Fish have evolved to survive at least briefly on land not just once or twice, but dozens of times, according to a new study published in August in the journal Evolution.

These land-dwelling blennies, whom I’ve featured on this blog before, had this to say when informed of the news:

OK, well, it’s hard to be interviewed when you’ve got a mouthful of delicious slime.

The new findings suggest that though the transition from water to land seems extreme, it does not seem particularly hard. Something else must prevent fish from evolving into full-fledged land-dwellers, as seems to have happened only once, giving rise to terrestrial vertebrates like ourselves. But what?

The authors of the study had originally set out to ask a fundamental question about biology: how do species establish themselves in new environments?

Based on studies of invasive species, it’s clear that in spite of the crushing success of a few of these (looking at you, dandelions), most species fail to establish themselves in new areas most of the time. So how often does a successful transition into a radically new habitat actually occur?

“Such ecological transitions have clearly occurred repeatedly in the history of life given that virtually every environment on Earth has some living organism occupying it,” the authors write. Good point.

To that end, they sought to see how often the move from water to land has occurred among fish, an iconic transition that seems like it should be hard and rare. To do so, fish must confront many challenges. Breathing, moving, and metabolizing all have to be modified in ways that make them possible in an environment of radically increased gravity and decreased wetness.

Yet there were hints that it might not be as hard as it seemed, as previous scientific reviews of the subject have found amphibious behavior in several different fish families. But no one had looked to see whether the families were closely related, suggesting just a few water-land transitions, or whether they were widely spaced on the fish family tree, implying more.

Terry Ord and Georgina Cooke of the University of New South Wales decided to do just that, and also to quantitatively examine what ecological factors seemed to favor fish out of water. They had several hyphotheses.

First, fish that live in the intertidal zone should be more likely to emerge onto land because the very nature of their habitat is a twice-daily transformation between water and land. Similarly, fish that live in freshwater ponds, puddles, or creeks that shrink or evaporate face the same challenge.

Fish that live in water that is prone to heating also have been known to leave it because warmer water loses oxygen and may become suffocating. In such cases, air may present the better option. If this were the case, it seems as if the tropics should support more amphibious fish because the incidence of warm water should be higher there than elsewhere. Tide pools, wherever they occur, also tend to heat up in the sun because they are small and shallow.

Fish that live on the bottom of their body of water seem like they would also be better candidates for land-dwelling, because they already have adaptations that favor life on the ground. These adaptations include a rounder, less vertically flattened body that makes walking easier and drying out less likely, and sometimes fins that are limb-like already.

Finally, since getting around on land is likely to be difficult for any fish, it might be easier for those that don’t have a diet that depends on catching “mobile prey”.

The scientists searched the scientific literature to identify every described amphibious fish and rate its level of amphibiousness. They then plotted their findings on a previously constructed fish family tree.

All told, they found reports of amphibious behavior in 130 fish species from 33 families spread right across the fish family tree, from the oldest branches to the youngest. Many, if not most of these 33 could be independent evolutionary instances of a transition from water to land.

Within just one family – the Blenny Family – the scientists could infer that there were at least three and perhaps as many as seven independent instances of evolution to a “highly terrestrial” lifestyle, and it’s possible, they say, something similar has happened in the Goby Family and other families as well.

A land-dwelling blenny from Guam that leaps around in the splash zone on intertidal rocks and hides in moist crevices when the tide is low. Credit: Dr. Georgina Cooke

It’s also possible there are a lot of fish out there who engage in amphibious behavior but simply haven’t been seen doing it yet, given that many only emerge for a short time each day. There may be many more amphibious species out there than we even realize.

Most of the family-level examples, however, were fish out of the water for only a few seconds or minutes, which may not meet the expectations of some regarding “amphibious” behavior. Four of the families did have species that spend hours or days out of water: mudskippers, rockskippers, eels, and four-eyed blennies each come from separate families, implying that at least four times, significant amphibious behavior has evolved on Earth, not including, of course, the crucial fifth instance that resulted in us. The closest living relatives of the lobe-finned fish that made this particular transition also include many species today that make at least occasional forays onto land.

What traits do modern amphibious fish share? More than anything, the scientist found that living in the intertidal zone promotes forays onto land. Fish living in the intertidal zone were amphibious much more frequently than would be expected given their overall numbers. With the twice-daily advance and retreat of the sea, and the ubiquity of tide pools that heat up in the sun, starving them of oxygen and forcing fish onto land, the zone seems to be particularly good at making fish terrestrial.

On the other hand, the tropics seemed to be no better than the polar regions, proportionately, at producing amphibious fish. Nor did freshwater systems produce more amphibious fish than would be expected.

There were, however, many fewer carnivorous amphibious fish than would be expected given their numbers, supporting the idea they are unlikely to become amphibious because catching prey on land is just too hard.

Crossing the land-water frontier is thus not, apparently, in itself difficult. So why have so few piscine pioneers gone onto terrestrial evolutionary glory? Breathing, walking, and reproducing do not seem to present the major barrier.

Instead, the authors theorize, drying out is the real challenge: specifically, the problems posed by the dessication of gills, which must remain moist for most fish to breathe. Our ancestors, such as the early tetrapod Tiktaalik, seemed to have been pre-adapted to overcome this obstacle because they possessed, in addition to gills, a primitive lung.

Without this trait or something like it, the forays of fish onto land seem destined to remain a flirtation. Though free to frolic, they must ultimately remain tethered to the water whence they came.


Ord, Terry J., and Georgina M. Cooke. "Repeated evolution of amphibious behavior in fish and its implications for the colonisation of novel environments." Evolution (2016).