One of the biggest mysteries in the history of life on Earth is how animals first made the incredible journey from water to land. Hundreds of millions of years ago, ancient fish gradually evolved the ability to move across solid ground, eventually giving rise to amphibians, reptiles, birds, and mammals—including humans.
Now, scientists have taken an unusual but fascinating step toward solving this mystery. Researchers at the University of Cambridge have developed a fish-like robot that can imitate how certain modern fish move on land. Their findings suggest that many different fish species independently evolved the same simple walking method, providing new clues about how the earliest vertebrates may have first learned to walk.
The study, published in Nature Communications, combines biology, engineering, computer modeling, and robotics to uncover one of nature's oldest movement strategies.
A Window Into Earth's Distant Past
The transition from water to land was one of the most important events in evolution. Before animals could colonize forests, deserts, and grasslands, some aquatic creatures first had to figure out how to move outside water.
Scientists have long studied fossil evidence and living fish species to understand this process. However, fossils can only reveal so much. They show body structures but often cannot tell researchers exactly how extinct animals moved.
To overcome this challenge, researchers turned to modern fish that can still walk on land today.
Several living species—including lungfish, bichirs, catfish, snakeheads, and sculpins—are capable of leaving the water and moving across land when necessary. While they are much more comfortable in aquatic environments, their ability to travel on land provides important survival advantages.
For example, a fish that can move across land may escape aquatic predators, search for food, or travel between isolated pools of water during droughts.
By studying these modern walking fish, scientists hoped to uncover clues about how ancient fish may have behaved millions of years ago.
Discovering a Shared Walking Strategy
The research team began by analyzing the movements of several walking fish species, especially Polypterus senegalus, commonly known as the gray bichir.
Using observations of real fish, the scientists created detailed computer models to examine how these animals move when out of water.
What surprised them was that many unrelated species appeared to use nearly the same movement pattern.
Although these fish evolved separately and belong to different branches of the evolutionary tree, they repeatedly arrived at a remarkably similar solution for moving on land.
The researchers named this movement the "undulating tripod gait."
Despite sounding complex, the idea is actually quite simple.
The fish anchors itself using its head or front fins while swinging its body from side to side. At the same time, the tail pushes the animal forward. Instead of using specialized legs, the fish relies on the same body undulations it normally uses for swimming.
In essence, it is swimming—but on land.
Lead researcher Dr. Michael Ishida explained that the movement looks somewhat awkward and clumsy, almost like a fish accidentally stranded on shore. However, beneath that appearance lies an efficient and surprisingly effective strategy.
Why This Gait Matters
The discovery is important because it reveals a common locomotion principle shared across multiple fish species.
Until now, scientists had studied walking fish individually, but no one had identified a unifying movement pattern that connected many different species.
The findings suggest that the undulating tripod gait may represent a simple evolutionary solution that naturally emerges whenever a fish needs to move on land without specialized limbs.
This is a classic example of convergent evolution, a process in which unrelated species independently develop similar traits or behaviors because they face similar challenges.
Nature often arrives at the same solution multiple times.
For example, wings evolved separately in birds, bats, and insects. Likewise, the new study suggests that different fish species independently developed the same basic walking strategy because it is mechanically efficient and relatively easy to achieve.
The repeated appearance of this gait across different species hints that similar movement patterns may also have been used by ancient fish during the earliest stages of land exploration.
Building a Robot Fish
To test their theory, the researchers went beyond computer simulations and built a real robotic fish.
The robot was designed to mimic the body movements observed in living walking fish. By experimenting with different movement patterns, the team could determine which gait worked best on land.
The results were striking.
The most efficient movement generated by the robot closely matched both the computer model predictions and the actual movements of real fish.
Whenever researchers altered the sequence of body bending or tried alternative walking styles, the robot became slower and less effective.
This consistency provided strong evidence that the undulating tripod gait is not merely one possible solution—it may be the optimal solution for a fish-like body attempting to move on land.
According to the researchers, this was one of the most surprising findings of the study.
The fact that computer simulations, robotic experiments, and observations of living fish all pointed to the same locomotion pattern suggests that this movement strategy is deeply rooted in basic mechanics.
Escaping Predators and Finding New Habitats
Walking on land may seem like an unusual skill for fish, but it offers significant survival advantages.
If a predator dominates the water, a fish capable of moving onto land gains access to an escape route unavailable to its enemies.
Similarly, temporary pools, marshes, and tidal environments often become disconnected. Fish that can travel short distances over land can reach new habitats and access additional food sources.
These benefits may explain why the ability evolved multiple times throughout fish evolution.
Even limited land movement can provide a major evolutionary advantage.
The study suggests that ancient fish facing similar environmental pressures may have gradually improved their terrestrial abilities over millions of years, eventually leading to the evolution of true limbs.
Looking Toward Ancient Fossils
Perhaps the most exciting implication of the research involves extinct species such as Tiktaalik, one of the most famous transitional fossils ever discovered.
Tiktaalik lived about 375 million years ago and possesses characteristics of both fish and early land vertebrates. It is often viewed as a crucial evolutionary link between aquatic and terrestrial animals.
Researchers believe that combining robotic models with computer simulations could help reveal how creatures like Tiktaalik actually moved.
Since fossils cannot demonstrate motion directly, robotics offers a powerful new way to test evolutionary theories.
By reconstructing ancient body shapes and experimenting with different movement patterns, scientists may be able to recreate the first steps taken by vertebrates on land.
A New Tool for Understanding Evolution
This study highlights how modern technology can help answer some of biology's oldest questions.
By bringing together engineering, robotics, paleontology, and evolutionary biology, researchers have uncovered a shared walking strategy used by multiple fish species and potentially by ancient vertebrate ancestors.
What appears to be a simple flopping motion may actually represent one of life's earliest and most important innovations.
The robot fish not only demonstrates how fish move across land today but may also provide a glimpse into a pivotal moment in Earth's history—when the ancestors of all land-dwelling vertebrates first ventured out of the water and began the journey that eventually led to life on land as we know it.
Reference: Ishida, M., Berio, F., Po, T. et al. The undulating tripod gait as a model of the locomotion of walking fish. Nat Commun 17, 4596 (2026). https://doi.org/10.1038/s41467-026-73111-2
Comments
Post a Comment