Nobody Has Ever Seen an Eel Reproduce: So We Tried to Find Out Where They Do

Here is a fact that should be more famous than it is: no human has ever observed a European eel reproduce, and no eel egg has ever been found in the wild. For most of recorded history this drove naturalists slightly mad. Aristotle concluded eels spontaneously generated from mud. A 19-year-old Sigmund Freud spent months in a Trieste lab dissecting hundreds of eels in search of their testes, found nothing conclusive, and left marine biology for, well, other kinds of repression.

The Sargasso clue

In the 1920s the Danish oceanographer Johannes Schmidt chased the trail the only way anyone has managed since: by catching the larvae. The smaller the leptocephalus, the closer to the birthplace, and the smallest ones all pointed to one place, the Sargasso Sea, a becalmed gyre of warm water in the middle of the North Atlantic. Every European eel, it seems, is born there, drifts thousands of kilometers to Europe, lives for years or decades in rivers and estuaries, and then swims all the way back to spawn and die. Nobody has ever watched the final act.

Why it suddenly matters

The European eel has collapsed by more than 90% and is now critically endangered. Climate change is the usual suspect, and the dominant story is specific: warming displaced the Sargasso spawning ground, lengthening or breaking the larval route to Europe, so fewer babies make it home. It is a tidy, intuitive narrative. We decided to test it.

What we did

We pulled 44 years of satellite and in-situ sea-surface temperature, ran three objective front-detection algorithms, built an eddy-resolving Lagrangian model of how larvae actually drift, added a five-model climate-projection ensemble, and used a Monte-Carlo back-tracking inversion (validated against the Atlantic eel leptocephalus database of Miller et al. 2015) to rewind the larvae toward their birthplace.

What we found

The textbook story is largely wrong. The spawning isotherms (the 22 to 24 degree C bands the eel spawns between) have migrated about 190 km poleward over four decades, exactly as warming predicts. The physical convergence front that organizes the spawning ground moves much more slowly. The warm water is climbing north through the spawning zone faster than the front it sits in, a geometric consequence of near-uniform warming (we do not claim a stationary front). The ground is not being displaced wholesale. It is warming in place.

And the larval-highway story does not hold either: delivery to Europe turns out to be controlled by how the larvae swim, not by the currents, and recruitment cannot be predicted from spawning-ground climate. So the historical collapse is not, primarily, a transport problem. What the projections do show is grimmer in a quieter way: the spawning ground already sits at the upper thermal limit for spawning and warms a further 1.7 to 3.3 degrees C by 2100. The warmth overtakes the eel where it spawns, with nowhere better to go.

Why the distinction matters

If the spawning ground were simply moving, you might hope the eel could follow it. "Warming in place" is different: the reference the animal is tethered to is heating up underneath it, with nowhere to go. It reframes the conservation question, and it is a sharper, more testable claim than the one it replaces.

We still cannot tell you where, exactly, an eel has sex. The act itself remains unobserved after 2,000 years of trying. But we can now say something concrete and validated about what is happening to the place it happens. The full notebook, the figures, and every number behind this are open.