The amazing mimicry of frogfish

I have a penchant for cases of mimicry, not only because they served as some of the earliest evidence for natural selection in Darwin’s time, but also because they show how far natural selection can achieve “perfection”—that is, how far do developmental and physical constraints prevent the evolution of an “optimum phenotype.” The answer is that constraints don’t matter much.

There are few cases in nature where one can judge how “optimum” an adaptation is, and mimicry is one of them. (Sex ratio is another.) What it shows, as this post demonstrates, is that it can be remarkably precise; that is, natural selection is pretty good at molding animals (and some plants) to hide their true nature by evolving to resemble either another organism or their environment. The resemblance can be astonishingly precise.

Finally, many examples of mimicry are simply unexpected, cool, and stunning. The first one below was sent to me by Matthew Cobb who got it from a tweet by M.J. Walker from the Blue Planet Society The photo is by Andrew Taylor.

One of these animals is a frogfish; the other is a sponge (yes, sponges are animals).  If you look closely you can see which is which, but it may not be so easy for a predator or a prey item. (Frogfish are almost all predators.)

Frogfish, sometimes known as “anglerfish” are in the order Lophiformes and the family Antennariidae; 47 species are recognized. 

Wikipedia has a good section on frogfish mimicry. I’ve reproduced it in the indented parts below, and inserted some pictures of different species of frogfish:

The unusual appearance of the frogfish is designed to conceal it from predators and sometimes to mimic a potential meal to its prey. In ethology, the study of animal behavior, this is known as aggressive mimicry. Their unusual shape, color, and skin textures disguise frogfish. Some resemble stones or coral while others imitate sponges, or sea squirts with dark splotches instead of holes. In 2005, a species was discovered, the striated frogfish, that mimics a sea urchin while the sargassumfish is colored to blend in with the surrounding sargassum.

Here’s the one, Antennarius striatus, that’s supposed to mimic a sea urchin:

Some frogfish are covered withalgae or hydrozoa. Their camouflage can be so perfect, that sea slugs have been known to crawl over the fish without recognizing them.

Here’s one that looks like an algae-covered rock:

Here’s another that looks like a sponge, hiding in a sponge:

For the scaleless and unprotected frogfish, the camouflage is an important defense against predators. Some frogfish can also inflate themselves, like pufferfish, by sucking in water in a threat display. In aquariums and in nature, frogfish have been observed, when flushed from their hiding spots and clearly visible, to be attacked by clownfish,damselfish, and wrasse, and in aquariums, to be killed.

Many frogfish can change their color. The light colors are generally yellows or yellow-browns while the darker are green, black, or dark red. They usually appear with the lighter color, but the change can last anywhere from a few days to several weeks. It is unknown what triggers the change.

If your appetite for mimetic frogfish isn’t sated, there’s a whole site devoted to them,, which has a great page showing many mimetic animals.

To show how the mimicry works, here’s a sponge-mimicking frogfish nomming a cardinal fish. It’s fast!

Finally, a short clip of a frogfish feeding. It was filmed at 1000 frames per second and played at 10 frames per second, so this whole 17-second video represents 0.17 seconds in real time. Note how opening the mouth creates a suction that draws the prey in:


  1. Alex Shuffell
    Posted November 29, 2012 at 5:25 am | Permalink

    Attacked by clownfish?! Having no scales really makes you vulnerable. I’ve always been suspicious of clownfish but I never expected them to be cruel, maybe it was their aquarium environment and they were trying to impress someone. In the first video there is a link to a frogfish eating a lionfish, those guys are fast.

    Thanks for sharing this, I’ve never heard of them before. I think I’m in love now.

    • gravelinspector
      Posted November 29, 2012 at 12:08 pm | Permalink

      Clownfish are very, very territorial. Which makes them easy to photograph, because they will keep on trying to drive away the invading camera.
      Which is why I’ve got some quite nice photos of “Nemo” from when I was snorkelling off the Tanzania coast. I don’t have super lung capacity, but even I could get them to rise to me on one lungful.

  2. Posted November 29, 2012 at 6:09 am | Permalink

    It’s now quite clear what Stephen Hillenburg was trying to accomplish.

  3. Stephen Barnard
    Posted November 29, 2012 at 6:35 am | Permalink

    Another example of mimicry is this website’s imitation of a blog. 🙂

  4. Karl Heinz
    Posted November 29, 2012 at 6:45 am | Permalink

    Probably old news here, but Science Friday some time ago had a nice little video on octopus camo:

  5. Posted November 29, 2012 at 7:47 am | Permalink

    Ha! You so got fooled.

    There is not a single photo of a frogfish on this page.

    There are, however, several photos of a mimic octopus.

    And, yes, before anybody asks: it’s the same octopus in each photo.



  6. Posted November 29, 2012 at 8:34 am | Permalink

    Shared/curated at G+ via #scienceeveryday. Thanks for a juicy science post.

  7. Paul S
    Posted November 29, 2012 at 9:02 am | Permalink

    Cool photos, unfortunately at a depth of 100M I don’t think I’ll get to see one in person. That’s a bit beyond my abilities.

    • gravelinspector
      Posted November 29, 2012 at 12:14 pm | Permalink

      I don’t see where you got the 100m from.
      Last year at a dive show I watched a presentation from an “in work” documentary on the weird and wonderful sea life of Lembeh in Indonesia. (They have probably finished it now, but I forget who was due to be showing it.) Frogfish were by no means the only stars, but they were popular targets.
      Typical filming depth were around 20m. Very definitely within amateur range.

    • Posted November 29, 2012 at 4:27 pm | Permalink

      I managed to see one in the Caribbean & I was only about 10-15m down. Apparently, the ones there (maybe all of them, I don’t know) find a spot of coral to attach to and stay there for months at a time.

      A person I was diving with (divemaster extraordinaire, with thousands and thousands of dives under his belt) screamed at me with exasperation when I told him I had seen a frogfish. It had been his #1 desire to see one of those things in the wild.

      I came upon it totally by chance (although I knew what to look for… the eyes, and in this case, a lure growing out of the forehead). I like staying motionless and waiting for the little critters to come out, so I spend a lot of time looking at really small stuff. In this occasion I was admiring a stand of yellow-green sponge growing off a knob of coral, looking for things like sea spiders and teeny swarms of juvenile fish, when I noticed the eyes on one of the sponges, and a telltale lure waving back and forth.

      • gravelinspector
        Posted November 29, 2012 at 7:03 pm | Permalink

        Hanging motionless in the water, not breathing more than necessary, waiting for the following diver to come along … really scared the wombats out of “Parrot”, the following diver.
        But I was just chilling (water temperature about 5C) and taking in the pretty patterns that my exhaust bubbles made against the cave roof. No reason for a “re-born” event.

  8. RFW
    Posted November 29, 2012 at 9:26 am | Permalink

    “constraints don’t matter much”

    Darwin comments in OtOoS that plant and animal breeders of his day viewed organisms as being plastic, capable of being molded in almost any way.

    However, there are certainly constraints. Though I know that it comes as a grave disappointment to PZ over at Pharyngula, it’s certainly not possible to breed a cat with tentacles instead of a tail or claws. At least, not without genetic engineering.

    Question about the taxonomic status of frogfish: why are so many of them in one zoological order? Have they been shown to be related, via anatomy or DNA or otherwise, or is that order in fact paraphyletic? Or to put this another way, is the order Antennariidae based on characteristics other than a talent for mimicry?

    • John Scanlon, FCD
      Posted December 2, 2012 at 8:58 am | Permalink

      Antennariidae is a ‘family’, not an ‘order’ (only animal family names, and all of them, end in -idae).
      Note that Anglerfish comprise numerous families in the order Lophiiformes; the Wikipedia article on ‘Anglerfishes’ is worth a look.
      There’s a recent molecular phylogenetic analysis of frogfish (, may be worth a look if you can access a copy (try the author’s email if it’s not open access).
      The antennariids featured in the pics’n’vids above look a lot like their relatives the Handfishes (Brachionichthyidae) so I think that ‘looking like a sponge’ may be the ancestral state for a more inclusive clade.

  9. Gregory Kusnick
    Posted November 29, 2012 at 2:02 pm | Permalink

    When I read the headline about “The amazing mimicry of frogfish”, I expected to see fish mimicking frogs, or maybe frogs mimicking fish. But what we have instead is fish mimicking stones, coral, sponges, sea squirts, sea urchins — everything, apparently, except frogs.

    So why are they called frogfish?

    • JLMelero
      Posted November 29, 2012 at 4:01 pm | Permalink

      I think it’s because of the mouth and general stance.

  10. marksolock
    Posted November 29, 2012 at 10:01 pm | Permalink

    Reblogged this on Mark Solock Blog.

  11. Craig
    Posted November 29, 2012 at 11:36 pm | Permalink

    “The unusual appearance of the frogfish is designed to conceal it from predators and sometimes to mimic a potential meal to its prey.”

    Something that always p’s me off when people/shows/book/whatever discuss adaptations in organisms is when they say ‘this adaptation is designed to …’.

    An adaptation is not DESIGNED to do anything, yet this phrase is often trotted out, even by respected naturalists and academics! It’s that sort of loose language that confuses the layperson about the basic principles of evolutionary change and adaptation, and there should be no place for it.

    • Stephen Barnard
      Posted November 30, 2012 at 4:21 am | Permalink

      I agree with your objection, but I can’t think of any suitable replacement phrase. Design implies intent. There should be another word that carries the meaning of “design” without the baggage of “intent”, but I can’t think of one. “Adapted” doesn’t do the job for me. Perhaps you have a suggestion?

      • Ichthyic
        Posted December 5, 2012 at 2:10 am | Permalink


        of course, one could just as well argue teleology for any of these.

        the point is, the baggage is just that… baggage. to remove it, one has to stress it is not necessary.

        this is our job, yes?

    • John Scanlon, FCD
      Posted December 2, 2012 at 9:10 am | Permalink

      Mimicry is the closest that nature comes to intentional ‘design’ in the sense people understand; and unless you believe in magical free will, ‘intentional’ design actually occurs through random generation and automatic selection (template-matching) of neuronal processes, so I don’t see such a big difference.

      Don’t forget that natural selection is much harder work than picking a few chips off a colour chart or twiddling a few ontogenetic knobs, it involves large numbers of lives and deaths and the constant operation of millions of intelligent agents. Don’t ever assume it’s less than design.

      • Craig
        Posted December 2, 2012 at 3:13 pm | Permalink

        Yes – the process can be thought of as design in some sense.

        But the phrase ‘designed to’ implies that the end result was a goal of the process from the outset. Sure, the fact that these fish mimic their surroundings was a RESULT of what could be described as a ‘design’ process, but it was certainly never the goal of ‘intelligent agents’ such as predators to design a well-camouflaged fish.

  12. KN
    Posted December 3, 2012 at 3:24 am | Permalink

    The interesting thing is that they can’t see the lower part of their body, or their behinds, but those areas also mimic. So do their skin “see”? Sort of like the sensory substitution cases? Has anyone looked at this?

    • Gregory Kusnick
      Posted December 3, 2012 at 10:50 am | Permalink

      They don’t need to see their own mimicry because they’re not the ones it’s meant to fool. If predators can see it, that’s all that matters.

      The frogfish aren’t sitting there thinking “Does my tail look enough like a sponge? What should I do about it?” They don’t have to think about it at all; natural selection takes care of it by weeding out the ones whose mimicry isn’t good enough.

      • KN
        Posted December 4, 2012 at 12:04 am | Permalink

        Well, I agree that natural selection could lead to good mimicry. And also that they don’t need to see their mimicry.

        What I wanted to know was the mechanisms involved, i.e. how do their skin manage to mimic? What is their brain telling the skin? What proteins are involved? What is the feedback loop, how does the system finetune, stop, start the mimicking process?

        • Gregory Kusnick
          Posted December 4, 2012 at 12:30 am | Permalink

          I think you’re misunderstanding what’s going on here. These fish aren’t chameleons. Their skin doesn’t dynamically change color to match their surroundings. They just grow up looking like sponges, or coral, or whatever, depending on species, in the same way that stick insects look like sticks.

          • Ichthyic
            Posted December 5, 2012 at 2:15 am | Permalink

            OTOH, the question is still worth answering if you take it to species that DO dynamically modify their colors or skin patterns to match backgrounds.

            In that case, yes, there is a lot of evidence indicating that visual systems are involved in regulating skin color in animals that can change their color.

            a classic paper was done to test this on flatfish, using different colors and types of substrates (even checkerboard). Let me see if I can find the reference for you.

  13. KN
    Posted December 5, 2012 at 12:48 am | Permalink

    Sorry, that was not clear from the text. Not sure whether that is entirely true, though:

    “Many frogfish can change their color. The light colors are generally yellows or yellow-browns while the darker are green, black, or dark red. They usually appear with the lighter color, but the change can last anywhere from a few days to several weeks. It is unknown what triggers the change.”

    Also, even if they grow up like that, the mechanism question still holds. What mechanism allows them to grow mimicking the surrounding? The system needs to perceive the surrounding to mimic it, and it needs to change the growth pattern in response to what is perceived, so that the resemblance/merging is good. How is this achieved?

    The chameleon example is interesting, because the mechanism of their mimicry is (somewhat) known:

    • Gregory Kusnick
      Posted December 5, 2012 at 1:08 am | Permalink

      I’m not an expert on frogfish, so I can’t help you with the specifics on that.

      But in general, mimicry does not require the organism or its developmental process to perceive the environment in order to mimic it. Natural selection does the job by rewarding those individuals who by chance happen to blend in, and punishing those who stand out. The perception is done by predators, not by the mimics themselves.

      • Ichthyic
        Posted December 5, 2012 at 2:36 am | Permalink

        I think he gets that, what he’s looking for is the specific physiological mechanism that frogfish utilize to match their surroundings.

        I linked above (scroll up a couple of comments) to some papers above looking at similar systems in flatfish, but I can’t recall any looking at frogfish specifically, and a quick glance through google scholar turns up some hints, but again, nothing specific past about 1970.

        it likely would be a worthy Masters or PhD project.

        • KN
          Posted December 5, 2012 at 5:56 am | Permalink

          Thank you Ichthyic. Actually, I have read the Ramachandran paper, and that is what provoked my question. Even in that case, the fish cannot see how the skin is responding, so how do they, or their brain, “know” that the skin has matched the environment, and stop tweaking the proteins?

          I suspect there is some kind of local sensing/perceiving going on at the level of the skin, maybe some kind of reflectance/wavelength tracking. I would think identifying this mechanism would give you a Nature paper, not to mention the Master’s/Ph.D.! 🙂

          BTW, I have an experiment in mind to test this idea, send me a mail ( if you are interested.

          • Ichthyic
            Posted December 6, 2012 at 10:43 pm | Permalink

            Even in that case, the fish cannot see how the skin is responding, so how do they, or their brain, “know” that the skin has matched the environment, and stop tweaking the proteins?

            well, going back up a level, they might not need to actually have any feedback perception. It could just as easily be:

            -sensory apparatus detects pattern
            -brain sends signal to chromatophores/melanophores
            -skin responds

            selection would favor a response that best matched the background.

            no need for direct feedback.

            that said, I think I understand that you are looking for a direct feedback mechanism, and if so then I would recommend you peruse the literature on this subject in cephalopods. Since they also use color change for communication, there most certainly is a feedback mechanism involved.

            • Ichthyic
              Posted December 6, 2012 at 10:44 pm | Permalink

              of course, one could also make the case that if there is sufficient selection pressure to “match” the background as closely as possible, then indeed it’s just as likely that it would favor a feedback mechanism to accomplish that task.

          • Ichthyic
            Posted December 6, 2012 at 10:46 pm | Permalink

            BTW, I have an experiment in mind to test this idea

            no time, I’m afraid you’re on your own.


            -find out who is working in this field by doing lit surveys on the subject at your closest uni library
            -contact them and ask for reprints, and ask questions after you have read their work.

            you never know where it might lead!

3 Trackbacks/Pingbacks

  1. […] one of 47 species of the weird, sponge-like frogfish. Here are all 47 species. Frogfish are really good at blending in with their environment, particularly with the sponges that they live on. Here’s […]

  2. […] on the ocean floor! They are totally innocuous. Just a pair of wild and crazy sponges! Oh, except the one on the left is a frogfish and if you’re a smaller fish, it will EAT your ass before you know what hit […]

  3. […] See on […]

%d bloggers like this: