Electric eels jump out of water to shock potential predators

Let’s start this summary of a new paper (by Kenneth Catnia in Proc. Nat. Acad. Sci. USA; reference and free download below) with a historical incident that helped motivate the work. This is how Catania’s paper begins:

In 1807, Alexander von Humboldt published his account of a battle between electric eels and horses. The stage for this event was set when Humboldt hired local fishermen to supply him with eels for research. Their method was to “fish with horses”. About 30 horses and mules were herded into a pool containing eels, which (according to Humboldt) emerged from the mud, swam to the surface, and attacked by pressing themselves against the horses while discharging. The fishermen kept the horses from escaping by surrounding the pool and climbing nearby trees with overhanging branches while crying out and waving reeds. Two horses drowned, and others stumbled from the pool and collapsed. Humboldt thought more horses would be killed, but the eels were exhausted before this happened. Five eels were then captured and Humboldt was able to conduct his experiments.

Here’s an illustration of that epic battle in which two horribly abused horses died so that Humboldt could satisfy his curiosity about electric eels. Ignore for a moment the photographs below the drawing:

F1.medium

(From paper): Fig. 1. Fishing with horses. (A) This illustration depicts the battle between eels and horses observed by Alexander von Humboldt in March 1800. It was published in 1843 as the frontispiece for The Naturalist Library, Ichthyology, Volume V, Part II, the Fishes of Guiana, authored by Robert H. Schomburgk, a friend and protégé of Humboldt’s . (B) Example of an eel leaping from the water to shock a simulated predator. LEDs are powered by the eel through a conductive carbon strip taped to the front of the plastic prop. See Movie S8.

The article continues:

This famous story has been illustrated and recounted many times (Fig. 1A). However, some have doubted its accuracy. Sachs suggested the story was “poetically transfigured,” Coates flatly considered it “tommyrot,” and Moller [and Catania] gently suggested Humboldt’s accounts were “tales.” The aggressive behavior of the eels, taking the offensive against horses, seems the most fantastic and questionable part of the story. Why would electric eels do this? No similar behavior has been reported since Humboldt’s  publication. Here I report that electric eels attack large, moving, partially submerged conductors by leaping from the water while pressing themselves against the threat and discharging high-voltage volleys (Fig. 1B). This behavior appears to be ubiquitous for comparatively large eels (over 60 cm).

But the story was not tommyrot, as Catania discovered as a byproduct of his other studies on predatory behavior of electric eels (Electrophorus electricus; note that electric eels aren’t eels, but bony fish). While watching them, he discovered that they can indeed jump out of the water if they’re disturbed by an intrusive object, and give a powerful shock to those objects.  The full-text version of the paper includes several movies showing this shocking behavior, but none is more compelling that this one (tweeted by science presenter Ziya Tong), in which a fake alligator head, containing LEDs to show any current, is lit up by a jumping eel:

In all of these videos, the eel leaps out of the water with its head and belly contacting the hapless intruder, and delivers a powerful shock.  This shock is more localized than if the eel stayed in the water where the electric impulse it emits is diluted throughout the water. When in contact with a large animal, the entire animal absorbs the shock, as shown in the diagram below (vertical arrow):

F2.large

(From the paper) Schematic of changes to the current path presumed to occur as an electric eel approaches and ascends a conductor. As the eel emerges from the water, two parallel current paths and corresponding resistances exist, representing a current divider. The eel’s high voltage is delivered in volleys of roughly 1-ms pulses, and eels do not modulate their total power output during a volley (see text).

It’s thus far more efficient for the eel to leave the water and deliver a direct shock. How strong is the shock? Very strong: here’s a diagram showing that it can exceed 200 volts (go here to read how it makes the shocks—a marvel of evolution):

Screen Shot 2016-06-09 at 12.43.30 PM

 

Some eels can deliver nearly 900 volts, and that’s enough to kill a person by stopping the heart or causing him to drown, though I haven’t found reports of any human being killed by an eel. But caiman have been killed when trying to nom an electric eel–you can see the video here.

I’ve already alluded to why the eels leave the water: they can deliver more voltage than if they remained in the water.The jumping behavior is almost certainly evolved rather than learned; it’s likely that eels that already could deliver shocks to stun their prey would be more likely to survive if they could jump out of the water to hurt potential predators.

Why did this behavior evolve? According to Catania, these eels often live in shallow water or even isolated pools that form during the dry season (electric eels are air-breathers and can survive in such situations). That makes them susceptible to predation by or injury from larger animals, so the shock, evolved to stun prey (which are swallowed whole), can be coopted—and intensified by selection—to protect the animal from being eaten itself.

The life of a naturalist interested in evolution is an exciting one, for fantastic stuff like this appears all the time. As the saying goes, “Evolution is cleverer than you are.” Who could have predicted that after all this time, Humboldt’s observation, dismissed as poppycock, would be substantiated?

You might have asked yourself why aren’t the eels stunned by their own shocks, especially in the water. As far as I know, this remains a mystery.

h/t: Ant

_________

Catania, K. C. 2016. Leaping eels electrify threats, supporting Humboldt’s account of a battle with horses. Proc. Nat. Acad. Sci. USA. P

68 Comments

  1. TomiP
    Posted June 9, 2016 at 1:24 pm | Permalink

    Real eels are bony fish as electric eel (But electric eel does not belong real eels).

  2. Mark Sturtevant
    Posted June 9, 2016 at 1:26 pm | Permalink

    Ok, if I ever visit the jungles of South America I knew i was not going to venture into the rivers because of piranhas. But now I am not even getting near the water, nosir, uh-uh.

    • Posted June 9, 2016 at 1:34 pm | Permalink

      Don’t forget the stingrays, caimans, and candiru fish…

      • Posted June 9, 2016 at 4:48 pm | Permalink

        From what I remember of reading Humboldt’s travel journal, despite surviving some extraordinary hardships close shaves with death, the only thing he really griped about was the mosquitoes.

      • Merilee
        Posted June 9, 2016 at 5:08 pm | Permalink

        Yeah, I’d worry about candirus. Most def do NOT pee in the water…

        • Mark Sturtevant
          Posted June 9, 2016 at 5:44 pm | Permalink

          Here it comes!!! 🐟

        • gravelinspector-Aidan
          Posted June 9, 2016 at 7:53 pm | Permalink

          FAR more entertaining is feeding a susceptible mind on horror stories of the urethra-ascending candiru (other names are available), then feeding them a mild diuretic, and sending them to hold “The Machine That Goes Ping” waist-deep in the river all day.
          It is definitely more entertaining than pulling the leeches off your skin, and marginally more ethically acceptable.
          Incidentally, Candiru may be an Amazonian legends, but that doesn’t even slow one down about transplanting the legend to Africa.
          Oh, the joys of green-hats!
          It slightly puzzles me why men are more cross-eyed (and cross-legged) over legends like Candiru urethrascens than women. After all, the surgical intervention for men (quoth QE1 : “off with his head!”) is a far more survivable than the equivalent female operation, even without anaesthetic or antisepsis.

          • infiniteimprobabilit
            Posted June 9, 2016 at 11:32 pm | Permalink

            If this is evolutionary behaviour, just what does the candiru get out of it?

            cr

            • Posted June 10, 2016 at 11:39 am | Permalink

              I think it is a gill parasite on bigger fish. So going up orifices is its thing.

              • infiniteimprobabilit
                Posted June 10, 2016 at 7:56 pm | Permalink

                OK, so the alarming behaviour in question is just a case of mistaken identity.

                In a way, that makes it worse. Getting nobbled by a stupid fish…

                cr

            • gravelinspector-Aidan
              Posted June 10, 2016 at 6:44 pm | Permalink

              The urethra-ascending behaviour in humans appears to be an urban legend. Still good for scaring the young’uns with, but legendary nonetheless.

              • infiniteimprobabilit
                Posted June 10, 2016 at 7:59 pm | Permalink

                It would be reassuring to think it was just a urban legend. Has anyone ever offered to dangle their, umm, dangly bits in a tank full of candirus to find out?

                (No, thought not…)

                cr

              • gravelinspector-Aidan
                Posted June 11, 2016 at 7:30 am | Permalink

                I look forward to reading your reports.
                And here is some light reading for your entertainment. I was actually visiting Munich at about the time this work was being done, and I didn’t see the expected smoking craters, so I’m less than surprised that at least one PhD candidate made it through to the end of the experimental programme.
                Anyway, good luck. I think, oat cakes and some of that fine blue cheese from Tain in preference to popcorn, but you get the message.

              • Posted June 10, 2016 at 9:51 pm | Permalink

                I wouldn’t be so quick to write it off. It’s true that there are no credible modern accounts of them going up men’s urethras. However there are a few credible accounts of it going up human vaginas. But with so much space available, it does not cause the problems it would cause if it went up penises:
                “…there has been a few credibly documented instances of them lodging in vaginas. However, in this case, while no doubt horrifying, it’s a fairly simple matter to get the fish out, such as in 1891 when Paul Le Cointe removed one from a vagina by simply pushing it forward slightly, turning it around, and pulling it out. You push forward and turn it around first so that its spines lie flat, rather than cause a bit of damage as they would if you just pulled it out backwards.”
                http://www.todayifoundout.com/index.php/2013/11/candiru-fish-dont-swim-stream-urine/

              • gravelinspector-Aidan
                Posted June 11, 2016 at 7:40 am | Permalink

                Lou, I suspect that you spent far too much f your youth reading Herriott’s “Vet” series of books. (which are full of passages like that involving him being armpit deep inside a cow).

  3. GBJames
    Posted June 9, 2016 at 1:26 pm | Permalink

    Wow. Remind me not to go taunting electric eels.

  4. Posted June 9, 2016 at 1:30 pm | Permalink

    When I lived in the Amazon the local Kichua people often accidentally caught electric eels on their fish hooks. They would have to paddle to shore and abandon their pole, line, and eel in the vegetation until the eel died, since they couldn’t take the eel off the hook without getting shocked.

    However, since an eel is basically a giant battery with a positive and negative end, I realized they could be handled without ill effect if the eel is grabbed through a continuous strip of conductive aluminum foil. All the current goes harmlessly through the foil. Using this technique I was able to remove even the largest eels from my friends’ hooks with no problem.

    • GBJames
      Posted June 9, 2016 at 1:36 pm | Permalink

      Can you expand on that? Do you mean that if you wrap the eel in foil you can handle it? (I think you mean something else but …?)

      • Posted June 9, 2016 at 1:49 pm | Permalink

        Yes, it shorts out the charge (rather than your body shorting the charge).

        The electric eel has a similar electrical signature to an defibrillator: High voltage, low current.

        If it had high current, it would vaporize the foil; but since there’s little current, the foil remains.

      • Posted June 9, 2016 at 1:55 pm | Permalink

        Current always follows Ohm’s law; if the path from one pole to the other includes a good conductor and a resistor mounted in parallel, virtually all of the current will go through the conductor. I am the resistor, the foil is the good conductor. I don’t wrap the eels in foil, I just lay a long sheet of foil over the eel and then grab the eel through the foil. Equivalently one could use foil gloves connected by a wire. But that would have been hard to make in the field.

        • Posted June 9, 2016 at 2:00 pm | Permalink

          We are fine conductors as well; but a sheet of foil, in contact with the skin of the eel is an even finer one.

          Minimum energy, shortest path (energy-wise; often geometrically as well).

          • Kevin
            Posted June 9, 2016 at 3:29 pm | Permalink

            Our conduction varies significantly. A dry hand with little oil is several MOhms. Sweaty skin can be less than an Ohm.

            • Torbjörn Larsson
              Posted June 9, 2016 at 4:15 pm | Permalink

              I don’t think we have evolved to survive lightning. But as long as we don’t get down on all fours hominins have a decent chance to get away a direct conduction event with some burns. (Sometimes superficial, sometimes lethal.) The same happens with trees, where the growth layer in the bark stands for the low resistance path.

              • Kevin
                Posted June 9, 2016 at 5:43 pm | Permalink

                Actually being fully drenched can help with lightning. First, lightening has plethora of frequencies, most in GHz and the skin depth is tiny. Wet clothes have been known to be blown right off victims with minor burns and ear damage, but heartbeat intact.

            • Posted June 10, 2016 at 8:38 am | Permalink

              Yes, of course the interface resistance is going to vary a lot. But the body conducts very well.

        • gravelinspector-Aidan
          Posted June 9, 2016 at 7:59 pm | Permalink

          Considering that our local maggot-smell-ologist wrote a book about “the Resistance”, I can only add “Ohm My Ceiling Cat!”

    • gravelinspector-Aidan
      Posted June 9, 2016 at 7:56 pm | Permalink

      I realized they could be handled without ill effect if the eel is grabbed through a continuous strip of conductive aluminum foil.

      Ready for the barbecue?

  5. Gamall
    Posted June 9, 2016 at 1:30 pm | Permalink

    “Some eels can deliver nearly 900 volts”

    Anybody can take that, is the current is weak … You’d also need the amps for that figure to become meaningful.

    • Posted June 9, 2016 at 1:58 pm | Permalink

      This is not correct.

      Defibrillators deliver a high voltage shock over a few milliseconds (~10ms) (little current). Very sufficient to either stop or start your heart. (I design defibrillators.)

      Check: EN 45502-2-2:2008 clause 20.2

    • Kevin
      Posted June 9, 2016 at 3:31 pm | Permalink

      Definitely not correct. About 30mA across the heart for >50V and if you have no AED in sight, you are a goner even if they pump you CPR for as long as they like…brain will be toast.

      • Gamall
        Posted June 9, 2016 at 10:58 pm | Permalink

        To rightly say “This is not correct” you’d need to show that some given voltage is harmful *no matter the amps*.

        Are your 50 V still harmful for 0.0001 mA ? Would it be just as harmful for 2A ? If that’s the case, then I retract my comment.

        Otherwise, I maintain that for every voltage, there is a current weak enough that you can take it harmlessly. (Given that I took something like 100000 volts once at some attraction geared at demonstrating that, I’m reasonably confident that this is true. Then it is possible that I’m misremembering this and other electricity-related things.)

        • infiniteimprobabilit
          Posted June 9, 2016 at 11:59 pm | Permalink

          I got a high-voltage tiny-current shock once off the outer insulator of a coaxial TV aerial cable. I worked out why – this cable was about 30 feet long at right angles to the 110kV power lines that run above our house.

          A digital voltmeter showed that the voltage was up in the kilovolts, but the current setting wouldn’t even register.

          cr

          • wonderer
            Posted June 10, 2016 at 8:01 am | Permalink

            Yes, in areas where there is low humidity, it is easy to get a static discharge shock in the thousands of volts just from rubbing the soles of your shoes on carpet and touching something grounded. (Or your little brother.)

            The electrostatic discharge (ESD) tester at work has an output setting of 16 KV, and I’ve deliberately shocked myself at the 16KV setting. (Needless to say, if you don’t understand the physics…)

            I’ve also been connected to 480 Vac from hand to hand and been unable to disconnect myself or even scream. I was able to contemplate the situation for a time, and my thinking did not seem to be any more scrambled by the alien electrical signals than by the panic hormones.

            I was fortunately standing at the time, and fell away breaking the contact.

            • Posted June 10, 2016 at 8:43 am | Permalink

              “480 Vac from hand to hand and been unable to disconnect myself or even scream.”

              You were lucky to survive that (as I’m sure you know). 220V (or 110V) is more than enough to kill you, given the right vector.

              • wonderer
                Posted June 10, 2016 at 11:41 am | Permalink

                Yeah, I was expecting to die, and had time to think, “I can’t let go. I can’t make a sound. This sucks. I had other plans. This really sucks for my kids.”, before falling away.

                It was only hours or day later that it occurred to me that the ideas of God, heaven, and hell hadn’t even entered my mind. Having grown up a devout preacher’s kid, it was kind of a big thing to me to realize that my thinking had been utterly god-free in such dire circumstance.

                “No atheists in foxholes.” Hmmph.

        • Posted June 10, 2016 at 8:41 am | Permalink

          At my company, we have tons of data (which I can’t share) that shows that a few 10s of mA is fully sufficient to stop/start your heart at voltages down to a couple of hundred V.

          The vector through the body is critical.

          And the energy delivered is also critical.

          Electric eels clearly deliver a similar shock to defibrillators and therefore can easily start/stop the heart.

          But: Point taken that voltage alone is not a sufficient formulation of the problem.

          • infiniteimprobabilit
            Posted June 10, 2016 at 7:02 pm | Permalink

            My father used to make studio flash units for commercial photographers. Boxes the size of small suitcases packed with electrolytic capacitors the size of drink cans, at 1000 volts, and up in the millifarads (not sure if they ever reached farads). Damn things used to scare the crap out of me, I’m a coward where energetic electricity is concerned. (I prefer it at 12v or 5v where I’m more danger to the circuit than it is to me…)

            cr

  6. Posted June 9, 2016 at 2:05 pm | Permalink

    A 900-volt shock would REALLY hurt. (I know from hard, personal experience with (MIS-)handling energized defibrillators.) Kind of like be wholloped with a baseball (or cricket) bat.

    However, it would not be likely to kill you unless the vector of the electricity through your body ran through your heart.

    Don’t lie down next to rivers in S. America (or otherwise get intimate with an electric eel). 🙂

    • Torbjörn Larsson
      Posted June 9, 2016 at 4:19 pm | Permalink

      A lab mate that didn’t know of the floating earth of oscilloscopes (and didn’t ask me) touched one that was hooked up to negative 1.1 kV… His muscle spasms threw him clear and against the next piece of plasma equipment.

      Some learning events you will remember better than others.

      • Torbjörn Larsson
        Posted June 9, 2016 at 4:23 pm | Permalink

        Sorry, *he* hooked the oscilloscope in his own equipment, I should add. Great experiment builder, not so great on measurements…

      • Merilee
        Posted June 9, 2016 at 5:06 pm | Permalink

        Earth=ground?

        • Kevin
          Posted June 9, 2016 at 5:46 pm | Permalink

          Typically.

      • Kevin
        Posted June 9, 2016 at 5:46 pm | Permalink

        A saw a lab mate touch a ~1.2kV piezo stack (low, low current <micro amps) but he still ripped the whole bloody grating off with one hand. Never thought he had it in him. 🙂

      • gravelinspector-Aidan
        Posted June 9, 2016 at 8:04 pm | Permalink

        Some learning events you will remember better than others

        Having just had another audiogram as part of a medical, “Pardon?”
        Halides of nitrogen are educational. Pretty much all of them.

        • Posted June 10, 2016 at 11:29 am | Permalink

          Halides of nitrogen are scary.

          I learned a few years back that nitrotriiodide seems to involve the triiodide ion, rather than three iodide ions. That makes it sound worse …

          • gravelinspector-Aidan
            Posted June 10, 2016 at 7:19 pm | Permalink

            The bond angle of sp3 orbitals in carbon is 106.something degrees, so you can’t form 4 carbon atoms into a square, because the strain is too high.
            And you certainly can’t make a cube of carbon atoms, with 90 degree bod angles, because that would be just too strained.
            And as for the oct-nitro derivative …
            “In The Pipeline” gets enough mentions here. Because it deserves them. This is one of my eye-watering favourites.

            • gravelinspector-Aidan
              Posted June 11, 2016 at 9:09 am | Permalink

              What’s that you say, Lassie?
              P3N21 down the well?
              Yes, Lassie, you go and deal with that ; I’ll just finish reading the Central European Journal of Energetic Materials (ISSN 1733-7178) here behind this reinforced concrete anti-tank defence, because it sounds a real “Bang Whizz” publication.

  7. Macha
    Posted June 9, 2016 at 2:26 pm | Permalink

    radio amateurs (hams) would say ..

    “It’s volts that jolts, but it’s mils that kills”

    So, how many mA do these creatures deliver?

    • Posted June 9, 2016 at 2:49 pm | Permalink

      Don’t know for sure but likely 10-ish. (several tens of Joules).

      • gravelinspector-Aidan
        Posted June 9, 2016 at 8:06 pm | Permalink

        So, energetically comparable with being hit by a …
        Introducing, stage right, the long quiet star of stage and screen …
        … a rock in a sock!

    • Kevin
      Posted June 9, 2016 at 5:54 pm | Permalink

      Energy is proportional to current. Energy is work for the organism. My guess is they deliver what they need. If my cat got a few microamps from a mouse at 900V, she would be a quick member of the “catch and release society” rather than a fellow of the “nom-a-mouse-a-day-society”.

      • phil
        Posted June 9, 2016 at 9:09 pm | Permalink

        “Energy is proportional to current.”

        That’s not entirely accurate. Energy is the integral of power, and power is the product of voltage, current, and power factor (which cane be a complicated, as distinct from complex, quantity).

        The issue of what constitutes a fatal shock, is not simply one of current or voltage, but has to consider the impedance and the broader circuit. In the human body all the current does not flow down a single well defined path, but will be distributed down every available path.

        However for relatively low currents it is probably only the current that flows through the heart or brain that is likely to prove fatal (if at all). To deliver a dangerous current to the heart (for example) from outside the body requires that enough current is delivered to service all the current paths so that the heart current reaches the danger level. For this reason I expect that external defibrillators must deliver more current than implanted defibrillators. External defibrillators also have to overcome skin resistance, so I expect their output voltage must also be greater.

        I recall reading that currents from our mains voltage (240 VAC) of about 100mA to 200mA are considered to be almost certainly fatal (assuming a significant portion goes through the heart, and not, say, just along one arm), whereas higher currents are likely to cause the heart to stop completely whereupon it may be restarted with a defib.

        I have an implanted defibrillator, and it has fired several times (it is not at all pleasant). It delivers about 350 joules from about 850V (IIRC).

        • Posted June 10, 2016 at 8:52 am | Permalink

          However for relatively low currents it is probably only the current that flows through the heart or brain that is likely to prove fatal (if at all). To deliver a dangerous current to the heart (for example) from outside the body requires that enough current is delivered to service all the current paths so that the heart current reaches the danger level. For this reason I expect that external defibrillators must deliver more current than implanted defibrillators. External defibrillators also have to overcome skin resistance, so I expect their output voltage must also be greater.

          This is correct. Well-played.

          Implantable defibs (the kind I design) deliver about 600V-800V and about 30J (or less, as programmed) at the heart tissue, to which they are connected by excellent conductors (leads, think expensive, highly reliable insulated wires) that have superbly conductive electrodes at the business ends.

          (They must be superbly conductive because one of the their other duties is often to sense the electrical activity of the heart. Often only because some electrodes are shock-only.)

          An external defib will typically deliver 3000V and something more like 100-200J (at the paddles). This is to ensure an always sufficient energy at the heart to convert any patient who can be converted (some can’t).

          Typical total impedance for the circuit through the heart (including all components) is in the range of 500 to 5000 Ohm.

        • Posted June 10, 2016 at 8:55 am | Permalink

          Your AICD probably delivers 35J (at the heart). That would be typical.

          350J would send you into orbit! 🙂

          • phil
            Posted June 10, 2016 at 8:11 pm | Permalink

            I was relying on memory, which isn’t always reliable. I think the shocks are more likely to send me to the floor.

      • Posted June 10, 2016 at 8:44 am | Permalink

        “My guess is they deliver what they need.”

        Exactly.

  8. rickflick
    Posted June 9, 2016 at 3:06 pm | Permalink

    Yikes! Quite frightening to observe and contemplate. Evolution has an impressive repertoire of killing methods – from deadly poison frogs, to snakes that compress your lungs, not to mention the usual suspects: tooth and claw. After this, I don’t feel like leaving the house today, and I’m in NY!

  9. Torbjörn Larsson
    Posted June 9, 2016 at 4:27 pm | Permalink

    A potentially useful behavior.

    • Kevin
      Posted June 9, 2016 at 5:48 pm | Permalink

      Yeah, current events may be valuable.

      • rickflick
        Posted June 9, 2016 at 8:23 pm | Permalink

        Positively electrifying.

  10. Merilee
    Posted June 9, 2016 at 5:04 pm | Permalink

    Fascinating ( but poor horsies)

    • gravelinspector-Aidan
      Posted June 9, 2016 at 8:10 pm | Permalink

      These days one would use post-grad students? “Pour encourager les undergrads.”

      • phil
        Posted June 9, 2016 at 9:16 pm | Permalink

        Honours students or undergrads are probably cheaper, and are more plentiful. I always try to stand behind a student when an experiment is powered up for the first time. 😉

        If you think that’s nasty and irresponsible, consider that the greatest threat to students is from themselves, followed by their supervisors.

  11. Posted June 10, 2016 at 11:30 am | Permalink

    Wow, lends new meaning to “torpedo fish”.

  12. Posted June 10, 2016 at 3:10 pm | Permalink

    Aw! You didn’t use the click-bait headline.

    /@


%d bloggers like this: