, given that there seems to be enough available space in most cases for them to spread out and have more tranquil lives?  The late George Barlow of UC Berkeley, and one of the greats of behavioral biology, coined the term ‘dear enemy effect’ to describe the tendency of territorial species to live beside one another, and to behave less aggressively to known neighbors than to strangers.  This is certainly the case for territorial damselfishes.  But that does not explain why they cluster together.  Nor does the need for accessible mates – they could space themselves apart and still come together to breed with two or three quick flicks of their tails.  We have to broaden the frame of reference and remember that these fish are defending their territories from myriad species of fish and some invertebrates – any herbivore or potential egg predator gets particular attention.  And when we do, we find a wonderful example in which group living by territory holders improves their ability to defend their homes, while group foraging by roving herbivores increases their capacity to invade and feed within those same homes.

Cooperative grazers – using win-win to succeed at win-lose

While there are many solitary herbivores, many species of parrotfish and surgeonfish tend to travel in groups, of one or several species.  These are not the highly-organized schools of herring, sardine or anchovy, maintaining rigid spacing one from another as they perform intricate group gymnastics rivalling those of flocks of birds and swarms of insects, while leaving human synchronized swimmers far behind in their dust.  The herbivore schools are more like herds of cattle or sheep as they spill across the reef, munching algae as they go.  When such a group moves over the territory of a damselfish the defender’s capacity to defend is swamped, and plenty of herbivores get to feed within the territory.  Work by Susan Foster in Panama in the mid-1980s demonstrated how damselfishes were more successful at defending their algal mats when in larger groups, and that surgeonfishes were not successful at all in entering damsel territories when alone, but could enter and feed when in groups.  She showed that the feeding rate of blue tang within damsel territories was directly related to the size of the foraging group.  What we have here is pugnacious territorial damsels banding together as a group because the group of contiguous territories is better defended when they act together than individual territories could be, and roving herbivorous parrot- and surgeonfishes banding together as an effective way of managing to get some feeding done within damsel territories – two win-win affiliative responses creating two opposing groups of allies engaged in a win-lose war for access to food.

A school of Manini, Acanthurus triostegus, foraging across a reef at Hanauma Bay, Hawaii.
© Hanauma Bay Snorkel Tours.

Cooperative hunting

It’s not only reef herbivores that sometimes band together in feeding.  In a post in July 2015, I described a study showing how the coral trout, an important serranid piscivore on the Great Barrier Reef, solicits the help of a moray eel when foraging for fish in complex reef habitats.  Many reef scientists have observed predators of different species apparently teaming up to hunt for prey from time to time, but the particular study that caught my eye was an experimental one, which asked the intriguing question, “Will a coral trout solicit the help of an eel when the prey would be otherwise inaccessible to it, but not share the hunt with an eel if the prey is more accessible?  The short answer is ‘yes’ (read the post if you want more), revealing that not only do reef fish cooperate across species in hunting, but that they decide, depending on the particular circumstances, whether to seek out potential partners or not.  This is surely learned behavior of an advanced kind, and I’d love to know whether young coral trout (and eels) learn to cooperate by watching more experienced members of their species, or if these rather solitary creatures have to learn this by trial and error.  I’d also like to know how general a trait this is both within locations and across reef regions.

Just like a small cantina

To those of us who remember the first screenings of Star Wars, episode IV, in 1977, two or three scenes stand out clearly.  The Mos Eisley cantina on Tatooine is invariably one of these.  An amazing collection of very different species, all gathered together, mostly though not entirely peacefully, enjoying their favorite libations while the bar tender served and the band played on.  There is tension in the air; brief savage fights break out, but there is also laughter at shared jokes.  Creatures that would not be friends at other times or places, come together in the cantina for enjoyment and deal-making.  For me, a cleaning station is the coral reef version of this cantina, particularly when it is busy and fish of many species are queuing up waiting their turn to be serviced.

Chalmun’s Cantina in Mos Eisley, Tatooine.  A little like a cleaning station with fish of many species lining up to be serviced?  Photo © Lucasfilm.

 As anyone with a rudimentary knowledge of coral reefs knows, certain reef fishes set up cleaning stations which are visited by a wide range of species of fish (the cleanees) seeking to have parasites removed.  As is often the case, this phenomenon is better (or more extravagantly) developed in the Indo-Pacific than in the Caribbean.  In the Indo-Pacific, the most prominent cleaning stations are maintained by species of the genus Labroides, a medium-sized wrasse reaching about 10cm in length.  Labroides is an obligate cleaner, in the sense that it feeds on ectoparasites throughout its life, so long as other species of fish are available to be cleaned.  A number of other species of wrasses, gobies and other fish, and a number of shrimps clean in both the Indo-Pacific and the Caribbean.  For most fish other than Labroides, this is a juvenile occupation, although it’s a whole life career for some gobies.  Cleaning stations in the Caribbean, which lacks Labroides, are a far less dramatic engagement than what takes place at Indo-Pacific sites, where large fish of many species can be seen lining up awaiting their turn to be cleaned.

Given that cleaning requires that a small, nutritious morsel – the cleaner – must approach a larger, often piscivorous, fish with big teeth, all the while dancing seductively, and then cruise about close to its surface, touching it intimately, and even wandering into the mouth and gill chamber in search of pesky parasites, the existence of cleaning behavior is an amazing example of why reefs are wondrous places.  Cleaning is not a close association between two individuals of different species who might have learned to recognize each other and recognize the benefits of helping each other.  This is an example where one small species of fish has set up shop offering a personal grooming service that any other species of fish is welcome to request (or other large creature, because Labroides will clean turtles and divers just as willingly as large fishes).  These are creatures like Winnie the Pooh; they are ‘of little brain’.  A single misstep means one fewer cleaner exists.  How do cleaners know that it is safe to approach a cleanee?  How do cleanees know that a cleaner is offering a service?  Why don’t potential cleanees ever decide to enjoy a cheap meal at a cleaner’s cost, and then go to a rival station to be cleaned?  Put another way, cleaner fishes put themselves at far more immediate risk of death than any street prostitute ever does, no matter how bad the section of town she/he patrols.  Their clients are often many times bigger than they are, they are expected to put themselves very much in harm’s way, and their interactions are always cross-species.  People often have difficulty training their dogs to behave!  Who tells the naïve young coral trout that visiting a cleaning station is fun, but there are certain unspoken rules you must obey?

A wrasse, Novaculiththys taeniourus, being serviced by two Labroides phthirophagus at a cleaning station on a Hawaiian reef near Kona.  Photo © Mila Zinkova.

A new review of cleaning symbioses by David Vaughn of James Cook University, Australia, and three colleagues, to appear shortly in Fish and Fisheries, makes clear that the cleaning phenomenon is geographically widespread in marine and freshwater environments.  It is particularly apparent on coral reefs but that may partly reflect differing levels of researcher attention.  They list 208 fish species and 51 shrimp species as reported to engage in cleaning behavior, but many of these are facultative cleaners performing only occasionally or only during juvenile life.  They also report that cheating by both cleaners and cleanees has been documented; indeed, cheating may be quite common in the cleaner wrasses, Labroides which frequently ingests mucus and scales while ostensibly removing parasites.  Indeed, George Losey, who did much of the pioneering work on cleaner behavior maintained that Labroides was an inveterate cheat, really out to seduce other species by giving them all the tactile stimuli they want, in order to get a meal of mucus or scales, or parasites if any were present.  Cheating by cleanees, by eating cleaners, appears to be a lot less common but still occurs, and my wonder at how cleaning symbioses evolved and how they are maintained remains.

Cooperation with corals

Until now I have avoided the corals and other sessile creatures.  I tend to think of them as part of the habitat, the backdrop to an exciting play involving the more mobile creatures.  Yet they too are reef creatures, and they are important in many cross-species interactions.  Many reef scientists would begin, and perhaps end, a discussion of symbiosis with the relationship between corals and their algal symbionts, the zooxanthellae.  Given the importance of coral bleaching these days, and the fact that bleaching is the breakdown of this very close relationship, I should not ignore it.  The coral-symbiont relationship is a crucial factor in permitting the existence of coral reefs.  But that is all I will say – I prefer cooperative associations in which behavioral decisions (rather than physiological or chemical ones) are more obviously in play.

Many reef creatures, such as the lemon damsels I began with, use corals as shelter sites.  In many cases, including the lemon damsel, living coral is so strongly preferred that these fish will leave a coral after it has been killed, running the risk of not finding another suitable refuge.  One should call these associations with living coral a cross-species affiliative response, but they are not terribly interesting ones.  Except in cases where they are.

Gobies of the genus Gobiodon are tiny obligate occupants of branching corals of the genus Acropora.  They are widely distributed through the Indo-Pacific, and show evident preferences for particular host species.  They settle from the plankton into living Acropora colonies and are relatively long-lived (four years) for small gobies.  They spend their entire lives among the branches of their home colony, feeding partly on coral mucus, but also on small invertebrates and algal cells.  They usually are found in pairs.  In a series of papers beginning in 1997, Phil Munday of James Cook University, Australia, and colleagues mapped out the use of corals by the eight species of Gobiodon present on the central Great Barrier Reef.  Each species is associated with from 3 to 10 species of Acropora, but shows evident preferences for certain species.  Gobiodon species overlap in the species of coral occupied, and there is evidence of competition for host colonies.  Apart from the fact that the gobies only occupy living corals, there is little about this relationship to suggest it is an affiliative response between fish and coral.  But in 2012, the story changed.

Gobiodon histrio, nestled among the branches of its Acropora host with some fronds of Chlorodesmis fustigiata to the right.  Photo © Danielle Dixson.

Danielle Dixson and Mark Hay, of the Georgia Institute of Technology, reported their studies of gobies and corals in Fiji in an article published in 2012 in Science.  They had done a series of field and laboratory experiments using Gobiodon histrio and Paragobiodon echinocephalus, a second coral-dwelling genus, the coral Acropora nasuta, and a seaweed, Chlorodesmis fastigiata.  Both gobies are common in A. nasuta colonies at Fiji.  Chlorodesmis is toxic, and coral surfaces in contact with its brilliant green fronds are damaged or killed.

Corals that were empty or housing only crabs suffered substantial impairment when Chlorodesmis fronds were placed in contact with their tissues.  Those housing either species of goby were scarcely affected by the alga.  Algal mimics of nylon threads placed in contact with the coral also had no effect on coral performance.  Image © D. Dixson & Science.

Dixson and Hay demonstrated that rather than just being a case of gobies choosing to live among the branches of live coral, like so many squatters in an abandoned building, both species of goby were actively protecting their home coral from contact with the alga.  And the coral, when it came into contact with the alga, was releasing chemicals which served to summon the fish to its defense.  Their series of simple, yet convincing experiments showed:

  • that corals were damaged when fronds of Chlorodesmis were brought into contact,
  • that corals occupied by Gobiodon or Paragobiodon, but not by two other species of common coral-sheltering fishes, were protected from contact with the alga, because Gobiodon ate the offending alga while Paragobiodon bit off fronds and removed them from the vicinity of the coral,
  • that chemicals released by coral tissue in contact with Chlorodesmis attract Gobiodon, while Chlorodesmis alone does not, and
  • that when the toxic hydrophobic chemical is extracted from Chlorodesmis, and applied to a nylon twine mimic of the alga placed in contact with a coral, it still attracts Gobiodon.

Naturally, Dixson and Hay described these results using verbs like ‘signal’ and ‘respond’, even referring in a press release to Gobiodon as ‘coming to the aid of’ its host coral.  The study got plenty of press at the time.  Still, it is a surprisingly intimate association between very different kinds of creatures, acting in ways that provide benefits to both.  Personally, I find the unanswered question about Paragobiodon one of the most interesting.  Gobiodon, when alerted chemically attacks the alga and eats it.  It gets some food, and perhaps bolsters its own chemical defenses (it is toxic itself).  One can easily imagine it learning to associate the particular chemical as a sign that there is food nearby.  But Paragobiodon does not eat the toxic Chlorodesmis.  It responds just as strongly to the chemical signal from the coral, but then bites off fronds and carries them away.  Why?  Is it just a neat freak?  Or is it altruistically caring for its coral home?

I admit to being biased against corals and other sessile creatures.  I have focused almost entirely on fishes.  I’ve managed to avoid any mention of Nemo and his anemone home, but I want to finish with an example involving another anemone and a crab.  My attention was drawn to it just this month, when Laurie Richardson of Florida International University told coral list about a new paper.  (Not the best way for a scientist to keep up with the literature perhaps, but, hey, I am retired!)

Pom-pom or boxer crabs are tiny members of the genus Lybia, one of the many small crabs that occur on coral reefs.   They occur across the vast Indo-Pacific from the Red Sea to the shores of Hawaii.  Lybia leptochelis is a Red Sea species; like all members of the genus, it carries a small anemone in each claw which is otherwise weak and not suited for defense.  The anemones provide defense and they also catch food some of which the crab steals for its own use.  In the Red Sea, L. leptochelis carries anemones belonging to an as yet unnamed species of Alicia.  Even the smallest, recently hatched Lybia have minute anemones in their claws, although immediately on hatching they are unarmed.  The Alicia species has only been found on Lybia claws.

The boxer crab, Lybia leptochelis with its two Alicia anemones, one in each claw.
© Yisrael Schnytzer, PeerJ.

Now, what kind of relationship is this?  The crab clearly benefits from carrying the anemones around, and the anemone perhaps benefits by being carried to new sources of food.  That the anemone is very rare if present at all away from the crabs suggests strongly that this is an obligate partnership for both partners.  The article that caught my eye was published by Yisrael Schnytzer, of Bar-Ilan University, Israel, and three colleagues in the journal PeerJ at the end of January.  By means of a combination of field observations, aquarium experiments and genetic analyses, Schnytzer established that crabs deprived of one anemone will usually split the remaining anemone in two longitudinally, thus mimicking or facilitating normal asexual reproduction via fission.  They do this by holding the body of the anemone with both claws and slowly stretching it apart over a period of 15 min to two hours.  The authors have a neat video of this in the supplemental materials accessible from the article.

Crabs that lack anemones will fight with crabs possessing anemones and will usually succeed in taking one of the anemones away.  In such cases, both crabs cause the fission of their single anemones over the next few hours or day, so that each then has two.  In some instances, fighting results in only part of one anemone being ‘captured’.  When this happens the crab with the partial anemone will attempt to split the part to provide one small anemone for each claw.  Genetic analysis revealed very little genetic variation among anemones collected from crabs within the Red Sea study location, and showed that every crab sampled carried a pair of genetically identical clones.

Putting these results with earlier results obtained by this team, it appears that Lybia leptochelis is an obligate carrier of the Alicia anemone, despite the fact that Triactis producta, an anemone commonly carried by other Lybia species is commonly present living freely in the environment.  Further, Lybia, by regulating the food available to its anemones practices a sort of bonsai, limiting anemone growth; anemones are closely sized to the size of the host crab.  When you add in the newly reported forced reproduction of the anemone, this relationship is looking a lot more like farming than a mutualism in which the partners gain almost equal benefits.  Does the crab view the anemone as a partner?  Well, even if the crab were capable of such thoughts, I doubt it would.  This is crustacean animal husbandry pure and simple, but it is still a great example of the complexity that is possible in the positive, affiliative interactions among creatures on a coral reef.  Win-win relationships, sometimes bizarre ones, are common on this planet.  Think about that next time you see a tweet from the Oval Office.  It IS possible to see things differently to the current view from there.  Even for little, orange crabs with weak hands.  Think of that next time you see the current occupant of the Oval Office.

2 thoughts on “The many forms of cooperation on coral reefs”

Comments are closed.