We humans are pretty smug about our large brains and sophisticated ways. But if there is one thing I have learned as a biologist, it is to never underestimate the abilities of animals that most people consider primitive and simple-minded. Usually mammals teach me this lesson. But recently the complexity of the behaviors I observed in a peculiar reptile known as the tentacled snake made my jaw drop in amazement.
The tentacled snake, Erpeton tentaculatus, is a fully aquatic serpent native to Thailand, Cambodia and South Vietnam. A relatively small snake (adults are about two feet long), it gives birth to live young and feeds exclusively on fish. The animal’s name refers to its most distinctive trait: the pair of tentacles that project from the sides of the snout. I first became interested in these creatures around a decade ago on a nostalgic visit to the National Zoo in Washington, D.C., where I had worked summers as an undergraduate. Walking through the reptile house, I came across an aquarium thick with vegetation where a tentacled snake was lying in wait. It hung motionless in the water trying hard to look like a stick, its body curved into the characteristic J shape that the snakes adopt when hunting.
As I watched the snake, I wondered what the tentacles were for. No other snake has anything quite like them. Because these animals feed on fish, it stood to reason that the tentacles might be fish detectors of some kind. But when I returned to my lab at Vanderbilt University and searched the scientific literature, I found that although tentacle theories, including this one, had been proposed, no one had tested them experimentally. So I set out to solve the mystery of the snake’s bizarre appendages once and for all.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
In my quest to discern the true purpose of the tentacles, I discovered that this animal was even more interesting than I had realized. It turns out that the tentacled snake uses an array of remarkably advanced attack strategies to capture prey. Furthermore, even newborns of this species possess these skills, revealing a dramatic example of nature, rather than nurture, molding behavior.
In the Blink of an Eye
Before i could test the theory that the tentacles function as fish detectors, I first had to carefully observe the snake’s hunting behavior. But watching snakes hunt is not as simple as it might sound. Tentacled snakes strike with incredible speed, and fish are equally fast. The entire contest between snake and fish plays out in about 40 milliseconds, or 1/25th of a second. To see these events, I recorded strike after strike for a number of snakes using a high-speed camera shooting at 500 to 2,000 frames per second and then played the video back in slow motion. As I watched the attacks, I noticed something very strange: seemingly suicidal fish.
In many instances, fish turned toward the approaching jaws of the striking snake, sometimes swimming straight into the snake’s mouth. This made no sense. Fish are a top menu item for many predators, and as a consequence, they are expert escape artists, having evolved rapid neural circuits and correspondingly swift behaviors to sense and evade enemies. When they detect the sounds and water motion generated by a predator, they can begin their flight to safety in just six to seven milliseconds—less than 1/150th of a second. This escape response, called a C-start because it starts with a C-shaped bend of the fish’s body, is supposed to propel the fish away from a predator on the hunt. Why, then, were fish moving toward the tentacled snake’s mouth?
The answer, I found, has to do with the unusual J-shaped hunting posture of the snakes, which forms a trap of sorts. These reptiles prefer to go after fish that have entered the concave area of the J-shaped region formed by their head and upper body. Careful examination of the slow-motion video revealed that just before attacking, the snake moved a portion of its body on the side of the fish farthest from the snake’s head, startling the fish toward the predator’s open mouth. When I filmed the strikes at 2,000 frames per second and simultaneously recorded sounds in the aquarium with an underwater microphone, I determined that the movement of the snake’s body just before the strike creates a propagating pressure wave strong enough to startle a fish.
The snake’s feint strategy is particularly insidious, because it taps into the neural circuitry that usually works in favor of fish. Fish have a pair of giant cells in their brain, one on each side, called Mauthner neurons. The neurons’ signal-carrying extensions, called axons, cross over to the opposite side of the body. A race between these two fast-conducting neurons determines the direction of the escape response. When a sound originates on the left side, for example, the ears excite the left Mauthner neuron first, which in turn carries a signal down its axon and stimulates motor neurons on the right side of the body, causing a massive muscle contraction that turns the fish to the right. At the same time, inhibitory neurons that cross back over to the left side prevent the muscles on the left from contracting, thus ensuring that nothing interferes with the all-important right turn. The result is an incredibly fast escape—unless the fish swims too close to a tentacled snake. In that case, the snake’s body feint usually sets in motion the cascade of neural events that leads to a turn in the wrong direction. And unfortunately for the fish, the simultaneous activation of the inhibitory circuitry that usually functions as a safety means there is no turning back.
The snake’s astonishing trick explains some previously puzzling observations. In 1999 John C. Murphy of the Field Museum of Natural History in Chicago reported that fish were eaten very quickly and sometimes disappeared completely during the snake’s strike, within one frame of his 30-frames-per-second video—much faster than expected. My high-speed videos reveal that even when the fish do not oblige the snake by swimming straight into its mouth, the turn they make toward the snake usually allows it to capture them headfirst, which is the quickest way for a snake to swallow fish. This fast eating not only allows the snake to eat more often, but it also helps to keep the predator’s identity under wraps (it is hard to look like a harmless stick if other fish have seen you devour their comrade). Furthermore, the snakes have their own predators and are most likely to be seen themselves when swallowing a fish, so rapid dining may reduce the hunter’s chances of becoming the hunted.
Making Predictions
Psychologist B. F. Skinner once said, “When you run into something interesting, drop everything else and study it.” In that spirit, I decided to put my interest in the tentacles aside temporarily and focus on the snake’s predatory behavior—a shift that turned up more tricks in the creature’s repertoire.
Although startling fish toward a strike is impressive, it works only when the fish is situated in the “sweet spot” between the snake’s head and neck and parallel to its jaws. What about fish in other orientations? Because the fish’s escape response propels it either to the left or the right, the snake cannot startle a fish toward its mouth if the fish is already facing its jaws. In this case, the tentacled snake employs another, even more impressive strategy: it predicts fish behavior. First it uses a body feint to startle the fish away from its body, sending the fish on a path parallel to the snake’s jaws. Then, before the fish even moves, the snake strikes toward the future location of the fish’s head, such that its jaws reach the spot just as the ill-fated fish arrives. The events occur far too quickly for the snake to use visual feedback to track the moving fish during the strike—it must plan ahead. In some experimental trials, the fish did not turn away from the body feint (the tactic is not foolproof), yet the snake still struck in the direction the fish should have moved in had it reacted in the usual way. This behavior confirmed that the snake strikes based on prediction, rather than tracking the fish as it moves.
Sometimes snakes simply struck at a fish even if they were unable to startle it in a particular direction. But for the most part, snakes patiently waited for fish to enter the trap formed by their J-shaped hunting posture. To my surprise, I observed yet more kinds of predictive strikes for fish in this zone, depending on the position of the fish. In one contortionistic variant, snakes curled their head under their own body to meet an escaping fish head-on. It seems tentacled snakes can choose from a range of attack strategies in their arsenal, depending on the situation at hand. These predictive strikes raised an interesting question: Do tentacled snakes learn to predict the movements of a C-starting fish from a lifetime of striking, or are they born with this ability? As luck would have it, several of our snakes gave birth. When we tested the newborns with live fish, they clearly struck for the future location of escaping fish (when fish were in the appropriate position), thus showing they were born knowing how a fish moves and how best to outsmart it.
Reporting our findings last year in PLoS ONE, we observed that this innate ability testifies to the long evolutionary history of tentacled snakes preying on fish and bears on one of the most fundamental questions in biology—namely, the relative roles of nature and nurture in the development of behavior. Tentacled snakes sit at the extreme nature end of this continuum, at least when it comes to strikes by newborns. The very reliable response of fish to a sudden water disturbance provided a framework for the evolution of one innate behavior (predictive strikes) that takes advantage of another innate behavior (fish-escape responses).
That the fish have not evolved a counterstrategy suggests that tentacled snakes are acting as what Stephen Jay Gould termed “rare enemies,” exploiting a behavior that is normally adaptive. Fish have many predators, and most of the time their best bet on detecting a sudden water disturbance is to flee in the opposite direction. It is the unlucky fish that encounters this snake and is tricked into turning toward its enemy rather than away from it.
Seeing in Darkness
As for the tentacles, my graduate student Duncan B. Leitch and my research assistant Danielle Gauthier and I conducted a series of investigations to determine their function. We published our results in 2010 in the Journal of Experimental Biology. By examining the anatomy of the nerve endings in these appendages, their responses to various stimuli and how they map into the brain, we were able to show that the tentacles are exceptionally sensitive touch organs that detect water movements generated by nearby moving objects. That is, the tentacles function exactly as would be expected for a fish-detecting organ in an ambush predator. We also filmed snakes under infrared illumination, which they cannot see, and demonstrated their ability to catch fish without using eyesight. Apparently the tentacles allow snakes to detect and capture fish at night or in murky water. Armed with a world-class motion detector and the ability to scare a target to its death, a tentacled snake is a fish’s worst nightmare.