The Confusing Evolution of Bat Echolocation

Bats! They yell at things, sometimes with their noses. What’s up with that? I guess we’ll talk about it and when it started happening. Can you tell I haven’t written an intro to a nonfiction post in a while?

Before we really get started, let’s have a brief primer on what echolocation actually is for those who don’t know. To put it simply, it’s using sound echoes to get information about what’s around you. Sound travels in invisible waves, like ripples on the surface of a pond, which are deflected by the presence of solid objects. A bat screams and then listens for these sound deflections- echoes- and is able to use the way they sound to determine things like the distance, texture, size, and direction of the object, among other things.

When I say the bat ‘screams,’ I’m not being facetious, by the way. The loudness of a bat’s echolocation call, surprisingly, doesn’t have much to do with the size of the bat. Even a tiny bat weighing four grams may be able to emit a shockingly loud 128 decibel sound. For comparison, a jet engine 100 feet away from the listener has a loudness of about 130 db. Many bat species have calls loud enough to damage human hearing- which makes it lucky for us that the frequencies are too high for us to hear.

Bats are able to yell this loud without damaging their own hearing only to a special adaptation that actually causes them to go deaf just before a call is emitted- the muscles in the middle ear actually pull apart the hammer, anvil, and stirrup bones so that sound can’t travel to the cochlea. The muscles relax to allow the bones to reconnect as the sound echoes back.

I wish I could have found a gif of this amazing adaptation, but our imaginations will do. The ossicles- ear bones- comprise of the aforementioned hammer, anvil, and stirrup, which are pulled apart by a muscle contraction to prevent the bat going deaf from its own screams.

Okay, now hoepefully everyone’s up to speed on just what echolocation is. So. The evolution of echolocation in bats is a point of contention among evolutionary biologists. The problem lies largely with a poor fossil record for bats, since their teeny low-calcium bones don’t preserve well. There are few transitional proto-bats in the fossil record, and even fewer fossilized structures that might indicate whether or not an ancient bat was able to navigate via scream. Add to that the fact that there are at least two origins for echolocation in modern bats, and you get a lot of phylogenetic fun.

But let’s start at the beginning. One of the earliest bat fossils ever found is the Eocene bat Onychonycteris. Now, prior to this discovery, there was a lot of debate as to whether bats evolved the ability to fly before the ability to echolocate or vice versa. (Echolocation would presumably still be helpful for a nocturnal gliding mammal, too.) Onychonycteris put that argument to rest because the skull was preserved well enough to show that it lacked some of the specialized hearing anatomy found in echolocating bats. So, as far as we know, flight came first.

Here is the holotype specimen for Onychonycteris. Note the long tail and the claws on the ends of the wingtips, both traits that modern bats have lost.

By the way, while googling for images of this specimen, I stumbled upon the ARK: Survival Evolved profile for this animal, and I’ve probably never been so quietly furious in my entire life. Spot the inaccuracies.

BATS DON’T DESERVE THIS, VIDEO GAME MAKERS.

Anyway, back to the origins of echolocation. So, we can settle at least one debate for the moment- the best evidence we have suggests that bats learned to fly before they learned to scream. But in this case, when did echolocation appear? Here again we have a lot of head-butting, not in the least because the relationships between bat families has traditionally been a massive headache to detangle.

Bats, you see, comprise about one-fifth of all mammal species, with over 1,200 species known, and more constantly being discovered. In that sense, they’re the rodents of the sky, except that they aren’t closely related to rodents at all. At one point it was thought that primates were some of bats’ closest living relatives, based on morphology studies comparing the anatomy of Pteropodids to monkeys. (That belief also assumed that Megabats and Microbats didn’t share a common ancestor, but we’re pretty sure they’re all in the family now.) But the monkey thing is bunk now thanks to genetic studies, which put bats on a different spot in the family tree: as a sister group to Feraeuungulata, a group comprised of all ungulates (hoofed animals, as well as whales) and all carnivores (dogs, cats, bears, etc). Basically, what you should take away here is that bats are a weird group kind of all on their own among the mammals. They’re separated from their closest living relatives by over 60 million years.

Bats and examples of their closest living relatives… really.

So the bat family tree itself took a lot of time to detangle. Even once molecular evidence sorted some things out, there were still more questions. The most recent grouping of bats discards the old terms ‘Megabat’ and ‘Microbat,’ used to refer to big fruit-eating bats and little shouty bats, respectively (though ‘megabat’ is still used to refer to Pteropodids), and adds the new terms Yinpterochiroptera and Yangochiroptera. Yinpterochiroptera, besides being a pain to spell, includes traditional megabats- Pteropodids, the flying foxes- as well as leaf-nosed bats, horseshoe bats, false vampire bats, and a few others, which used to be considered microbats. Yangochiroptera is easier to spell and includes all the rest of the bats traditionally called microbats.

Are you tired yet? I’m very tired.

Now that we have that out of the way, we have to consider which bats in this modern grouping echolocate, and which bats don’t. That should give us a clue as to when the ability evolved- for example, if all bats have it, it must be an ancestral trait; or if one branch of the tree has it, it must have evolved at the beginning of that branch; or if all but one branch has it, it must have been lost in that branch. Easy-peasy, right?

Here’s a phylogeny of bats that can and can’t echolocate, with boxes referring to echolocating groups colored in green.

You’re about to see this poorly-constructed diagram a lot.

Oops! We have a problem. There’s no clean echolocation monophyly here- that is, no easy branches of this tree we can ‘trim’ to collect all the echolocating bats in one group. The big problem seems to be that Rousettus genus up there, which evolved from a group of bats whose other living descendants don’t have echolocation but evolved from another group of bats that do. We’re talking about a gain/loss/regain situation here, which evolutionary biologists absolutely hate. (But it isn’t impossible.)

In fact, there’s another wrinkle to the matter which either clears everything up or throws everything into more chaos, depending on your point of view. The three green groups of bats all have somewhat different ways of echolocating, which I’ll go into more detail about later. This actually led some biologists to suggest that bats evolved echolocation three different times, like so:

Pluses indicate echolocation was gained, minuses indicate that the ability was lost.

But three separate events is a bit much for other evolutionary biologists, who postulate that no, echolocation only evolved twice, damnit, and the common ancestor of all modern bats had the capability. Like so:

Checkmate… ish?

By now you’re probably as sick of looking at that phylogeny as I am, but let me finish up by saying that more evidence points to the second theory than the first, which is nice, because gaining a trait twice is still more parsimonious than gaining it three times, and it lets evolutionary biologists sleep a little better at night. Currently. Probably. Let’s move on.

As I mentioned earlier, the three groups of bats that can echolocate all have slightly different means of doing so, which was the source of some of that phylogenetic kerfuffle. The Yangochiropterans, i.e. the little guys, echolocate using their larynx. They open their itty bitty mouths to do it, like so:

A little brown bat, screaming.

Leaf-nosed bats, horseshoe bats, and other Yinpterochiropteran bats in the Rhinolophoidea superfamily (are you sick of bat phylogeny yet) also echolocate using their larynx to produce sound. However, the sound doesn’t come out of their mouth, but their nose. So, bats in this superfamily have weird, horrible noseleafs specially shaped to shoot out sound. Because they use their noses and not their mouths, they can actually echolocate continuously without having to pause to, say, take a breath or eat an insect.

A horseshoe bat.

One family among Yangochiroptera, the New World leaf-nosed bats (Phyllostomidae), also echolocates through the nose- an ability that evolved separately from that of the Yinpterochiroptera nose-yellers. They have convergently evolved similar noseleaves.

A carefully curated selection of leaf-nosed bats.

Old world vs new world leaf-nosed bats: a lesser false vampire bat (Megaderma spasma) compared with a spectral bat (Vampyrum spectrum). Though they look similar, they are from opposite sides of the bat family tree and evolved their nose-echolocation separately. Confusingly, the spectral bat is also sometimes called the false vampire bat. Neither species drinks blood.

Knowing this, the theory that nose-echolocation in Yinpterochiroptera evolved independently from mouth-echolocation in Yangochirpotera makes more sense, even though that’s probably not what happened. However, we’re fairly certain that the third group of bats, the Rousettus fruit bats, DID evolve their echolocation completely separately from the other two groups. Why? Because they don’t echolocate with their larynx, or throat. They echolocate by clicking their tongues. Hence, their faces aren’t a horror show.

An Egyptian fruit bat (Rousettus aegyptiacus) at the Omaha Zoo.

The particularly interesting thing about Rousettus bats is that they are nested within Pteropodidae. Nearly all Pteropodids are crepuscular fruit-eating bats, who neither need to chase insects nor navigate in the dead of night. Hence, the loss of echolocation in the group. However, despite the fact that they eat fruit, Rousettus bats are nocturnal, and live in caves during the day. This is likely why they regained the ability to echolocate, albeit in a new way.

So: these are the three main ways to echolocate (though there might be a fourth secret way… wait for it). To review the facial adaptations of the bats that use each: bats with large ears and small eyes are probably mouth-echolocators from the superorder Yangochiroptera, while bats with large ears and weird nose ornaments are probably nose-echolocators from the Rhinolophoidea superfamily of Yinpterochiroptera (or Yangochiropteran New World leaf-nosed bats), and bats with large eyes, small ears, and adorable ickle faces either don’t echolocate or happen to be tongue-clicking Rousettus.

Here we see examples of yellers, snootlers, and clickers: Eastern red bats, Honduran white bats, and, uh, Rousettus. Together, they form the most amazing bat musical group you’ve probably never heard of.

Now that we’ve gone over the different forms of evolution, we need to turn to how echolocation is used. The one thing that all bats with echolocation use it for is navigation through dark places, of course, but secondary functions have developed as well. For most echolocating bats, that secondary function is to detect bugs and eat them. Here’s a video with visual representations of how a bat ‘looks’ around an area with sound, comparing echolocation for navigation to echolocation used to chase a target.

The sound a bat makes to echolocate is altered based on almost innumerable factors. The frequency of a bat’s call, for example, corresponds to the size of its prey- smaller insects get higher frequencies. (The reason bats use relatively high wavelengths to hunt for insects is actually because the sound wavelength has to be shorter than the insect’s wing to be effective!) However, in areas with multiple bat species hunting in the same area, each species may stick to a narrow bandwidths of sound to avoid competition, sort of like having their own species-specific radio station. On the other hand, bats who fly in cluttered areas like forests rather than open areas are forced to use broader bandwidths with shorter call lengths in order to be able to differentiate the sound of echoes bouncing off of things like leaves and trees from the echoes returning from insects, in order to both chase prey and not crash into things.

These are just a few examples of how bat calls can differ; in fact, some near-identical species of bat can only be differentiated by listening to their hunting calls. Collecting bat calls is one of the most common ways to survey which species of bat are in an area, other than actually catching them.

Here’s a video of a bat using echolocation to hone in on the location of insect prey. Note how the bursts of sound come faster as the bat draws closer and needs more detailed information. (And listen for the distinct crunching noise of the bat capturing the moth.)

Insects, however, are not passive actors in all this. Indeed, bats and their prey have been involved in a technological arms race for millenia. Most moths and butterflies don’t have ears, and can’t hear. But certain species of moths happen to be the favorite food of certain species of bats, who hunt by echolocation. This drove the moths to evolve- ironically, in multiple different lineages- simple ears, so that they can hear the bats echolocation calls approaching and dodge.

In a surprising twist, moth ears evolved on their waists (thorax), not on their heads. In an even more surprising twist, tiny ear mites have evolved to attack the moth’s tiny ears; however, they only attack one of the moth’s ears so that the moth can still hear enough to save itself and its tiny hitchhikers from being eaten by a bat. That’s cool, I guess.

In response to the appearance of moth ears, some bats have ramped up their echolocation abilities in different ways. For example: some bats have simply changed the frequency of their echolocation to one the moths can’t hear. Other bats have evolved ‘whisper’ echolocation, i.e., echolocation so quiet that the moth’s simple ears can’t detect it. The ear shapes of these bats have changed to better reflect their own teeny cries back to them.

The “just turn your entire face into an ear, just do it” strategy, employed by this barbastelle bat.

The reason for this ghost-faced bat’s bizarre face flaps has not been studied, but it’s possible it evolved for a similar whispery purpose.

Other bats, though they use echolocation to detect the general location of their prey, actually shut up as they get close so the moths can’t detect where they are. Instead, these bats use super-sensitive hearing- even more sensitive than ordinary bat hearing, which is pretty damn sensitive- to pick up the tiny sounds of a moth’s wings moving, or even their feet moving along a surface like a leaf. The bats who use this strategy tend to be the ones with the most ridiculous ears.

Townsend’s big-eared bat. I mean, what else could you say about this guy. Ears.

One can assume that the weirder a bat’s face, the more specialized its insect prey has gotten, with a few exceptions. On the other hand, generalist beetle-eaters (most beetles can’t hear) like this big brown bat get away with relatively normal faces.

And look at those cute beetle-crushing teefers!

Bats that eat fruit or nectar only need to use echolocation for navigational purposes if they use it at all, so their equipment tends to be less specialized as well. However, a lot of herbivorous bats have still managed to evolve bizarre faces. Mainly for sex purposes. Because apparently, there’s nothing sexier than a faceful of horrific flesh flaps.

Wrinkle-faced bats; a rare member of the nose-echolocator group that doesn’t have a noseleaf. Not only do these bats have faces that look like like flesh-colored cottage cheese- possibly having something to do with shaping the sound of mating calls- the males have ‘skin masks’ that they can pull up over their chins, as seen in the photo. Maybe this is to better distribute the musky odor from their chin glands onto their face. Maybe it’s just to distress me. I don’t know.

Male hammerhead bats, which are non-echolocating Pteropodids, have enlarged snouts with weird nose flaps used for producing buzzes and honks that are apparently irresistible to females.

In any case, the ways that the weird faces of bats reflect and shape sound is surprisingly understudied, given how much diversity of form there is among the faces of bats. What we do know is that one of the most important tools for echolocation is a bat’s tragus, which is a fleshy protrusion within the ear. Humans have tragi as well, which some of you may already know if you have a tragus piercing.

Human and bat tragi.

Interestingly, non-echolocating Pteropodids don’t have tragi (see the ears of the hammerhead bat above), and instead rotate their ears to determine sound direction. But echolocating bats (sans Rousettus) keep their ears fixed in position, as seen in the video below.

Humans, too, have mostly immobile ears, which is why we have convergently evolved our own tragii to alter sounds based on the direction they’re coming from relative to our heads. But the variation in the shape of bat tragii- so diverse that a tragus shape can be used to identify differences between species with nearly identical characteristics- suggests that bats use them in much more complex ways than we do.

Diversity among bat tragii. Unfortunately, not much has been studied regarding how tragus shape affects sound.

Humans mainly use their tragii to determine whether a sound is coming from in front or behind us on a horizontal plane- the shape of the tragus causes sounds coming from behind to be slightly delayed. However, bats also use their tragii while echolocating to determine the vertical position of objects that are in front of them, based on the angle the sound bounces back at.

In one study, big brown bats were trained to choose between pairs of marbles suspended at different heights- the pair that was closer together would earn them a reward. However, when their tragii were glued down, the bats had a much harder time determining which pair was the correct one. Other studies have confirmed that altering the tragus makes bats struggle to hunt via sound, though they are surprisingly good at adjusting their behavior.

Other research has determined that wrinkles in the pinna (the rest of the bat’s outer ear) and on other parts of the face help bats detect the general direction of sound, though exact specifications on how all this works aren’t yet well-studied. What has been studied a bit more is how nose-echolocating bats use their noseleafs to shape sound as it comes out. This was discovered, funnily enough, when scientists thought it would be hilarious to cover the noseleaf of a horseshoe bat in petroleum jelly and record changes in the sound quality of its echolocation.

A horseshoe bat, like the one in the study. Side note, please don’t cover bats in jelly. Unless it’s a fruit bat. A fruit bat’d probably enjoy it. They’re kinky.

What the noseleaf appears to do is to split the bat’s echolocation beam into two parts- a broad, general spread that lets the bat have a wide field of sound-based ‘vision,’ as well as a narrow, focused beam that lets the bat focus in on things directly ahead. This enables the bat to retrieve information on the general environment it’s flying through at the same time as it is honing in on insect prey- important for horseshoe bats in particular, since they tend to fly very close to the obstacle-filled ground. Abilities like these are probably why bats with noseleafs are considered among the most sophisticated and skilled echolocators among all bats, able to detect tiny details like changes in the movement of an insect wing through sound.

Also, their echolocation sounds really cool on a bat detector.

There is a lot more stuff that’s been studied regarding echolocation in bats that I could discuss, but we’ve already gone on a bit long here. So let’s close up with a very recent discovery regarding bat echolocation that could possibly shake up the entire field: Pteropodids other than Rousettus species might actually be able to echolocate.

Eh? I can hear you saying. Didn’t I devote a good portion of this article to saying how Pteropodidae can’t echolocate? The fact that they rely on eyesight and are crepuscular, not nocturnal? How can this be a thing that was only recently discovered (in 2014)?

Well, as previously discussed, there are three well-documented ways that bats echolocate: laryngeal echolocation through the mouth, laryngeal echolocation through the nose, and tongue-clicking. Pteropodids definitely don’t use any of these methods. However, they are capable of navigating decently well in complete darkness, albeit not as gracefully as other bats, suggesting that they do have some form of aural navigation.

Like many ‘novel’ animal discoveries, this one came from local knowledge- a man on a bus in Indonesia mentioned to two visiting Israeli scientists that he knew of fruit bats that made clicking sounds with their wings. The scientists, upon trawling the available scientific literature, only found one reference to the behavior in pteropodids from a 1988 paper suggesting one species of bat (the cave nectar bat, a close cousin to Rousettus bats) ‘clapped’ its wings to make clicking sounds, but drew no conclusions as to how those sounds were used.

The scientists tested the cave nectar bat and two other pteropodid species- the lesser short-nosed fruit bat and the long-tongued fruit bat- via recording the sounds they made while flying in a small, completely dark room. Much to their surprise, all three species made audible short click-like sounds as they flew, and were able to avoid large obstacles, though they tended to crash into smaller ones like cables. When tested in light conditions, the bats clicked less than they had in the dark.

Considering the relationship these bats had to Rousettus, the authors wanted to ensure that the bats weren’t producing the sound with their tongues, so they sealed their mouths in some trials and even anesthetized their tongues. The bats still clicked. Video showed that the clicks matched perfectly with the bats’ wingbeats.

The authors were able to prove that this behavior was not, as the earlier study suggested, caused by wing ‘clapping’- when the wingtips of the bats were padded, the clicking sound could still be heard. The exact mechanism of how these clicks are produced is not yet known.

While none of the bats were particularly awesome at using this form of echolocation (they tended to need multiple tries to land using click navigation versus a single try to land using vision) they were definitely able to use it to differentiate between things like a solid block and a soft cloth. Interestingly, these bat species were all from different spots in the Pteropodid family tree, as shown in this figure from the paper:

The authors suggest that this could mean that all pteropodids could feasibly have this rudimentary form of echolocation, which would have evolved after the loss of laryngeal echolocation in this group. Or, some echolocating ability could be so useful for bats that they simply re-evolve it spottily based on their particular behavioral needs. As far as I know, no further studies have been published yet on this remarkable discovery, and the initial study was based only on nineteen different animals, so as yet it all remains up in the air. But it could change- again– the whole paradigm that scientists had finally started to settle on about how echolocation evolved in Chiroptera.

Bats, man.

References and Further Reading

Boonman, A., Bumrungsri, S., & Yovel, Y. (2014). Nonecholocating fruit bats produce biosonar clicks with their wings. Current Biology, 24(24), 2962-2967.

Chiu, C., & Moss, C. F. (2007). The role of the external ear in vertical sound localization in the free flying bat, Eptesicus fuscus. The Journal of the Acoustical Society of America, 121(4), 2227-2235.

Denzinger, A., Siemers, B. M., Schaub, A., & Schnitzler, H. U. (2001). Echolocation by the barbastelle bat, Barbastella barbastellus. Journal of Comparative Physiology A, 187(7), 521-528.

Henson, O. W. (1965). The activity and function of the middle‐ear muscles in echo‐locating bats. The Journal of physiology, 180(4), 871-887.

Houston, R. D., Boonman, A. M., & Jones, G. (2004). Do echolocation signal parameters restrict bats’ choice of prey. Echolocation in bats and dolphins, 339-345.

Jones, G., & Teeling, E. C. (2006). The evolution of echolocation in bats. Trends in Ecology & Evolution, 21(3), 149-156.

Jones, G., & Holderied, M. W. (2007). Bat echolocation calls: adaptation and convergent evolution. Proceedings of the Royal Society of London B: Biological Sciences, 274(1612), 905-912.

Li, G., Wang, J., Rossiter, S. J., Jones, G., Cotton, J. A., & Zhang, S. (2008). The hearing gene Prestin reunites echolocating bats. Proceedings of the National Academy of Sciences, 105(37), 13959-13964.

Müller, R., Lu, H., & Buck, J. R. (2008). Sound-diffracting flap in the ear of a bat generates spatial information. Physical review letters, 100(10), 108701.

Nikaido, M., Harada, M., Cao, Y., Hasegawa, M., & Okada, N. (2000). Monophyletic origin of the order Chiroptera and its phylogenetic position among Mammalia, as inferred from the complete sequence of the mitochondrial DNA of a Japanese megabat, the Ryukyu flying fox (Pteropus dasymallus). Journal of Molecular Evolution, 51(4), 318-328.

Simmons, N. B., Seymour, K. L., Habersetzer, J., & Gunnell, G. F. (2008). Primitive Early Eocene bat from Wyoming and the evolution of flight and echolocation. Nature, 451(7180), 818-821.

Spangler, H. G. (1988). Moth hearing, defense, and communication. Annual review of entomology, 33(1), 59-81.

Springer, M. S., Teeling, E. C., Madsen, O., Stanhope, M. J., & de Jong, W. W. (2001). Integrated fossil and molecular data reconstruct bat echolocation. Proceedings of the National Academy of Sciences, 98(11), 6241-6246.

Teeling, E. C., Scally, M., Kao, D. J., Romagnoli, M. L., Springer, M. S., & Stanhope, M. J. (2000). Molecular evidence regarding the origin of echolocation and flight in bats. Nature, 403(6766), 188-192.

Tsagkogeorga, G., Parker, J., Stupka, E., Cotton, J. A., & Rossiter, S. J. (2013). Phylogenomic analyses elucidate the evolutionary relationships of bats. Current Biology, 23(22), 2262-2267.

Wotton, J. M., & Simmons, J. A. (2000). Spectral cues and perception of the vertical position of targets by the big brown bat, Eptesicus fuscus. The Journal of the Acoustical Society of America, 107(2), 1034-1041.

Yovel, Y., Geva-Sagiv, M., & Ulanovsky, N. (2011). Click-based echolocation in bats: not so primitive after all. Journal of Comparative Physiology A, 197(5), 515-530.

Zhou, X., Xu, S., Xu, J., Chen, B., Zhou, K., & Yang, G. (2011). Phylogenomic analysis resolves the interordinal relationships and rapid diversification of the Laurasiatherian mammals. Systematic biology, syr089.

Zhuang, Q., & Müller, R. (2006). Noseleaf furrows in a horseshoe bat act as resonance cavities shaping the biosonar beam. Physical review letters, 97(21), 218701.