Why Can Some Rocks Float on Water? The Strange Physics of Pumice

Drop a granite pebble into a bucket and it sinks before you can reach for a towel. Basalt sinks. Sandstone sinks. Even a fist sized chunk of obsidian, black and glassy, disappears beneath the surface in an instant. Rocks sink. That has been the rule for as long as anyone has been throwing stones into rivers.

Then someone hands you a piece of pumice. It looks like ordinary volcanic rock, gray or tan, rough to the touch, sometimes stained rust colored by iron. You toss it into a sink full of water expecting the usual splash and disappearance. Instead it bobs, sitting there indifferent to gravity, drifting across the surface like a bath toy.

The question of why can some rocks float on water sounds almost like a trick, since "rock" and "floats" seem to belong in different sentences. Pumice is not cheating physics, though. It is revealing a part of physics that a granite pebble never gets the chance to show.

Density Decides, Not Geology

A rock does not sink because it is a rock. It sinks because of a number: its density, the mass packed into a given volume, calculated as density equals mass divided by volume. Whether an object floats or sinks in water depends on how that number compares with the density of water itself, close to 1,000 kilograms per cubic meter at ordinary temperatures.

This comparison is the heart of Archimedes' principle, worked out more than two thousand years ago and still doing all the real work here. Any object submerged in a fluid experiences an upward push, a buoyant force, equal to the weight of the fluid it displaces. If the object weighs less than the water it displaces, buoyancy wins and it floats. If it weighs more, gravity wins and it sinks.

Nothing in that rule mentions granite, basalt, or "rock" as a category. Whether something floats has nothing to do with being classified as stone, and everything to do with bulk density, the average density of the whole object including any empty space trapped inside it. That single fact, more than anything about geology, is the real answer to why can some rocks float on water while most never stand a chance.

Two Different Densities Hiding in One Stone

Here is where pumice gets interesting. The solid material that makes it up, a volcanic glass usually rich in silica, is not light at all. Measured with no gaps or bubbles, that glass typically comes in around 2,350 to 2,450 kilograms per cubic meter, well over twice the density of water. Compress a piece of pumice down until every internal space disappears, and the resulting glass pebble would sink immediately.

Nobody is handed compressed glass, though. They are handed pumice, and pumice is mostly not glass at all. It is glass wrapped around an enormous volume of empty space, and that space, not the glass, decides whether the whole object floats.

This is the distinction that separates a real explanation from a shrug. The density of the solid material and the bulk density of the finished rock are two different numbers, and only the second matters for buoyancy. Pumice floats not because volcanic glass is light. It floats because so little of a piece of pumice is glass at all.

Building a Rock Out of Bubbles

To see where all that empty space comes from, go back to where pumice is born: inside a volcano, in magma still rising toward the surface. Silica rich magmas often carry large amounts of dissolved gas, mostly water vapor and carbon dioxide, held in solution by the enormous pressure of the rock above.

As magma rises, that pressure drops. Gas that was perfectly stable at depth can no longer stay dissolved, in much the same way carbon dioxide leaves soda the moment a cap comes off and pressure suddenly falls. Bubbles nucleate within the magma and begin to grow as the ascent continues.

If the magma were thin and runny, those bubbles would rise out and escape, the way bubbles leave a glass of soda within minutes. The magmas that produce pumice, though, are usually thick and viscous, and viscosity fights the gas trying to leave. Bubbles get packed against each other, deforming and multiplying, until the magma is more foam than liquid.

Eruption fragments this frothy magma violently, and the fragments cool so fast that the glassy structure locks in place before the bubbles can collapse or merge. This is essentially how pumice forms: a solid record of a foam that never had time to settle.

Life Inside the Vesicles

Volcanologists call these frozen bubble cavities vesicles, and the fraction of a rock's volume they occupy is its vesicularity, one way of describing pumice porosity. A rock with 70 percent vesicularity is, by volume, seventy percent hollow space and only thirty percent solid glass.

Measurements of pumice produced by dry, gas driven explosive eruptions commonly cluster in a striking range, roughly 60 to 93 percent vesicularity. Not every sample sits at these extremes, and vesicularity can vary considerably even within a single fragment, but the figures show how close pumice gets to the physical limit of how many bubbles a foam can hold before it falls apart.

Not all volcanic rock reaches anywhere near those numbers, and this is worth saying plainly: it is not true that every volcanic rock floats. Basalt, andesite, and pumice's coarser, denser relative scoria typically carry far less pore space, or pore space arranged very differently, and usually stay dense enough to sink exactly the way granite does. Pumice vesicles, in both their abundance and their arrangement, are unusual even among volcanic rocks.

That abundance of vesicles is the part of the story most people already sense: a rock riddled with holes has a low enough average density to beat water's roughly 1,000 kilograms per cubic meter, even though the glass itself never stopped being dense. Explaining why pumice is full of bubbles, though, only opens the harder question.

The Part That Air Bubbles Alone Cannot Explain

Vesicularity by itself does not fully explain why does pumice float for as long as it typically does. Vesicles are not sealed capsules stacked neatly inside the rock. Many are connected to each other, threading pathways through the entire stone, a property called vesicle connectivity or connected porosity, as opposed to isolated porosity, where a cavity has no path to the outside at all.

Volcanic glass is also readily wetted by water, meaning water spreads into and along glass surfaces rather than beading away from them. Put those two facts together and a real puzzle appears: if most of the cavities in pumice are connected, and water wets glass easily, why doesn't water simply invade the whole network and sink the rock within minutes of it touching the surface?

That is the question a simple "it has air trapped inside" explanation glosses over. It is also the one that actually explains how does pumice float for months rather than minutes.

Gas That Gets Trapped, Not Just Enclosed

The answer, worked out through laboratory studies using X-ray microtomography to look inside partially waterlogged pumice, involves something more subtle than sealed air pockets. As water moves into a connected pore network, it does not flood every passage evenly. Narrow constrictions between vesicles, called pore throats, can pinch off pockets of gas before water fully reaches them.

Surface tension at the boundary between water and gas creates a resisting pressure inside these narrow throats, a phenomenon known as capillary gas trapping. Where the throats are small enough, the trapped gas becomes cut off from the atmosphere and from other gas elsewhere in the rock, even though the pore network, on paper, is technically connected the whole way through.

That is essentially what happens inside pumice when it absorbs water: some passages flood completely, others get pinched off mid flood, and whatever gas survives stays trapped rather than simply enclosed. Studies have shown that the sizes of these trapped gas clusters follow patterns predicted by percolation theory, a branch of physics describing how connectivity emerges or breaks down across disordered networks. As long as enough gas stays trapped, the pumice's overall density, water and glass and gas combined, can remain below that of the surrounding water.

Why a Floating Stone Eventually Sinks

Floating pumice does not float forever, and the reason it eventually gives up is not that "the bubbles fill with water" in one simple step. Water keeps creeping into whatever connected pore space stays accessible, slowly raising the fraction of the rock's interior occupied by liquid rather than gas.

The gas trapped in isolated pockets is not perfectly sealed either. It diffuses outward through the surrounding water, molecule by molecule, escaping toward the atmosphere over a timescale set by how far it must travel and how much resistance the wet, porous rock puts in its way. As trapped gas dwindles, water saturation inside the pumice climbs, and the bulk density of the waterlogged rock climbs with it.

Eventually the balance tips. Once the combined mass of glass, residual gas, and absorbed water exceeds the weight of the water the object displaces, buoyancy runs out and the pumice sinks, sometimes abruptly. Why does floating pumice eventually sink on one particular afternoon rather than another comes down to a tangle of factors: total porosity, the degree of vesicle connectivity, the geometry and width of individual pore throats, how much gas got trapped in the first place, current water saturation, the size of the fragment, and the temperature of the surrounding water. Two pieces of pumice that look nearly identical from the outside can carry completely different internal architectures, and therefore completely different fates once they hit the water.

A Slightly Different Bar in the Ocean

The threshold for flotation shifts a little depending on where a piece of pumice ends up. Seawater is denser than fresh water because of its dissolved salts, commonly cited around 1,027 kilograms per cubic meter under typical oceanographic conditions, against roughly 1,000 for fresh water. A piece of pumice riding right at the edge of sinking in a lake would have a little more cushion in the open ocean, though the difference is modest next to the gap between solid glass and either kind of water.

When an Eruption Turns the Ocean Into a Conveyor Belt

A single floating stone explains one clast's behavior. How can volcanic rock float on the ocean at the scale of an entire eruption's worth of debris is really a story about timing and volume. Explosive volcanoes, especially those erupting underwater or near coastlines, can produce enormous volumes of pumice almost simultaneously, and when enough of it reaches the surface at once it can coalesce into a pumice raft, a floating mat of rock fragments sometimes stretching for kilometers.

Once formed, these rafts drift with ocean currents and wind, and because individual clasts can stay buoyant for a very long time, they can travel extraordinary distances. Pumice from the 1952 eruption of Volcán Barcena, off the coast of Mexico, was documented floating for at least 560 days and covering more than 8,700 kilometers before finally coming ashore or sinking. The 2012 eruption of Havre volcano, more than 900 meters below the surface in the southwest Pacific, produced a raft over 400 square kilometers that was tracked by satellite and later identified as the largest deep ocean eruption of its kind recorded in the past century.

These rafts of floating rock are not lifeless. Pumice floating on water offers a temporary surface that algae, barnacles, corals, and gastropods can settle on, letting small marine organisms cross stretches of open ocean they could never otherwise manage under their own power. It is a strange, accidental ferry service, run by rocks that should not be floating at all.

What Happens After the Rock Finally Loses

Sinking is less an ending than a change of chapter. Once pumice becomes waterlogged enough to lose its buoyancy, it starts behaving much more like ordinary sediment, settling through the water column and accumulating on the seafloor the way sand or ash normally would, though its comparatively low density and irregular shape can still shape how far and how gently it settles compared with denser mineral grains.

That transition matters to geologists piecing together a volcano's history from what it left behind. Layers of buoyant pumice mixed unevenly with denser material can complicate the record of how far an eruption's debris actually traveled, since some of it floated away entirely before settling somewhere far from the vent that produced it.

What still resists easy prediction is exactly when any particular piece of pumice will give up and go under. Researchers can measure porosity, map connectivity with X-ray imaging, and calculate roughly how fast trapped gas should diffuse away, yet the internal pore architecture of a given fragment, its private maze of connected channels, dead ends, and pinched throats, is invisible from the outside and different in every sample pulled from the same eruption. A rock that seems to defy gravity for a year, then vanishes beneath the waves on an ordinary afternoon, is still keeping some of its structure to itself.

YOU MAY ALSO LIKE

Does a Shadow Have Weight? The Strange Physics of Light and Darkness

Scientists Are Building Harry Potter's Invisibility Cloak

Scientists Found a Hard Drive That Survived a Million Years It Wasn't Made of Metal

Meet the Computer That Has Never Touched an Electron