Lost and found in the museum
How a jar of “unidentified” tadpoles became a new genus
Not every new species gets discovered in the field. Some are discovered in a cabinet, in a jar, under a label that just says “unidentified,” years after whoever collected it walked back out of the jungle.
This is one of those stories.
Summer 2023, at the Smithsonian
I was in Washington, DC, at the Smithsonian’s National Museum of Natural History, doing something that sounds a lot less exciting than it turned out to be: going through the museum’s backlog of unidentified tadpole lots. Every natural history collection has a version of this shelf. Adult frogs get identified quickly, because someone can flip through a field guide and match the pattern. Tadpoles are harder. A lot of species’ tadpoles have never even been described, so a jar of larvae that doesn’t obviously match anything often just gets a label reading “Bufonidae, gen. sp.” and gets shelved, sometimes for decades, waiting for someone who works specifically on tadpoles to come along.
A very ordinary-looking shelf Somewhere in rows like this one sits the next new species, already collected, already preserved, and simply still waiting on the right person to take a second look.
That’s exactly the kind of jar I go looking for.
Flashback to 2006
To explain what I found, I have to back up almost twenty years, to a biologist named Dr. Bruce Means.
Bruce is a naturalist who has spent decades working in Guyana, particularly in the Pantepui, the remote highlands and table-mountain (“tepui”) country of the Guiana Shield. It’s some of the least-visited terrain in South America, and it keeps producing new species because so few biologists have ever had the chance to look closely at it. In 2006, during fieldwork on the slopes of Mount Kopinang, Bruce collected some odd tadpoles from a fast-flowing mountain stream. He didn’t know exactly what they were. He preserved them anyway, and deposited them at the Smithsonian, the way careful field biologists do with anything that looks like it might matter later, known or not.
FIGURE PLACEHOLDER: portrait of Dr. Bruce Means in the field (you used one in the talk, but double-check you have the rights/his OK before it goes on the public site, since it’s a photo of a specific named person).
Here’s roughly what I looked like that same year, for reference, and for what it’s worth I was not yet doing anything useful with tadpoles:
Also 2006 While Bruce was in a Guyanese stream collecting the tadpole that would eventually get its own genus, I was doing this.
Then the jar sat there. Not lost, exactly. Just unidentified, and waiting.
Some very strange tadpoles
Seventeen years later, that jar is one of the ones I pulled.
The tadpoles inside were striking: banded in dark chocolate brown and cream, with an oral disc clearly built for gripping, not just grazing. Genetic work placed them, somewhat surprisingly, within Rhaebo nasicus, an obscure toad species first named in 1903 from very little material and rarely collected since. Case closed, more or less. A known species, just from a stream nobody had sampled before.
The tadpoles from the jar Striking banded coloration, and a mouth clearly built for clinging to rock in current, not for the ordinary business of scraping algae off a leaf.
Except it didn’t quite sit right.
Really, Rhaebo?
A few things kept nagging at us. The color pattern was unlike any other Rhaebo tadpole on record. The mouth and throat anatomy looked wrong for the genus. The skull, once we cleared and stained specimens and ran others through a micro-CT scanner, was missing bone features that are supposed to be diagnostic for Rhaebo. Even the ecology was off: Rhaebo species are generally unremarkable pond-and-stream generalists, and this animal was a specialized, sucker-mouthed torrent tadpole clinging to boulders in a waterfall-fed creek.
The skull bone that gave it away Fair warning: frog palate bones are a hard sell as a visual. But look at the yellow shape (the sphenethmoid) in the bottom-right panel, “Rhaebo” nasicus, next to Rhaebo haematiticus just to its left. Narrow versus wide. That kind of difference, repeated across several bones, is what a taxonomist actually means by “doesn’t fit the genus.”
Individually, none of these differences would necessarily mean much. Animals are variable, and a weird-looking population doesn’t automatically deserve its own genus. But once color, oral anatomy, skull anatomy, and ecology all point the same direction, and once the molecular data confirms it’s not simply a variable population of an existing species, that combination is usually a signal that you’re not looking at Rhaebo at all. You’re looking at something Rhaebo has been mistakenly holding onto.
Adhaerobufo gen. nov.
So we gave it, and a close relative from the upper Amazon called Rhaebo ceratophrys, a new genus: Adhaerobufo.
The name comes from the Latin adhaerens, meaning adherent or clinging, and bufo, toad. An adherent toad, named for the one trait that gives the whole genus away: a tadpole mouth modified into a suction cup, built to hold on in water that would otherwise sweep it downstream. It’s a genus named after its larvae, not its adults, which is unusual, but fitting: the adult frogs are fairly unremarkable toads. Almost everything that makes this group distinctive is happening in the tadpole stage.
Not exactly a subtle difference, once you line them up Adhaerobufo nasicus (top) next to two ordinary Rhaebo tadpoles, R. caeruleostictus (middle) and R. haematiticus (bottom). Same general toad-tadpole body plan, completely different mouth, body shape, and color.
Where Adhaerobufo lives The genus spans two disjunct areas: the Guyana tepui highlands (including Kaieteur Falls, upper right) and lowland Amazon forest in Peru, Ecuador, and Brazil, hundreds of kilometers apart.
Adhaerobufo now joins a surprisingly long list: sucker-mouthed, current-clinging tadpoles have evolved independently at least ten separate times just within the toad family Bufonidae, and dozens of times across frogs more broadly. It’s one more entry in the same story I told in the last post, wearing a different costume: fast water keeps handing tadpoles the same problem, and evolution keeps arriving at strikingly similar-looking solutions to it.
Ten separate inventions of the same trick Every colored branch on this tree is a genus that independently evolved a suctorial or gastromyzophorous (belly-sucker) tadpole. Adhaerobufo is the green branch at lower right, sitting right next to its ordinary, non-suctorial relatives.
Lost and found in the museum
Here’s the part of this story I actually want people to take away from it.
Nobody set out to find a new genus of frog. Bruce Means didn’t know what he’d collected in 2006, and said so, honestly, by leaving it labeled “unidentified” rather than guessing. That’s the unglamorous, essential half of taxonomy: collecting and preserving things properly even when you can’t identify them yet, and trusting that eventually the right person will be looking for exactly that jar.
The other half is being that person. It took someone who spends enough time with tadpole anatomy specifically, mouths, skulls, muscles, to look at an “unidentified Bufonidae” label and recognize that this one didn’t just need a name. It needed a new genus. Knowing your organism well enough to notice when something isn’t just unidentified, it’s new, is its own kind of expertise, and it’s exactly why natural history collections keep paying off decades after the fieldwork that filled them.
Dias, PH, JR Phillips, MO Pereyra, DB Means, A Haas, and PJR Kok (2024). The remarkable larval morphology of Rhaebo nasicus (Werner, 1903) (Amphibia: Anura: Bufonidae) with the erection of a new bufonid genus and insights into the evolution of suctorial tadpoles. Zoological Letters 10, 17. Open access.