In the distant past, between 350 and 400 million years ago, a group of our fishy ancestors started crawling up on land. The fins that propelled them through the water gradually evolved into sturdy, weight-bearing limbs. Their hind legs connected directly to their hips, which became bigger. Swimming fish became walking, four-legged tetrapods, such as amphibians, reptiles, and mammals.
Scientists have studied the evolution of tetrapod limbs and skeletons in incredible detail, but other aspects of our invasion of land are less clear. How, for example, did our pioneering ancestors eat?
Many fish feed by sucking. As they open their jaws, a horseshoe-shaped bone called the hyoid pushes down on the floor of the mouth, expanding it, and creating a flow of water that draws prey inside. Even species that take bites and nibbles rely on a similar suction to swallow food once it’s inside their mouths.
This technique works because fish are constantly surrounded by water. It doesn’t work on dry land. Fortunately, tetrapods solve that problem with a muscular tongue, which helps to move food from the mouth to the throat. Once again, the hyoid is involved—it’s the bone that the tongue is attached to. But how did this structure evolve? How did the hyoid go from being a bone that creates suction to one that moves a tongue? How did the first tetrapods swallow?
These questions have been hard to answer because very few fossils of early tetrapods contain decent traces of the hyoid. But Krijn Michel from the University of Antwerp tried a different tactic: he studied a delightful fish called the Atlantic mudskipper. This tiny creature looks like a tiny doorstop with a pair of fins and googly eyes, and it lives throughout the mangrove swamps of eastern Africa, the Indian Ocean, and the western Pacific. It spends a surprising amount of time on land. It hauls itself about on its fins, fighting, mating, and foraging in the open air.
Michel filmed Atlantic mudskippers with high-speed cameras as they sucked up pieces of shrimp that had been placed on dry surfaces. As he reviewed the videos, he noticed something odd. In the moments after a mudskipper leans forward and opens its mouth, a small bubble of water protrudes from its open jaws. The water spreads over the morsel of food, which the mudskipper envelops with its mouth. It then sucks both morsel and water back up.
The water acts like a tongue—a “hydrodynamic tongue”, in Michel’s words. It allows the fish to lap up its food and then swallow it.
Michel showed how important the ‘tongue’ is by placing morsels of shrimp on an absorbent surface and filming the mudskippers with X-ray video cameras. This time, as the mudskippers leant in, their watery tongues were drained away. They could still grab the shrimp in their jaws but they couldn’t swallow. On 70 percent of their strikes, they had to return to water before they could gulp down their mouthfuls.
This explains why mudskippers almost always fill their mouths with water before they come out on land. By keeping that watery tongue, they can swallow several mouthfuls before having to return to the water. By contrast, the eel-catfish, which also ventures onto land but doesn’t use the same trick, must always return to water after it has grabbed its prey.
“These findings suggest that swallowing food in air may have been a substantial problem for the transition from water to land during vertebrate evolution,” says Beth Brainerd from Brown University. “When early tetrapods started feeding on land, they had to evolve a new way to move food to the back of the throat for swallowing.”
Did they use a watery tongue, a la mudskippers? Perhaps, but it’s important to remember that these are modern fish, and not tetrapods-in-the-making. Hundreds of millions of years of evolution separate them from our land-colonising ancestors. At most, they can hint at the kinds of strategies that early tetrapods might have used when they moved onto land.
A watery tongue, for example, could have provided a workable interim solution, allowing the animals to feed successfully while their hyoids changed and they developed muscular tongues. Indeed, when Michel trained his X-ray cameras on a fish and a newt, to watch how their hyoids moved when they ate, he found that the newt’s movements more closely resembled those of the mudskipper. The bone’s making the right sort of movements, even if there’s no muscular organ attached to it.
Reference: Michel, Heiss, Aerts & Van Wassenbergh. 2015. A fish that uses its hydrodynamic tongue to feed on land. Proc Roy Soc B http://dx.doi.org/10.1098/rspb.2015.0057
PS: Fish do have “tongues” but the term is a loose parallel; unlike our muscular organs, these tongues (usually) can’t stick out of the mouth, and they don’t help with swallowing. They can, however, help with chewing.