Koumiss (pasteurized mare milk)

A milk drink with more than sourness

Koumiss is fermented mare milk, also written kumys or kumiss. It is a drink rather than a drained curd: pale, frothy and acidic, with alcohol and gas contributing to its character. Lactic acid bacteria and yeasts participate in the fermentation, making it a useful contrast with a dairy food described only as sour milk. Its texture, aroma and strength depend on the milk, culture and process used.

The pale appearance can hide substantial change. Sugar is consumed while acids, ethanol and other compounds develop. Froth can come from handling as well as gas, so the visible surface does not reveal how far each biological process has progressed. Nor does a mild taste demonstrate an absence of alcohol. The name covers drinks whose histories and production conditions differ, rather than one fixed strength.

The practical preparation here follows a technical handbook's flow diagram for pasteurized mare milk. It belongs beside the history of household koumiss, while remaining a separate manufacturing method. That choice makes the account specific: the reader can explore the wider food without silently combining a skin-vessel tradition, a laboratory study and a packaged-dairy schedule into an invented universal recipe.

A travelling witness describes milk, foals and a skin vessel

William of Rubruck's account of his thirteenth-century journey through Mongol lands describes a drink he calls cosmos. In the translated passage, foals are brought near the mares to facilitate milking. Milk is collected in a large skin vessel and repeatedly beaten with a wooden implement. He describes foaming and souring, then a drink with a pungent bite and an aftertaste he compares with almond milk.

The account ties the drink to a seasonal horse economy. Rubruck presents it as especially important in summer and describes milk supplies reaching an elite household. Production required animals, access to their milk, vessels and repeated human work. Fermentation was one part of that system, rather than a container left apart from the daily management of horses.

His language also records an encounter between the food and a visitor. Wine and almond milk supplied comparisons familiar to him; they are descriptions, not modern analytical results. The text demonstrates a recognisable fermented-mare-milk practice in the world he visited. It cannot establish the drink's first invention, and his physiological explanations do not become current dairy science.

Earlier mare-milk evidence asks a different question

Herodotus describes Scythian mare milking and milk worked in wooden vessels. His account includes agitation and removal of the upper portion, which was valued above what settled. It is evidence for milk handling and separation. It does not name koumiss, describe a fermentation period or explicitly say that the milk soured into an alcoholic drink.

The distinction is small in wording but important to the history. “People used mare milk” and “people made this fermented drink” are different claims. Archaeological milk evidence, a written description of skimming and an eyewitness account of souring answer different questions. Placing them together can show the depth of human engagement with horse milk without pretending that each supplies a recipe for koumiss.

Rubruck's later description is more explicit about a fermenting drink, while Herodotus' passage provides an earlier dairy scene. Neither should be stretched into a precise birthplace. The food's history becomes more legible when the operations actually described—milking, agitating, skimming, souring and drinking—remain visible, rather than disappearing behind a single confident origin date.

Starter-sharing connected households to the next batch

Historical accounts from Siberia describe another way milk, access and fermentation were linked. G. F. Müller's notes, reproduced in Science First Hand, describe fresh mare milk entering a leather vessel with starter from old koumiss, followed by repeated stirring. The account also records repeated additions of fresh milk and drawing off fermented drink. This is a continuing household process, unlike a completely isolated batch with a fixed beginning and end.

Access to starter could be social as well as practical. The reproduced notes describe households with fewer animals obtaining ferment from wealthier neighbours, and differences in how long households could continue production. Milk supply and an active starter were both resources. The vessel, the next milking and the people sharing culture helped connect one preparation with another.

Modern research calls this carryover technique back-slopping. A portion of an earlier fermentation seeds new milk with a microbial community. That explains its biological function without assuming that every starter is identical. A culture carried through repeated batches responds to its surroundings and handling. Household inheritance can preserve a useful process while allowing variation in the organisms and flavours that develop.

Bacteria and yeasts change different parts of the drink

Lactic acid bacteria use milk sugar and produce acid. Yeasts contribute alcoholic fermentation, and carbon dioxide helps give koumiss its fizz. These activities can occur together, so a single vessel holds several routes of transformation. Increasing sourness does not quantify ethanol, and a lively surface cannot measure acidity. The sensory result reflects their combined effects.

A study of traditional and industrial koumiss in Inner Mongolia followed microbial communities and the changing milk during fermentation. Lactose decreased while lactic acid, ethanol and measured flavour-related amino acids increased. The changes explain why fermented milk is more than the original milk with an acidic ingredient added: organisms consume substrates and produce a different mixture of compounds over time.

The researchers also found differences between their production modes. The traditional batches had greater microbial counts and diversity and a different final chemical profile from the industrial batches. Several conditions differed together, including starting milk, temperature, scale and starter, so the result cannot be attributed to temperature alone. It demonstrates that “koumiss fermentation” describes an interacting system. Naming one organism or one incubation setting cannot fully specify the taste and composition of the finished drink.

Mare milk gives the culture a particular starting material

The choice of mare milk is central to the food. Milk from different animals differs in sugar, fat and protein composition, including the balance between casein and whey proteins. Mare milk has a lower viscosity than cow milk, making a direct substitution a poor way to reproduce its behaviour. A thin fermented drink and a firm cultured cow-milk gel do not begin with identical materials.

Lactose supplies a fermentable substrate, while the milk proteins and fat contribute to body and flavour. In the Inner Mongolia study, lactose remained after fermentation even as its concentration fell. Fermentation should therefore not be treated as a guarantee of lactose removal. Its extent depends on the particular community and conditions, not simply on whether the finished drink tastes sour.

Cow-milk koumiss adaptations are separate formulations. They adjust composition rather than merely relabel cow milk as mare milk, and some introduce other dairy components or sugar. Those procedures help explain the technical problem of substitution. They do not justify adding their ingredients to the mare-milk flow diagram used here. Preserving the starting milk's identity makes the practical recipe clearer and keeps distinct adaptations available for honest comparison.

Agitation belongs to both the vessel and the fermentation

Repeated movement is conspicuous across accounts of koumiss. A paddle in a skin vessel, a wooden shovel in an older account and controlled stirring in a technical method are different tools for working a liquid. Agitation distributes milk and starter and can introduce air. Its physical effects overlap with the conditions experienced by the organisms, but the movement itself is not fermentation.

Aeration supports yeast growth. This technical method divides stirring into an initial period and later intermittent movement. This makes attention part of the method: the maker is not merely waiting for a clock to run down. A different vessel or larger volume changes how milk moves, how air reaches it and how evenly conditions are maintained.

The historical descriptions should not be converted into a required number of paddle strokes. Their value is showing the work and the equipment involved. Likewise, foam is a visible consequence of a changing and agitated liquid, not an analytical endpoint. The souring, alcohol production and physical mixing must be understood together while keeping their measurements distinct.

A packaged process has more than one fermentation stage

In the technical flow used for this item, inoculation is followed by warm incubation and stirring, then a further stage after the milk has been packaged. Cooling comes later. The sequence matters because packaging does not automatically stop microbial activity. A filled container can still hold a changing drink, with acid, alcohol and gas continuing to develop.

Cooling changes the conditions for that activity; it does not erase what has already occurred. Acid and alcohol can continue to increase during storage and the product can become stronger. This connects quality at drinking with the history of the batch. A drink tasted earlier and the same batch held longer need not have the same balance.

The separate incubation stages also explain why a single total fermentation time would conceal useful information. The bulk vessel and the package are different settings, with different assigned conditions. Stirring in the former and subsequent fermentation in the latter give the process its sequence. The package must accommodate a drink that can continue producing gas; pressure and closure specifications are not given.

Household and industrial koumiss share a principle, not every condition

The Inner Mongolia comparison used back-slop starter in both traditional and industrial production. Industrial manufacture therefore need not mean abandoning previous-batch culture for a single pure organism. Conversely, using an inherited starter does not make two production modes identical. Milk treatment, environment and handling still shape the community that enters and develops in the vessel.

Source-tracking in that study associated the industrial community mainly with starter, while the traditional mode showed contributions from both starter and environment. This helps explain why cleaning, vessels and surroundings belong to an account of fermentation even when the recipe includes a deliberate culture. Those observations concern the sampled batches; the bulk starter in this preparation remains unspecified.

Kazakhstan's dairy sector provides a broader setting for coexistence. Household, small-enterprise and industrial dairying coexist, with seasonal supply and market access affecting production. A traditional fermented drink can remain associated with a region while also entering modern packaged production. Neither a factory line nor a household vessel alone represents every way people make and obtain koumiss.

Drinking, medicine and the stories attached to the food

Koumiss has been described as both an everyday food and an important drink offered within social life. Rubruck's account places it prominently in summer and describes its place around an elite household; Müller's reproduced notes connect production to household resources and season. These are particular observations, rather than a single serving custom shared by every population that ferments mare milk.

The drink also acquired a medical reputation. George L. Carrick's book, published in Edinburgh in 1881, is titled Koumiss, or fermented mare's milk, and its uses in the treatment and cure of pulmonary consumption and other wasting diseases. Its title records that historical medical framing. The presence of such claims in a period book does not establish that koumiss treats or prevents disease.

For the reader, the useful appreciation is culinary and historical: an acidic, gas-producing milk fermentation linked to horses, repeated work, seasonal supplies and shared cultures. Its strength and character can change with its making and keeping. That is enough to distinguish it as a food without attaching an unsupported promise of health or a universal ritual of consumption.

Limits of the practical flow

The retained method is a technical manufacturing chart using pasteurized mare milk. It supplies no absolute batch size, measured endpoint, package pressure specification, defined cooling rate or storage duration; the starter percentage has no stated basis. These omissions prevent it from becoming a complete validated home process merely by copying its temperatures. Keep these unresolved controls separate from the schedules of raw-milk traditions, cow-milk adaptations and laboratory batches.