A mangrove swamp may not seem like an ideal hangout – after all, it's often dirty and smelly – but to the small mangrove rivulus fish it's home sweet home. Nonetheless, at times even this well-adapted fish can find the swamp uninhabitable, for example when rotting vegetation causes sulphur levels to spike or when oxygen levels dip. To escape, they will often venture on to land, where they can survive for up to 2 months at a time, breathing through their skin instead of their gills. What's more, they can travel a fair distance inland, as the recent discovery of wooden log, several metres from the water's edge, crammed full of these tiny fish indicates. When Benjamin Perlman, a PhD student in Miriam Ashley-Ross' lab, at Wake Forest University, USA, heard about this at one of the lab's weekly meetings it made him wonder: how on earth did they get there? After all, these fish don't have limbs or specialised appendages like other amphibious fish, and just look like any other bony fish. When undergraduate Alexander Pronko joined the lab and wondered the same thing, the duo decided it was time to investigate how these fish can move on to land (p. ).
The duo began by training their fish to traverse a small clay barrier in the middle of the fish's tank, as Perlman explains: ‘We'd put a little cricket on the other side of the tank and we conditioned these fish to go over this barrier to feed. Initially, the barrier was flush with the water surface, so it barely had to get out of the water, it just had to glide over it. After a number of weeks we kept increasing the width of the clay barrier until eventually the barrier took up most of the tank and was sloped at 15 deg to represent the slope at the land–water interface in the mangrove swamps.’
Having accustomed their fish to crossing the barrier, the duo then set about patiently waiting for the fish to perform their movements in front of two high-speed cameras. After months of filming, they found that the fish were able to perform three types of manoeuvres – the launch, the squiggle and the pounce. The duo found that the fish predominantly used launches to get over the barrier, propelling themselves at 27 body lengths per second (1 m s−1), but in other cases, they squiggled their way up on to land. The squiggle is almost like an army crawl, as Pronko explains: ‘Their tail meets their head and then as they push off the ground, their body moves through a S-shape, and [the tail] then comes back to the other side of head in a C-shape. As this happens, they're propping themselves up on their pectoral fin using it as a pivot point, while the other fin arcs up into the air as the fish yaws a bit.’ The third movement, the ‘pounce’, was used to capture crickets placed near the water–land interface. After pouncing they would then drag their dinner underwater. Perlman explains that, during eating, the gill filaments become exposed and by dining underwater the risk of them drying out is avoided.
By using these different modes of locomotion, the mangrove rivulus can move across land without the need for specialised appendages, and this has important implications, as Perlman points out: ‘It makes you think, what about all these other animals in the fossil record that have a certain structure but you wouldn't necessarily think that they would be able to move into a different habitat?’ And it just goes to show – you can't judge a fish by its cover.
