Skip to content
NatureHQ

Senses and abilitiesability

Feeling the ground

An elephant call is low enough that much of its energy goes into the ground, where it travels faster and further than it does in air.

Sound travels through the ground as well as the air, and a great many animals read it there — spiders through their legs, elephants through their feet and trunks. Ground waves travel at a different speed from air waves, so the same event arrives twice.

Vibration is the sensory channel that gets left out of the list, and it may be the most widely used one on the planet. An insect on a plant stem lives in a world of vibration: courtship signals, rival signals, the approach of a predator and the struggle of trapped prey all arrive as waves through the stem, and the great majority of insect species that signal to each other at all signal this way rather than acoustically. A spider in a web is doing something similar with a structure it built, tuned by how it tensions the threads. The mammal cases are rarer and better known. An elephant’s calls are so low that a substantial part of their energy goes into the ground, where the wave travels faster than in air and further before it fades; elephants respond to played-back seismic calls with no airborne component, and they adopt a characteristic posture — leaning forward, feet spread, trunk laid on the ground — that would be a reasonable way to improve coupling if you were designing an animal to do this. What remains genuinely open is how the signal reaches the ear, whether by bone conduction up the leg or by receptors in the foot, and how much of the elephant’s response is to the ground wave rather than to the airborne part of the same call.

Developed coverage · 57% complete · reviewed 2026-09-02

What this page covers

Substrate vibration is used across an enormous range of animals — insects and spiders most of all, and among mammals in elephants, several rodents and the golden mole. It is probably the least studied major sensory channel relative to how widely it is used.

Often confused with: Hearing, which some of the same animals also do, through different organs and a different medium; Infrasound, which is airborne low-frequency sound and often travels alongside the ground wave

Quick facts

How common
The dominant signalling channel in insects, and under-studied everywhere
Why two arrivals
Ground waves travel faster than air waves, so one event arrives twice
Elephant posture
Leaning forward, feet spread, trunk on the ground — consistent with coupling
Touch, used like vision
The star-nosed mole examines a target in about eight milliseconds

Where this appears

Every organism below has been linked to this page because the evidence links them. Each one carries its own evidence, and its own limits.

The ground is a second channel

A vibration source on or near the ground puts energy into two media at once. In air the wave travels at about 340 metres per second and spreads in three dimensions, losing intensity quickly. In the ground it travels several times faster, spreads mostly in two, and — for low frequencies in the right substrate — carries further before it fades. For an animal able to detect both, that is not redundancy: the two arrivals are separated in time by an amount that grows with distance, which is in principle a rangefinder.

Elephant calls are low enough to travel through the ground as well as the air, and elephants given only the ground-borne version stop, turn towards it and lean forward.

Well supported

Good evidence backs this, though some details remain open.

Low-frequency elephant vocalisations couple into the substrate and propagate detectably over kilometres. Seismic-only playback elicits freezing, orientation and forward weight-shifting, consistent with detection through pedal mechanoreceptors and bone conduction.

Who this applies to
African elephants; the Asian species is less studied in this respect.
Studied in
Loxodonta africana
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

Seismic-only playback is the right manipulation and the responses are consistent. What holds confidence at moderate is that the behavioural measures are attention rather than comprehension, and the receptor pathway is inferred from anatomy.

Caveats

  • "Hearing through their feet" compresses a detection result into a claim about comprehension.
  • The receptor pathway is inferred from anatomy rather than recorded.
  • How much ordinary communication uses this channel rather than the air is unknown.

Still unanswered

  • Whether elephants extract identity or message from a seismic signal, or only presence and direction.

Last reviewed 2026-09-02

The evidence (2 studies)

How we know

Playing an elephant a call through the ground instead of the air

Elephant calls travel through the ground as well as the air. Do elephants notice the part that arrives underfoot?

Recorded elephant calls were reproduced not through a loudspeaker but through a device that shook the ground, so the vibration travelled through the substrate with no airborne sound to accompany it. Wild elephants at a distance were then watched for the behaviours that normally follow a call — freezing, turning towards the source, and the characteristic forward lean that shifts weight onto the front feet.

What happened

The elephants responded: they stopped, oriented towards the source, and leaned forward onto their front feet.

What it shows

That a communication channel exists which does not involve the air at all. Low-frequency calls couple into the ground and remain detectable kilometres away, and elephants act on the ground-borne version — which is the evidence behind the phrase about hearing through their feet.

What it does not show

It does not show that elephants extract the same information seismically as acoustically, or that they identify individuals or messages this way; the responses recorded are attention and orientation, not comprehension. The receptor pathway is inferred from anatomy rather than recorded. And nothing here establishes how much ordinary elephant communication uses this channel rather than the air.

The controls — what makes this evidence rather than a story
  • Seismic transmission only, so an airborne route is excluded rather than assumed away.
  • Control vibrations that were not conspecific calls, so a response cannot be to shaking as such.
  • Behavioural measures chosen in advance from what elephants do when they hear a call normally.

From Wild elephant (Loxodonta africana) breeding herds respond to artificially transmitted seismic stimuli

The honest limit on this material is the delivery. Playing a seismic signal without any airborne component is difficult — a shaker in the ground radiates into the air as well, and demonstrating that an animal responded to the ground wave alone requires showing the airborne part was below its threshold. The better experiments measure both and report both, and the ones that do not are the reason this subject is more often asserted than established.

Words used here
Substrate vibration
Mechanical waves travelling through a solid — ground, a plant stem, a web — rather than through air or water.
Seismic communication
Deliberate signalling through the substrate. Distinct from incidental vibration, which animals also read but nobody produced on purpose.

The animals that live in vibration

The mammals are the famous cases. The insects are the ones doing it constantly.

  • A spider in an orb web reads position and size from vibration, and adjusts the web’s tension — which changes what it transmits, making the structure part of the sensory system.
  • Treehoppers, leafhoppers and many other bugs court entirely through plant stems, in songs a person can hear only with a contact microphone clipped to the plant.
  • Some parasitoid wasps locate larvae inside wood by the vibration of their chewing.
  • Kangaroo rats and several other desert rodents drum with their feet, in patterns that are individually distinctive.
  • The golden mole, blind and living in sand, orients towards grass hummocks by the vibration the wind produces in them.

And at the far end of what mechanoreception can be pushed to do:

The star-nosed mole examines the world by touch through an organ built like an eye: a high-resolution region it aims at things, rapid movements to bring targets into it, and a huge share of its brain.

Well supported

Good evidence backs this, though some details remain open.

The mole’s twenty-two appendages carry approximately 25,000 Eimer’s organs, with one appendage pair functioning as a tactile fovea onto which objects of interest are brought by rapid movements analogous to saccades. The star occupies a disproportionate area of somatosensory cortex, and prey identification and consumption are completed in well under 250 milliseconds.

Who this applies to
One highly specialised species.
Studied in
Condylura cristata
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

Anatomy, cortical mapping and high-speed behaviour all agree, and all come from essentially one research programme on one species.

Caveats

  • The comparison with vision is an analogy supported by structure, not an identity.
  • One species, so specialised that generalising from it would be unwise.
  • Behavioural timings come from small numbers of individuals in one laboratory.

Still unanswered

  • Whether the tactile fovea arrangement occurs in any other tactile specialist, or is unique to this animal.

Last reviewed 2026-09-02

The evidence (2 studies)

The star-nosed mole belongs here because it makes the general point about touch that the rest of this page makes about vibration: a mechanical sense can be built into something that functions the way vision does in another animal — fast, spatial, with a high-resolution region it aims at whatever it is examining. The star is not a nose. It is a touch organ with a fovea.

Where else this appears

What is still unsettled

  • How does the vibration reach an elephant’s ear?

    Why it matters: Bone conduction up the leg and specialised receptors in the foot are both proposed, and they predict different sensitivities and different postures.

    What would settle it: Recording from the auditory pathway during controlled substrate stimulation, which the size of the animal makes very difficult.

  • Can an animal use the delay between the ground wave and the air wave to judge distance?

    Why it matters: The information is unambiguously present. Whether anything uses it is a separate question that has barely been tested.

    What would settle it: Playback with the two arrivals artificially separated by delays corresponding to different distances, and measuring where the animal goes.

  • How much insect communication is vibrational and simply unrecorded?

    Why it matters: Estimates suggest the majority of signalling insect species use substrate vibration, and it is invisible to anyone not holding a contact microphone.

    What would settle it: Systematic vibrational survey work, which is being done in a small number of groups and nowhere near broadly enough.

Claims about this, checked

Things people have heard, and what the evidence actually supports.

The research behind this page

4 studies, newest first. Each one has a page explaining what it found and what it could not show.

2011Philosophical Transactions of the Royal Society B

The sense of touch in the star-nosed mole: from mechanoreceptors to the brain

The star is a tactile organ of extreme resolution, with one small pair of appendages functioning as a tactile fovea that the animal moves onto anything of interest, and prey identification and consumption completed in well under a quarter of a second.

2006Behavioral Ecology and Sociobiology

Wild elephant (Loxodonta africana) breeding herds respond to artificially transmitted seismic stimuli

Herds responded to the seismic-only stimulus with alerting and defensive bunching.

2006Nature

Underwater "sniffing" by semi-aquatic mammals

The animals exhaled air bubbles onto objects and re-inhaled them at rates of up to about ten times a second, and followed scent trails underwater successfully; performance dropped when bubble behaviour was prevented.

1986Behavioral Ecology and Sociobiology

Infrasonic calls of the Asian elephant

Elephants produce powerful calls with fundamental frequencies below about 20 hertz, at the edge of or below human hearing, and these calls accompany social behaviour that had appeared silent.

This page is a stop on a longer route

A guided journey reads several subjects in a deliberate order, with an argument for why one follows another. You can join in the middle.

Where to go from here

Each of these follows from something on this page — a relationship in the evidence, a claim people ask about, or the next mechanism along.

How complete this page is, and what it is still missing

NatureHQ publishes its own gaps. This page is at 57% completeness against what we would call a finished subject, and was last reviewed on 2026-09-02. It carries 2 claims and answers 20 mapped search questions.

  • no research from the last few years is attached — check for newer work
  • The receptor biology — campaniform sensilla, subgenual organs, Pacinian corpuscles — is not described at all, and the mechanisms differ substantially between groups.
  • Vibrational signalling in spiders is summarised in one line despite being the best-studied case in the subject.
  • Anthropogenic vibration, which is a real conservation question, is absent.