Skip to content
CiderHQ
Search

Fermentation

Keeving

Forming a floating pectin gel that lifts nutrients and solids out of the juice, so that the ferment starves before it finishes and leaves natural sweetness behind.

Known as Défécation where it originates. Also called Defecation, Keeve.

Stage
Fermentation
Traditional in
Pays d’Auge, Normandy, Cotentin, Domfrontais and 2 more
What it most changes
Sweetness up, alcohol down
Safety
Carries a safety consideration — see below
Safety — read this before doing it

A keeved cider is bottled deliberately incomplete: it holds fermentable sugar and a viable, starved yeast population, and it is stable only for as long as the yeast stays starved. Warmth, a nitrogen source or agitation can restart it. The pressures involved are not trivial and are now written down: the Traditional Welsh Cider specification puts a bottle-conditioned cider at 1.5 to 3.0 bar at 0 °C, and requires that any bottle intended for a secondary fermentation be heavy-duty, punted, closed with a 29 mm crown cap or a cork under a wire cage, and able to withstand a sustained internal pressure of 6 to 12 bar at 25 °C. A restarted keeve in a bottle built for still cider can exceed what the glass will hold. Producers use pressure-rated glass and secure closures, keep bottled stock cool, and confirm the ferment has genuinely arrested — a fall of under a point of gravity a week — before bottling rather than assuming it from the taste.

What it is

Keeving — défécation in French — is a juice clarification carried out before fermentation, in which the fruit’s own pectin is converted into an insoluble calcium gel that floats to the surface as a firm brown cap, carrying the suspended solids, colloids and much of the juice’s nutrient content with it. The clear juice between that cap and the sediment beneath is drawn off and fermented on its own. Because it has been stripped of nitrogen and of the solids yeast depends on, the fermentation that follows is extremely slow and eventually stops of its own accord with sugar still in the drink. Keeving is therefore not a clarification technique that happens to affect sweetness: it is the mechanism by which the classic sweet ciders of Normandy and Brittany are made without adding any sugar at all.

Keeving, step by stepHow pectin chemistry lifts a brown cap, strips the juice of nitrogen and leaves a fermentation that stops while sugar remains.Pectin methylesterase actsPectin loses its methyl groupsCalcium binds the bare pectinA calcium pectate gel formsThe gel traps yeast and riseschapeau brunclear juicelow in nitrogenlees fall outracked offSlow ferment stops sweet
How pectin chemistry lifts a brown cap, strips the juice of nitrogen and leaves a fermentation that stops while sugar remains.
Described in full
Step one: the enzyme
Pectin methylesterase, present naturally in apple tissue and sometimes added, acts on the pectin released by milling. Cool temperatures and a slow start are what allow it time to work before yeast takes over.
Step two: de-esterification
The enzyme strips methyl groups from the pectin chains. What was a soluble, methylated pectin becomes a bare polygalacturonic acid backbone carrying free carboxyl groups.
Step three: calcium bridges the chains
Calcium ions — from the fruit, or added as calcium chloride or carbonate — bind between adjacent bare chains. The chains cross-link into a network instead of staying dissolved.
Step four: the gel forms and rises
The cross-linked network becomes calcium pectate, a gel. It traps yeast cells, solids and the fine bubbles of the earliest fermentation, and the trapped gas floats the whole mass to the surface.
The *chapeau brun*
The floating gel is the brown cap. Below it the juice runs clear, and a layer of heavier solids falls to the bottom as lees. The vessel separates into three layers without filtration.
The nitrogen-stripped middle
The clear middle layer is the point of the whole operation. Rising and falling solids carry most of the assimilable nitrogen out of it, so what is racked off is juice that yeast cannot ferment quickly.
Arrested fermentation
Starved of nitrogen, the ferment runs slowly and stops of its own accord with fermentable sugar still present. That is how a cidre doux reaches natural sweetness and low alcohol without back-sweetening or sterile filtration.
Why it fails
Too warm, too much nitrogen, or fruit low in pectin and the gel never forms — the juice simply ferments out dry. Keeving is not a step that can be added to an ordinary ferment part way through.

Why it is used

How it works

What it changes

The direction this step pushes the finished drink in, dimension by dimension. A direction, not a measurement: how far it moves depends on the juice, the temperature and how the step is carried out.

Effect on each sensory dimension. Hover or focus a dimension name for what that dimension means on CiderHQ.
DimensionDirectionWhy
SweetnessRaisesNitrogen starvation halts the ferment before the sugar is used up, so the residual sweetness is sugar the yeast was never able to reach rather than sugar added afterwards.
AlcoholLowersA ferment that stops early converts less sugar to ethanol, which is why keeved ciders characteristically sit at a lower strength than fully fermented ones from the same juice.
AstringencyLowersThe floating gel removes oxidised and polymerised phenolic material along with the solids, so the juice that ferments carries less of the fraction that binds salivary protein.
BodyRaisesResidual sugar and the low alcohol of an arrested ferment together give a rounder, weightier palate than a dry cider of the same origin.
Fruit characterRaisesA very slow, cool ferment strips far less volatile aroma than a vigorous one, so more of the fruit’s own character survives into the finished cider.

The chemistry and the organisms

What is actually being changed, and by what. Each entry says what that compound or organism does in cider generally; this page is one place it does it.

Compounds involved

Organisms involved

Where it is traditional

The places this step belongs to as a matter of practice. It is not a claim of exclusivity — a method can be traditional in one region and perfectly ordinary in another.

What it is done with

What can go wrong

Faults that begin at this step, or that this step makes more likely. Each one is set out with its causes and whether it can be put right.

Styles it produces

Categories in which this step is characteristic or required. Some name it in their definition; for others it is simply how they have always been made.

More on keeving

The chain of events in a keeve is worth following link by link, because each one is doing something specific and the process fails if any of them is missing. Apple juice carries pectin, a long chain of galacturonic acid units whose carboxyl groups are mostly capped with methyl esters. In that esterified state the pectin is soluble and does nothing. The fruit also carries pectin methylesterase, an enzyme that strips those methyl caps off. As it works, free carboxyl groups appear along the chain, and these are negatively charged sites that a divalent cation can bridge. Calcium — present in the juice, and in traditional French practice supplemented as calcium carbonate or a prepared sel de défécation — links adjacent chains through those sites, and the whole juice sets into a soft, continuous gel of calcium pectate. That gel is not a precipitate; it is a network occupying the entire volume, and it entrains everything suspended in the juice.

What lifts it is fermentation. The juice’s own yeasts begin working slowly at cellar temperature, and the carbon dioxide they produce cannot escape freely through the gel. Bubbles lodge in the network, the gel becomes less dense than the liquid, and the whole mass rises to the surface as a firm cap several centimetres thick — the chapeau brun, brown because oxidised phenolic material has been carried up in it. Heavier debris drops out below as the vase. Between the two lies a bright, pale juice, and the maker’s job is to remove it without disturbing either layer, usually through a tap set at the right height and drawn off gently over hours. Timing is everything: too early and the cap has not formed properly, too late and it waterlogs, breaks up and falls back through the juice, undoing the whole thing.

The clarified juice is now deliberately impoverished. Most of its assimilable nitrogen — ammonium and amino acids alike — has gone up with the cap or down with the sediment, along with the suspended solids that yeast uses as nucleation sites and as a source of sterols and lipids. The removal is measurable rather than notional: published work puts the reduction in pectin and in amino nitrogen at no less than half, and the yeast population falls too, since cells are physically trapped in the rising gel. The mechanism behind the nitrogen loss is electrostatic, and it is the part most accounts of keeving leave out — pectin carries a negative charge while thiamine and asparagine, the major amino acid in apple juice, carry a positive one, so the nutrients are drawn into the gel and leave the system inside it. Yeast in that juice can ferment, but only slowly, and as the small remaining nitrogen pool is exhausted the population cannot maintain itself. Fermentation slows to a crawl over months and then stops with a substantial part of the original sugar unconsumed. This is the whole point. The sweetness of a cidre doux is sugar the yeast could not reach, not sugar returned to a dry cider afterwards, and the difference is audible on the palate: the drink is sweet at low alcohol, with the aroma of a ferment that was never hot or fast enough to strip it.

The preconditions are unforgiving, and they reach back into the orchard. Keeving needs juice that is low in nitrogen to begin with, which means fruit from unfertilised or lightly fertilised trees — nitrogen applied to the orchard floor arrives in the juice and defeats the mechanism directly. It needs pectin and active pectin methylesterase, which means fruit that has not been treated with a commercial pectinase, since those preparations depolymerise the pectin backbone rather than de-esterifying it and are therefore exactly the wrong enzymes. It needs cool temperatures, because the gel has to form before fermentation becomes vigorous enough to tear it apart, and it needs patience: keeved ciders take months, and traditional Norman practice adds successive gentle rackings to keep the yeast population suppressed all the way to bottling.

When a keeve fails it usually fails in one of three recognisable ways. The gel does not form at all, because the fruit was low in pectin or the enzyme was inactive or the juice too warm; the maker is left with an ordinary juice and can only ferment it out dry. The cap forms and then sinks, redistributing everything it had removed and leaving a murky juice with its nitrogen restored. Or — the most consequential failure — the keeve works, the cider is bottled sweet, and the arrest turns out to have been a pause rather than a stop. A keeved cider is a metastable thing by design, and its safety margin is the yeast’s hunger. In England the technique nearly died out in the twentieth century and has been revived since, largely from the French model, which is why English keeved cider today owes more to Norman practice than to any continuous domestic tradition. It is worth being precise about that, because keeving is often described as a French technique England borrowed. Seventeenth-century English writers knew the cap perfectly well and had their own name for it — the flying lees. What crossed the Channel in the modern revival was not the idea but the method, worked out again by people who had the French example in front of them.

Related processes

Steps that sit alongside this one, replace it, or depend on it having been done.

What people ask next

Questions readers ask about the things this page mentions. Each one goes to the section that answers it rather than to a page written to receive the question.

Where to go next

Sources

What this page rests on. Where a source is marked as registered rather than read, CiderHQ is recording that the body is authoritative on the subject without claiming to have worked through the document itself. See our evidence policy for what each state means.