The Loom

A Hairy Archaeopteryx?

i-bbe6764e51861d81c8f373365f980029-onychonycteris.jpg I guess it’s only appropriate that the week of Darwin’s birthday is seeing a bunch of new reports about evolutionary transitions. On Monday there was news about how ancient whales with teeth turned into whales with baleen–thanks to the discovery of a fossil of an ancient whale that appears to have had both teeth and baleen. Today’s news takes us from the sea to the trees–the fossil of a primitive bat. The transition that the ancestors of bats made from scampering shrew-like mammals to masterful flyers has remained particularly mysterious. Today’s new fossil lets us look back further than ever into this transition.

The origin of bats is one of those big questions that attracts scientists like…I suppose like bats to a swarm of mosquitoes. Here we have an incredibly successful group of animals. They make up a quarter of all mammal species on Earth. They are one of only three lineages of vertebrates that have evolved powered flight (birds and pterosaurs being the other two). Many of them have an awesome power to see with sound, shrieking into the dark to hear the echoes deformed by passing prey. How did they make this transition, scientists have wondered. Did they start as gliders in trees? Did they start echolocating before they began to fly, or after?

Ten years ago I wrote an article about these questions for Discover. Normally when I look at articles from a decade ago, I cringe at how obsolete they are, how they’ve been transformed into quaint historical artifacts thanks to the relentless progress of science. But this one remains a pretty good background piece, I think. It’s not that I was able to see into the future in 1998. It’s just that the puzzle of the origin of bats has budged very little in the ten years. The puzzle is this: a couple centuries of fossil hunting have yielded fossils of bats dating back 50 million years ago so. Found in Wyoming, they are full-fledged bats, with arms three times longer than their torsos, with long fingers that supported a flight membrane. The shape of their earbones also suggests that they could echolocate. It wasn’t clear which of the living groups of bats was most closely related to those fossil groups. One group, the megachiropterans, includes fruit bats and lacks the ability to echolocate. The other bats, the microchiropterans, can all echolocate. So how do all these clues fit together?

One particularly powerful way to get to the answer to that question is to find a new fossil. But for the past ten years, paleontologists continued to search in vain. Of course, they did not throw up their hands and say that the earliest bats must have popped out of nowhere. And while it took a long time, today a team of scientists reports in Nature, that patience paid off.

They describe a recently discovered fossil dubbed Onychonycteris finneyi. It comes from the same Green River formation in Wyoming where other old bat fossils have been found. (The photograph above shows one specimen the scientists studied; a partly crushed skull and other bones were also found.) The scientists compared the anatomy of Onychonycteris to other fossil and living bats to draw a new evolutionary tree. On their tree, Onychonycteris occupies the deepest branch, making it the most primitive bat yet known.

Onychonycteris had long arms and fingers, demonstrating that it, too, could fly. (It also had other equipment essential for flying, such as a breastbone that could anchor powerful flight muscles and shoulder bones that could rotate through an entire flight stroke.) But Onychonycteris also has a combination of traits found in no other known bat–traits that help reveal how the bat body plan evolved step by step.

Before its discovery, no living and fossil bats had ever been found with claws on all its fingers. Most bats have just a single claw left; fossil bats have tiny bumps on a couple other fingers. Onychonycteris, on the other hand, had well-formed claws on all five digits–big ones on that first and middle fingers, smaller ones on the other. Its arms and legs were also peculiar for bats. Its arms were relatively short and its legs were relatively long–in other words, its limbs were a bit more like those of walking mammals. The scientists suggest that this primitive set of limbs and claws allowed Onychonycteris to clamber around in trees with more agility than later lineages of bats.

Scientists have found a way to predict from the shape of a bat’s wings alone the style in which it flew. In the case of Onychonycteris, its short, small wings suggest that it alternated between gliding and fluttering. This is a rare kind of flight today, seen in mouse-tailed bats. Other bats of the same size never glide. Given the fact that Onychonycteris was the most primitive bat yet found, the scientists propose that early bats started out as gliders, and then gradually added fluttering to their flight. Later, some lineages of bats evolved a more powerful flight stroke without the gliding. Onychonycteris comes from that transition.

It also looks as if echolocation evolved only after Onychonycteris branched off from other bats. The ear bones in its head don’t have the distinctive shape found in living bats that echolocate, suggesting that it had to rely on sight and sound to catch prey–insects, judging from its teeth. Flight evolved first in bats, the scientists argue, and echolocation only came later. It’s possible that flight actually made the evolution of echolocation possible. It takes a lot of energy for bats to issue their shrieks. It may have been less expensive for full-powered fliers to echolocate, because they were already flapping their wings up and down forecefully enough to squeeze air out of their lungs.

There are plenty of other intriguing details to the Onychonycteris fossil that will keep scientists busy for a long time. Most living bats have a membrane that stretches from their back legs to their tail. Some of them use it as a net; they zero in on their prey with echolocation and then snatch up prey in the tail membrane, reacing their head down to snap up their victim. Onychonycteris couldn’t hunt like that because it didn’t have echolocation. But it did have a tail membrane. So it’s possible that the tail membrane evolved to help in flight, allowing bats to turn or brake, for example. Only in some younger lineages of bats did it evolve into a hunting net.

The discovery of Onychonycteris is electrifying. It reminds me of the discovery of another early flier, Archaeopteryx. In 1861 quarry workers in Germany discovered the fossil of a 150-million year old bird that was a combination of new and old: it was covered in feathers and had wings capable of flight, but it also had a long bony tail, teeth, claws, and other reptilian features not seen in living birds. Archaeopteryx was for a long time the best fossil for understanding the transition from ground-running dinosaurs to birds. Now, thanks to the perseverence of paleontologists, it sits on one of many, many branches of related lineages–branches occupied by early birds as well as by bird-like dinosaurs. Let’s hope that Onychonycteris gets some company of its own, and soon.

Update: In response to John’s comment below, I thought I’d add the picture of the skull material shown in the paper…

i-32203fbcfefd649da164f1d49bf55c15-bat skull.jpg


  1. #1 John Conway
    February 13, 2008

    Wow, this is big. Shame it’s head is missing, but the claws on it’s wings are a nice touch.

    Now, if only similar fossils could be found for pterosaurs, we’d be set.

  2. #2 carl Zimmer
    February 13, 2008

    John–They do have some of the skull, just not on this particular specimen.

  3. #3 Paul
    February 13, 2008

    Now this is what Scienceblogs is all about. Solid science blogging.

    Great stuff.

  4. #4 John Conway
    February 13, 2008

    Carl — thanks for posting that. The skull looks fantastic. I did notice you referred to the ear bones, and was a little confused, I though maybe they were disarticulated. I’ll RTFP next time.

  5. #5 Larry Ayers
    February 13, 2008

    Thanks, Carl, that’s just good solid science-blogging!

  6. #6 Jason J Brunet
    February 13, 2008

    The Creationist coffin is running out of room for nails…

  7. #7 Zach Miller
    February 13, 2008

    Beautiful fossil! Does this new bat reveal anything about where bats split off among placentals?

  8. #8 George
    February 13, 2008

    I’ve always liked bats. In the summer I like to go out at sunset and watch them fly out of their hiding places. I can usually tell them from birds easily from the shape of the wing and the distinctive flight pattern you describe. I like that we’re learning more about how that flight pattern evolved from these new fossils.

  9. #9 John S. Wilkins
    February 13, 2008

    Some fruitbats do echolocate, indicating that it is a plesiomorphy of the clade (now that bats are regarded as monophyletic).

  10. #10 Johnx
    February 15, 2008


    Articles such as this are the reasons that I keep coming back to you for more. You took this news and put it into its proper perspective. Good work.

  11. #11 Thomas Eiting
    February 15, 2008

    John S. Wilkins: You may be right that echolocation is plesiomorphic (see a good review in Jones and Teeling, 2006, TREE 21:149), though not really for the reasons you specify, as the echolocation calls of Rousettus (the megabat to which you refer) are produced in a fundamentally different manner from those of all other echolocating bats and likely represent a reacquisition of echolocation. In any case, echolocation is a plesiomorphy apparently only shared by the clade including Icaronycteris and its sister taxa (all other bats except for Onychonycteris). Importantly, the new bat, Onychorhynchus, falls outside the clade for which echolocation can be considered primitive, which gets to the heart of why this find is so intriguing.

  12. #12 Doug
    February 15, 2008

    Hi Carl,

    I was surprised to read on Wiki Animal echolocation thatapparently not all bats have this ability.

    On 9 AUG 2006, ABC News had this story Blind People Who Interact With the World Like Dolphins and Bats.
    “Fourteen-year-old Ben Underwood of Sacramento, Calif., is one of the few people known to use echolocation as a primary means of navigating the world on land. There’s not even a hint of light reaching his brain. His eyes are artificial, but his brain has adapted to allow him to appraise his environment. He makes a “clicking” sound to communicate with objects and people around him.”
    “Forty-year-old Daniel Kish of Long Beach, Calif., also uses echolocation, and has become an expert on it, founding the World Access for the Blind, an organization that teaches others how to echolocate.”
    “Peter Scheifele, who studies hearing and sound production in animals and people at the University of Connecticut, analyzed samples of the clicks that Ben and Kish make
    “Ben clicks, looks to me like once every half second, whereas a dolphin is actually making 900 clicks per second.”

    Wiki does have an entry on Human echolocation.

  13. #13 DDeden
    February 25, 2008

    Onychonycteris had long arms and fingers…

    Its arms were relatively short…

    I’m relatively confused, can you clarify a bit?

  14. #14 Carl Zimmer
    February 25, 2008

    DDeden [#13): Long compared to other mammals, short compared to other bat species.

  15. #15 David Peters
    June 13, 2008


    The origin of bats has been an enigma. My own phylogenetic studies indicate that bats evolved from primitive civets, one of which, the extant Ptilocercus, has been misidentified as a tree shrew.

    And to John Conway’s comment about pterosaurs, there are five genera that increasingly approach the pterosaurs. The first has been recognized as a basal lizard. The latter four have that unusual fifth toe and reduced chevrons among many other characters. 1) Huehuecuetzpalli. 2) Langobardisaurus. 3) Cosesaurus. 4) Sharovipteryx. and 5. Longisquama. As in birds, the wings came last.

    David Peters

New comments have been disabled.