Fish Fingers and Life on Land

When a roulette ball lands on its wheel, its fate is not absolutely random. It does not bounce off the wheel and stick to the ceiling. It does not end up perched on the border between two numbers. The force of gravity, the energy of the throw, and the instability of the wheel’s borders push the ball onto one of the numbers. Its fate is constrained, although it remains unpredictable.

The same holds for evolution. It is channeled within certain constraints, but that doesn’t mean that its transformations unfold with steady, predictable progress. The internal forces of evolution–the way genes interact to build an organism–meet up with the external forces of climate, geography, and ecology, like advancing weather fronts. When they collide, they produce evolutionary tornadoes and hurricanes. As a result, scientists have to be on their guard when they try to reconstruct how evolutionary transformations took place, because it is easy to impose a simple story on a counterintuitive reality.

If 530 million years ago marks one major milestone in our evolution–the dawn of vertebrates during the Cambrian explosion–the next must be 360 million years ago, when vertebrates came on land. During those intervening 180 million years, vertebrates evolved into a vast diversity of fishes–including the ancestors of today’s lampreys, sharks, sturgeon, and lungfish, as well as extinct forms, such as the jawless, armor‑plated galeaspids and placoderms. But during that entire time there was not a single back‑boned creature walking on dry land. Only 360 million years ago did vertebrates finally emerge from the ocean. From them, all terrestrial vertebrates (known as tetrapods) are descended–everything from camels to iguanas to toucans to ourselves.

Early descriptions of this transition were infused with a heroic tone, as if it were part of some foreordained step toward the rise of humanity. From the squirming fish of the sea, the story went, pioneering species emerged onto dry land, struggling with their fins and evolving lungs and legs to let them conquer dry land, rising high and standing tall. In 1916, the Yale paleontologist Richard Lull wrote, “The emergence from the limiting waters to the limitless air was absolutely essential to further development.”

In fact, the origin of tetrapods was a far different story, one that paleontologists themselves did not even begin to understand properly until the 1980s. Before then, evidence about what the earliest tetrapods were like was hard to come by. Researchers knew that of all fishes, the ones most closely related to tetrapods were an ancient lineage known as lobe‑fins. The living lobe‑fins include lungfishes, which live in Brazil, Africa, and Australia. These freshwater fishes can breathe air if their ponds dry up or if the oxygen levels of their water drop dangerously low. The other lobe‑fin is the coelacanth, a hulking, widemouthed creature that lives hundreds of feet below the ocean’s surface off the coasts of southern Africa and Indonesia.

The skeletons of lobe‑fins bear some special similarities to those of tetrapods. Their stout, muscular fins, for example, have the same basic arrangement as legs and arms: a single bone closest to their body, which connects to a pair of long bones, which in turn connect to a group of smaller bones. Although lungfish and coelacanths are the only lobe‑fins alive today, 370 million years ago lobe‑fins were among the most diverse groups of fish. And paleontologists discovered that some of those extinct lobe‑fins were even more like tetrapods than living lobe‑fins are.

As for the oldest tetrapods, paleontologists knew of only one species: a 360‑million‑year‑old creature called Ichthyostega. Discovered in the mountains of Greenland in the 1920s, this 3‑foot‑long, four‑legged creature was clearly a tetrapod, but it had a flat‑topped skull that looked more like lobe‑finned fish than it did later tetrapods.

Paleontologists concluded that Ichthyostega was the product of a long struggle to adapt to dry land. The American paleontologist Alfred Romer sketched out the most thorough scenario for this origin for tetrapods. Their lobe‑fin ancestors lived in freshwater rivers and ponds, but a change in the climate brought on seasonal droughts that made their homes evaporate every year. The fish that could drag themselves to the next pond survived, while the ones that were stranded died. The more mobile lobe‑fins were more likely to survive, so over time their fins evolved into legs. Eventually these fish became so good at moving on land that they could hunt the insects and other invertebrates crawling around on the ground, and they gave up life in the water altogether.

Romer’s scenario seemed logical enough, at least until a second early tetrapod was discovered in Greenland. In 1984 Jennifer Clack, a paleontologist at the University of Cambridge, was perusing the notes from an expedition of Cambridge geologists in the 1970s. They had discovered Ichthyostega‑like fossils and had simply stored them away right under Clack’s nose. Clack returned to their site in 1987 and found the complete skeleton of another 360‑million‑year‑old tetrapod, named Acanthostega.

Acanthostega had all the required hallmarks of a tetrapod, such as legs and toes, but it was an animal that could only have lived underwater. For one thing, Clack and her colleagues discovered bones in its neck that supported gills. For another, its legs, shoulders, and hips were all far too weak to hold up its weight on dry land.

Acanthostega made no sense in Romer’s scenario, but paleontologists were realizing that some of his assumptions were wrong. Acanthostega and other early tetrapods did not in fact live in harsh, drought‑plagued habitats. Instead they lived in lush coastal wetlands, a habitat that was coming into existence for the first time on Earth as large trees began to grow along coasts and rivers. There were no droughts to drive fish toward a tetrapod’s body.

Clack and other paleontologists now argue that fish evolved legs and toes not to walk on land but in order to move underwater. With a few minor changes to the ways in which master‑control genes built fins, evolution rearranged the bones into toes. These tetrapod‑like fish could then have used their fins to clamber through reedy marshes, over fallen logs and other debris. They could grip on to rocks to lie still as they waited to ambush passing prey. While this sort of locomotion may seem strange to us, some living fish do much the same thing. Frogfish have finger‑like projections on their fins that they use to walk slowly over coral reefs.

Whatever they were originally used for, feet and toes didn’t evolve in response to the demand for walking on land, their current use. All told now, paleontologists have discovered a dozen or so species of early tetrapods, and they all appear to have been aquatic. (Clack and her colleagues have taken a fresh look at Ichthyostega and have concluded that it may have been able to drag itself around on dry land like a seal.) Ted Daeschler of the Academy of Natural Sciences in Philadelphia has even found the fossil of a separate lineage of lobe‑fins that evolved finger‑like bones independently of our own ancestors. Between about 370 and 360 million years ago, it seems, walking lobe‑fins went through an underwater evolutionary explosion–cichlids with legs, as it were. Only later did one branch of tetrapods move on land, their legs now taking on a new function.

Evolution often borrows things adapted for one function to perform a new one (a process known as “preadaptation” or “exaptation”). As is so often the case with evolution, this tendency was first noticed by Darwin. “When this or that part has been spoken of as adapted for some special purpose, it must not be supposed that it was originally always formed for this sole purpose,” he wrote in 1862. “The regular course of events seems to be, that a part which originally served for one purpose, becomes adapted by slow changes for widely different purposes.”