In the Sandhills of North Carolina, scrubby pine and oak forests now cover what used to be sand dunes twenty million years ago, when the sea came 150 miles further inland. Coral snakes used to hide in the tall grasses here, their red and black bands inspiring rhymes that warn hikers to avoid the United States’ most toxic snake. No one has seen a coral snake in the Sandhills since 1960.
But in a way, the coral snake is still present here. An impostor, the relatively harmless scarlet kingsnake, flashes its own red and black rings from its hiding places among pine needles and rotting logs. The kingsnakes’ borrowed disguise is an evolutionary memory, a reminder of the venomous coral snake that used to live in the Sandhills. Most memories wander from their originals and eventually fade—but the scarlet kingsnakes’ mimicry has been getting stronger and more faithful in the decades since the original disappeared.
It’s an evolutionary enigma that David Pfennig is struggling to decipher. “Mimicry has always fascinated me; it’s one of the classic problems of evolution,” says Pfennig, a biologist at the University of North Carolina. It’s one that’s been studied over and over, but never under circumstances quite like this. Something is driving the kingsnake to look more—not less—like the vanished coral snake it impersonates. And Pfennig thinks that something is lurking as a ghostly imprint in the genes of the raccoons and birds and bears that hunt the Sandhills’ snakes.
A Long History
The study mimicry in evolution dates back to the field’s earliest days, when Charles Darwin was home in England writing up the notes that became On the Origin of Species, and naturalist Henry Walter Bates was trekking through the Amazon. Bates was financing his travels by collecting some of the rainforest’s many species of exotic butterflies to sell to collectors. Knee-deep in butterflies, Bates noticed that a group of brightly-colored, smelly (and, he reasoned, presumably poisonous) butterflies never seemed to get eaten by birds—and that a non-smelly butterfly in a totally different family looked very similar. A few hundred miles up the river, when the smelly butterflies looked different, the tasty ones did, too. Bates started to wonder if the resemblance wasn’t accidental.
Darwin published On the Origin of Species in 1859. Two years later, Alfred Russell Wallace, who had arrived at many of the same conclusions about natural selection as Darwin, wrote Darwin with a puzzle: he kept running across groups of unrelated butterflies that looked uncannily similar. “And he said he just couldn’t understand that, it couldn’t come under sexual selection, it couldn’t come under any adaptive story that he could think up,” says Harvard biologist Jim Mallet, who spent his childhood collecting butterflies and other insects and today supplements his studies of butterfly mimicry with an extensive knowledge of the history of evolution. “The person who solved it was Bates.”
Bates also wrote Darwin, explaining that the non-toxic butterflies were mimicking the poisonous butterflies to trick predators into avoiding them. Darwin was thrilled. “Believe it or not,” Mallet explains, “Darwin didn’t have any good examples which proved natural selection.” A nontoxic species imitating a toxic one, and changing its appearance when its model did, provided a perfect case study of a population evolving in response to the environment. This phenomenon, which came to be called “Batesian mimicry,” was, Mallet says, “probably the first good example of natural selection.”
Eye of the Beholder
The evolutionary strategy Bates discovered in butterflies is played out in nature again and again. Hoverflies sport wasp-like yellow and black stripes, and they sometimes even go so far as pretending to sting. Certain caterpillars’ backsides are decorated with eyespots designed to make them look like snakes. The mimic octopus, a particularly virtuosic practitioner, as its name implies, impersonates at least fifteen 15 other animals, including venomous lionfish and sea snakes. Coral snakes, whose mimics Wallace first noticed during his own travels in South America, inspire imitation more often than any other animal: more than 200 other snakes try to outwit hungry predators by flashing coral snakes’ colors.
These flashy mimics, and their toxic models, are the characters who catch our attention. But, Pfennig says, “in a way, they’re passive actors in this play. It’s really the predators that are driving it.” Whether a mimic’s flamboyant ruse will work hinges on a cost-benefit analysis by a hungry predator: are the consequences of a bad guess outweighed by the nutritional value of the meal? Johanna Mappes, a professor at the University of Jyvälskylä in Finland, points out that “All prey, even if they are really toxic, are also a piece of meat.”
Figuring out which variables influence that calculation gets tricky, though, and scientists hoping to probe the evolution of mimicry have had to find some creative ways to observe predators in a controlled environment. Mappes, for example, created an aviary filled with simple black-and-white symbols, allowing her to test how quickly birds learned to interpret warning signals without interference from biases that exist with natural prey. Other groups run computer simulations with digital organisms, which can develop, learn, and reproduce far faster than their flesh-and-blood counterparts.
These studies tell us that poison figures heavily in a predator’s risk analysis. Really noxious animals tend to accumulate mimics, Pfennig explains, because predators are strongly motivated to learn to avoid toxic meals. The more toxic the model is, the more quickly mimicry will evolve, and the more protection the mimics will enjoy. This helps us understand why mimicry is so common among snakes. “There are venomous mammals, right, there are a few venomous birds, but not like in snakes. I mean, there are really venomous snakes,” Pfennig says.
But predators always have to weigh nutritional value against any potential toxicity, and an animal’s size also helps determine whether mimicry will be a good strategy. For really tiny animals, the best bet is just to blend in, while for larger animals—which can be particularly rewarding as meals—some kind of physical defense, like horns or claws, may be more useful in deterring a predator. In that sweet spot in the middle—where large insects and small reptiles live—mimicking something toxic can convince a predator you’re not worth it. (Wallace, writing in his Contributions to the Theory of Natural Selection, also proposed that snakes are similar enough to each other to make mimicry easy, and that insects’ hard exoskeletons make it possible to change their external appearance without affecting their internal organs. On the other hand, he wrote, “We can hardly see the possibility of a mimicry by which the elk could escape from the wolf, or the buffalo from the tiger.”)
A predator’s calculations apparently consider the abundance of toxic and nontoxic prey, too: Mimicry is more likely to evolve if the model is common. In Florida, for example, where coral snakes are abundant, “there’s really strong selection on predators to just avoid anything that looks like a coral snake,” Pfennig explains, even the mediocre impersonators. But once mimicry evolves, mimics will be more accurate if their model is rare. In the Sandhills, where coral snakes were always scarce, a predator is more likely to take a chance on something with only a passing resemblance—statistically, it’s probably not a coral snake. Only very accurate mimics will convince a bird or a bear to stay away.
But even when all the conditions are right—a toxic, medium-sized model abundant enough that predators know to avoid it—it’s still not obvious how mimicry would evolve. “It is fascinating; it’s a simple idea on the surface, and textbooks are full of just the simple explanations,” says Tom Sherratt, a professor of biology at Carleton University in Canada. “But when you dig a little bit deeper there’s all sorts of interesting questions there that still have to be resolved.”
Imagine a deadly species, with bright colors warning predators away (itself an counterintuitive development, since conspicuousness is usually a direct route to a predator’s stomach). It seems natural that other, tastier species would adopt those protective patterns, but getting there isn’t straightforward. For one thing, most prey species, which rely on camouflage to escape predators, start out looking nothing like brightly-colored toxic species: different color, different pattern, even different behavior. “How on Earth,” Sherratt asks, “would you gain all of these features in a relatively short period of time?” To go from zero to mimic without going extinct in an awkward in-between phase—where you’re different enough to be conspicuous but not enough like the model to scare predators—seems like a pretty tough evolutionary needle to thread.
The key to this conundrum is realizing that mimicry, as Sherratt points out, “is all about deception,” which means that all that matters is a predator’s perception. A mimic doesn’t need to deceive a taxonomist—although many eventually do—it just needs to deceive a predator. And it turns out that even a minor costume change may be enough. A recent study in Finland showed that birds pick up on color much more quickly than shape or pattern. (Humans do, too, Sherratt says.) “It suggests that when mimicry evolves, all you really need, to begin with at least, is the color,” Sherratt explains. “Those other details only matter when predators start to pick up on potential differences and start to learn that not every yellow is bad to eat, and so on.” Like a novice art critic who starts off not being able to tell the difference between a Degas and a Monet, the predator can be initially fooled by a sloppy mimic, but slowly becomes more and more discriminating, gradually driving the evolution of more accurate mimicry by learning to avoid the really toxic critters and eat the forgeries.
In the world of mimicry, though, the coral snake is stands apart, and it presents an additional mystery. “The coral snake, it’s not like in other mimicry complexes, like in certain butterflies where the consequences of a predator making a mistake might just be that it gets a tummy ache,” Pfennig says. “Here there’s a really high risk of death to a predator that makes a mistake.” In a sense, coral snakes are too toxic: A predator that chows down on a coral snake won’t learn to avoid one for its next meal—it won’t have a next meal. That throws a wrench into understanding mimicry in terms of predator learning.
So how did avoidance of coral snakes develop so strongly that hundreds of other snakes can hide under its umbrella? In the 1970s, Susan Smith, a biology professor at Mount Holyoke, ran a series of experiments with birds that had been raised in captivity and had never seen a coral snake in their lives. When these birds were presented with a wooden stick painted with red and yellow rings, they fled to the other side of their cage. Coral snakes, apparently, have cultivated an innate, genetically-programmed avoidance in their would-be predators.
It looks like the same thing is happening in the Sandhills: predators who were born long after the last coral snake died are still avoiding the lookalikes. “Selection hasn’t broken down yet, even though there’s been no coral snakes that any herpetologist has collected since 1960,” Pfennig says.
Smith’s experiments weren’t perfect. Some birds simply avoid attacking anything unfamiliar, so it’s not clear that the coral-snake pattern, in particular, was what spooked them. The study could stand to be repeated, and while Pfennig says he would love to do it himself, regulations for experiments involving vertebrates have grown stiffer since the ’70s, and the red tape is daunting. Instead, researchers now leave painted clay models in the wild and tally up bite and beak marks over subsequent days and weeks. Without being able to monitor the interaction between predator and prey, we can’t fully understand the mix of perception and psychology that helps a predator decide whether to pounce or turn away. “Maybe predators in these areas, they have increased sensitivity to these ringed patterns, or maybe to the really bright colors or a combination of those two, or it could be, you know, decision rules in their brain,” Pfennig speculates. “It’s a fascinating question.”
But we are starting to accumulate a few clues: Recent evidence suggests that red might be such a common warning color because, to the eye of a bird, it provides a high contrast to the (often green) background and looks the same, even when the light varies. If birds’ visual systems are primed to see red, and toxic animals take advantage of that sensitivity, birds that are especially jumpy around red prey would be more likely to survive and reproduce. But we still don’t know. “I think the psychology of the predators is a real black box to us, and I think that’s actually where the answers lie,” admits Butch Brodie, a professor at the University of Virginia. “Is there just something nasty about a red belly, an animal that curls up and shows you its red belly, that tweaks some part of the nervous system that says ‘don’t eat me’?”
Whatever it is, the Sandhills’ kingsnakes are still exploiting it. And what’s more, Pfennig says, is that they’ve gotten even better at it. In the decades since coral snakes left the Sandhills, the kingsnakes’ appearance has gotten even more coral-snake-like. (But in Florida, where there are still plenty of coral snakes, kingsnake appearance has stayed the same).
“It was definitely not what I expected,” Pfennig admits, although he says that the fact that mimicry tends to get better as the model gets more rare could have led them to predict it. The ingrained avoidance of coral snakes is apparently so strong that it created what Pfennig calls “evolutionary momentum:” a trait that persists after the environmental conditions that molded it—in this case, the presence of coral snakes in the Sandhills— have changed.
But of course, the real trigger for coral snake mimicry isn’t the coral snakes at all—it’s the way the predators respond to them. The predators will determine how long that momentum lasts. If food gets scarce in the Sandhills, even genetically-programmed wariness can be overridden by hunger. “No defense,” Mappes, the Finnish professor, reminds us, “is perfect.”