If you need to identify a snake, try the Snake Identification Facebook group.
For professional, respectful, and non-lethal snake removal and consultation services in your town, try Wildlife Removal USA.

Tuesday, September 29, 2015

Can snakes hear?

Last month I wrote about whether snakes sleep, a topic that is far more interesting than the minuscule amount of research devoted to it. Another common question is whether snakes can hear, since they don't have external ear openings. The short answer is yes, snakes can hear, but the long answer is (as usual) more complicated. Happily, there is a good deal of research on this question, including a recent review. In general, many popular sources and some scientific ones have incorrectly claimed snakes to be deaf, whereas a plethora of behavioral, neurological, and physiological experiments, particularly those performed by the eminent Princeton hearing researcher Ernest Glen Wever in the 1960s and 70s, by UC-San Diego neurologist Peter Hartline in the 1970s, and by herpetologist and anatomist Bruce Young from the 1990s to the present, have conclusively shown that snakes can detect and respond to sounds.

Anatomy of the human ear
Most tetrapods have a three-part ear (outer, middle, and inner) that is useful for detecting airborne sounds. The boundary between the outer and middle ear is called the tympanic membrane or "ear drum", and its function is to convert airborne sounds from the outer ear into fluid-borne ones in the inner ear1, by way of one or more middle ear bones. Sounds are ultimately converted by auditory hair cells called stereocilia into nerve impulses, which travel to and are interpreted by the brain. At many stages along the way, the sounds are amplified by the vibrations they produce in the different parts of the ear, including the middle ear bones (more on these in a minute). It's been suggested that this three-part system evolved (possibly multiple times) around the beginning of the Triassic Period, in concert with the evolution of sound production in insects, the probable prey of many early amniotes. Many modern animals, such as songbirds, bats, dolphins, humans, frogs, and crocodilians, have very sensitive hearing that can detect extremely quiet airborne signals in spite of the presence of other competing noises.

Micro-CT scan of a ball python's skull and ear.
Red: mandible; dark blue: quadrate;
green: columella; purple/light blue: inner ear chambers

From Christensen et al. 2012
Click here for an interactive 3-D model.
You're probably familiar with the three bones of the middle ear in mammals, the malleus, incus, and stapes (also known as the hammer, anvil, and stirrup). Snakes and other reptiles have only a single middle ear bone, which is usually called the columella, although it is homologous with the mammalian stapes. The malleus and the incus evolved from the articular and quadrate bones in the lower jaw of early mammal-like reptiles, leaving modern mammals with a single lower jaw bone, the dentary. Modern reptiles still have three bones in their lower jaws, where they play a role in detecting vibrations, particularly those propagating through the ground. Most modern lizard ears are essentially like those of modern mammals, with a small external ear leading to a large ear drum close to the body's surface, which passes sound from the air (or the jawbones) to the columella and thence to the inner ear. In contrast, snakes lack all traces of an outer ear as well as an ear drum. Instead, a snake's columella is in direct contact with, and picks up vibrations from, its quadrate bone (the dark blue bone in the diagram above). You might suspect that this arrangement would only be useful for detecting ground-borne vibrations, and you'd be partially right: snakes are exquisitely sensitive to ground-borne vibrations. But, they can also detect airborne sounds.2

Diagram of the ear of a watersnake (Nerodia)
Modified from Wever 1978
Both older and several more recent experiments suggest that snakes can hear the vibrations produced by airborne sounds. Physiological data suggest that they are able to detect certain airborne frequencies directly using the inner ear, although the specific bioacoustic mechanisms remain poorly known. Instead, most airborne sounds are probably detected in using "somatic hearing". This happens when airborne sound waves strike a snake's body  and some of their energy is transferred to its bones, tissues, and organs, particularly the head and lung. The snake's vibration-sensitive hearing system can then pick up on and translate the vibrations from the rest of its body into fluid-borne vibrations and, ultimately, nerve impulses. So a snake probably can't hear, say, most music3 or human speech directly, but it can hear the sound of its own body vibrating in response to those sounds. So, instead of being deaf, snakes essentially have two auditory systems that are at least peripherally distinct. Whether signals from these two systems are integrated into a single neural pathway, as is the case for the eye and the pit organ, or whether they serve different functions, remains to be studied and determined.

The length and arrangement of the auditory hairs in the inner ears of snakes appears to be fairly uniform across species, at least relative to the variation seen in lizards, which can have very different auditory hair anatomy among families and often even among closely-related species. Snakes mostly have simple, tuatara-like papillae, which suggests that they have secondarily lost a more complex type of auditory organ. This might be due to the aquatic or burrowing lifestyle of their ancestors and/or to specializations of their lower jaws in response to their unusual eating habits. There is some variation in inner ear anatomy (and presumably in hearing capacity) among snakes: burrowing snakes have the longest papillae, arboreal snakes the shortest, and terrestrial snakes have papillae of intermediate length. Many mammals have over 10,000 auditory hair cells, whereas most snakes have only about 250 (although acrochordids have nearly 1,500). Supporting cells of unclear function are relatively more numerous in snakes and these cells have ultrastructural features that suggest that they are more specialized than those of other reptiles.

Hearing range of various animals, not including snakes
The louder and lower frequency airborne sounds are, the more easily a snake can detect them. This isn't entirely unlike our own hearing—although we do hear high-pitched airborne sounds directly more easily than snakes do, we also rely on amplification provided by our ear drums, inner ear hairs, and other parts of our bodies. Studies have shown that snakes can hear sounds in the 80-600 Hz range optimally, with some species hearing sounds up to 1000 Hz (for comparison, the range of human hearing is from 20-20,000 Hz). This means that a snake could hear middle C on a piano, as well as about one octave above and two below, but neither the lowest key (which is 27.5 Hz) nor the highest (which is 4186 Hz). The average human voice is around 250 Hz, which means that snakes can hear us talking as well. Of course, there is likely a lot of variation among snake species, and the hearing of most species has not been examined, so these are generalizations.

Use the player above to hear how the airborne parts of Led Zeppelin's classic "Good Times, Bad Times" would sound to a snake. Parts of the song below 80 Hz (some bass & drums) or above 600 Hz (almost all guitar, vocals, and cymbals) have been muted. This doesn't include their sensitivity to the groundborne vibration parts of the song, which you could simulate by turning the bass on your speakers all the way up.

Audibility curves for living reptiles, including birds (left). The lower
the curve, the quieter a sound can be detected at a given frequency.
You can see that snakes cannot hear very quiet sounds, but
otherwise are not that much worse than other reptiles
(although their hearing sucks compared to, say, owls).
Note the different y-axes. From Dooling et al. 2000.
What do snakes do with their hearing? Unlike frogs, birds, and insects, snakes don't seem to use sound for communication with each other. Although many snakes hiss and some use tail rattling, growling, scale rubbing, or cloacal popping to send messages to their would-be predators, these sounds are mostly above 2,500 Hz, so the snakes themselves cannot hear them. Some species are capable of producing sounds whose frequency overlaps with their hearing range, such as the loud, robust hisses of pinesnakes and gophersnakes (Pituophis), the bizarre and intimidating growling sounds of king cobras (Ophiophagus), and the famous rattles of some large rattlesnakes (Crotalus). Some people have suggested that rattlesnakes find their hibernacula by following the rattling sounds of other rattlesnakes, but this idea has been disproven because the power output of rattling is insufficient to serve as a long-distance signal, and playback experiments have not yielded a behavioral response to rattling.

Snakes might eavesdrop on the alarm calls of other, more vocal animals, as some lizards do with bird alarm calls, but probably not since most of these calls are between 2,500 and 10,000 Hz, well above their optimal frequency range. Most likely, snakes use their hearing to monitor their environment for sounds produced by approaching predators or prey, many of which are ground-borne vibrations. Snakes can hear in stereo and can use their hearing to determine the directionality and thereby the sources of sounds. One genus of snakes that probably relies quite heavily on vibration to hunt are Saharan sand vipers (Cerastes). These snakes ambush lizards and rodents from a position partially or completely buried in sand. Experiments have shown that their reliance on chemosensing and thermal cues was minimal and that, although snakes with their eyes obscured had altered strike kinematics, they were still able to capture prey.

1 This is necessary because "hearing" evolved under water. Many fishes and fully aquatic amphibians (such as amphiumas) have a network of hair-like cells all over their body, which is called a lateral line system. The lateral line allows them to sense water-borne vibrations using their entire body like one big eardrum. When early amniotes emerged onto land, the inner ear was still adapted to detecting fluid-borne vibrations, and the eardrum and outer ear evolved to facilitate collection of airborne sounds and translation of them into fluid-borne ones. These adaptations were further refined as amniotes began to hold their bodies off the ground (lizards, mammals) or fly (birds), minimizing their ability to pick up ground-borne vibrations with their ears. Snakes probably have a better capacity to pick up ground-borne vibrations than most amniotes, since at least some part of their body is in contact with the ground (or a tree) most of the time. To date, no one has examined hearing in fully aquatic snakes.

2 Many burrowing and aquatic amniotes have lost their external ear opening, because their need to detect airborne sounds is minimal, they can rely mostly on ground-borne vibrations, and their middle/inner ear could be damaged during burrowing or swimming if it was exposed. 
Amphisbaeneans and other lizards lacking external ears hear mostly ground-borne vibrations, which makes sense considering that many of them are fossorial and spend most of their lives with most of their bodies in contact with the ground. Amphisbaeneans have lost more of their airborne sound detection capacity than most burrowing lizards, in that, like snakes, they have also lost their tympanum and have their columella connected directly to their lower jaw (some naked mole rats have a similar jaw-middle ear connection and rely heavily on vibrational communication). One leading hypothesis suggests that snakes evolved from burrowing ancestors, and another suggests that they evolved from aquatic ancestors, so perhaps snakes lost and then regained an ability to hear airborne sounds. Other limbless squamates, such as pygopod geckos, specialize in making high-frequency vocalizations and have sensitive hearing to match.

3 At least two studies have investigated whether cobras can hear the music played by snake charmers, and concluded that cobras are responding to tactile and visual stimuli, not auditory.


Christensen, C. B., J. Christensen-Dalsgaard, C. Brandt, and P. T. Madsen. 2012. Hearing with an atympanic ear: good vibration and poor sound-pressure detection in the royal python, Python regius. The Journal of Experimental Biology 215:331-342 <link>

Clack. J.A. 1997. The evolution of tetrapod ears and the fossil record. Brain, Behavior, and Evolution 50:198-212 <link>

Dooling, R.J., R.R. Fay, and A.N. Popper. 2000. Comparative Hearing in Birds and Reptiles. Springer, New York, NY, USA <link>

Dooling, R. J., Lohr, B., & Dent, M. L. 2000. Hearing in birds and reptiles. Pp. 308-359 in Comparative Hearing in Birds and Reptiles. Ed. by Robert J. Dooling, Richard R. Fay, and Arthur N. Popper. Springer New York <link>

Friedel, P., B. A. Young, and J. L. van Hemmen. 2008. Auditory localization of ground-borne vibrations in snakes. Physical Review Letters 100:48701 <link>

Fuong, H., Keeley, K. N., Bulut, Y., & Blumstein, D. T. 2014. Heterospecific alarm call eavesdropping in nonvocal, white-bellied copper-striped skinks, Emoia cyanura. Animal Behaviour, 95:129-135 <link>

Hartline, PH. 1971. Physiological basis for detection of sound and vibration in snakes. Journal of Experimental Biology 54:349-371 <link>

Ito, R., & Mori, A. 2010. Vigilance against predators induced by eavesdropping on heterospecific alarm calls in a non-vocal lizard Oplurus cuvieri cuvieri (Reptilia: Iguania). Proceedings of the Royal Society of London B: Biological Sciences, 277:1275-1280 <link>

Köppl, C., Manley, G. A., Popper, A. N., & Fay, R. R. 2014. Insights from Comparative Hearing Research. Springer New York <link>

Manley, G. A. 2012. Peripheral hearing mechanisms in reptiles and birds (Vol. 26). Springer Science & Business Media <link>

Manley, G. A., & Fay, R. R. (Eds.). 2013. Evolution of the Vertebrate Auditory System. Springer Science & Business Media <link>

Wever, E. G. 1978. The Reptile Ear: Its Structure and Function. Princeton: Princeton University Press <not available online>

Wever, EG and JA Vernon. 1960. The problem of hearing in snakes. Journal of Auditory Research 1:77-83 <not available online>

Young, B. A. 1997. A review of sound production and hearing in snakes, with a discussion of intraspecific acoustic communication in snakes. Journal of the Pennsylvania Academy of Science 71:39–46 <not available online>

Young, B. A. 2003. Snake bioacoustics: toward a richer understanding of the behavioral ecology of snakes. The Quarterly Review of Biology 78:303-325 <link>

Young, B. A., & Aguiar, A. 2002. Response of western diamondback rattlesnakes Crotalus atrox to airborne sounds. Journal of Experimental Biology 205:3087-3092 <link>

Young, B. A., & Morain, M. 2002. The use of ground-borne vibrations for prey localization in the Saharan sand vipers (Cerastes). Journal of Experimental Biology 205:661-665 <link>

Young, B. A., N. Mathevon, and Y. Tang. 2014. Reptile auditory neuroethology: What do reptiles do with their hearing? Pages 323-346 in C. Köppl, G. A. Manley, A. N. Popper, and R. R. Fay, editors. Insights from Comparative Hearing Research. Springer, New York <link>

Creative Commons License

Life is Short, but Snakes are Long by Andrew M. Durso is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.


Dale Hoyt said...

Do you know if there is a way for the columella to be "uncoupled" from the quadrate when a snake is feeding? I would think that the movement of the jaws during ingestion of prey must set up a tremendous racket unless the columella didn't move as the quadrate. In humans the stapedial muscle pulls the stapes away from the inner ear while we are chewing our food, otherwise the sound would be deafening.

Very nice post!

Andrew Durso said...

Wow, great question Dale. It does seem likely that there is some sound-dampening mechanism during feeding, but I've been unable to find any research explicitly addressing the topic. The best I can find is that the connection between the quadrate and the columella is loose, and the quadrate is surrounded by muscles that move it during feeding. It's possible that some of these muscles move the columella away from the quadrate during feeding, but it doesn't seem that anyone's looked. You can see a 3-D model of a snake skull at rest here, but it can't be manipulated to show how the bones move during feeding.