column spacer Defenses

Fish defenses


Defenses of fishes can be categorised into an heirarchical "cascade" beginning with:

1) avoid detection
2) take evasive action if spotted
3) prevent capture
4) prevent being eaten if captured
5) escape

Of these, the last is obviously the most important, but success in any of the others will aid in achieving it. Note that costs and risks to the prey escalate from first to fourth. The fifth and ultimate category, that of escape, can come at any time and won't be considered further here. The first task of any potential prey fish is to avoid detection. Helfman et al. 1997 The diversity of fishes. Blackwell Sci Publ.


hot buttons for defensive "tasks" for fishes

Fish defenses: prevent being eaten if captured: poisonous flesh

  If captured by a predatory fish, the "task" of the intended prey fish is to prevent being eaten. This may include having poisonous flesh, poisonous secretions, or electrical shock, topics considered here, or having ERECTABLE SPINES, a topic dealt with in another section.
seahorse dive leader for Biology of Caribbean Coral Reefs website photograph of pufferfish taken from a video "Whoa up, damselfish. Take care. Those pufferfishes and this one probably have toxic livers, skin, and gonads, which can cause sickness or death when eaten. In fact, the notorious fungu cuisine in Japan is the flesh of a pufferfish." - Bonaire 2003

The peaceful demeanour of a Caribbean sharpnose pufferfish belies the deadliness of its flesh. If similar to Indo-Pacific counterparts, this pufferfish's skin, liver, and gonad will contain tetrodotoxins that are 25 times more potent than curare. Now, if a predatory fish gets poisoned by a prey pufferfish at the moment of capture, the lesson is simple and quick for the predator...don't eat photograph of a sharpnose puffer courtesy Anne Dupont, Florida another one of those particular fishes! If the attacker's bite is non-lethal to the intended prey, and if the predator has learned its lesson, then the prey and its conspecifics theoretically have gained a measure of protection from that particular predator. A noxious taste of a prey when bitten is similar to getting stung by a wasp; that is, the message is clear and instant. However, in the case of a predator that takes some time to swallow a prey (perhaps attempting to swallow it whole) and to commence digestion, the message may not be received until some time later, that is, post-ingestively. This involves a time-delay, and here the lesson can only be learned if the predator can associate the bad taste of the present with the past action of capturing and eating the prey. Whether learning occurs will depend upon several factors, chief of which is the time delay between the predator attacking and consuming the pufferfish, and the predator's later perception of the toxin during digestion. Photograph courtesy Anne Dupont, Florida.

NOTE lit. "toad toxin". The toxin, originally isolated and characterised from toads, is a nerve-impulse inhibitor and is highly potent

Sharpnose pufferfish
Canthigaster rostrata


  The cartoon series below shows the outcome of the interaction as mentioned above, and there are other possibilities.
cartoon 3 of a series of a grouper meetin a pufferfish cartoon 3 of a series of a grouper meetin a pufferfish cartoon 3 of a series of a grouper meetin a pufferfish cartoon 3 of a series of a grouper meetin a pufferfish
A naive predator encounters a pufferfish... ...eats it, either whole or in part... ...and later gets sick, and perhaps dies... ...and perhaps the pufferfish dies.

In fact, a number of different scenarios can be generated from this encounter, and the challenge for us here is to determine which of them, if any, is evolutionarily beneficial to the pufferfish. Five possibilities are shown in the cartoons below. Think about each in the context of the question posed, then read the explanations to follow.

NOTE that is, favouring the continued survival of pufferfishes, defined as the continuance of gene transfer to future generations

cartoon 1 of a series on evolution of pufferfish toxins cartoon 2 of a series on evolution of pufferfish toxins cartoon 3 of a series on evolution of pufferfish toxins cartoon 4 of a series on evolution of pufferfish toxins cartoon 5 of a series on evolution of pufferfish toxins
1. grouper gets sick but lives, pufferfish dies 2. grouper gets sick but lives, pufferfish lives 3. gouper gets sick and dies, pufferfish lives 4. grouper enjoys its meal, doesn't get sick, pufferfish dies 5. gouper gets sick and dies, pufferfish dies

Now, which is evolutionarily beneficial to the pufferfish?

1. Yes. A sickening meal can be a clear lesson for a predator not to attack another member of a certain noxious species. Note that avoidance of other pufferfishes must be learned by each predator and, in this scenario, each lesson is paid with a pufferfish life. So, for this to be advantageous for pufferfishes, there must be more pufferfishes born than predators, or older experienced predators must educate naive younger predators not to eat pufferfishes. While this is known for other vertebrates, it is not known for fishes.

2. Yes. Here, the pufferfish species benefits by the lesson presumably learned by the predator but without loss of a pufferfish life. The caveat is that the predator must take enough of a bite out of the pufferfish to get sick, but not enough to kill the pufferfish. This scenario is similar to #1 in terms of its evolutionary implications.

3. Yes. This is highly advantageous for the pufferfish species, especially as it assumes that the same protection will work equally well on other predatory species.

4. No. Here, a certain type of predator, groupers, are for some reason unaffected by pufferfish toxins and can attack them at will. This could set up a "see-saw" evolutionary battle, with each species evolving new and better defenses - on the one hand, selection for more elaborate and potent toxins aimed at a certain predatory species by the pufferfishes and, on the other, selection for greater chemical immunity by the predator. In this respect, an "ultimate" toxin would be one manufactured with least metabolic cost by the pufferfishes, but potent enough to kill a predator after a single, tiny, non-life-threatening nibble. This is not the case with pufferfishes, nor with any other vertebrate with toxic flesh.

5. No. This is an evolutionary dead-end, as neither species benefits.


According to the foregoing discussion, with their known content of potent neurotoxic tetrodotoxins in their mucus, skin, and liver, pufferfishes Canthigaster rostrata are unlikely to be a staple diet of any predator. But, in fact, certain birds and barracuda fishes are known to eat this species, and to this short list we may now add peacock flounders Bothus lunatus. In what must have been a fascinating experience, scientists in Exuma Cays, Bahamas witnessed a flounder attack and consume a pufferfish. In reaction to the attack the pufferfish inflated, which made it difficult for the prey to be swallowed by the flounder. In response, the flounder banged the prey several times on the sea bottom. Once the prey was ingested, the flounder, itself now slightly inflated, seemed none the worse for wear. Gochfeld & Olson 2009 Coral Reefs 28: 155.

photograph of a peacock flounder Bothus lunatus in the act of swallowing a pufferfish Canthigaster rostrataNOTE the authors describe the flounder’s banging behaviour somewhat anthropomorphically as “an apparent effort to daze or deflate its prey”. If true, it does raise the question as to whether this particular flounder had learned from experience how to handle a puffed-up pufferfish

NOTE what about post-ingestive effects of the pufferfish’s tetrodotoxin? The authors remark that they observed the flounder for “several minutes” after the event and saw no “adverse neurological responses” but, as the principal location of the toxins is in the liver, it may have taken more than just a fewl minutes for this organ to be digested. The following may be germane to this discussion. While analyses of tetrodotoxin levels in Caribbean pufferfishes seem not to have been done, those in related species in the Gulf of California show two orders of magnitude greater concentration in Sphoeroides spp. (represented in the Caribbean by bandtail and checkered puffers) than in Canthigaster sp. (Caribbean sharpnose puffers). In other words, sharpnose puffers may be potentially less toxic that other Caribbean pufferfish species. Nunez-Vazquez et al. 2000 Toxicon 38: 729.

Partially inflated sharpnose puffer Canthigaster rostrata
being ingested by a peacock flounder Bothus lunatus


photograph of a honeycomb cowfish Lactophrys polygonia
Several other Caribbean reef fishes possess tetrodotoxins in their flesh, including triggerfishes, porcupinefises, and some boxfishes, but not to the same extent as pufferfishes.

NOTE boxfishes, along with cowfishes, and trunkfishes are closely related in the Family Ostraciidae




Honeycomb cowfish Lactophrys polygonia 0.2X


photograph of mat tunicate Tridedemnum solidumDo any reef chordates other than fishes employ toxic flesh in defense? Studies have shown that organic extracts of certain tunicates when incorporated into artificial food-strips at natural concentrations can deter feeding by predatory reef fishes.

NOTE the Phylum Chordata includes animals such as fishes, reptiles, birds, and primates that possess a notochord (a supporting rod along the back), a dorsal tubular nerve cord, and pharyngeal gill slits for at least part of their life cycle. TUNICATES or sea squirts are sessile chordates in which only the larvae possess these typical chordate features




Mat tunicate Trididemnum solidum 2X


photograph of tadpole larva of a tunicateThe larvae of tunicates are apparently also distasteful to reef fishes through the presence of unpalatable chemicals. This protection allows comparatively large, well-provisioned (yolky) larvae to be released safely from the adults during daylight periods when they have the greatest chance of using light cues to select appropriate settlement sites. So even though they have the appearance of being tasty morsels to the many species of planktotrophic fishes over the reef, they may have at least some chemical protection.


Two stages of development of a tunicate: tadpole larva (below) and partially metamorphosed
(above). The tail is used for propulsion by the larva but, just after settlement, is
resorbed during metamorphosis to the adult. The larvae are non-feeding and, partly for
this reason, are extremely short-lived in the plankton (minutes to hours only) 100X

  Interestingly, only a few vertebrates, such as toads, pufferfishes, and some sculpins, have evolved chemical defenses in their eggs and embryos. In the most that haven't, predation on eggs and embryos may be high. The reason for the rarity of such defenses is not clear, but one suggestion is that such chemical defenses may be incompatible with the actively developing tissues in the young stages. Orians & Janzen 1974 Amer Nat 108: 581; Gladstone 1987 Copeia: 227.
Prevent being eaten if captured: poisonous secretions

photograph of seahorse dive leader for Biology of Caribbean Coral Reefs website photograph of a soapfish taken from a video

"Oh, who's this nondescript fish? It's a soapfish. Apparently it has noxious chemicals in its skin mucus. Watch what it does now. See, it flops over. That's the way you usually find them, under a rock, lying on their side, maybe to expose as much of their stinkyness as possible." - St. Maarten 2006

NOTE Rypticus saponaceus


photograph of a soapfish Rypticus saponaceus resting under an overhang
Several other common species of reef fish, such as trunkfishes, toadfishes, soles, and gobies also produce mucus containing distasteful bitter or stinging chemicals that are thought to repel predators. Much of a soapfish's daylight hours appears to be spent resting under overhangs.

NOTE the secretion contains a polypeptide that is slippery to the touch, foams up when disturbed, and is known to be toxic to guppies in physiological concentrations. Maretzki & del Castillo 1967 Toxicon 4 (4): 245.photograph of soapfish Rypticus saponaceus being cleaned by several gobies


Soapfish Rypticus saponaceus 0.2X

It is interesting
that cleaning gobies Gobiosoma sp. are able to do their cleaning with no effect from the saponin toxins


Prevent being eaten if captured: electrical shock


drawing of electric ray showing location of electric organs
Electric rays form an electrical field around themselves and produce an electric shock when touched. The shock is likely not for stunning the ray's primarily sand-dwelling invertebrate prey, but rather for defense against photograph of electric ray Narcine brasiliensispredators. The electrical field generated, however, may also allow for communication between conspecifics regarding size, sex, distance, and motivational state.

NOTE are electric rays dangerous to humans? A casual touch produces a shock that varies from mild discomfort to downright unpleasantness (40-220V depending upon species). The Brazilian ray shown here produces about 40V, which is certainly a "wake-up call" but not much more. In Roman times gout was thought to be cured by standing on a ray in shallow water until such time as the entire lower leg became numb. This apparently removed the pain and delayed its return. Now for a cure from the ray! Moller 1995 Electric fishes: History & behaviour. Chapman & Hall, London

Electric ray Narcine brasiliensis 0.3X The accompanying
drawing shows the location of the electric organs.
Shocks can be delivered from any part of the ray's body

hot button for take evasive action if spotted part of fish defenses hot button for prevent capture part of fish defenses hot button for prevent being eaten if captured part of fish defenses