Redox banding in sediment

Why do the microbes in a jar of pond mud sort themselves into coloured bands that nobody arranged?

Fill a glass tube with pond mud, add a little shredded newspaper and a lump of chalk, leave it on a windowsill for a month, and it stops looking like mud. It develops bands: black low down, a red or purple layer above it, sometimes a green one, and a brown film at the top. Nothing sorted them. The mud was stirred to uniformity on day one, and the organisms in it do not move to appointed places. So what puts a boundary where a boundary is?

Begin with what every one of these organisms is doing for a living. Respiration means moving electrons from a donor — here, the organic matter in the mud — to an acceptor. The donor is shared. The acceptors are not all alike: oxygen, then nitrate, then manganese and iron oxides, then sulfate, then carbon dioxide, in decreasing order of energy released per electron delivered. That ordering is thermodynamic bookkeeping, not biology; it would hold in a sterile flask.

Now ask what competition does with that ordering. Two organisms in the same mud draw on the same organic carbon. The one coupling it to oxygen conserves the most energy per unit of substrate, so it can keep growing at donor concentrations at which the sulfate reducer can no longer conserve anything at all. It does not politely go first — it draws the donor down below the level its rival needs. So the acceptors are consumed in sequence rather than in parallel, and the sequence is imposed by substrate competition, not by preference or by any signal passing between them.

Where does depth come in? Only through supply. Oxygen enters from the water above, and its only route downward is diffusion, which is a poor way to move anything far. The flux delivered across a layer of thickness L falls off as one over L, while the demand accumulated within that layer grows with L — so there is a depth at which supply and demand meet and below which no oxygen exists, and it scales roughly as the square root of the ratio of supply to consumption. Push the numbers with plausible values and you land in millimetres; push a microelectrode into real sediment and you measure millimetres. In a jar of nutrient-rich mud, oxygen is gone a few grains down.

So the top of the column is oxic, everything below is not, and the acceptor sequence maps onto depth. Sulfate reduction dominates the deep mud, and it produces hydrogen sulfide, which now diffuses upward. That is the second gradient, running the other way, and its meeting with the descending oxygen gradient is what makes the bands sharp rather than smeared: sulfide and oxygen do not coexist, so their overlap is a narrow front, and any organism whose living depends on having both must sit inside it. Colourless sulfur bacteria do exactly that, forming the pale film near the surface.

The coloured bands need a third ingredient the mud does not supply: light. Purple and green sulfur bacteria are photosynthetic, but with sulfide as the electron donor rather than water, so they need sulfide and light and, for the green ones, no oxygen. Only a jar on a windowsill has all three anywhere. The split between purple above and green below is not arbitrary either: the purple layer absorbs first, and the green bacteria beneath carry pigments that work on what is left. Each band shades the one under it and thereby defines it.

One more feature makes the pattern self-sharpening. Every population consumes the thing that limits it, which steepens the gradient it sits on, which narrows the region where anyone else can do the same job. The bands are not lines drawn in the mud; they are the thin regions where a particular combination of donor, acceptor and light is momentarily available, held in place by the consumption of the organisms occupying them.

The tidy ladder is a tendency, not a law, and three qualifications matter.

The zones overlap more than the textbook figure suggests. Sulfate reduction is routinely measured inside nominally iron-reducing sediment, and the thresholds at which one process actually gives way to another depend on the mineral form of the oxidant and on how easily the organic matter can be broken down — often more than on the free-energy ranking itself.

Much of the cycling is invisible in a standing profile. Sulfide produced at depth can be re-oxidised so fast that it never accumulates, so a chemical profile showing almost no sulfide is entirely compatible with vigorous sulfate reduction. Rates and standing stocks are different quantities, and only rate measurements distinguish them.

And the geometry is not always diffusive. Cable bacteria form filaments centimetres long that carry electrons from sulfide oxidation deep down to oxygen reduction at the surface, coupling two zones that never touch chemically. Where they are active the neat sequence is short-circuited outright — a reminder that "the bands are set by diffusion" is a statement about the usual case.

The account is testable and cheaply so. If sulfide from below is what the coloured bands live on, a column run on sulfate-free water should never develop them, however much light and organic matter it gets. If the oxygen gradient sets the upper boundary, bubbling air through should drive the sulfide front and the purple layer downward. The refuting observation is clean: coloured phototrophic bands appearing at the same depths in a sulfate-free column would mean the sulfur cycle is not what positions them, and the whole explanation would have to be rebuilt.

• How cable bacteria conduct electrons across centimetres of mud, and what that does to a standard depth profile.
• Why the same layering appears in stratified lakes at metre scale, and what changes when it does.

Every term the collection defines is gathered in the glossary.