What is actually fermenting in a cacao heap

Three broad groups take turns — yeasts, then lactic acid bacteria, then acetic acid bacteria — each changing the conditions in ways that favour the next. The beans are not fermenting; the pulp around them is, and the beans are being cooked and chemically altered by what that produces.

intermediate

If you read nothing else

Nobody adds anything. The succession happens because each group of organisms consumes something and produces something that suits the next group, and the bean is transformed by heat and acid arriving from outside it.

The thing being fermented is the pulp

This is the single most useful correction, and almost everybody gets it the wrong way round. The microorganisms are not working on the bean. They are working on the sweet, acidic pulp surrounding it.

What reaches the bean is the CONSEQUENCE: heat generated by microbial activity, acids produced in the pulp and soaking inward, and — critically — the death of the bean's own embryo. A living seed keeps its compartments separate. Once it dies, enzymes and substrates that were kept apart can meet, and the reactions that create flavour precursors begin inside the bean while the microbial activity continues outside it.

So fermentation does two jobs at once. It removes the pulp, which is a practical necessity, and it drives a chemical transformation inside a bean that no organism ever enters.

The succession, in order

YEASTS go first. The fresh pulp is sugary, acidic and low in oxygen — conditions that suit yeasts and exclude most competitors. They consume sugars and produce ethanol and carbon dioxide, and they secrete enzymes that break down the pulp's structure so that it begins to drain away as liquid. That drainage is what lets air into the mass.

LACTIC ACID BACTERIA follow as sugars fall and the environment changes, converting remaining sugars and citric acid to lactic acid. Their activity reduces the pulp's acidity in a specific way and further changes what is available.

ACETIC ACID BACTERIA arrive once air can reach the mass, which is why turning matters so much. They oxidise the ethanol the yeasts made into acetic acid, and that reaction releases a substantial amount of heat. The temperature of the mass climbs markedly, and the combination of heat plus acetic acid penetrating the bean is what kills the embryo and drives the internal changes.

Other organisms, including spore-forming bacteria and filamentous fungi, appear when a fermentation runs long or goes wrong, and their products are the off-flavours associated with over-fermentation.

What this explains about practice

Almost every fermentation technique makes sense once the succession is clear.

Turning introduces air, which is what the acetic stage needs — an unturned mass stays anaerobic in the middle and the succession stalls. Heap size and box depth control heat retention, and a mass too small loses heat faster than the reaction generates it. Banana leaves over a heap both insulate and admit some air. Draining pulp deliberately, before fermentation, changes how much sugar the whole process has to work with.

It also explains why fermentation is not a fixed recipe. The pulp's sugar content, the ambient temperature, the microbial population resident on the equipment and in the environment, and the bean genetics all differ by place, which is why durations that work in one region produce under- or over-fermented beans in another.

Under and over, and what each tastes like

UNDER-FERMENTED beans have not had enough heat and acid for long enough. The embryo may survive, the internal reactions do not complete, and the result in the finished chocolate is flat, excessively bitter and astringent, often with a raw or 'green' character. The classic visual sign is a slaty purple interior when a bean is cut, rather than the brown, fissured interior of a well-fermented one.

OVER-FERMENTED beans have gone past the useful window. Acetic character becomes harsh and vinegary, unwanted organisms produce ham-like, putrid or mouldy notes, and the beans can be visibly damaged.

Neither is recoverable downstream. Roasting can moderate volatile acidity slightly and conching can drive off some of it, but no factory step reverses an incomplete or spoiled ferment. This is why fermentation is repeatedly described as the stage where most of chocolate's flavour is decided.

Covered in this guide

Sources

  • Chocolate Science and TechnologyWiley-Blackwell(citation identity confirmed; passage not re-read)
  • The Science of ChocolateRoyal Society of Chemistry(citation identity confirmed; passage not re-read)