In February 2005, John Lundberg, an evolutionary biologist at the Academy of Natural Sciences in Philadelphia, obtained molecular data from a tissue sample that he just couldn't believe. The sample came from a bizarre species of Mexican catfish that no one could identify. But the data said the fish was from Africa. He thought, "Wait, maybe someone switched the jars." To be sure of the findings, he wanted a second tissue sample. So, in May 2005, Lundberg headed approximately 2000 km south to the Lacantún River along the Mexico-Guatemala border. There, he found another fish, which confirmed the original conclusion: This catfish was, indeed, of African descent. So how on earth did it end up in Mexico?
The story began nine years...
Lundberg then enlisted the help of his postdoc, John Sullivan, who was already carrying out a phylogenetic analysis of all the major catfish lineages based on a 3.6 kb stretch of nuclear DNA spanning the rag1 and rag2 genes, involved in immune cell maturation. Sullivan tossed the Chiapas catfish sequence into the mix, and ran the analysis again.
"I can still recall with that first sequence when I put it in and got the result after doing the Bayesian analysis," says Sullivan. "I was quite surprised." The analysis showed almost unequivocally that the Chiapas catfish was nested in among the African species, not the North or South American species as the researchers had anticipated. Through fossil-calibrated molecular clock dating, the researchers concluded that the Chiapas catfish branched off from its African relatives between 75 and 94 million years ago (Proc Acad Nat Sci Phila, 156:39-53, 2007).
After the findings were confirmed, the researchers set about determining how the fish had traveled from Africa to Mexico. Nearly all catfishes are freshwater species, so a marine route was highly improbable. The simplest solution would be for the Chiapas catfish to diverge from the African catfishes when South America and Africa separated. But the last time the two landmasses were combined was in the southern supercontinent of Gondwana, which broke apart an estimated 100 million years ago, before the split of the Chiapas catfish.
According to the researchers, that leaves only two scenarios where the geological record provides evidence for a plausible land-locked freshwater connection: Either the Chiapas catfish traveled through Asia, across the Bering Strait, down through North America, and into southern Mexico; or, it traversed Europe, journeyed along an ancient North Atlantic land bridge, through Eastern Canada, and further on to Mexico. "Any of these scenarios seem somewhat outlandish, and it's a matter of choosing between hypotheses," says Sullivan.
Joseph Nelson, a fish systematist at the University of Alberta in Edmonton, Canada, thinks the North Atlantic land bridge explanation is more likely. "To me, it's the simplest, the most direct, and the shortest, so I would tend to favor it - but you never know," he says. With no living relatives or fossil remains, there's no way to say one way or the other, he admits. "The most direct way of solving it will be to find fossils."
Sullivan optimistically thinks that's a real possibility. "The discovery of Lacantunia in the first place was a huge surprise, and there could be other surprises in the future."