Summary

What the science shows

Honey is not a substance with a fixed composition, but the product of a biological process. What it contains depends on the resource collected — nectar or honeydew —, on the transformations carried out by the bees, on maturation within the colony, and subsequently on harvesting, processing and storage. Two perfectly authentic honeys can therefore differ substantially without one necessarily being of better quality than the other.

Its remarkable stability is based on well-understood physicochemical mechanisms: a high sugar concentration that reduces water availability, an acidic pH and several antimicrobial factors. Crystallisation is a natural process and changes this balance by increasing water availability in the remaining liquid phase. Ageing follows kinetics governed jointly by time and temperature, not by crossing a single threshold: there is no biological boundary at 40 °C.

From a nutritional perspective, honey is primarily a source of free sugars. General health benefits attributed to its consumption are not firmly established, and honey-specific health claims have not been accepted by EFSA. Some clinical applications are nevertheless documented in clearly defined situations, particularly for certain wounds or burns and for the prevention or reduction of oral mucositis in some oncological settings. These uses belong to a controlled therapeutic context and cannot be extrapolated to ordinary food-grade honey. Cough provides another example of why caution is needed: trials using a sweet placebo with sensory characteristics similar to honey suggest that the specifically pharmacological component of the effect is probably smaller than comparisons with no treatment might imply. One safety recommendation, however, is unambiguous: no honey before the age of twelve months.

Authenticity is likewise not a property that can be revealed by a single test. Each analytical method answers a defined question, and adulteration strategies adapt precisely to the detection methods currently in use.

What this means for beekeeping practice

Capping remains an excellent biological pre-screening indicator, but it does not mark the chemical end of maturation. When the situation is uncertain, measuring moisture content provides the most directly useful information. Three different levels must not be confused: 20% is the general legal maximum, below 17.5% is a quality target recommended by Agroscope, and 18.5% is the limit set by the apisuisse quality label. Moisture measured at extraction is not permanently fixed either: honey can reabsorb moisture from ambient air, and the extent of this effect depends in particular on the exposed surface area, ambient humidity and duration of exposure.

Crystallisation should be managed rather than fought, and with melezitose-rich honey decisions need to be taken early. Heat should be applied according to a clear purpose: for as short a time and at as low a temperature as the process allows, bearing in mind that every heating episode adds to the previous thermal history. Packaging also matters: container tightness, lid quality, material and protection from external odours all contribute to preserving the product.

Residue management starts before harvest — with authorised products, appropriate treatment timing and sound wax management — and communication should claim only what origin, analysis and the applicable legal framework can genuinely establish. Good beekeeping practice therefore does not consist of accumulating empirical rules, but of knowing which mechanism, measurement or legal framework is relevant to the decision at hand.

1. Honey – a biological product, not a fixed substance

Understand why honey is a variable biological matrix, and distinguish the biological, the analytical and the regulatory perspectives.

Honey is often described as a concentrated sugar solution produced by bees from floral nectar. This definition is useful, but it remains too simple to account for its actual nature. Honey is not a standardised substance with a fixed composition: it is the result of a collective biological process applied to a raw material that is itself highly variable. Its chemical profile and its physical properties depend on the botanical origin or on the honeydew source, on environmental conditions, on the processing work carried out by the bees and, after harvest, on the product's storage and treatment history. Two perfectly authentic honeys may therefore differ appreciably in water content, sugar profile, acidity, electrical conductivity, mineral composition, enzymes, aroma compounds or tendency to crystallise, without either being of better quality than the other (De-Melo et al., 2018; Almeida-Muradian et al., 2020).

1.1 Two resources, two signatures

The first source of this diversity lies in the material collected. Most honeys come from nectar, a sugary solution secreted by the nectaries of plants, whose composition varies with the plant species, the flowering stage and environmental conditions. Other honeys come largely from honeydew, that is, from sugary liquids associated with the activity of hemipteran insects feeding on plant sap.

This difference in origin is not merely botanical: it also influences the final composition of the honey. Honeydew honeys show, on average, distinctive sugar and oligosaccharide profiles, a higher electrical conductivity and often more minerals than many nectar honeys, even though the values overlap substantially between the groups (Pita-Calvo & Vázquez, 2017). From its very origin, therefore, honey carries the signature of the resource exploited by the colony.

But this raw material is not simply concentrated and then stored. During collection, transport and exchanges between workers, it undergoes a series of enzymatic and physical transformations. Part of the sucrose is hydrolysed into glucose and fructose, other sugars are modified or formed, the acidity changes and a large quantity of water is progressively removed. Glucose oxidase in particular contributes to the formation of gluconic acid and, under certain conditions, of hydrogen peroxide. The final product therefore has a composition and properties that no longer correspond to those of the starting nectar or honeydew. Speaking of honey as merely "concentrated nectar" thus overlooks an essential part of the biological work accomplished by the bees.

1.2 A store before it is a food

This transformation only makes full sense when the function of honey within the colony is considered. For the bees, it is above all an energy store intended to allow survival when outside resources become scarce or temporarily unavailable. A solution very rich in water would be poorly suited to this role: it would take up more volume and would be far more readily usable by microorganisms. The sugar concentration, the reduced availability of water, the acidity and several antimicrobial mechanisms instead turn the collected resource into a much more stable food. Several familiar properties of honey — its high viscosity, its hygroscopicity, its tendency to crystallise or its capacity to limit the growth of many microorganisms — thus make sense once they are placed within this storage function.

This history does not end at harvest: time, temperature, humidity, crystallisation and any liquefaction operations may still modify certain properties. A honey sample must therefore always be understood as the result of a continuous chain: origin of the resource → transformation by the bees → ripening → harvest → treatment and storage. This logic explains why the major European studies do not describe a universal average honey, but distributions of values and profiles characteristic of different honey types (Persano Oddo & Piro, 2004).

1.3 Biology, analysis and law: three ways of talking about honey

Several ways of talking about honey must also be distinguished. Biology seeks to understand how the colony transforms and preserves a food resource. Laboratory analysis measures characteristics such as water content, sugar profile, enzyme activity, electrical conductivity or hydroxymethylfurfural content. The law, finally, sets the conditions under which a product may be placed on the market as honey and defines certain composition or quality criteria. These three levels overlap, but they are not equivalent: a typical value is not necessarily a legal limit, a regulatory limit does not necessarily correspond to a biological boundary, and a quality marker is not automatically an indicator of food safety.

The same caution applies when speaking of "honey quality". The term may refer to its biological ripeness, its physical and microbiological stability, its chemical and technological quality, its sensory characteristics, its authenticity or indeed its food safety. A crystallised honey may be perfectly authentic and of high quality; a honey that complies with regulatory limits may have lost part of its aromatic freshness after prolonged storage; an unusual botanical profile is not in itself an indication of adulteration. There is therefore no single parameter capable of summarising every dimension of quality.

The thread running through this article starts from this simple idea: to understand honey, one must understand how it is produced and why its matrix behaves as it does. The following chapters therefore examine in turn its ripening, its composition, the methods used to measure it, the role of water and sugars in its stability and its crystallisation, the influence of botanical or honeydew origin, then the effects of storage, the microbiological properties, authenticity, nutritional questions, health risks and therapeutic applications. Honey then appears less as a chemically uniform product than as the final expression of a complex biological, physicochemical and environmental system.

2. From nectar or honeydew to ripe honey

Follow the transformation of nectar or honeydew into ripe honey and understand the role of enzymes, water removal and capping.

When the forager returns to the hive with nectar or honeydew, she is not yet bringing back honey. The material collected generally contains far more water, its sugar composition is different and its microbiological stability is markedly lower. The transformation that follows is a collective process during which the colony simultaneously alters the chemical composition and the physical properties of the resource. The workers therefore do not merely evaporate water: they progressively turn a perishable sugar solution into an extremely concentrated and durable food store. This ripening begins as soon as the material is collected and continues during exchanges between bees, deposition in the cells and the successive phases of concentration (Nicolson, 2009; Wright et al., 2018; Almeida-Muradian et al., 2020).

2.1 What the enzymes change

A first transformation already takes place during transport and trophallactic exchanges. Enzymes of mainly bee origin are incorporated into the collected material. Invertase, or α-glucosidase, notably hydrolyses sucrose into glucose and fructose. Other enzymatic reactions modify the sugar profile in parallel and contribute to the formation of di- and oligosaccharides. The final result is therefore not simply a more concentrated version of the initial nectar: the composition of the sugar mixture has changed. Glucose oxidase plays another important role by transforming glucose and contributing to the formation of gluconic acid and hydrogen peroxide. Gluconic acid contributes to the characteristic acidity of honey, while hydrogen peroxide will become, under certain conditions, one of the elements of its antimicrobial activity. These mechanisms will be taken up again later, since their importance depends strongly on the concentration of the honey and on any dilution.

2.2 The removal of water

The most striking physical transformation, however, remains the removal of water. A very dilute plant resource could not constitute a durable store: it would take up a great deal of volume and would offer favourable conditions for the development of microorganisms. The bees accelerate concentration by handling small quantities of liquid and thereby increasing their exchange surface with the air. Once deposited in the combs, the material continues to lose water from the open cells, while ventilation of the hive promotes the removal of water vapour. This process depends on the climate of the colony, on ambient humidity, on the volume of incoming nectar, on the comb surface available and on the activity of the workers. Ripening cannot therefore be reduced to a fixed duration: it results from a dynamic balance between intake, transformation, evaporation and storage.

In experimental MiniPlus colonies fed with controlled sucrose solutions, Eyer, Neumann and Dietemann (2016) showed that storage is dynamic and that material may be moved during ripening. The liquid removed from a cell could be less concentrated than the liquid subsequently deposited there, which indicates that these transfers take part in the concentration process. The authors also showed that a comb simultaneously contains cells at different stages: ripening honey is therefore a spatially heterogeneous matrix. These results illuminate the mechanism of ripening, while recalling that the experiment did not directly reproduce a natural nectar or honeydew flow.

2.3 Capping is not a chemical boundary

This observation is particularly important for understanding capping. In beekeeping practice, a high proportion of capped cells is rightly a useful indicator of advanced ripening. But capping does not correspond to a perfectly sharp chemical boundary. Eyer et al. (2016) observed that the final phase of ripening may still continue once capping has begun; conversely, honey held in some cells that are still open may already be highly concentrated. A distinction must therefore be drawn between the biological signal provided by the bee and the analytical measurement sought by the beekeeper, and it should not be assumed that every part of the comb has exactly the same water content or the same physicochemical state.

This distinction explains why the "ripeness" of honey may refer to several different realities. From the colony's point of view, it corresponds to the functional completion of a storage process. From the physical point of view, the quantity and availability of water are of particular interest. From the chemical point of view, the transformations of the sugars, the acidity or certain enzyme activities can be followed. Finally, from the beekeeper's point of view, the central question is often more pragmatic: can the honey be harvested and stored without excessive risk of fermentation or loss of quality? These dimensions are linked, but none of them alone constitutes a universal definition of ripeness.

This is why caution is needed with overly simple rules of the type "capped honey = ripe honey" or, conversely, "uncapped honey = unripe honey". Capping remains an excellent field criterion for assessing the general progress of ripening, but it works better as a biological pre-filter than as an analytical measurement. When conditions are atypical — a strong flow, high humidity, an early harvest, large proportions of open cells or visible differences between frames — a measurement of the water content provides information that is more directly usable. The ripening of honey is therefore better understood as a continuum: the bees progressively transform, concentrate and redistribute the resource until they produce a matrix stable enough to fulfil its storage function.

This dynamic conception of ripening forms the basis of the following chapters. It makes it possible to understand why two honeys harvested at the same moment may have different compositions, why water content alone is not sufficient to explain microbiological stability, and why the sugars determine not only the energy value of honey but also its viscosity, its crystallisation and its behaviour in storage. Ripe honey is therefore not simply "nectar without water": it is the result of a collective biological process that profoundly transforms the raw material even before the beekeeper intervenes.

3. The chemical matrix – a great deal of sugar, but not only sugar

Identify the main constituents of honey and distinguish their analytical presence from their quantitative, functional and nutritional importance.

The composition of honey reflects both the raw material collected and the transformations carried out by the bees. In quantitative terms, the picture is simple: honey is above all a matrix of sugars and water. Carbohydrates generally account for around four fifths of its mass, while water makes up most of the remainder. This apparent simplicity nevertheless conceals a great chemical diversity. Several dozen different sugars are found in it, together with organic acids, proteins, enzymes, amino acids, minerals, phenolic compounds and a multitude of volatile molecules. Their importance is not, however, proportional to the number of substances identified: a molecule detectable by a highly sensitive analytical method may be biologically interesting without representing a nutritionally significant quantity (Doner, 1977; da Silva et al., 2016; De-Melo et al., 2018).

3.1 The major sugars

Fructose and glucose dominate the carbohydrate fraction by a wide margin. In many honeys, fructose is the most abundant sugar, followed by glucose, but their proportions depend strongly on the botanical origin. This relationship matters well beyond flavour. Fructose is more soluble in water than glucose; the ratio between the two therefore helps to determine the tendency of a honey to crystallise. The sugars also influence the viscosity, the hygroscopicity and the glycaemic response of the product. Sucrose is usually far less abundant in a ripe honey, notably because it is hydrolysed by the bees' enzymes during ripening. A relatively high sucrose content may have several explanations — a particular botanical origin, incomplete ripening or, in certain contexts, an authenticity problem — but it does not on its own allow a decision between these hypotheses.

The carbohydrate fraction is not, however, limited to fructose, glucose and sucrose. Honey also contains various disaccharides, trisaccharides and oligosaccharides. Some already come from the nectar or the honeydew, while others may be formed or modified during transformation by the bees. Their profile sometimes constitutes a genuine signature of the honey's origin. This is notably the case for certain honeydew honeys, in which the oligosaccharides are generally more significant than in floral honeys. Melezitose is the most spectacular example: this trisaccharide can become abundant enough in certain honeydew flows to alter the physical behaviour of the honey radically. We shall return to it in the chapter devoted to honeydew honeys and melezitose.

3.2 Proteins, amino acids and enzymes

Alongside the sugars are compounds present in much smaller quantities but of biological or analytical importance. Proteins generally represent only a small fraction of the total mass. Part of them corresponds to enzymes introduced by the bees, notably invertase, glucose oxidase and diastase. Their interest is above all functional: they take part in the transformation of the nectar or serve as indicators of the ripening and the thermal history of the honey. Their presence does not mean that honey is an important nutritional source of protein.

Even taking a relatively high content of the order of a few grams of protein per kilogram, a 20 g portion of honey would supply only a few dozen milligrams. By way of comparison, the daily protein requirement of an adult runs to tens of grams. The enzymes of honey are therefore biologically interesting without being nutritionally significant as dietary proteins.

The same distinction applies to the amino acids. Proline is generally the dominant free amino acid in honey, and its determination has long been used as complementary information on ripening or authenticity. The values observed nevertheless vary with the botanical origin and the production conditions, and an empirical threshold must not be turned into a biological constant. From a nutritional point of view, the quantities of free amino acids remain low. Their main interest lies more in the characterisation of the honey and in understanding its origin than in their contribution to human protein intake.

3.3 Acids, minerals and phenolic compounds

The organic acids constitute another small fraction of particular importance for the properties of the product. Gluconic acid is generally the principal one. It results largely from the action of glucose oxidase on glucose and contributes to the acidity of the honey. Other organic acids are present in varying proportions and take part in the sensory profile and in chemical stability. Here too, their functional importance far exceeds their importance by weight: a few compounds present at low concentration can appreciably alter the pH, the aroma or certain chemical reactions without thereby constituting a significant fraction of the diet.

The mineral elements follow the same logic. Potassium generally dominates the mineral fraction, accompanied notably by calcium, magnesium, sodium, phosphorus and various trace elements. Honeydew honeys contain more of them on average than many floral honeys, which contributes to their higher electrical conductivity (Pita-Calvo & Vázquez, 2017; Solayman et al., 2016). But the expression "rich in minerals" would be misleading for most honeys. If a honey contains, as an order of magnitude, 0.1 % mineral matter, a 20 g portion supplies about 20 mg in total; even doubling this value, the contribution remains low relative to daily requirements, which run to hundreds or thousands of milligrams for the main minerals. The analytical presence of mineral elements is therefore important for characterising honey, but generally secondary in the diet.

Phenolic compounds and flavonoids attract particular attention because they may take part in colour, in certain antioxidant properties measured in the laboratory and in the botanical differentiation of honeys. Their concentration varies considerably from one honey to another, with dark honeys often showing higher contents than very pale ones. But the results are generally expressed in milligrams per kilogram of honey: even a honey relatively rich in phenolic compounds therefore supplies only a few milligrams in an ordinary portion.

This quantity may be chemically measurable and interesting without being comparable to the cumulative intake obtained over a day from coffee, tea, fruit, vegetables or other plant foods. Above all, a high polyphenol content or a strong activity in a DPPH or FRAP test does not in itself constitute proof of a clinical benefit in humans.

Vitamins are even less abundant. Several water-soluble vitamins can be detected in honey, but generally in trace amounts or at low concentrations. They testify to the complexity of the matrix and to its biological origin, without allowing honey to be presented as an important source of vitamins. Lipids too are present only in very small quantities and play practically no role in its energy value.

The volatile substances are an exception to this hierarchy. Quantitatively minute, they are essential in sensory terms: hundreds of aroma molecules have been identified in different honeys, and some contribute to recognising a floral origin or to distinguishing sensory profiles. Their low mass therefore says nothing about their importance — a few micrograms of an odorous compound can strongly influence the aroma, whereas several milligrams of a mineral remain imperceptible to taste.

The difference between analytical presence and quantitative importance will be essential when we come later to the effects of honey on health. The chemical richness of the product is real, but it must not be confused with nutritional richness across every category of nutrient. The mass of honey remains dominated by sugars and water; the minor components largely give the product its botanical, sensory, enzymatic and biological diversity. Understanding this hierarchy makes it possible to avoid two opposing simplifications: reducing honey to "sugar and water" or, conversely, attributing major nutritional importance to every molecule that modern analysis is capable of detecting in it.

4. How is honey measured – and what do the laboratory values really mean?

Understand what the main laboratory parameters measure, how they are obtained and what limits govern their interpretation.

The figures used to describe honey are not simply properties that could be read directly from the product: they depend on a measurement method, on a protocol, on a temperature, on a sample preparation and on a convention of interpretation.

This methodological dimension matters, because a water content, a diastase activity or a hydroxymethylfurfural (HMF) concentration only make sense if one knows how they were obtained and against what they are being compared. The BEEBOOK devoted to standard methods for Apis mellifera honey research today provides a particularly broad methodological synthesis, while the methods of the International Honey Commission (IHC), the Codex Alimentarius and the AOAC remain essential references for the practical analytical determinations (Almeida-Muradian et al., 2020).

4.1 What each parameter measures

Water content is a good example. In beekeeping practice and in many laboratories, it is generally determined indirectly using a refractometer: the refractive index of the honey is measured, then converted into a water content by means of an experimentally established relationship. The measurement appears simple, but its accuracy depends notably on the temperature, on the calibration of the instrument, on the homogeneity of the sample and on the quality of the sampling. A difference of a few tenths of a per cent may become important when a value lies close to a quality target or a regulatory limit. This explains why the correct use of the refractometer is a practical skill in its own right, developed in the dedicated article on this measurement.

Other parameters say less about the immediate physical state of the honey than about its history. HMF forms progressively during ageing and under the effect of heat; it is therefore used as a marker of storage and heat treatment. Several methods exist for determining it, notably spectrophotometric methods and liquid chromatography. The Winkler method, historically important, uses p-toluidine and is today less desirable for reasons of analytical safety; the White and HPLC methods are alternatives commonly used in the methodological reference works. A high HMF concentration may signal prolonged storage or substantial heat exposure, but it does not allow a precise past temperature to be reconstructed. Two honeys that have undergone different combinations of time and temperature may show comparable contents.

Enzyme activity provides another kind of information. Diastase, or amylase, has long been used as a quality indicator, since its activity decreases with time and heat. Invertase is often more heat-sensitive and can supply complementary information. But these parameters are not biological thermometers. Initial activities naturally differ according to the origin of the honey, and a low activity does not necessarily mean that a product has been strongly heated. Conversely, a compliant value does not on its own guarantee an unblemished history. Erban et al. (2021) even showed that foreign amylases originating from microorganisms could be detected in certain honeys: adding an exogenous enzyme can artificially increase the diastase activity without restoring the other characteristics of a degraded honey. A classic indicator therefore remains useful, but it is not on its own proof of authenticity or freshness.

Proline, the principal free amino acid of many honeys, is also used as a complementary parameter for characterisation. As with diastase, however, its interpretation requires knowing the nature of the value under consideration: a concentration frequently observed in certain honeys is not automatically a biological minimum, still less a universal legal limit.

Electrical conductivity illustrates another important methodological development. Older work often described ash content as an indicator of the mineral fraction. Conductivity has progressively taken a central place, because it is easier to measure and closely reflects the concentration of ions, acids and mineral substances. It is particularly useful for distinguishing many floral honeys from honeydew honeys, without being an infallible criterion in every case.

Determining the botanical origin likewise rarely rests on a single measurement. Melissopalynology examines the pollen grains present in the honey and remains a fundamental method for assessing floral origin. But the proportion of pollen does not directly reproduce the proportion of nectar: some plants produce a great deal of pollen that is readily represented in the honey, others far less. Interpretation therefore ideally rests on the combination of pollen, physicochemical and sensory data, sometimes complemented today by molecular or spectrometric methods. The harmonised work of the IHC on European honeys has shown precisely the value of this combined approach (Persano Oddo & Piro, 2004; von der Ohe et al., 2004).

4.2 Five types of numerical value

A general principle then emerges: the same numerical value may have very different meanings depending on the context in which it is used.

Confusing these categories is one of the main sources of error in discussions about honey. An average value is not a standard. A regulatory limit is not necessarily the boundary between a "good" and a "bad" honey. A quality target stricter than the law does not imply that a product falling between the two is dangerous. And a limit set to monitor freshness or technological treatment must not be turned into a toxicological threshold.

To these five categories a final source of misunderstanding must be added: measurement uncertainty. A laboratory result is never an infinitely precise representation of reality; there is a variability linked to the instrument, to the method, to the sampling and to the sample preparation, and this uncertainty becomes particularly important near a decision limit. The consequence for the beekeeper coincides with the distinction set out above: handling the instrument, calibration and the representativeness of the sample belong to practice, but the values from which a honey is considered compliant, stable or problematic must come from scientific or regulatory references, and not from professional habit.

5. Water, water activity and crystallisation – the physics of honey stability

Relate water content, water activity, hygroscopicity, viscosity and crystallisation to the physical and microbiological stability of honey.

The stability of honey depends closely on water, but not solely on the total quantity of water it contains. Two honeys with the same water content may behave differently according to their sugar composition, their botanical origin, their liquid or crystallised state and the way the water is bound within the matrix. This is why two complementary notions must be distinguished: water content, which expresses the total proportion of water present in the product, and water activity, which indicates what fraction of that water is effectively available for chemical reactions and above all for the growth of microorganisms. Water content remains the most practical parameter for the beekeeper and for quality control, but water activity makes it possible to understand more precisely why a honey is stable or, on the contrary, becomes vulnerable to fermentation.

Water activity is a thermodynamic quantity between 0 and 1. In simplified terms, it corresponds to the ratio between the vapour pressure of water above a food and that of pure water at the same temperature. In a honey highly concentrated in sugars, a large part of the water interacts with the dissolved molecules and is therefore not as available as it is in free water. This is one of the main reasons for the remarkable microbiological stability of ripe honey: most bacteria cannot multiply in it, and even many yeasts and moulds are strongly limited.

Certain osmophilic yeasts, notably of the genus Zygosaccharomyces, are nevertheless capable of growing at exceptionally low water activities. Values around 0.61 to 0.62 are often cited as the lower zone in which the most tolerant species can still develop, but this is not a universal boundary: the outcome depends on the yeast species and strain, on the temperature, on the initial cell count, on the storage time and on the composition of the honey (Beuchat, 1983; Chirife et al., 2006; Snowdon & Cliver, 1996). Sorption trials carried out on pine and citrus honeys illustrate this gap: visible yeast growth at the surface appeared there only above a water activity of 0.7, that is, well above the theoretical minimum for growth (Arslan & Turhan, 2022).

This distinction also explains why water content does not allow a single water activity to be predicted mechanically. The two parameters are closely correlated, but the relationship varies between honeys. Gleiter, Horn and Isengard (2006), working on a series of 249 samples, notably observed that at comparable water content, liquid honeydew honeys could show a higher water activity than floral honeys. This does not mean that forest honey is generally more exposed to fermentation. The finding is more precise: at equal water content and under comparable conditions, the composition of a honeydew honey may leave a slightly greater fraction of water available. Water content therefore remains an excellent practical indicator, but it is a proxy for a more complex physicochemical phenomenon.

5.1 Hygroscopicity and viscosity

Water is also involved in the hygroscopicity of honey. Honey can absorb or lose water until it tends towards equilibrium with the relative humidity of the surrounding air, and this equilibrium lies lower than is often assumed. A honey containing about 82.4 to 82.5 % dry matter equilibrates with air at about 58 % relative humidity at 21 °C, and at 54.4 % at 34.5 °C (Doull & Mew, 1977). Australian unifloral honeys absorbed water from about 68 % relative humidity and lost it below about 51 % at 30 °C (Yao et al., 2003). Since the air in a dwelling frequently exceeds these values, honey left in the open tends to gain water; no temperature difference is necessary for this.

Sorption begins at the surface, after which the water diffuses slowly inwards until the water activity equalises (Arslan & Turhan, 2022). The effect is therefore greatest where exchange with the atmosphere is most direct, and it depends strongly on the exposed surface area and on the duration of exposure. This property explains the importance of storage in well-sealed containers and of careful handling of supers and extracted honey. It also helps to explain why an initially stable honey may become locally more humid without water having been directly added.

The very high sugar concentration also explains the exceptional viscosity of honey. Viscosity increases sharply as temperature falls and as water content decreases. A cold, dry honey may become extremely difficult to pump, strain or mix, whereas a moderate increase in temperature makes it much more fluid. Viscosity is not, however, merely a question of handling: it also acts on the diffusion of molecules and therefore on the rate of crystallisation processes. At low temperature, the solubility of glucose decreases and supersaturation increases, which thermodynamically favours crystallisation; but at the same time, viscosity increases and slows molecular movement. The rate of crystallisation therefore results from the combination of these two opposing effects.

5.2 A solution supersaturated with glucose

Liquid honey is indeed generally a solution supersaturated with glucose: it contains more dissolved glucose than could remain durably in solution at equilibrium. This state is metastable — the honey may remain liquid for a certain time, but crystallisation becomes thermodynamically favourable.

It begins with nucleation, that is, with the formation or presence of minute structures capable of serving as a starting point for the crystal. Microcrystals already present, but also certain suspended particles, can facilitate this process: pollen can thus provide nucleation surfaces, but it would be too simple to state that "pollen causes crystallisation". The final rate depends mainly on glucose supersaturation, on temperature, on viscosity and on the number of crystal nuclei available.

5.3 Why crystallisation may favour fermentation

When glucose crystallises, it forms mainly glucose monohydrate. Part of the glucose then leaves the liquid phase and organises itself into a crystal lattice, incorporating one water molecule per glucose molecule. The composition of the remaining liquid phase changes at the same time: it becomes relatively richer in fructose and in water compared with the glucose still in solution. This is an essential point, because crystallisation does not only alter the visible texture of the honey; it also alters the properties of the liquid fraction remaining between the crystals. Zamora and Chirife (2006), and subsequently Gleiter et al. (2006), showed that water activity generally increases after crystallisation, often by a few hundredths of a unit. In the study by Gleiter et al., the mean difference was about 0.04 between the crystallised state and the same honey after complete dissolution of the crystals.

This increase does not automatically turn a stable honey into a fermentable one. It becomes important above all for honeys that are already close to a critical zone. A honey with a very low water activity may crystallise while remaining well below the conditions favourable to osmophilic yeasts; in a more humid honey, by contrast, crystallisation may reduce the safety margin of the liquid phase. This mechanism helps to explain certain cases of fermentation observed in partially crystallised honeys or in honeys showing separation between a crystallised layer and a liquid phase richer in water. Crystallisation is therefore not the general cause of fermentation; it may, under certain conditions, create a liquid phase more favourable to yeasts.

5.4 What governs the rate of crystallisation

The tendency to crystallise depends strongly on the sugar profile. The fructose/glucose ratio, often written F/G, is an empirical indicator: the higher the proportion of fructose relative to glucose, the more the honey generally tends to remain liquid. Thresholds such as 1.1 or 1.3 appear frequently in the literature, but they must not be interpreted as natural boundaries. The glucose/water ratio, or G/W, can also provide useful information: a honey rich in glucose and relatively low in water has more glucose in excess of its solubility and will in principle crystallise more readily. Here too, the values proposed in the literature — for example around 1.7 or 2.0 — are empirical categories and not absolute rules. The botanical origin, the oligosaccharides, the temperature, the suspended particles and the history of the product also contribute to the outcome.

Temperature plays a particularly interesting role because its effect is not linear. At relatively high temperature, glucose remains more soluble and the driving force of crystallisation decreases. At very low temperature, supersaturation increases but the high viscosity slows molecular movement and crystal growth. Between these two extremes lies a zone in which nucleation and growth are particularly favoured: for many honeys, crystallisation is rapid within an approximate range of 10 to 15 °C, and several experimental studies place an optimum around 14 °C. This value must not, however, be turned into a universal constant, since different sugar profiles and different water contents shift the kinetics.

The difference between fine and coarse crystallisation is precisely what the production of creamed honey relies on. Rather than seeking to prevent indefinitely a thermodynamically natural process, the beekeeper can seek to control it. Introducing a small quantity of honey that already has fine crystals provides numerous nucleation seeds; crystallisation is then distributed over a large number of sites, which favours the formation of smaller crystals and a more homogeneous structure. Temperature and stirring subsequently influence the growth and distribution of these crystals.

Taken together, these phenomena show why a single water content value cannot on its own describe the stability of honey — without this taking anything away from its practical usefulness. Measuring water activity directly requires specific equipment and is generally not necessary on a beekeeping operation. Water activity provides above all the mechanistic understanding: it explains why two honeys with the same water content may not be exactly equivalent, why crystallisation alters the liquid phase and why microbiological stability never rests on a single figure.

The physics of honey can thus be summarised as a system in permanent equilibrium between water, sugars, temperature and phase state. The quantity of water determines a large part of the stability, but its availability depends on the composition of the matrix. Glucose largely determines the tendency to crystallise, but the rate of this process depends on temperature, on viscosity and on the seeds present. Finally, crystallisation in turn transforms the composition and the water activity of the liquid phase. Understanding these interactions makes it possible to move beyond simplistic rules and to explain why one honey may remain stable for years, while another, seemingly close to it, crystallises rapidly, separates or begins to ferment.

6. Floral honeys, honeydew honeys and melezitose – origin also determines behaviour

Understand how floral or honeydew origin influences the composition and the behaviour of honey, with melezitose as a special case.

The origin of honey does not determine only its colour, its aroma or its taste. It also influences its chemical composition and, consequently, its physical behaviour. The most important distinction sets honeys derived mainly from floral nectar against honeydew honeys. In the first case, the bees collect directly a plant secretion produced by the nectaries; in the second, they exploit sugary liquids associated with the activity of hemipteran insects feeding on plant sap. Both routes lead to honey after transformation by the bees, but they do not start from the same raw material. The initial differences in sugars, minerals and other constituents are therefore found again, at least in part, in the final product (Pita-Calvo & Vázquez, 2017; Seraglio et al., 2019).

6.1 Two groups that overlap

At the scale of large sets of samples, honeydew honeys generally show a higher electrical conductivity than many floral honeys. This property is linked to their greater content of ionic substances, notably minerals and organic acids, and explains why conductivity is a useful analytical criterion for characterising them. Their pH is also often higher, even though they remain acidic foods. Honeydew honeys are frequently darker and may show higher contents of certain phenolic or nitrogenous compounds. These tendencies are well documented, but they do not form absolute boundaries: some dark floral honeys, such as chestnut honeys, may show characteristics close to those of certain honeydew honeys, while the distributions overlap for several parameters (Pita-Calvo & Vázquez, 2017).

The sugar profile is particularly important, because it directly links the origin of the honey to its physical properties. Floral honeys are generally dominated by fructose and glucose, but their proportions vary strongly with the plant species. Honeydew honeys also contain these two monosaccharides, but often show a higher proportion of di-, tri- and oligosaccharides. Their composition reflects at once the plant sap, the metabolism of the honeydew-producing insect and the subsequent transformations carried out by the bees. This succession of stages explains why certain sugars may become useful markers of origin while having direct consequences for the viscosity, the solubility and the crystallisation of the product.

The major harmonised studies of European honeys show that it is therefore more appropriate to speak of characteristic profiles than of fixed values. A unifloral honey is not defined by a single molecule or a single figure, but by a coherent combination of pollen, physicochemical and sensory criteria. The International Honey Commission programme thus showed that the main European unifloral honeys present characteristic distributions for parameters such as conductivity, sugars, acidity or certain enzyme activities, with nevertheless zones of overlap between honey types (Persano Oddo & Piro, 2004). Botanical origin therefore cannot be read like a simple chemical barcode: it is reconstructed through the convergence of several indications.

The case of melezitose illustrates this relationship between origin and behaviour particularly well. Melezitose is a trisaccharide encountered above all in certain honeydews. Its presence is not characteristic of all honeydew honeys, and its concentration may vary strongly with the producing insects, the host plants, the season and the conditions of the flow. It would therefore be incorrect to present honeydew honey as a honey systematically containing a high proportion of melezitose. In certain situations, however, this sugar becomes abundant enough to alter the crystallisation of the honey profoundly.

This particularity stems from its solubility properties and from its interaction with the rest of the sugar matrix. When the proportion of melezitose is high, crystallisation may begin very rapidly, sometimes already in the cells of the comb. The honey then loses its fluidity before the beekeeper can extract it normally. The crystals form throughout the mass of the honey and make the contents of the cells extremely difficult to expel by centrifugation. It is this situation that is commonly designated by the terms "cement honey" or melezitose honey. It shows in particularly spectacular fashion that a change in a few elements of the carbohydrate profile can transform not only the texture of the finished product, but also the entire logic of harvesting and processing.

6.2 Melezitose – when sugar chemistry becomes an operational problem

The causal chain can be summarised as follows: honeydew source → distinctive sugar profile → high melezitose content → altered solubility and supersaturation profile of the matrix → strong tendency to rapid crystallisation → crystallisation in the combs → difficult or impossible extraction. The relatively limited solubility of melezitose and its interactions with the other sugars contribute to this behaviour, but the kinetics depend on the matrix as a whole. The presence of melezitose is neither an adulteration nor a defect of ripening: it is a natural characteristic of certain honeydew flows. For the beekeeper, the problem is first of all technological — a perfectly natural honey of good quality may become practically impossible to extract by the usual methods — and, as we shall see later, stores very rich in melezitose may in addition pose a physiological problem for the bees during overwintering. Chemical composition therefore determines not only what honey is, but also what the beekeeper can do with it.

This crystallisation must not, however, be confused with that of an ordinary honey rich in glucose. In both cases, the solubility of the sugars and the state of supersaturation play a central role, but the distinctive carbohydrate profile of honeydews rich in melezitose strongly alters the kinetics and the structure of the phenomenon: the texture becomes very compact and setting may occur while the honey is still in the combs. Melezitose thus stands exactly at the hinge of the preceding chapters, between trophic origin, sugar chemistry and the physics of crystallisation.

It also shows why a simple classification into "floral" versus "honeydew" is not always sufficient to anticipate the behaviour of a honey. Two honeydew honeys may show very different profiles depending on the plants and the insects involved. One will remain relatively fluid and easy to extract; the other will crystallise rapidly. Likewise, some floral honeys may crystallise extremely quickly because of their richness in glucose without containing melezitose in any significant quantity. Origin therefore provides essential information, but it does not replace analysis of the sugar profile and observation of the actual behaviour of the flow.

Melezitose is not, however, only a technological problem for the beekeeper. Experimental data indicate that a diet rich in melezitose may also directly affect the physiology of the bees. In three independent cage feeding experiments, the bees receiving melezitose consumed more food, showed a higher gut weight and increased mortality compared with the controls receiving a control carbohydrate solution without melezitose. Marked symptoms, notably a swollen abdomen, an abnormal position of the abdomen and locomotor disorders, were also observed. The composition of the intestinal lactic acid bacteria was moreover altered (Seeburger et al., 2020).

The results suggest that melezitose is only partially and relatively slowly metabolised. Part of it may be transformed by the bee or by its microbiota, but this capacity appears insufficient when intake is high: the sugar may then accumulate in the hindgut (Seeburger et al., 2020). This property becomes particularly relevant during overwintering. Bees normally avoid defecating in the hive and depend on cleansing flights to void their gut contents. When long periods of cold prevent these flights, food that is difficult to digest increases the gut load and may favour digestive disorders, dysentery and increased mortality.

Data at colony level are less direct than the cage experiments. An Austrian survey covering 33,651 colonies nevertheless showed that the presence of a melezitose flow was significantly associated with higher winter losses, whereas a honeydew flow in general did not show this association (Oberreiter & Brodschneider, 2020). This distinction suggests that the overwintering problems traditionally attributed to honeydew honey as a whole cannot be explained solely by its higher mineral content. Melezitose probably has a physiological effect of its own, even though no controlled trial at the scale of whole colonies has yet made it possible to quantify its contribution precisely.

The melezitose problem thus has two distinct but linked dimensions: during the harvest, its distinctive solubility and supersaturation profile is associated with rapid crystallisation in the combs and turns sugar chemistry into a technological problem; during overwintering, its poor digestibility may turn that same chemical particularity into a physiological constraint for the bees.

Managing melezitose is particularly relevant in the forested regions of central Europe, where honeydew flows may account for a substantial part of production. It also shows the limits of an approach that would consider honey quality solely through a few regulatory parameters. A honey may comply with the usual chemical criteria while posing a major technological problem because of its sugar profile. Conversely, its rapid crystallisation says nothing negative about its authenticity: it may be precisely the consequence of its natural origin.

The study of floral and honeydew honeys thus reveals a more general principle. The raw material determines part of the composition, that composition influences the physical properties, and those properties in turn determine how the honey can be harvested, stored and processed. Colour and aroma are the most immediately perceptible part of origin; the behaviour of the honey is another expression of it, less visible but just as important. Melezitose is the extreme case: it makes the direct link between the ecology of the flow, sugar chemistry and the physics of honey observable right down to the beekeeper's daily work.

7. How time, heat and storage transform honey

Understand how time, temperature, storage and processing operations progressively alter the quality of honey.

A large part of the discussion about heating honey centres on a supposedly critical temperature: 35 °C, 40 °C or 45 °C depending on the source. This way of reasoning is attractive because it provides a simple rule, but it does not correspond to the way chemical and enzymatic reactions actually proceed. The quality of honey is not determined by the mere fact of having passed a given temperature; it depends on the combination of time, temperature, the composition of the honey and the parameter one is seeking to preserve. A few minutes at a relatively high temperature do not necessarily produce the same effect as several days at a lower temperature, and two different honeys may react differently to the same thermal exposure. The thermal history of honey must therefore be thought of as a cumulative load rather than as the crossing of a single boundary (Tosi et al., 2002, 2008; Manickavasagam et al., 2024).

7.1 HMF: what the figure tells us

Hydroxymethylfurfural, or HMF, illustrates this principle particularly well. It forms notably from the sugars during ageing and under the effect of heat, in an acidic medium such as honey. Its concentration therefore tends to increase with time and with temperature, but the rate of formation also depends on the pH, on the sugar profile, on the water content and on the botanical origin. A very fresh honey generally contains little HMF; a honey stored for a long time or subjected to a high thermal load contains more. But an HMF measurement does not allow the exact history of the product to be reconstructed. The same concentration may result from relatively long storage at moderate temperature or from shorter, more intense heating. HMF is thus a marker of thermal history and of ageing, not a thermometer capable of indicating the maximum temperature reached.

7.2 The enzymes move in the opposite direction

Enzymes provide complementary information because they move in the opposite direction. Diastase has long been used as a quality indicator: its activity decreases progressively during storage and under the effect of heat. Invertase is often even more sensitive to thermal exposure and may, in certain contexts, detect a change earlier. It would nevertheless be incorrect to regard these enzymes as universal chronometers. Their initial activity varies strongly between honeys, and certain types naturally show a lower diastase activity than others. A low value may therefore reflect an unfavourable thermal history, but equally a natural characteristic of the product. Conversely, a still-high activity does not guarantee that all the other properties of the honey have been preserved.

Glucose oxidase deserves particular attention because it links thermal history directly to the antimicrobial properties of honey. As with many enzymes, its activity may decrease with heat and prolonged storage. The capacity of the honey to form hydrogen peroxide after dilution may then be reduced. But this response varies between honeys and cannot be summarised by a single temperature at which the enzyme would suddenly be destroyed. The loss of activity is progressive and dependent on time and on the matrix. This is why statements of the type "above 40 °C, all the enzymes in honey are destroyed" are not scientifically defensible.

7.3 Aroma and colour

The earliest changes are, moreover, not necessarily those measured by HMF or diastase. The volatile compounds responsible for the aroma are sensitive to temperature, to oxygen and to the duration of storage: some evaporate, others degrade or take part in new reactions. A honey may therefore lose part of its aromatic freshness while its regulatory parameters remain perfectly compliant. The distinction matters, because legal conformity is not equivalent to maximum preservation of sensory quality: a honey heated just enough to remain within the regulatory limits is not necessarily identical, in aromatic terms, to a freshly harvested and little-handled honey.

This internal evolution must be distinguished from the absorption of external odours. The greater part of the aromatic changes observed during storage comes from the loss of compounds belonging to the honey itself and from the formation of degradation products, not from contamination by the environment (Castro-Vázquez et al., 2008, 2012; Manickavasagam et al., 2024). An unusual note in a stored honey is therefore more often a marker of ageing than an odorant picked up from outside. The documented case of transfer from a container is dealt with in Chapter 13.

With time and heat, the colour may also change. Maillard reactions, which involve notably reducing sugars and nitrogenous compounds, contribute to the formation of new pigments and aroma molecules. The honey may progressively darken, while certain natural substances degrade and others are formed. These reactions explain an apparent contradiction often encountered in studies: after heating, certain chemical tests of "antioxidant activity", such as DPPH or FRAP, sometimes decrease, but may also increase. A rise in these values does not necessarily mean that the honey has become more beneficial to health; it may simply reflect the formation of Maillard products capable of reacting in the analytical test. DPPH and FRAP assays measure a chemical capacity under laboratory conditions, not a clinical effect in humans.

Light is a distinct factor, often confused with heat because sun exposure supplies both at once. Where the two have been separated, temperature and duration remain the main drivers of HMF formation and of the loss of diastase activity. The best-documented effects specific to light concern colour, phenolic compounds and antioxidant activity assays: a few days of sun exposure reduces total phenols, flavonoids and several measures of antioxidant activity (Yalçın, 2021), and a year of storage in clear glass under natural light caused these same parameters to fall sharply, particularly in an acacia honey (Šarić et al., 2012).

By contrast, no study documents a degradation of aroma compounds attributable specifically to light, independently of heat. The common claim that light "destroys the aromas" of honey is therefore not established; what is established is the loss through heat and time described above.

Packaging comparisons give contradictory results. Opaque containers reduced colour variation and the general drift of physicochemical parameters over twelve months compared with transparent bottles (Idris et al., 2021), whereas another twelve-month comparison between clear glass, dark glass and a metal tin found no significant difference in HMF, diastase, total phenols or catalase activity, despite a general degradation over time (Yiğit et al., 2024). These studies rarely measure the actual light dose or the temperature reached inside the container, which limits their value as a test of photochemistry. Protecting honey from light remains prudent, but the expected benefit is probably smaller than that obtained by controlling temperature and storage time.

This logic also applies to storage without deliberate heating. A jar of honey kept for several months in a cool, dark room does not follow the same trajectory as a honey stored in a warm room, in a vehicle or near a heat source. Even temperatures that seem moderate may, when they act for long enough, accelerate the formation of HMF, reduce certain enzyme activities and alter the aroma profile. Time is therefore a component of quality in its own right. A honey that has never been "heated" in the beekeeping sense may have undergone a high thermal load simply because it was stored too long at an unfavourable temperature.

Liquefying a crystallised honey is the operation through which the beekeeper intervenes most directly in the thermal history of the product. The purpose is technological: to make the honey fluid enough to be transferred, packed, strained or worked. The more it is heated, the faster liquefaction proceeds — but accelerating the process also accelerates the reactions of deterioration. The challenge therefore consists in finding a compromise between technological efficiency and preservation of the product: a lower temperature requires more time, a higher temperature reduces the duration needed but increases the rate of chemical and enzymatic reactions.

The same principles apply to repeated heating. A first, moderate liquefaction cycle may have only a limited effect, but each new exposure adds to the previous history. There is no universal scientific number of "permitted" liquefactions before degradation. The relevant notion is that of cumulative thermal load. This argues for an organisation of work that avoids heating an entire stock every time a small quantity has to be packed. Managing honey in batches, with volumes matched to needs, naturally limits repeated exposures.

Microwaves are another subject laden with beliefs. Experimental studies do not show that microwave heating is systematically more damaging or more protective than conventional heating. Depending on the power, the duration, the volume, the type of honey and the temperature actually reached, short and optimised treatments may limit certain thermal damage, while other protocols accelerate the formation of HMF or reduce sensitive constituents. Heating may moreover be uneven in a viscous product and create hot spots. The technology therefore cannot be assessed independently of the time–temperature profile actually applied and of the uniformity of the heating (Kowalski, 2013; Al-Ghzawi et al., 2026).

7.4 From the apiary to the retail shelf – what does processing actually change?

Between extraction and the jar sold at retail, honey may follow very different paths. A beekeeper may let it settle, strain it coarsely to remove fragments of wax and other particles, then put it into jars with relatively little intervention. On a larger scale, batches often have to be made fluid enough to be pumped and mixed, then homogenised, clarified or filtered, possibly heated in order to facilitate packing or to delay crystallisation, before being transported and stored, sometimes for long periods. It is therefore more accurate to speak of lightly processed honey and of industrially packed honey than simply to set a "beekeeper's honey" against a "commercial honey", since both are products placed on the market.

These operations do not generally turn honey into a chemically different food: its dominant fraction still consists of fructose, glucose and water, and moderate treatments alter the overall sugar composition relatively little. The most sensitive constituents are rather the enzymes, certain aroma compounds, the colour and various minor components. When the thermal load increases, the activity of diastase, invertase or glucose oxidase may decrease while the HMF concentration rises, and volatile compounds may be lost or transformed. These effects depend, as explained above, on the combination of time × temperature × composition of the honey, and not on the mere fact that a product has been "industrially treated" (Subramanian et al., 2007; da Silva et al., 2016; Manickavasagam et al., 2024).

Filtration calls for a comparable distinction. Coarse filtration or straining mainly removes fragments of wax, debris and other unwanted particles while retaining most of the pollen. Much finer filtration, and above all ultrafiltration, may by contrast strongly reduce the quantity of pollen and suspended matter. This scarcely alters the main sugars, but it may complicate the determination of botanical or geographical origin by melissopalynology. To say that "filtering honey removes its beneficial substances" is therefore too general: the effect depends strongly on the level of filtration and concerns above all the particles retained, including pollen (Subramanian et al., 2007; Soares, Amaral, et al., 2017).

Blending several batches is another change that is less chemical than sensory and informational. Homogenising honeys from different apiaries, years or sometimes countries makes it possible to obtain a more constant colour, flavour and consistency. This standardisation may, however, attenuate the particularities of a terroir or of a specific floral origin. It may also complicate the interpretation of certain markers of origin, since the final product represents the average of several different profiles. The honey remains authentic if each component meets the definition of honey and if the labelling is correct; blending is not in itself an adulteration (Subramanian et al., 2007; Soares, Amaral, et al., 2017).

The heat treatments used on a large scale also have legitimate technological objectives. Reducing viscosity facilitates pumping and jarring; dissolving microcrystals may delay visible crystallisation; a sufficiently intense heat treatment may reduce the load of osmophilic yeasts and improve the stability of certain batches. The price to pay is an additional thermal load. The stronger or the more frequently repeated it is, the higher the probability of losing aromas and enzyme activity and of forming HMF. Industrial processing therefore represents a compromise between ease of handling, uniformity, commercial stability and maximum preservation of the characteristics of fresh honey (Subramanian et al., 2007; Soares, Amaral, et al., 2017).

The time spent in the commercial chain may ultimately be as important as the packing operation itself. A honey that has been little heated but transported or stored for long periods at high temperature may accumulate more HMF and lose more enzyme activity than a honey briefly heated and then kept in good conditions. Freshness analyses thus tell us about a cumulative history of processing and storage, and not about a simple opposition between "industrial" and "natural" honey (da Silva et al., 2016; Manickavasagam et al., 2024).

7.5 Can you cook with honey?

The claim that honey should "never be heated" conflates two different questions: the optimal preservation of the characteristics of a fresh honey and the safety of a food used in cooking. Heat progressively alters honey, but the available data do not show that normal culinary use turns it into a food specifically toxic to humans. What changes are above all certain quality markers and sensitive constituents: enzyme activity decreases, aroma compounds are lost or transformed and the HMF concentration increases with the intensity and duration of the heat treatment (Subramanian et al., 2007; da Silva et al., 2016; Scepankova et al., 2021).

HMF deserves an important distinction here. Its formation increases when sugars, particularly fructose, are heated in an acidic medium such as honey; a high content is therefore a useful indicator of ageing or of a substantial thermal load in the product placed on the market. But the regulatory limits used to monitor the quality of honey are not thresholds beyond which a cooked food would become toxic, as the box devoted to this marker explains. HMF is moreover formed in many other heated foods: coffee, bakery products, dried fruit or caramelised preparations. To date, no specific toxicity linked to the normal culinary use of honey has been demonstrated in humans (Abraham et al., 2011; Faustino et al., 2026; Scepankova et al., 2021).

The most perceptible transformation often concerns the aroma. Honeys owe a substantial part of their botanical identity to very small quantities of volatile compounds: terpenes and their derivatives, certain aldehydes, alcohols and other compounds responsible for floral, fruity or herbaceous notes decrease under increasing thermal exposure. Studies carried out on different honeys show that heating or prolonged warm storage progressively attenuates part of the original aromatic signature (Castro-Vázquez et al., 2008, 2012; Escriche et al., 2009; Machado et al., 2020).

But heating does not simply mean "removing" the aroma. As thermal exposure increases, sugar degradation, caramelisation and the Maillard and Strecker reactions generate new odorous molecules, notably furan compounds, pyranones and pyrazines. Caramelised, toasted, malty or cooked notes may then replace part of the initial floral character (Serra Bonvehí & Ventura Coll, 2003; Vázquez et al., 2007; Starowicz & Zieliński, 2019). This transformation is not necessarily undesirable: in a pastry, a sauce, a marinade or a glaze, these new aromas may be precisely part of the intended result.

Honey does not, moreover, behave exactly like sucrose when used in cooking. It already contains a great deal of glucose and fructose, whose reducing power favours browning reactions, and its water as well as its hygroscopic character also influence texture and moisture retention. Replacing granulated sugar with honey is therefore not simply replacing one source of sweetness with another: colour, moisture, aroma and the behaviour of the preparation may all change as well (De-Melo et al., 2018; Subramanian et al., 2007).

The practical consequence thus depends on the culinary objective. If the aim is to preserve the floral or varietal character of a particular honey, it is preferable to limit its exposure to heat and, where the recipe allows, to add it late. This recommendation is strongly plausible given the volatility of the aroma compounds, even though trials directly comparing the same recipe with honey added before and after cooking remain rare. Conversely, when caramelised, toasted or cooked notes are sought, heating the honey may be an integral part of the intended culinary transformation.

It is therefore not justified to claim that a heated honey automatically becomes "toxic": it progressively becomes a different ingredient in sensory and functional terms. The more intense the cooking, the more its role rests on the sugars, the sweetness, the colouring, the texture and the thermal aromas, and the less on the enzymes and the aromatic signature of fresh honey.

The quality of honey after harvest is thus the result of a history, not of a single event. Ageing begins as soon as the product exists, even when it is never deliberately heated. Temperature accelerates or slows this development, while the initial composition of the honey determines its sensitivity. HMF, diastase, invertase, glucose oxidase, colour and aroma compounds each give access to a different facet of this history, but none of them summarises it alone. To preserve the product as well as possible, one must therefore abandon the search for a magic figure and reason in terms of time, temperature, technological objective and the quality one wishes to preserve.

8. Microbiology and antimicrobial activity – stable, but not sterile

Distinguish microbiological stability, the presence of microorganisms and antimicrobial activity, then understand the main mechanisms involved.

Honey presents a microbiological situation that may seem paradoxical. It is capable of strongly inhibiting many microorganisms and possesses, in laboratory assays, an antibacterial activity that is at times remarkable; yet it is not sterile. Yeasts, moulds, various bacteria and above all bacterial spores may be detected in it. This coexistence is explained once the presence of a microorganism is distinguished from its capacity to multiply. The physicochemical properties of ripe honey create an environment highly unfavourable to many species, but they do not necessarily destroy all the microorganisms introduced into it before, during or after harvest (Snowdon & Cliver, 1996; Luca et al., 2024).

8.1 The microbiota of honey

The microbiota of honey comes from multiple sources: nectar and honeydew, pollen, the digestive tract and body surface of the bees, dust, soil, air, beekeeping equipment and handling after harvest. The quantities of culturable microorganisms are generally low in a ripe honey correctly harvested and stored, precisely because its high sugar concentration and its low water activity limit their multiplication. The most important microorganisms in technological terms are the osmotolerant yeasts, notably certain species of Zygosaccharomyces, capable of developing at water activities much lower than most bacteria. When enough water becomes available, they may ferment the sugars and produce notably ethanol and carbon dioxide. The microbiology of honey is therefore directly linked to the physical mechanisms described in the preceding chapter: water content, water activity, crystallisation and phase separation are not phenomena independent of fermentation (Snowdon & Cliver, 1996; Luca et al., 2024).

8.2 Spores: a case apart

Bacterial spores represent a different case. A spore is an extremely resistant survival form that must not be confused with a growing bacterial cell. The low water activity of honey may prevent the germination and multiplication of a bacterium without thereby eliminating its spore. The principle is fundamental: inhibition of growth does not mean destruction of spores. Honey may therefore be microbiologically very stable while containing dormant forms capable of surviving for long periods. This distinction will take on particular importance in the chapter devoted to food safety; at this stage, it is enough to note that an antimicrobial product is not necessarily a sterile product (Snowdon & Cliver, 1996).

8.3 Mechanisms that act together

The antimicrobial activity of honey itself rests on several mechanisms acting simultaneously. In concentrated honey, the high osmotic pressure and the low water activity play a major role. Microbial cells placed in such an environment lose water and their metabolic functioning is severely disrupted. Acidity constitutes a second barrier: the low pH of many honeys creates conditions unfavourable to many bacteria. These effects are intrinsic to the concentrated matrix and explain a large part of the natural stability of the product. But they are not sufficient to explain all the activity observed. When Kwakman et al. (2010) compared a medical-grade honey with a sugar solution of equivalent composition, the honey retained a bactericidal activity at concentrations at which osmolarity alone was no longer sufficient. Other antimicrobial components were therefore necessary to explain the phenomenon.

Among them, the glucose oxidase–hydrogen peroxide system occupies a historically and mechanistically central place. White, Subers and Schepartz (1963) showed that honey contains glucose oxidase and that this enzyme can produce hydrogen peroxide when the honey is diluted. In highly concentrated honey, the activity of glucose oxidase is strongly limited. Adding water increases its mobility and allows the oxidation of glucose, with simultaneous formation of gluconic acid and H₂O₂. Bang, Buntting and Molan (2003) studied eight honeys from six floral origins and observed that maximum accumulation of H₂O₂ generally appeared when the honey was diluted to about 30–50 % by volume. The relationship was not, however, identical between samples, which already shows how greatly antimicrobial activity varies from one honey to another.

This variation depends notably on the balance between the production and the destruction of hydrogen peroxide. Glucose oxidase favours its formation, while catalase breaks it down. Catalase activity is present to varying degrees in honeys and is notably associated with components of plant origin, in particular with pollen. Two honeys with a comparable capacity to produce H₂O₂ may thus accumulate different concentrations if their catalase activity differs. The quantity of peroxide actually measured is therefore the result of a dynamic balance, and not the direct measurement of a fixed quantity of "antiseptic" contained in the honey (White et al., 1963; Weston, 2000).

Light constitutes a third factor of fragility. White and Subers (1964) showed that visible radiation can destroy the peroxide accumulation system, with maximum efficiency between 425 and 525 nm, provided a photosensitising substance is present and the medium remains acidic. This is the source of the historical description of "inhibine" as a factor labile to light as well as to heat.

This reading nevertheless calls for a qualification. Recent work shows that antibacterial activity tracks the concentration of H₂O₂ and of polyphenols better than the quantity of glucose oxidase itself, and that peroxide production may be mediated by compounds of plant origin independently of the enzyme (Bucekova et al., 2018; Brudzynski, 2020). The chain "light destroys glucose oxidase, therefore antibacterial activity disappears" is thus plausible for certain honeys, but it is not established as a universal mechanism.

Hydrogen peroxide is not, however, the only factor. Kwakman et al. (2010) showed experimentally, in one particular medical-grade honey, that bactericidal activity resulted from the combined action of the sugars, H₂O₂, methylglyoxal (MGO), bee defensin-1 and acidity. Bee defensin-1 is an antimicrobial peptide produced by the bee and may contribute to the activity of certain honeys. Its importance is not, however, uniform.

The presence or concentration of the various factors varies sufficiently for two honeys with comparable antibacterial activity to achieve it through different mechanisms. In a direct comparison of two medical-grade honeys, Kwakman et al. (2011) thus showed that H₂O₂ and bee defensin-1 contributed strongly to the rapid activity of one of the honeys studied, whereas the manuka honey tested relied much more heavily on a high MGO concentration.

Methylglyoxal is indeed a particular mechanism associated above all with certain honeys derived from Leptospermum, including manuka. It may contribute strongly to their so-called non-peroxide antibacterial activity. This particularity has sometimes led to MGO being presented as the "active principle" of honey in general, which is incorrect. Many honeys possess substantial antibacterial activity with very little MGO, while in them the H₂O₂ system plays a much more important role. Even in manuka, neutralising the MGO does not necessarily abolish all activity, a sign that other factors remain involved (Kwakman et al., 2011). It is therefore more accurate to speak of different antimicrobial profiles than simply to set "peroxide" honeys against "non-peroxide" ones.

Phenolic compounds and other substances of plant origin may also contribute to antimicrobial activity, but their quantitative role varies strongly with the botanical origin and cannot be generalised to all honeys. The same caution applies to bee defensin-1. The value of modern mechanistic work lies precisely in having shown that antibacterial activity is not attributable to a single molecule. The factors reinforce or compensate for one another: a strong osmotic effect may coexist with a low production of H₂O₂, a high MGO content with a small contribution from defensin-1, or the reverse. It is this combination that partly explains the sometimes very large differences observed between honeys in microbiological tests (Kwakman et al., 2010, 2011; Luca et al., 2024).

This variability finally requires three frequently confused questions to be distinguished. The first is the microbiological stability of the honey itself: can microorganisms multiply in it and cause it to ferment or spoil? The second concerns its experimental antimicrobial activity: at what concentration and under what conditions does a given sample inhibit or destroy a microorganism in the laboratory? The third is that of its clinical efficacy, which requires knowing whether applying or consuming the product actually improves a relevant outcome in a patient. A positive answer to the second question does not automatically provide a positive answer to the third. Medical applications must therefore be assessed separately, indication by indication.

Honey thus offers a particularly instructive example of the difference between stability and sterility. The mechanisms that make it inhospitable do not destroy all microbial forms, do not carry the same weight in all honeys and do not necessarily retain their importance when the biological medium is changed entirely. Honey may therefore be strongly antimicrobial without being sterile, and an antimicrobial activity demonstrated in the laboratory is not sufficient to establish therapeutic efficacy.

9. Quality, origin and authenticity – a race between detection and circumvention

Understand the different dimensions of honey authenticity, the possibilities and limits of analytical methods and the interpretation of control results.

The quality, the authenticity and the safety of honey answer three different questions. A honey may be authentic yet have lost part of its quality after prolonged storage; it may meet the usual composition criteria while being the subject of an incorrect declaration of origin; and economic fraud does not necessarily imply a health hazard. Authenticity therefore concerns first of all the correspondence between what the product actually is and what is claimed about it: is it really honey, have foreign sugars been added, is the stated botanical or geographical origin plausible, and was the product obtained in accordance with the definition of honey? Once these questions are distinguished, one conclusion follows: there is no single chemical property called "authenticity" and, consequently, no isolated test can demonstrate all its dimensions (Almeida-Muradian et al., 2020; Walker et al., 2022a, 2022b). The recent history of detecting added sugars shows particularly well why.

9.1 Isotope analysis and its limits

For a long time, carbon isotope analysis was one of the most robust tools for revealing the addition of syrups derived from C4 plants, principally maize and sugar cane. These plants fix atmospheric carbon by a photosynthetic pathway different from that used by most melliferous plants, which are C3 plants. This difference leaves a measurable isotopic signature. The reference method based on the analysis of stable carbon isotope ratios, notably AOAC Official Method 998.12, thus makes it possible to compare the isotopic signature of the honey's sugars with that of its protein fraction and to detect certain additions of C4 sugars. This principle was a major advance because it did not look for a molecule specific to the syrup: it exploited a fundamental difference between photosynthetic pathways.

But an effective method also changes the behaviour of those seeking to circumvent it. Maize syrups are today less attractive for adulteration precisely because they are relatively easy to detect by the classic isotopic methods. The Joint Research Centre of the European Commission notes that syrups derived from rice, wheat or sugar beet are now used more frequently; yet these plants use, like the majority of melliferous plants, the C3 photosynthetic pathway. Their isotopic signature is therefore much closer to that of the sugars naturally present in honey (Joint Research Centre, 2023). A negative result in the classic C4 sugar test may thus demonstrate that one particular form of adulteration has not been detected, but it does not on its own demonstrate that the honey is authentic.

9.2 Why does a negative C4 test not prove authenticity?

Classic isotope analysis answers a precise question: is an abnormal proportion of sugars from C4 plants found? If the result is negative, this hypothesis becomes less likely — but it is only one of the possible adulterations, and rice, wheat or beet syrups escape this question by construction.

A negative result therefore means: "this anomaly was not detected", and not: "all possible forms of fraud have been excluded". This is a general principle of authenticity analysis: each method answers a defined question and has a domain within which it is informative.

More elaborate isotopic techniques have been developed to overcome this difficulty. Liquid chromatography coupled to isotope ratio mass spectrometry — LC-IRMS or EA-LC-IRMS depending on the configuration — makes it possible to examine different sugars separately rather than only the overall isotopic signature of the honey. It is a well-established advanced method, which notably improves the detection of certain adulterations with syrups derived from C3 plants. Walker et al. (2022a, 2022b) nevertheless stress that even a very high-performing technique answers only those forms of adulteration for which its performance has been established: analytical sophistication does not remove the need to define the question posed precisely.

9.3 Global fingerprints and machine learning

Other approaches give up looking for a specific adulterant and seek instead to determine whether the overall profile of the honey resembles that of a reference population. Nuclear magnetic resonance (NMR), chromatography coupled to high-resolution mass spectrometry (LC-HRMS), metabolomics and various spectroscopic methods can produce fingerprints comprising hundreds or thousands of variables simultaneously. A sample can then be compared with a database of known authentic honeys. This strategy is powerful, but it introduces a new dependency: the quality of the result depends directly on that of the reference database.

A honey may appear "atypical" because it is adulterated, but also because its botanical origin, its region, its year of production or its environmental conditions are poorly represented in the database used. NMR and HRMS are therefore not machines capable of recognising "real honey" intrinsically; they make it possible to measure similarities and divergences relative to populations whose construction must be sufficiently transparent and representative (Biswas et al., 2023; Walker et al., 2022a).

The same logic applies to machine learning models. Very high classification performance can be obtained when the algorithm is trained and evaluated on samples from the same database. It becomes much less convincing if the model has never been tested on genuinely independent honeys from other regions, other years or other laboratories. An accuracy of 95 or 99 % in internal validation is therefore not in itself proof that the method will correctly identify an unknown honey on the market. For food authentication, external validation and the representativeness of the reference population are at least as important as the power of the algorithm.

9.4 Botanical origin: pollen and DNA

Botanical origin poses a comparable difficulty. Melissopalynology remains a fundamental method: the pollen present in honey provides a great deal of information about the plants visited by the bees. But the percentage of a given pollen in the sample does not correspond directly to the percentage of nectar coming from the same plant. Some species are over-represented because they produce a great deal of pollen or because it readily enters the nectar, while others are naturally under-represented.

This is why the identification of European unifloral honeys ideally rests on the convergence of pollen analysis, sensory characteristics and specific physicochemical parameters rather than on a universal pollen percentage (Persano Oddo & Bogdanov, 2004; von der Ohe et al., 2004).

DNA metabarcoding offers another way of detecting the plants associated with a honey. Hawkins et al. (2015) showed that this approach could recover a large part of the dominant taxa identified by microscopy and sometimes detect additional taxa. But the results are influenced by DNA extraction, the choice of primers, amplification efficiency, reference databases and bioinformatic processing. Above all, the number of sequences obtained for a species does not necessarily correspond to the quantity of nectar it supplied. Metabarcoding is therefore a promising complement to melissopalynology, but it does not currently make it possible to replace pollen and physicochemical interpretation generally for the regulatory establishment of botanical origin.

9.5 Geographical origin and colony feeding

Determining geographical origin is more demanding still. Isotopic, mineral, pollen, volatile or metabolomic signatures may indeed differ between regions. But a geographical classification works only by comparison with samples of known origin. It therefore answers less the abstract question "where does this honey come from?" than a more limited one: "among the regions correctly represented in this reference database, which does this sample most resemble?". The more one seeks to distinguish neighbouring regions and production subject to strong annual variability, the more a vast, durable and regularly updated reference database becomes necessary.

Authenticity may moreover be compromised even before the honey is extracted. Substantial feeding of colonies with sugar solutions during a flow intended for harvest poses a particular problem: the sugars are ingested, transported and transformed by the bees before entering the combs. They therefore no longer necessarily present themselves as a simple syrup added after harvest. The European control action carried out in 2015–2017 had already noted the difficulty of distinguishing certain exogenous sugars when these are chemically close to the natural sugars of honey. Analysis here has to reconstruct a history of production and not merely look for a contaminant added to the finished product.

9.6 When the marker itself becomes the target

The same race between control and circumvention appears with the enzymes. Diastase has long been used as an indicator of freshness and heat treatment. But Erban et al. (2021) identified by proteomics foreign amylases, originating notably from microorganisms, in several honey samples. Adding an exogenous amylase can artificially increase diastase activity and make a thermally degraded or manipulated honey appear more compliant on a classic parameter, without restoring the other properties that have been lost. Proteomics then makes it possible to investigate the origin of the enzyme itself rather than merely measuring its activity. The example illustrates an important principle: as soon as a parameter becomes a known control criterion, it may itself become a target for manipulation (Erban et al., 2021).

This difficulty appears clearly in the European action "From the Hives", coordinated by the Directorate-General for Health and Food Safety and carried out with the support of the Joint Research Centre. Between November 2021 and February 2022, 320 consignments of imported honey from 20 countries were sampled at the European borders. The JRC considered 147 samples, or 46 %, to be suspicious of non-compliance because at least one marker of a foreign sugar source had been detected. In an earlier action carried out in 2015–2017, the corresponding proportion was 14 % (European Commission, 2023; Joint Research Centre, 2023).

9.7 46 % suspicious does not mean 46 % of honeys shown to be adulterated

The 147 samples from "From the Hives" were classified as suspicious because the methods used had detected at least one marker compatible with the presence of exogenous sugars. The techniques provided mainly qualitative information and did not make it possible to determine the proportion of any syrup added.

The European Commission further notes that the advanced methods used for this operation fell outside the scope of accreditation of the activities concerned at the JRC laboratory. This circumstance had to be taken into account before any regulatory measure. Of the 44 European operators subsequently subject to investigation, seven had been sanctioned at the time of the review published by the Commission (European Commission, 2023).

The correct interpretation is therefore as follows: 46 % of the consignments tested showed analytical signals justifying further investigation. To say that "46 % of honeys were proven to be adulterated" would turn a screening result into a final verdict.

This case connects directly with the analysis by Walker et al. (2022a, 2022b). These authors examined complex authenticity reports in which different techniques could produce partially divergent results. Their conclusion is not that modern methods should be abandoned, but that interpretation must be as transparent as the measurement itself. A report should state the data that led to the conclusion, the known performance of the methods, the references against which the sample is compared and the degree to which the results actually support the hypothesis of adulteration. In their analysis of several certificates used to accuse British commercial honeys of adulteration, the data available ultimately provided only weak or limited support for some of the unfavourable interpretations published. The expression "suspicious honey" is therefore an evaluative conclusion, not a chemical substance that the instrument has directly measured.

This methodological caution is now reflected in European law. Directive (EU) 2024/1438 has strengthened transparency about the origin of honey blends: since its application from 14 June 2026, the countries in which the honey was harvested must in principle appear in the principal field of vision in descending order of their proportion, with the corresponding share and a tolerance provided for the declaration. But this same directive implicitly acknowledges that the analytical toolkit is not yet fully harmonised.

It instructs the Commission to adopt, by 14 June 2028, harmonised methods of analysis for detecting adulterated honey; in the meantime, Member States must use, where possible, validated and internationally recognised methods, notably those of the Codex Alimentarius (European Union, 2024).

It would therefore be paradoxical to claim today that an instrument or an analytical platform is capable on its own of certifying the absolute authenticity of a honey when the European legislator is still providing for the harmonisation of the methods needed for that task. Analytical progress is considerable, but it works increasingly through the convergence of indications: isotopes, sugar profiles, marker compounds, pollen, DNA, spectra, metabolites, geographical data and traceability can reinforce one another. The more sophisticated the fraud, the more the conclusion must itself rest on several independent elements.

The authenticity of honey thus appears as a permanent race between detection and circumvention. Isotopic methods have made the addition of C4 syrups more difficult; fraudsters may turn to C3 syrups. Enzyme profiles may reveal a degraded product; foreign enzymes may be added. Fingerprinting methods detect atypical profiles; their interpretation then depends on the quality of the reference databases. Each new method closes off certain routes to fraud without necessarily closing all the others. The relevant scientific question is therefore not "which test proves that this honey is genuine?", but rather: what claim about this honey is one seeking to verify, which methods can actually test it, and what degree of certainty does their combination allow?

10. Honey as a food – sugars, nutrition, allergies and health claims

Situate honey within human nutrition, assess the data on its metabolic effects and distinguish scientific evidence from health claims.

Discussing the effects of honey on health exposes one to two opposing simplifications. The first consists in taking its chemical diversity to make it necessarily a food endowed with substantial protective properties; the second in reducing it to "coloured sugar" whose effects would all be identical to those of sucrose. The available data justify a more nuanced position. As we saw in Chapter 3, honey does indeed contain organic acids, enzymes, minerals, phenolic compounds and numerous aroma molecules, but these minor constituents are consumed at doses far below those of the sugars. In nutritional terms, honey therefore remains above all a source of simple carbohydrates, principally fructose and glucose, even though its matrix is chemically more complex than that of a refined sugar (da Silva et al., 2016; De-Melo et al., 2018).

This distinction appears clearly in public health recommendations. The World Health Organization explicitly classifies the sugars naturally present in honey among free sugars, in the same way as those of syrups and fruit juices. It recommends that these free sugars represent less than 10 % of total energy intake, in adults as in children; a further reduction to below 5 % is proposed in order to obtain additional benefits, notably with regard to dental caries. The first recommendation is regarded as strong, the second as conditional (WHO, 2015).

This arithmetic matters when interpreting clinical trials. The experimental doses commonly used – 40 or 50 g of honey per day – are not nutritionally trivial: on their own they represent a large share of the free sugar intake corresponding to the 10 % threshold and clearly exceed that of 5 %. One must therefore always consider not only the biological result measured in a study, but also the quantity of honey needed to obtain it and what it replaces in the diet.

10.1 What do 50 g of honey represent in daily sugar consumption?

A honey contains approximately 80 to 82 % sugars. For a diet of 2,000 kcal per day, the WHO benchmarks correspond approximately to less than 50 g of free sugars per day in order to stay below 10 % of energy, and to less than 25 g per day to reach the level below 5 %.

A 20 g portion of honey supplies about 16 g of free sugars. It therefore already represents about a third of the first benchmark and close to two thirds of the second.

A dose of 50 g of honey supplies approximately 40 to 41 g of sugars. On its own, it corresponds to about 80 % of the 50 g level and far exceeds that of 25 g.

These figures do not mean that 50 g of honey constitutes a threshold of danger. They show why a study using 40 or 50 g of honey per day must be interpreted in the context of the diet as a whole.

10.2 Glycaemic response and metabolic effects

The immediate effect on blood glucose is not strictly identical for all honeys. Their relative content of fructose, glucose, sucrose and oligosaccharides varies with the botanical origin, and this composition may alter the postprandial response. In a crossover trial covering six Greek honeys, Gourdomichali and Papakonstantinou (2018) observed different glycaemic indices depending on the honey, some falling into the medium category and others into the high category. Conversely, Ischayek and Kern (2006) had observed no significant difference between four American honeys despite differing fructose/glucose ratios. The glycaemic index is therefore a variable and informative property, but it does not constitute a general measure of the "health value" of honey: it describes the rise in blood glucose after ingestion of a standardised quantity of carbohydrate, not the consequences of habitual long-term consumption (Ischayek & Kern, 2006; Gourdomichali & Papakonstantinou, 2018).

Some acute studies have shown a smaller rise in blood glucose after ingestion of honey than after a comparable quantity of pure glucose. This result is physiologically plausible, notably because a substantial part of the sugars in honey consists of fructose, which does not raise blood glucose directly in the same way as glucose. But this comparison does not demonstrate that honey protects against diabetes, nor that it constitutes a treatment for the disease. When honey is compared over a longer period with other caloric sweeteners, the differences may become much more modest. In a controlled crossover trial, Raatz et al. (2015) administered 50 g of carbohydrate per day for two weeks in the form of honey, sucrose or high-fructose corn syrup to people with normal or impaired glucose tolerance. The three sweeteners produced similar effects on blood glucose, insulin, lipids and several inflammatory markers; triglycerides increased with all three treatments (Raatz et al., 2015).

A small Swiss study illustrates the same question from another angle. Despland et al. (2017) studied eight healthy men in whom a controlled diet supplied 25 % of energy either as Robinia honey or as a mixture of fructose and glucose reproducing the ratio of the main sugars in honey. After eight days, no convincing difference emerged in glucose tolerance or hepatic insulin sensitivity between the honey and the sugar mixture. The study was very small and of short duration, but it does not support the idea that a "honey matrix" necessarily produces a substantial metabolic advantage when the sugar dose is equivalent (Despland et al., 2017).

Trials in people with type 2 diabetes are particularly instructive because their results are contradictory. Bahrami et al. (2009) observed, after eight weeks of increasing honey consumption, a decrease in weight and in several lipid parameters, but also a significant increase in HbA1c; the dose had been progressively raised from 1 to 2.5 g of honey per kilogram of body weight per day, which represents a considerable quantity in an adult (Bahrami et al., 2009). In a later crossover trial using 50 g of honey per day for eight weeks, Sadeghi et al. (2019) observed no convincing improvement in glycaemic control; HbA1c changed unfavourably compared with the control period, even though certain anthropometric parameters changed in the other direction (Sadeghi et al., 2019). These results therefore do not allow honey to be recommended as a therapeutic strategy for controlling diabetes.

The review literature reflects this heterogeneity. Ahmed et al. (2023) pooled 18 controlled trials representing 33 comparisons and 1,105 participants. The median dose was 40 g of honey per day and the median duration eight weeks. The meta-analysis found small improvements in certain parameters, notably fasting blood glucose and several blood lipids, but the certainty of the evidence was generally low; some results varied with the type of honey, its treatment or other study characteristics, while certain inflammatory markers moved in an unfavourable direction (Ahmed et al., 2023). A more recent synthesis incorporating dose–response analyses has reinforced this caution further: depending on the parameter, honey was associated with small improvements, with an absence of effect or with unfavourable changes; several estimates rested on evidence of low or very low certainty and some results disappeared when a single influential study was removed (Norouzzadeh et al., 2025).

This literature therefore does not provide a simple cardiometabolic profile. The available data show no robust and general reduction in weight, blood pressure, blood glucose, lipids or inflammation that could be attributed to honey independently of the dose, the type of honey, the population and the comparator. This does not mean that all the differences observed are necessarily null: small specific effects are plausible and some deserve further study. But they do not today justify turning a food rich in free sugars into a general preventive or therapeutic intervention.

An essential distinction here concerns substitution. Adding 40 g of honey daily to an existing diet is not the same intervention as replacing 40 g of another sweetener with 40 g of honey. Several trials in the meta-analysis by Ahmed et al. (2023) used honey as an additional intake, while others studied an energy replacement. These situations do not permit the same conclusion. If certain properties of honey gave it a small advantage over sucrose or another sweetener, that advantage would be relevant above all when it replaces those sugars, and not as an argument for increasing the total quantity of sugars consumed. This interpretation is consistent with the public health recommendations aimed at reducing overall exposure to free sugars (WHO, 2015; Ahmed et al., 2023).

10.3 Honey in endurance sport

The fact that honey supplies mainly free sugars does not mean that these are undesirable in every situation. During prolonged endurance exercise, the nutritional objective changes: it may become useful to supply carbohydrate rapidly in order to maintain blood glucose, partly spare endogenous reserves and provide substrate to the working muscles. In this context, honey may constitute a functional source of carbohydrate.

Describing it simply as a source of "fast sugars" would, however, be reductive. Honey contains mainly glucose and fructose, two monosaccharides that do not follow exactly the same pathways of intestinal absorption and metabolism. This combination is particularly interesting during prolonged efforts: when carbohydrate intakes become high, combining glucose and fructose makes it possible to use several intestinal transporters and to increase the oxidation of the carbohydrate ingested compared with the use of glucose alone (Currell & Jeukendrup, 2008; Stellingwerff & Cox, 2014).

Requirements depend strongly on the duration of the effort. For brief exercise, ingesting carbohydrate brings little metabolic advantage. When the effort continues beyond about one to two hours, intakes of the order of 30 to 60 g of carbohydrate per hour frequently become useful; for efforts exceeding two and a half to three hours, higher intakes may be beneficial if the athlete tolerates them, and glucose–fructose mixtures then become of greater interest (Stellingwerff & Cox, 2014).

Honey appears able to fulfil this function without possessing any specific ergogenic effect. In a recent crossover study of trained cyclists, ingesting honey supplying 90 g of carbohydrate per hour during three hours of cycling produced responses comparable to those of a conventional sports nutrition product: carbohydrate oxidation, gastrointestinal symptoms and subsequent exercise capacity did not differ significantly between the two treatments (Fortis et al., 2025). An earlier systematic review had already reached a similar conclusion: when compared with an absence of carbohydrate, honey may improve certain performance parameters, but it shows no convincing superiority when compared with other sources of carbohydrate (Hills et al., 2019).

Before a prolonged effort, the situation is more nuanced. Carbohydrates with a lower glycaemic index may produce a more moderate rise in blood glucose and insulin and reduce certain glycaemic fluctuations before exercise. These metabolic differences do not, however, translate consistently into better performance. Honey itself does not, moreover, have a single glycaemic index: its profile of fructose, glucose and other sugars varies with its botanical origin (Notbohm et al., 2021; Stellingwerff & Cox, 2014).

After exercise, honey may likewise contribute to replenishing carbohydrate reserves. The joint presence of glucose and fructose is physiologically interesting, fructose notably favouring the restoration of liver glycogen. But the available data do not show that honey restores energy reserves better than an equivalent quantity of appropriately formulated carbohydrate. Studies reporting favourable effects of honey compared with water or a placebo do not make it possible to distinguish a specific effect of honey from the effect expected of any carbohydrate intake (Décombaz et al., 2011).

Honey may therefore be a practical source of carbohydrate for endurance, but its value in sport rests on its sugars. The available data do not demonstrate that its enzymes, its polyphenols or its other minor constituents confer any advantage in performance or recovery beyond that provided by a comparable quantity of carbohydrate. And here as elsewhere, using honey in place of another sugar does not have the same dietary meaning as adding it to a ration that already contains enough of it.

10.4 What the law allows one to say

This caution becomes particularly important when moving from research to advertising. In 2011, EFSA assessed several claims proposed specifically for honey, notably protection against oxidative damage, defence against pathogens and maintenance of normal blood cholesterol. The panel concluded that honey was not sufficiently characterised in relation to these effects and that the data presented did not make it possible to establish a cause-and-effect relationship (EFSA NDA Panel, 2011). This conclusion does not mean that honey has no biological activity: the antimicrobial mechanisms described in Chapter 8 are a clear demonstration of it. It means that a chemical or microbiological activity is not sufficient to justify a general promise of health benefit to the consumer.

Swiss law applies the same distinction between scientific knowledge and commercial communication. Health claims may be used only if they are among those authorised by the Ordinance concerning food information (OIDAl) and their conditions are met, or if they have been the subject of an authorisation from the Federal Food Safety and Veterinary Office. A general, non-specific claim relating to health must be accompanied by an authorised specific claim. The FSVO further recalls that claims suggesting properties of preventing, treating or curing a disease are prohibited for foodstuffs (FSVO, n.d.-b). The general prohibition on presenting a food as a remedy derives moreover from the Ordinance on foodstuffs and utility articles (ODAlOUs).

The question of allergies deserves finally to be separated from supposed anti-allergic effects. Allergy to honey is possible, but it appears rare, and the allergens involved may come from pollens as well as from components of bee origin. In a double-blind study conducted among people allergic to pollens, Kiistala et al. (1995) observed no serious reaction, or none clearly attributable to 30 g of honey, despite demonstrable allergenic activity in the samples. Documented cases nevertheless show that systemic reactions can genuinely occur in certain sensitised individuals (Helbling et al., 1992; Kiistala et al., 1995). Being allergic to pollens therefore does not automatically mean having to avoid all honey, but neither does the expression "natural product" mean "free of allergens".

The nutritional place of honey becomes clearer once what is well established is separated from what remains hypothetical. Its chemical matrix is unquestionably more complex than that of a refined sugar; certain honeys may produce slightly different metabolic responses and a few trials have identified favourable changes in biomarkers. But these signals are small, variable, often obtained with substantial doses and do not currently translate into a demonstrated general clinical benefit in the long term. For everyday eating, the most robust fact remains that the sugars in honey count among free sugars. Honey may therefore be appreciated as a distinctive, complex and sensorially rich food without it being necessary to attribute to it health properties that current evidence does not allow to be established.

11. Safety – botulism, natural toxins and residues

Identify the main health risks associated with honey and distinguish spores, natural toxins, residues and environmental contaminants.

The analytical presence of an undesirable substance in honey does not automatically mean that consuming it presents a risk to health. Toxicological assessment distinguishes at least three levels: the hazard, that is, the intrinsic capacity of an agent to cause a harmful effect; the exposure, which depends on its concentration in the food and on the quantity consumed; and the risk, which combines these two dimensions with the sensitivity of the person exposed. This distinction is essential for honey, because modern analytical methods make it possible to detect substances at extremely low concentrations. A pesticide, a metal or a plant toxin may thus be measurable without the dose actually ingested reaching a level of concern. Conversely, certain particular situations – infant botulism or certain honeys containing plant toxins – justify very clear preventive measures precisely because the combination of hazard, exposure and vulnerability is unfavourable there (Bogdanov, 2006; Nowak & Nowak, 2023).

Infant botulism is the most important case. As explained in Chapter 8, the antimicrobial activity of honey does not imply the destruction of all bacterial spores. In adults and older children, the intestinal ecosystem normally constitutes an effective barrier against colonisation by the clostridia responsible for botulism. In infants, by contrast, the gut microbiota and several mechanisms of colonisation resistance are still developing. Spores of Clostridium botulinum that have been ingested may then germinate, multiply in the intestine and produce botulinum neurotoxin there. The mechanism therefore differs from classic foodborne botulism: the infant does not necessarily ingest a toxin preformed in the honey; it may ingest spores, from which the toxin is subsequently produced in its intestine (Arnon & Chin, 1979; Aricò et al., 2026).

The link with honey is solidly established. In the first epidemiological investigations, spores of C. botulinum were found in some samples of honey consumed by affected infants and in a proportion of the honeys studied independently (Chin et al., 1979). This does not mean that honey accounts for the majority of cases of infant botulism. Spores of C. botulinum are widely present in the environment, notably in soils and dust, and the source of exposure remains unknown in a large proportion of cases. But honey is an identifiable and avoidable dietary source; international recommendations therefore converge on a simple rule: do not give honey to children under 12 months (Koepke et al., 2008; Aricò et al., 2026).

11.1 Pyrrolizidine alkaloids and viper's bugloss

Pyrrolizidine alkaloids (PAs) illustrate an entirely different form of hazard. Several plants produce these substances as a defence against herbivores. Among them, certain 1,2-unsaturated PAs are of concern because their metabolism may form reactive derivatives capable of damaging the liver and genetic material; chronic exposure must therefore be kept as low as reasonably possible. When bees collect nectar or pollen from PA-producing plants, some of these molecules may end up in the honey.

The presence of such plants in the landscape does not, however, make a honey problematic in itself. The risk depends on their abundance, on their attractiveness at the time of the flow, on the other resources available and on the proportion their nectar ultimately represents in the honey. A study of Swiss honeys showed that contamination varies strongly with botanical and geographical origin: average exposure is generally low, but particular situations may lead to markedly higher concentrations when colonies intensively exploit large populations of PA-producing plants. The presence of PAs is therefore not a general property of Swiss honey; it depends on the plant resources actually exploited by the colonies (Kast et al., 2014).

Viper's bugloss (Echium vulgare), known in German as Natternkopf, is particularly instructive in this respect in Switzerland. This melliferous plant contains several PAs and can be highly attractive to bees when it is abundant. Work carried out by Agroscope and the University of Neuchâtel determined by which route these alkaloids reach the honey: although PA concentrations are far higher in the plant's pollen than in its nectar, the alkaloid profile found in the honey more closely resembles that of the nectar. The Echium-type PAs present in honey therefore come principally from floral nectar, and not from contamination by pollen (Lucchetti et al., 2016).

The right conclusion is therefore not "viper's bugloss makes honey toxic", but that a locally dominant plant resource can alter exposure to a natural contaminant.

11.2 Grayanotoxins and tutin

Other plants produce natural toxins far more directly responsible for acute poisoning. The best-documented case is that of "mad honey", produced in certain regions where bees exploit rhododendrons containing grayanotoxins. These diterpenes hold certain sodium channels in an activated state and may cause bradycardia, hypotension, dizziness, nausea, cardiac conduction disturbances and, in severe cases, loss of consciousness. The documented poisonings are associated above all with certain honeys from the Black Sea region, notably in Turkey, as well as with a few other regions possessing a corresponding flora. This is therefore a botanically and geographically specific risk, and not a characteristic of European honeys in general (Jansen et al., 2012).

Tutin offers a comparable example with a different ecological mechanism. In New Zealand, leafhoppers feed on plants of the genus Coriaria, notably Coriaria arborea, which contain this neurotoxin. Their honeydew may contain tutin; the bees collect it and the substance subsequently passes into the honey. Human poisonings have been documented, including one outbreak in which 22 possible or probable cases were identified. New Zealand applies a specific risk management system and the maximum level authorised for tutin in honey and comb honey is 0.7 mg/kg (Beasley et al., 2018; New Zealand Ministry for Primary Industries, n.d.). This case again illustrates the importance of context: a molecule constituting a concrete risk in one specific ecological region may be practically irrelevant for a Swiss beekeeper.

11.3 Pesticide residues

Residues of agricultural pesticides raise a different question. Bees may come into contact with insecticides, fungicides or herbicides when foraging in agricultural environments, and certain substances or their metabolites may be detected in pollen, wax or honey. Two toxicological assessments that are often confused must, however, be distinguished: the risk to the bee and the risk to the consumer of honey. A concentration capable of altering the behaviour or the survival of an insect weighing a few tens of milligrams does not mean that the quantity transferred into a portion of honey represents a significant dose for a human being weighing several tens of kilograms.

Assessment for the consumer must take account of the actual concentration in the honey, the quantity consumed, the toxicology of the substance and the applicable regulatory values. The available reviews show that residues of pesticides and metals may be detected in bee products, sometimes with high values in particular samples, but that the levels measured in honey generally lead to low dietary exposures (Bogdanov, 2006; Nowak & Nowak, 2023).

11.4 Varroa treatments and the role of wax

Some of the most directly controllable forms of contamination do not, moreover, come from the environment, but from beekeeping itself. Treatments against Varroa destructor are the classic example. Lipophilic acaricides such as coumaphos, tau-fluvalinate or bromopropylate have a strong affinity for wax: they may accumulate in it during repeated treatments, persist there long after their use has ceased, and then diffuse into the honey in contact with the combs (Wallner, 1999; Bogdanov, 2006).

This property explains why wax plays a particular role as a historical reservoir. The monitoring programme conducted for nearly three decades by Agroscope on commercial Swiss beeswax shows that lipophilic molecules used in the past sometimes remain detectable many years after the corresponding products were abandoned. Kast, Kilchenmann and Charrière (2021) thus still found residues of bromopropylate and tau-fluvalinate in recycled wax although their use had long since ceased. Recycling wax may therefore prolong the presence of a contaminant independently of a beekeeper's current practice. Earlier work from Liebefeld and Hohenheim had already highlighted this fundamental difference between hydrophilic and lipophilic substances (Wallner, 1999; Bogdanov, 2006; Kast et al., 2021).

This observation does not mean that every varroa treatment constitutes a risk to the honey. It underlines above all the importance of the chemical nature of the product, of its authorisation, of the dose and of the timing of application. Hydrophilic substances such as formic and oxalic acid do not accumulate in wax in the same way; they may, on the other hand, pass more readily into the aqueous compartments and, if used at the wrong time or at inappropriate doses, alter the composition or the sensory characteristics of the honey. Use in accordance with authorised products and treatment recommendations is aimed precisely at separating varroa control from the harvest period and at keeping residues at levels compatible with the quality of the honey (Bogdanov, 2006; Kast et al., 2021).

In Switzerland, since bees are regarded as farm animals, the use of veterinary medicines is regulated and must be documented. In principle, the medicines authorised for the species and indication concerned must be used; the derogations provided for in the Ordinance on veterinary medicinal products (OMédV) fall under a specific veterinary framework (FSVO, n.d.-a).

11.5 Heavy metals and emerging contaminants

The detection of heavy metals such as lead, cadmium or mercury in honey must be interpreted according to the same logic. Bees and their products are sometimes used as tools of environmental biomonitoring because they integrate information from a vast foraging area. An analytical signal may thus be of interest for documenting local pollution without honey necessarily constituting a quantitatively important source of that substance in the human diet. Reviews of the literature show that concentrations may be higher near certain industrial or urban areas, but that most exposure assessments conclude that honey makes a low contribution to total dietary exposure; occasional exceptions nevertheless justify monitoring (Bogdanov, 2006; Nowak & Nowak, 2023).

A second route of entry, distinct from the environment, deserves mention: the equipment itself. Because honey is acidic, utensils, containers and processing surfaces may release metals, and some work interprets abnormally high aluminium contents in honey compared with the bees as secondary contamination rather than as transfer from the foraged resource (Dżugan et al., 2018; Solayman et al., 2016). This route is rarely quantified directly in honey, but it has a property that the environment does not: it is entirely controllable by the beekeeper, through the choice of materials (Chapter 13).

The distinction also matters because metals present in honey may alter its chemistry. Spiking trials show that lead, nickel, chromium or cadmium accelerate the formation of HMF during storage (Birhanu & Tolcha, 2023), and copper and iron may inhibit glucose oxidase while favouring a pro-oxidant chemistry of the flavonoids (Sanhueza & Fuentes, 2025). The point is not to demonstrate that ordinary processing in stainless steel degrades honey placed on the market, but to recall that metallic contamination, once present, is not neutral.

Finally, analytical progress is bringing to light a growing category of emerging contaminants: microplastics, plasticisers, per- and polyfluoroalkyl substances, pharmaceutical residues or other organic contaminants can be sought at ever lower concentrations. Their detection constitutes interesting environmental information, but the shortcut "detected = dangerous" must be resisted. For several of these substances, the data are still insufficient to quantify specifically a health risk linked to habitual honey consumption. Multi-residue techniques using chromatography and mass spectrometry are becoming extremely sensitive, whereas toxicological characterisation and exposure assessment do not always advance at the same pace (Nowak & Nowak, 2023; Słowik-Borowiec et al., 2026).

The safety of honey cannot therefore be summarised by a list of "good" or "bad" substances. Some rules are extremely robust and deserve to be simple: no honey before 12 months because of infant botulism. Certain natural risks are real but essentially regional, such as grayanotoxins or tutin. Others, such as pyrrolizidine alkaloids, require an assessment of botanical origin and exposure. Pesticides, veterinary medicines and metals must finally be assessed according to their concentration, consumption and toxicology, and not on the basis of their analytical presence alone. The general rule thus remains the same: an identified hazard is not yet a quantified risk; the risk depends on the dose actually received and on the vulnerability of the person exposed.

12. Honey as a treatment – what clinical trials actually allow us to conclude

Assess what clinical trials actually allow us to conclude about the therapeutic uses of honey according to the indication studied.

Eating honey, taking a spoonful to soothe a cough and applying a honey-based medical product to a wound are three distinct interventions. They do not expose the same tissues to the same concentrations, do not necessarily mobilise the same mechanisms and do not rest on the same level of standardisation. This distinction is essential for interpreting the clinical literature: the existence of antibacterial activity in vitro does not prove that a spoonful of honey exerts an anti-infective effect in the body and, conversely, the absence of a general metabolic benefit from honey as a food does not exclude the possibility that a local application is useful in a particular indication. The mechanisms described in Chapter 8 – osmolarity, acidity, the glucose oxidase–H₂O₂ system, methylglyoxal and other factors – provide biological plausibility, but a therapeutic indication is ultimately judged on results obtained in patients.

The distinction is particularly important for wounds. Medical Grade Honey (MGH) is not simply a food honey chosen for its botanical reputation: products intended for medical use are selected and prepared according to criteria of reproducibility, microbiological safety and quality appropriate to clinical application. Food honey, as we saw in Chapter 8, is not sterile. This property is acceptable for normal consumption, but it becomes a problem as soon as a product is applied directly to a wound.

Postmes et al. (1995) showed experimentally that gamma irradiation could sterilise honeys artificially contaminated with spores of Clostridium botulinum and Bacillus subtilis. At 25 kGy, the heavily inoculated samples were sterilised while overall antibacterial activity was preserved; amylase activity nevertheless decreased to varying degrees depending on the honey. The value of the process is precisely that it reduces microbiological risk without subjecting the product to intense heating liable to alter other constituents (Postmes et al., 1995).

12.1 What the clinical literature allows us to compare

This standardisation does not remove all the difficulties of interpretation. The products studied differ in their botanical origin, their composition and sometimes through the addition of other constituents. Moreover, part of the recent literature devoted to MGH involves researchers employed by manufacturers or concerns products supplied by those companies. These links are explicitly declared in several publications and do not render the results invalid, but they reinforce the importance of independent replication, particularly for case series and for studies concerning one specific commercial product. Caution is all the more necessary in that some recent proposals for defining and standardising MGH themselves include authors affiliated with a manufacturer (Peters et al., 2025).

12.2 Cough: a small effect, but what exactly is being measured?

Acute cough in children is an excellent example of the difficulty of distinguishing pharmacological, physical and contextual effects. The Cochrane review by Oduwole et al. (2018) concludes that honey may reduce certain symptoms of acute cough in children compared with no treatment, with a placebo or with diphenhydramine, but that the quality of the evidence varies and that the differences from dextromethorphan are much less clear. The trials were of short duration and assessed essentially the night-time symptoms reported by parents (Oduwole et al., 2018).

The British NICE recommendation reflects exactly this level of evidence. It does not present honey as a demonstrated treatment for the cause of the cough, but indicates that people over one year of age may wish to try honey to relieve the symptoms of an acute cough, while specifying that the evidence of benefit is limited. NICE also stresses that the clinical significance of the differences observed remains uncertain and that the sweet, viscous consistency of cough products may itself exert a soothing effect (NICE, 2019).

This reservation has become particularly interesting with the multicentre trial by Nishimura et al. (2022). One hundred and sixty-one children aged one to five years with an upper respiratory tract infection received, in double blind, either acacia honey or a honey-flavoured placebo syrup. Night-time symptoms improved in both groups over the two nights of treatment, but no statistically significant difference was observed between honey and placebo (Nishimura et al., 2022).

12.3 What happens when the placebo really tastes of honey?

In several older trials, honey was compared with no treatment or with a product that did not fully reproduce its taste, its viscosity and the experience of taking it before bedtime. Parents therefore sometimes knew, or could guess, that their child was receiving honey.

The trial by Nishimura et al. (2022) corrected precisely this point by using a honey-flavoured syrup, much harder to distinguish from the product being tested.

This result does not demonstrate that every effect of honey on cough is null. The trial was not designed as an equivalence study capable of excluding any small effect. It does show, however, that part of the improvement observed may be explained by non-specific factors: the soothing effect of a sweet, viscous liquid on the throat, salivation, the expectation of benefit, the bedtime ritual and the spontaneous course of the infection.

Honey therefore remains a reasonable option for symptomatic relief in children over one year of age, but the data do not justify making it an antitussive with a solidly demonstrated specific pharmacological action.

A frequent extrapolation must also be avoided: the majority of these trials concern children with acute respiratory infections. They do not automatically demonstrate equal efficacy in adults, in chronic cough, in asthma or in other respiratory diseases. The clinical signal is therefore narrow: possible short-term symptomatic relief in certain acute paediatric coughs, and not a general property of "honey against cough".

12.4 Wounds: results that depend on the type of lesion

The topical application of honey benefits from a much larger clinical literature, but one that is also highly heterogeneous. The Cochrane review by Jull et al. (2015) identified 26 trials totalling 3,011 participants and covering burns, acute wounds, venous ulcers, diabetic foot ulcers, pressure ulcers and several other types of wound. The authors concluded that it was impossible to formulate a general statement about "honey for wounds", precisely because the conditions, the comparators and the methodological quality differed greatly (Jull et al., 2015).

The most favourable signal concerned certain partial-thickness burns. For the comparison with several conventional dressings, the Cochrane review estimated that honey reduced healing time by an average of about 4.7 days, on the basis of two trials totalling 992 participants, and classified this evidence as high quality for that specific outcome (Jull et al., 2015). This conclusion must nevertheless remain tied to its comparators and to its historical body of evidence: a large part of the trials on burns comes from a limited number of teams, notably the work of Subrahmanyam, and the comparators included very diverse treatments – polyurethane film, paraffin gauze, antibiotic gauze or exposure of the burn. The same body of evidence moreover shows that early excision followed by grafting was faster than honey in the burns for which that strategy was indicated. It would therefore be an abuse to turn the Cochrane result into the general claim that "honey is the best treatment for burns" (Subrahmanyam, 1991; Jull et al., 2015).

For chronic wounds, the situation is more uncertain. An updated meta-analysis of eight studies covering 906 participants found in 2024 a shortening of mean healing time with honey dressings, but with very high heterogeneity; the overall quality of the evidence was assessed as very low, notably because of risks of bias, inconsistency and publication bias (Tang et al., 2024). The authors themselves therefore call for cautious interpretation despite favourable quantitative estimates (Tang et al., 2024).

Venous leg ulcers show why the underlying condition must never disappear behind the dressing used. In a large randomised trial of 368 patients, all treated with compression – the central causal treatment of venous ulcers – adding a honey-impregnated dressing did not significantly increase the proportion of ulcers healed after twelve weeks and was associated with more adverse events (Jull et al., 2008). The Cochrane review therefore considered that the data did not demonstrate a robust advantage of honey for this indication (Jull et al., 2015).

For the diabetic foot, the data have evolved. The 2015 Cochrane review had only two small trials totalling 93 participants and concluded that the evidence was insufficient (Jull et al., 2015). Further studies have since been published. A 2026 meta-analysis pooling 16 randomised trials and 1,423 participants reports a higher rate of complete healing and a shorter healing time with honey dressings compared with the comparators. The authors nevertheless assess the certainty as moderate for the healing rate and low for the healing time (Yao et al., 2026). These data therefore strengthen the plausibility of a benefit in the diabetic foot without justifying the replacement of the fundamental elements of treatment: mechanical offloading, control of infection and ischaemia, debridement where necessary, metabolic control and multidisciplinary management.

The literature on wounds as a whole therefore leads less to asking "does honey heal wounds?" than to specifying which wound, which product, which comparator and which underlying treatment. A local intervention may improve a wound environment without treating the cause of the ulcer, and the efficacy demonstrated for a superficial burn cannot be transposed to an arterial ulcer or an infected diabetic foot.

12.5 Oral mucositis: a more specific indication

Oral mucositis caused by radiotherapy for head and neck cancers is another indication in which a relatively coherent clinical signal has emerged. Trials have studied the oral application of honey before or after radiotherapy sessions, sometimes combined with its ingestion. The results are not perfectly homogeneous, but several syntheses show a decrease in the frequency of severe mucositis, as well as favourable signals for pain, weight loss or interruptions of oncological treatment (An et al., 2021; Anshasi et al., 2025).

This literature led the MASCC/ISOO group to issue in 2020 a suggestion in favour of honey administered topically and systemically to prevent oral mucositis in patients with head and neck cancer receiving radiotherapy, with or without chemotherapy (Yarom et al., 2020). The choice of word is important: in the MASCC/ISOO methodology, a suggestion corresponds to a lower level of support than that required for a recommendation. It is therefore neither a universal remedy for all mucositis, nor a strong recommendation applicable to all chemotherapies (Yarom et al., 2020).

An overview of reviews published in 2025 reinforces this nuanced reading. Among twelve systematic reviews assessed, the most consistent results concerned the reduction of severe grade III–IV mucositis in adults receiving radiotherapy. The data were far less conclusive for all grades taken together and for mucositis linked to chemotherapy or chemoradiotherapy, because of the heterogeneity of the studies, the products and the protocols (Anshasi et al., 2025).

It would also be premature to conclude that conventional honeys are clinically superior to manuka honey because some reviews found more favourable results in the trials using the former. The studies do not constitute balanced direct comparisons between types of honey: populations, radiotherapy modalities, doses, administration protocols and comparators all differ. A difference between sets of studies may therefore generate a hypothesis, but it cannot establish a therapeutic hierarchy between botanical origins.

These examples make it possible to draw a fairly clear line between what clinical science supports and what it does not. The antimicrobial activity of honey is well demonstrated in the laboratory, but it does not authorise presenting its consumption as a general treatment for infections. Taking honey may bring modest relief in certain acute coughs in children, even though the trial with a flavoured placebo shows how difficult the specific effect remains to isolate. Topical applications give genuinely favourable results for certain wounds, whereas other indications remain uncertain and the medical product cannot be replaced by a food honey. Finally, severe mucositis linked to radiotherapy of the head and neck is the only indication for which specialist recommendations today acknowledge a benefit credible enough to justify a clinical suggestion.

The therapeutic efficacy of honey is therefore real in certain contexts, but it is specific to the indication, the product, the route of administration and the comparator. It is precisely this specificity that distinguishes an evidence-based medical use from the general claim that "honey heals".

13. From knowledge to beekeeping practice – what actually follows from it

Translate the relevant mechanisms and thresholds into beekeeping decisions on harvest, crystallisation, feeding, storage, residues and communication.

The science of honey is of practical interest only if it makes it possible to take better decisions. Yet a large part of management errors stems less from a lack of rules than from applying a good rule to the wrong question. Capping tells us about the progress of ripening, but does not directly measure water; the refractometer measures water content, but not the whole of microbiological stability; crystallisation tells us about the physical state of the honey, but not about its authenticity; HMF tells us about part of the thermal history, but its concentration is not on its own a measure of toxicological risk. Good practice therefore consists in first identifying the problem to be solved, then choosing the mechanism, the measurement or the legal rule that is actually relevant. This logic can be summarised as: practice → evidence → recommendation.

13.1 Harvest and ripeness

For deciding when to harvest, capping remains an excellent first indicator: a largely capped super signals that the colony has taken the ripening process a long way. But, as we saw in Chapter 2, capping is not an analytical threshold, and an intense flow may lead to situations in which even capped honey still contains more water than desired. Agroscope therefore recommends complementing observation of the combs with a refractometric measurement when the situation is uncertain (Kast & Ritter, 2014). In practice, it is not necessary to measure every cell: capping serves as a biological pre-filter, then a few representative samples make it possible to check the risk situations. An early harvest, a strong flow, humid weather or very heterogeneous frames justify increased vigilance. The operational rule is therefore simple: first observe ripening in the colony, then measure when observation is not sufficient to make the decision safe.

Water content is precisely one of the few parameters for which a simple instrument gives the beekeeper immediately usable information. The refractometer is therefore the practical reference tool, provided it is correctly calibrated, the sample is representative and the temperature and measurement conditions are controlled.

Three levels must nevertheless be distinguished which practice often conflates. Annex 7 of the FDHA Ordinance on foodstuffs of animal origin (ODAlAn) sets a maximum water content of 20 % for ordinary honey: this is a Swiss regulatory requirement. Agroscope recommends ideally aiming for less than 17.5 % in order to have a good margin against fermentation: this is a quality target. The regulations of the apisuisse quality label — a voluntary programme of the Swiss beekeeping associations, presented in 13.8 — set a maximum of 18.5 % for its part: this is a requirement specific to that programme, and not the general legal limit (ODAlAn, Annex 7; Kast & Ritter, 2014; apisuisse, 2024). These figures are not contradictory: they answer different questions. The underlying mechanism remains the water activity described in Chapter 5: the less water is available to osmophilic yeasts, the greater the margin of stability. On calibration and sampling, see Calibrating the refractometer and Controlling the water content of honey.

One qualification is nevertheless needed: the water content measured at extraction is not definitively fixed. As Chapter 5 shows, the equilibrium humidity of honey lies at around 50 to 60 % relative humidity, that is, below that of an ordinary room, and sorption begins at the exposed surface. The critical moments are therefore not those of the sealed jar, but those in which the honey presents its largest exchange surface: supers waiting to be extracted in a humid room, an extractor or a settling tank left open, honey stirred for several days to obtain a creamed texture.

A honey measured at 17.5 % at the time of extraction may thus cross a quality target before jarring, without any water having been added. The practical consequence is twofold: control the humidity of the honey house and limit the time during which the honey remains exposed to the air. The same mechanism moreover works in the other direction; air that is sufficiently dry causes honey to lose water, which explains the use of a dehumidified room for supers awaiting extraction.

This possibility nevertheless calls for a distinction. Drying the air in a room to prevent supers from absorbing moisture while awaiting extraction is not the same operation as dehydrating an already extracted honey in order to lower its water content. As Chapter 9 shows, early harvesting followed by technical drying is precisely one of the quality problems identified in world production. The reason lies in what Chapter 2 explains: ripening is not only a loss of water, but also an enzymatic and compositional transformation. Drying corrects the figure, not the ripeness. A honey harvested too early and then dehydrated may show a compliant water content without having undergone all the transformations that make ripe honey.

13.2 Crystallisation: managing rather than submitting

Crystallisation calls for a comparable change of perspective. Seeking to keep every honey liquid indefinitely amounts to fighting a thermodynamically normal property of a solution supersaturated with glucose. The practical question is therefore not how to prevent crystallisation, but what texture one wishes to obtain. When a honey crystallises spontaneously and homogeneously, it loses nothing of its quality.

If the objective is a creamed honey with fine crystals, seeding, temperature and stirring make it possible to guide the process. The scientific principle is robust: multiplying nucleation sites favours a more distributed and generally finer crystallisation. The operational details – frequency of stirring, quantity of seed, final consistency – belong on the other hand to a body of know-how that varies with the type of honey and the equipment used; they must be presented as such and not as universal physicochemical laws. The practical recommendation is therefore to manage crystallisation when texture is a commercial objective, rather than to regard all crystallisation as a defect. The methods are developed in Crystallisation of honey.

13.3 Melezitose: deciding early

Melezitose is the case in which early observation can radically alter the options subsequently available, both at the time of harvest and when preparing for overwintering. If a honeydew honey begins to crystallise rapidly in the combs and becomes increasingly difficult to extract, waiting until the cells have turned into "cement honey" greatly reduces the options. A harvest that is still extractable can be worked quickly; when a substantial part is already crystallised in the combs, specific strategies become necessary. The practical recommendation is therefore to recognise the change in behaviour early, to check extractability and to intervene before crystallisation in the combs makes conventional harvesting impossible.

The question does not, however, stop at the harvest. As explained in Chapter 6, feeding experiments show that a diet rich in melezitose increases the gut load and the mortality of bees, while field observations associate melezitose flows with an increased risk during overwintering (Seeburger et al., 2020; Oberreiter & Brodschneider, 2020). The Agroscope Swiss Bee Research Centre consequently advises against letting colonies overwinter exclusively on stores rich in melezitose. When substantial quantities of forest or melezitose honey are still present in the brood combs at the end of the season, Agroscope recommends replacing several heavily filled combs with empty combs or foundation before feeding, in order to allow the storage of a better-suited winter food (Agroscope, n.d.-a).

The practical consequence is therefore twofold: during the flow, detect melezitose early enough to preserve the harvesting options; before overwintering, prevent a high proportion of these stores from constituting the colony's only food.

The various practical options are developed in Melezitose: origins, risks and practical management in the apiary.

13.4 Honey or syrup: which is the better winter food?

The idea that honey is necessarily far more beneficial to bees than a feeding syrup seems intuitive: it is their natural food and it contains, besides the sugars, organic acids, enzymes, minerals and various substances of plant origin. These constituents may have biological effects. But for winter stores, the dominant quantitative function remains the supply of energy.

Honey itself consists mainly of fructose and glucose. The comparative data available nevertheless remain limited and do not demonstrate a general and robust superiority of appropriate blossom honey stores over a correctly prepared sucrose syrup (Quinlan et al., 2023).

Sucrose is not absorbed directly in its initial form: it must first be hydrolysed into glucose and fructose. Bees are, however, very well equipped for this reaction thanks to their α-glucosidase, often called invertase. In the industrial inversion of a syrup, this same hydrolysis is carried out before the product is given to the bees. It may be achieved enzymatically or chemically. The fact that a syrup already contains more glucose and fructose does not, however, mean that it is biologically superior: in a field trial covering 70 colonies, an enzymatically inverted syrup did not significantly improve overwintering, spring development or subsequent production compared with a sucrose syrup (Přidal et al., 2023).

The quality of the syrup nevertheless counts as much as its composition. Syrups may contain hydroxymethylfurfural, whose content increases with the heat of manufacture and the duration of storage. Now this compound, which for humans is merely a quality marker (Chapter 7), is not harmless for the bee: experimental work shows that HMF affects caged bees, and that its formation in high-fructose corn syrups is accompanied by toxicity to the bee (LeBlanc et al., 2009; Gregorc et al., 2020). More broadly, sugar feeding may have side effects on colony health that do not reduce to its energy contribution alone (Frizzera et al., 2020). A syrup made with care and kept cool is therefore preferable to an overheated or long-stored syrup.

There is therefore no scientifically established optimal proportion between monosaccharides, disaccharides and more complex sugars. The degree of inversion is not, on its own, a criterion of feed quality. What matters more is that the sugars can be efficiently hydrolysed, absorbed and metabolised by the bee, that the stores remain physically usable and that the product does not contain problematic quantities of poorly assimilable sugars or of degradation products (Přidal et al., 2023; Szczęsna et al., 2021).

This comparison also shows why the opposition between "natural food" and "artificial sugar" is too simple. Not all honeys constitute an equally favourable winter food, and not all syrups are equivalent. A honey rich in melezitose may be less suited to overwintering than a simple sucrose syrup, whereas a poorly formulated, strongly heated or acidified syrup may introduce risks that do not exist with correctly prepared sucrose.

The practical recommendation is therefore to favour carbohydrate stores that are readily usable by the bees, stable in the combs and free of undesirable degradation products. Sucrose meets these criteria well and constitutes a solidly established practical standard. An already inverted syrup may be convenient, but "more inverted" does not mean "better for the bee".

13.5 Heat, packing and storage

Heat too must be managed according to an objective rather than according to a temperature elevated into a symbol. Liquefying a honey, facilitating pumping or making jarring possible may require an input of heat. Chapter 7 showed that there is no magic biological boundary at 40 °C: the formation of HMF, enzyme inactivation and the loss of aroma compounds obey kinetics dependent on time, on temperature and on the honey itself (Tosi et al., 2002, 2008; Manickavasagam et al., 2024). This does not mean that temperature is unimportant; on the contrary, reaction rates increase sharply as it rises. The most robust practice therefore consists in heating only to the level necessary for the process, for the shortest possible time, and then not keeping the honey warm unnecessarily. Large containers require particular attention, since the poor thermal conductivity of honey may create substantial differences between the centre and the heated surfaces: an acceptable average temperature does not protect against local overheating. A well-regulated warming cabinet, appropriate stirring where the technique allows it and the avoidance of hot spots therefore count for more than confidence placed in an isolated figure.

The same logic continues during storage. Once extracted and packed, honey continues to change slowly. An airtight container limits exchanges of moisture with the air; a dry environment reduces the risk of water absorption at the surface; darkness limits the degradation of colour, phenolic compounds and the peroxide system (Chapter 7); a moderate temperature slows the reactions of ageing. For quality-preserving storage, Agroscope recommends airtight, dry and dark storage in a cool area of about 10 to 16 °C (Agroscope, n.d.-b).

This range is not optimal for any of the parameters taken individually, and that is precisely what makes it a practical recommendation. It does not minimise chemical ageing: as the box in Chapter 7 shows, storage at 4 °C or freezing preserves HMF, enzyme activity and colour better. Nor does it minimise crystallisation, since it largely covers the zone in which crystallisation is most rapid (Chapter 5). What it offers is a compromise: a temperature clearly below that of a heated room, hence a substantial slowing of ageing; a honey that remains immediately workable, without refrigerated equipment or the constraint of a cold chain; and a range achievable in a cellar or an ordinary honey house.

The compromise is all the more acceptable in that crystallisation is not a quality defect. A stock intended to be sold crystallised or worked into creamed honey loses nothing by being kept within this range. Conversely, if the objective is to preserve for as long as possible the characteristics of a particular liquid honey — a honey fresh from the harvest intended for a competition, a reference sample, a batch kept for several years — cold or freezing become the better option. The temperature must therefore be chosen according to what one wishes to preserve, and not by seeking a single value.

Whatever temperature is chosen, it is the tightness of the container that decides. Reviews of low-temperature storage generally show no significant variation in water content when the honey is kept in an airtight container (Manickavasagam et al., 2024). The risk comes from exposure to humid air, and it exists just as much at room temperature: as Chapter 5 shows, the equilibrium humidity of honey is lower than that of an ordinary dwelling. What counts is therefore the exposed surface and the humidity of the air, not the difference in temperature.

The container decides something else as well. Tananaki et al. (2005) stored honeydew honey in containers previously used for spices, coffee or tea, and found in the honey volatile compounds derived from the previous contents: a single substance with coffee, two with peppermint, three with sage, four with cinnamon, eight with oregano and up to nine with basil, whereas tea produced no detectable effect. The transfer was perceptible to tasters, and its source was not the wall of the container but the plastic seal of the lid (Tananaki et al., 2005).

The rule that follows is more precise than the adage that "honey takes up odours". What is documented is transfer from a container or a lid that has already held an aromatic product, with an intensity that depends strongly on that product. By contrast, the absorption of odours from the ambient air of a room, as well as the influence of the exposed surface and the duration of exposure, is practically unmeasured; the usual caution on this point is a matter of professional experience rather than of demonstration. The practical consequence is simple: do not reuse containers or lids that have held an odorous product, and pay as much attention to the seal as to the container itself.

The material counts too. Honey is naturally acidic, and acidity mobilises certain metals in contact with food surfaces; release increases as the pH falls, the temperature rises and contact is prolonged. Most of the measurements nevertheless come from acidic food simulants and kitchen utensils rather than from experiments conducted in honey itself (Koo et al., 2020; Solayman et al., 2016). Reactive surfaces — copper, iron, zinc, galvanised equipment — are therefore unsuitable for storage. Food-grade stainless steel releases mainly chromium and nickel during the first uses, then markedly less thereafter (Koo et al., 2020), and a twelve-month comparison between a metal tin, clear glass and dark glass showed no significant difference in pH, proline, diastase, HMF, phenolic compounds or catalase activity (Yiğit et al., 2024). It was storage time, not the material, that governed degradation. A clean stainless steel spoon may therefore be used without reservation: contact that brief does not alter the honey.

In all cases, each subsequent episode of heating adds to the previous thermal history. It is therefore preferable to liquefy the quantities actually needed rather than to reheat an entire stock several times.

For the beekeeper who packs their own honey, the same logic leads to favouring the minimum treatment necessary for the objective sought: strain sufficiently to remove undesirable particles, heat only when a technological operation requires it and avoid repeated thermal cycles. Limited processing helps to preserve as well as possible the aromatic characteristics, the enzyme activity and certain sensitive constituents of the honey. It does not, however, allow it to be claimed that a "raw", "unheated" or lightly processed honey is generally better for health than a honey correctly packed on a larger scale: this clinical superiority is not currently demonstrated (Subramanian et al., 2007; Manickavasagam et al., 2024; Majtán, 2024).

13.6 Residues: what is decided in the apiary

Controlling residues begins before the harvest. Work from Liebefeld and Hohenheim has long shown that certain lipophilic acaricides may accumulate durably in wax, which then becomes a reservoir liable to pass part of the residues on to the honey (Wallner, 1999; Bogdanov, 2006). Swiss monitoring confirms that substances used historically remain detectable for years in recycled wax circuits (Kast et al., 2021).

The practical consequence is not to give up controlling Varroa destructor, which would be far more damaging biologically. It is to use, in principle, the veterinary medicines authorised for the species and indication concerned, according to the doses, periods and methods prescribed; the derogations provided for in the OMédV fall under a specific veterinary framework. The choice of substance and the treatment schedule must moreover be considered in relation to the production cycle: a method that is useful against varroa is not automatically appropriate when the supers intended for harvest are in place.

Wax management is an integral part of this control. Documenting treatments and renewing combs on a considered basis therefore bear directly on the food quality of the honey.

13.7 Labelling and communication

The final stage is the one the consumer sees: the label and the communication. Swiss law allows the designation of honey to be supplemented by a floral or plant origin if the honey comes principally from the source indicated and displays its organoleptic, physicochemical and microscopic characteristics. A regional, territorial or topographical indication presupposes that the honey actually comes from the stated origin (ODAlAn, Art. 98). Agroscope moreover provides a practical guide on the correct labelling of honey (Agroscope, 2025).

Nutrition and health claims, by contrast, fall under the Ordinance concerning food information (OIDAl). A nutrition claim must be among those provided for and must comply with their conditions; a health claim must be authorised by the OIDAl or have been the subject of an authorisation from the FSVO. Promises of preventing, treating or curing a disease are not permitted for a foodstuff. The ODAlOUs finally adds the general principle of protection against deception and prohibits attributing to a food properties that are not sufficiently established scientifically (ODAlOUs, Art. 12; OIDAl, Art. 29–35; FSVO, n.d.-b).

13.8 What can be said about one's honey – and what is better left unsaid?

Since 14 June 2026, the European Union has moreover applied more detailed rules to honey blends: the countries of harvest must in principle be indicated in the principal field of vision, in descending order, with their proportion by weight and a tolerance of 5 % for each declared share; certain simplifications are possible for blends comprising more than four origins (European Union, 2024). The directive further obliges the Commission to establish, by 14 June 2028, harmonised methods for detecting adulterated honeys. These provisions are rules of the European Union and do not apply automatically to marketing within Switzerland. Swiss law as currently applicable continues to govern the origin and labelling of honey notably through the ODAlAn and the OIDAl. The consolidated versions of these texts consulted on 29 August 2026 do not take up the European obligation to indicate the percentages of each country in a blend. A honey intended for the EU market must, by contrast, comply with the requirements applicable to that market (European Union, 2024; ODAlAn; OIDAl).

These new European requirements have not been transposed as such into Swiss law as at 30 August 2026. Art. 15 para. 4 OIDAl continues to treat honey blends as processed foodstuffs, for which a wider geographical area — for example "EU" or "South America" — may be indicated in place of the countries of production (OIDAl, Art. 15 para. 4). This relaxation had been introduced precisely in order to align with the European regime then in force; the current divergence is therefore the product of a unilateral development of European law, and it could close again. A honey intended for the Union market must in any case comply with the requirements applicable to that market.

The transfer from science to beekeeping can finally be summarised in a few principles.

• Observe before measuring, but measure when observation is not sufficient.
• Manage crystallisation rather than treating it as a defect.
• React early when the behaviour of a honeydew honey changes.
• Heat for a precise objective, as briefly and as moderately as possible.
• Store in a way that slows the thermal history and prevents the uptake of moisture.
• Treat colonies against varroa with authorised products and include wax in residue management.
• Communicate only what origin, analysis and the law actually allow to be asserted.

Good beekeeping practice therefore does not consist in accumulating empirical rules: it consists in knowing which mechanism, which measurement or which legal framework is relevant to the decision to be taken.

---

See also:

• Controlling the water content of honey
• Crystallisation of honey
• The honey harvest
• Melezitose: origins, risks and practical management in the apiary
• Beekeeping and honey law in Switzerland
• Palynology and pollen analysis

References

• Abraham, K., Gürtler, R., Berg, K., Heinemeyer, G., Lampen, A., & Appel, K. E. (2011). Toxicology and risk assessment of 5-hydroxymethylfurfural in food. Molecular Nutrition & Food Research, 55(5), 667–678. https://doi.org/10.1002/mnfr.201000564
• Agroscope. (2025). Correct labelling of honey. Swiss Bee Research Centre. https://www.agroscope.admin.ch/fr/miel
• Agroscope. (n.d.-a). Melezitose and late flows. Retrieved 29 August 2026 from https://www.agroscope.admin.ch/fr/melezitose-et-miellees-tardives
• Agroscope. (n.d.-b). Honey. Retrieved 29 August 2026 from https://www.agroscope.admin.ch/fr/miel
• Ahmed, A., Tul-Noor, Z., Lee, D., Bajwah, S., Ahmed, Z., Zafar, S., Syeda, M., Jamil, F., Qureshi, F., Zia, F., Baig, R., Ahmed, S., Tayyiba, M., Ahmad, S., Ramdath, D., Tsao, R., Cui, S., Kendall, C. W. C., de Souza, R. J.,… Sievenpiper, J. L. (2023). Effect of honey on cardiometabolic risk factors: A systematic review and meta-analysis. Nutrition Reviews, 81(7), 758–774. https://doi.org/10.1093/nutrit/nuac086
• Al-Ghzawi, A. A., Zaitoun, S., Rababah, T., Samarah, N., Bin Zaidan, A., Yücel, S., & Fish, H. (2026). Microwave liquefaction of crystallized Jordanian honeys and its impact on quality parameters. ACS Omega, 11(13), 20520–20533. https://doi.org/10.1021/acsomega.5c11767
• Almeida-Muradian, L. B. de, Barth, O. M., Dietemann, V., Eyer, M., de Freitas, A. da S., Martel, A.-C., Marcazzan, G. L., Marchese, C. M., Mucignat-Caretta, C., Pascual-Maté, A., Reybroeck, W., Sancho, M. T., & Sattler, J. A. G. (2020). Standard methods for Apis mellifera honey research. Journal of Apicultural Research, 59(3), 1–62. https://doi.org/10.1080/00218839.2020.1738135
• An, W., Li, S., & Qin, L. (2021). Role of honey in preventing radiation-induced oral mucositis: A meta-analysis of randomized controlled trials. Food & Function, 12(8), 3352–3365. https://doi.org/10.1039/D0FO02808H
• Anshasi, H., Abufarsakh, B., Alkhawaldeh, J. M., & Al Halalmeh, S. (2025). Honey for managing oral mucositis induced by cancer therapies: An overview of systematic reviews. Complementary Therapies in Medicine, 92, 103197. https://doi.org/10.1016/j.ctim.2025.103197
• apisuisse. (2024). Reglement zur Individualmarke Honig-Qualitätssiegel: Ausgabe 2024. https://bienen.ch/wp-content/uploads/2024/05/Honigreglement_apisuisse_2024_DE.pdf
• Aricò, M. O., Caselli, D., Stefanizzi, P., Tafuri, S., & Aricò, M. (2026). Infant botulism and honey exposure: Global epidemiology, prevention policies, and communication strategies. Acta Paediatrica, 115(7), 1429–1433. https://doi.org/10.1111/apa.70538
• Arnon, S. S., & Chin, J. (1979). The clinical spectrum of infant botulism. Reviews of Infectious Diseases, 1(4), 614–624. https://doi.org/10.1093/clinids/1.4.614
• Arslan, K., & Turhan, M. (2022). Static water vapor sorption properties of honey. Journal of Apicultural Science, 66(1), 121–132. https://doi.org/10.2478/jas-2022-0010
• Asha'ari, Z. A., Ahmad, M. Z., Wan Din, W. S. J., Che, C. M., & Leman, I. (2013). Ingestion of honey improves the symptoms of allergic rhinitis: Evidence from a randomized placebo-controlled trial in the East coast of Peninsular Malaysia. Annals of Saudi Medicine, 33(5), 469–475. https://doi.org/10.5144/0256-4947.2013.469
• Bahrami, M., Ataie-Jafari, A., Hosseini, S., Foruzanfar, M. H., Rahmani, M., & Pajouhi, M. (2009). Effects of natural honey consumption in diabetic patients: An 8-week randomized clinical trial. International Journal of Food Sciences and Nutrition, 60(7), 618–626. https://doi.org/10.3109/09637480801990389
• Baloš, M. Ž., Popov, N., Jakšić, S., Mihaljev, Ž., Pelić, M., Ratajac, R., & Pelić, D. L. (2023). Sunflower honey — Evaluation of quality and stability during storage. Foods, 12(13), 2585. https://doi.org/10.3390/foods12132585
• Bang, L. M., Buntting, C., & Molan, P. (2003). The effect of dilution on the rate of hydrogen peroxide production in honey and its implications for wound healing. Journal of Alternative and Complementary Medicine, 9(2), 267–273. https://doi.org/10.1089/10755530360623383
• Beasley, M., Hood, D., Anderson, P., Reeve, J., & Slaughter, R. J. (2018). Poisoning due to tutin in honey—a report of an outbreak in New Zealand. New Zealand Medical Journal, 131(1473), 59–71. https://pubmed.ncbi.nlm.nih.gov/29649198/
• Beuchat, L. R. (1983). Influence of water activity on growth, metabolic activities and survival of yeasts and molds. Journal of Food Protection, 46(2), 135–141. https://doi.org/10.4315/0362-028X-46.2.135
• Bio Suisse. (2026). Standards for the production, processing and trade of Bud products (version of 15 April 2026). Bio Suisse.
• Birhanu, T., & Tolcha, T. (2023). Influence of heavy metal ions and storage time on hydroxymethylfurfural formation in honey collected from Ilu Woreda, Oromia Regional State, Ethiopia. Bulletin of the Chemical Society of Ethiopia, 38(1), 1–12. https://doi.org/10.4314/bcse.v38i1.1
• Biswas, A., Naresh, K. S., Jaygadkar, S. S., & Chaudhari, S. R. (2023). Enabling honey quality and authenticity with NMR and LC-IRMS based platform. Food Chemistry, 416, 135825. https://doi.org/10.1016/j.foodchem.2023.135825
• Bogdanov, S. (2006). Contaminants of bee products. Apidologie, 37(1), 1–18. https://doi.org/10.1051/apido:2005043
• Bowie, A. (2020). The ritual role of honey in ancient Egypt, Hatti and Greece. Istraživanja / Journal of Historical Researches, 31, 7–23. https://doi.org/10.19090/i.2020.31.7-23
• Brudzynski, K. (2020). A current perspective on hydrogen peroxide production in honey. A review. Food Chemistry, 332, 127229. https://doi.org/10.1016/j.foodchem.2020.127229
• Brudzynski, K., & Sjaarda, C. (2021). Colloidal structure of honey and its influence on antibacterial activity. Comprehensive Reviews in Food Science and Food Safety, 20(3), 2481–2505. https://doi.org/10.1111/1541-4337.12720
• Bucekova, M., Buriova, M., Pekárik, L., Majtan, V., & Majtán, J. (2018). Phytochemicals-mediated production of hydrogen peroxide is crucial for high antibacterial activity of honeydew honey. Scientific Reports, 8, 9061. https://doi.org/10.1038/s41598-018-27449-3
• Castro-Vázquez, L., Alañón, M. E., González-Viñas, M. A., & Pérez-Coello, M. S. (2012). Changes in the volatile fractions and sensory properties of heather honey during storage under different temperatures. European Food Research and Technology, 235, 185–193. https://doi.org/10.1007/s00217-012-1756-1
• Castro-Vázquez, L., Díaz-Maroto, M. C., González-Viñas, M. A., de la Fuente, E., & Pérez-Coello, M. S. (2008). Influence of storage conditions on chemical composition and sensory properties of citrus honey. Journal of Agricultural and Food Chemistry, 56(6), 1999–2006. https://doi.org/10.1021/jf072227k
• Charrière, J.-D., & Würgler, O. (2024). Beekeeping in Switzerland and in international comparison (Agroscope Transfer No. 528). Agroscope. https://ira.agroscope.ch/fr-CH/Page/Publikation?einzelpublikationId=60412
• Chin, J., Arnon, S. S., & Midura, T. F. (1979). Food and environmental aspects of infant botulism in California. Reviews of Infectious Diseases, 1(4), 693–697. https://doi.org/10.1093/clinids/1.4.693
• Chirife, J., Zamora, M. C., & Motto, A. (2006). The correlation between water activity and % moisture in honey: Fundamental aspects and application to Argentine honeys. Journal of Food Engineering, 72(3), 287–292. https://doi.org/10.1016/j.jfoodeng.2004.12.009
• Currell, K., & Jeukendrup, A. E. (2008). Superior endurance performance with ingestion of multiple transportable carbohydrates. Medicine & Science in Sports & Exercise, 40(2), 275–281. https://doi.org/10.1249/mss.0b013e31815adf19
• da Silva, P. M., Gauche, C., Gonzaga, L. V., Costa, A. C. O., & Fett, R. (2016). Honey: Chemical composition, stability and authenticity. Food Chemistry, 196, 309–323. https://doi.org/10.1016/j.foodchem.2015.09.051
• De-Melo, A. A. M., Almeida-Muradian, L. B., Sancho, M. T., & Pascual-Maté, A. (2018). Composition and properties of Apis mellifera honey: A review. Journal of Apicultural Research, 57(1), 5–37. https://doi.org/10.1080/00218839.2017.1338444
• Décombaz, J., Jentjens, R., Ith, M., Scheurer, E., Buehler, T., Jeukendrup, A., & Boesch, C. (2011). Fructose and galactose enhance postexercise human liver glycogen synthesis. Medicine & Science in Sports & Exercise, 43(10), 1964–1971. https://doi.org/10.1249/MSS.0b013e318218ca5a
• Despland, C., Walther, B., Kast, C., Campos, V., Rey, V., Stefanoni, N., & Tappy, L. (2017). A randomized-controlled clinical trial of high fructose diets from either Robinia honey or free fructose and glucose in healthy normal weight males. Clinical Nutrition ESPEN, 19, 16–22. https://doi.org/10.1016/j.clnesp.2017.01.009
• Doner, L. W. (1977). The sugars of honey-A review. Journal of the Science of Food and Agriculture, 28(5), 443–456. https://doi.org/10.1002/jsfa.2740280508
• Doull, K. M., & Mew, P. (1977). The hygroscopic properties of different dilutions of honey. Apidologie, 8(1), 19–24. https://doi.org/10.1051/apido:19770102
• Dżugan, M., Wesołowska, M., Zaguła, G., Kaczmarski, M., Czernicka, M., & Puchalski, C. (2018). Honeybees (Apis mellifera) as a biological barrier for contamination of honey by environmental toxic metals. Environmental Monitoring and Assessment, 190, 101. https://doi.org/10.1007/s10661-018-6474-0
• EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). (2011). Scientific opinion on the substantiation of health claims related to: dairy products (ID 1140, 1141, 1191), raw or processed food products of animal origin, plus bread and panification products (ID 1193, 1194), herbal yeast plasmolysate (ID 1815, 1816), apple polyphenols (ID 2713), rye flour (ID 1266), tomato juice (ID 1202), whey protein and alphalactalbumin (ID 424, 430, 432, 725, 1433), 'brocco shoots', 'broccoli sprout powder' and Brassica oleracea var. italica (broccoli) (ID 1362, 1481, 2844, 2845), honey (ID 1159, 1160, 1318, 4678, 4679), and Cucurbita pepo L. (pumpkin) seeds and seed extracts (ID 2029, 2365) pursuant to Article 13(1) of Regulation (EC) No 1924/2006. EFSA Journal, 9(6), 2243. https://doi.org/10.2903/j.efsa.2011.2243
• Erban, T., Shcherbachenko, E., Talacko, P., & Harant, K. (2021). A single honey proteome dataset for identifying adulteration by foreign amylases and mining various protein markers natural to honey. Journal of Proteomics, 239, 104157. https://doi.org/10.1016/j.jprot.2021.104157
• Escriche, I., Visquert, M., Juan-Borrás, M., & Fito, P. (2009). Influence of simulated industrial thermal treatments on the volatile fractions of different varieties of honey. Food Chemistry, 112(2), 329–338. https://doi.org/10.1016/j.foodchem.2008.05.068
• European Commission. (2023). EU coordinated action 'From the Hives' (Honey 2021-2022). https://food.ec.europa.eu/food-safety/eu-agri-food-fraud-network/eu-coordinated-actions/honey-2021-2022_en
• European Commission. (2026). Honey market overview (January 2026). Directorate-General for Agriculture and Rural Development. https://agriculture.ec.europa.eu/farming/animal-products/honey_en
• European Union. (2024). Directive (EU) 2024/1438 of the European Parliament and of the Council of 14 May 2024 amending Council Directives 2001/110/EC relating to honey, 2001/112/EC relating to fruit juices, 2001/113/EC relating to fruit jams, and 2001/114/EC relating to certain partly or wholly dehydrated preserved milk for human consumption. EUR-Lex. https://eur-lex.europa.eu/eli/dir/2024/1438/oj
• Eyer, M., Neumann, P., & Dietemann, V. (2016). A look into the cell: Honey storage in honey bees, Apis mellifera. PLOS ONE, 11(8), e0161059. https://doi.org/10.1371/journal.pone.0161059
• Fallico, B., Arena, E., & Zappalà, M. (2009). Prediction of honey shelf life. Journal of Food Quality, 32(3), 352–368. https://doi.org/10.1111/j.1745-4557.2009.00253.x
• Faustino, C., & Pinheiro, L. (2021). Analytical rheology of honey: A state-of-the-art review. Foods, 10(8), 1709. https://doi.org/10.3390/foods10081709
• Faustino, D. C., Santos, L. O., Feitosa, B. A. C., Cedro, M. A., Borges-Silva, W., Nonaka, C. K. V., Souza, B. S. F., Daleprane, J. B., Gondim, L. F. P., Guedes, C. E. S., & Matos, L. C. P. (2026). From cellular toxicity and inflammation to human safety: An integrated in vitro-in silico assessment of 5-hydroxymethylfurfural (HMF). Toxicology Letters, 419, 111892. https://doi.org/10.1016/j.toxlet.2026.111892
• Federal Department of Economic Affairs, Education and Research. (2026). EAER Ordinance on organic farming (SR 910.181), status as at 1 January 2026. Fedlex. Retrieved 30 August 2026 from https://www.fedlex.admin.ch/eli/cc/1997/2498_2498_2498/fr
• Federal Department of Home Affairs. (2016a). FDHA Ordinance concerning food information (OIDAl; SR 817.022.16), status as at 1 July 2025. Fedlex. Retrieved 29 August 2026 from https://www.fedlex.admin.ch/eli/cc/2017/158/fr
• Federal Department of Home Affairs. (2016b). FDHA Ordinance on foodstuffs of animal origin (ODAlAn; SR 817.022.108), status as at 26 August 2026. Fedlex. Retrieved 29 August 2026 from https://www.fedlex.admin.ch/eli/cc/2017/152/fr
• Federal Food Safety and Veterinary Office. (n.d.-a). Bees. Retrieved 29 August 2026 from https://www.blv.admin.ch/fr/abeilles
• Federal Food Safety and Veterinary Office. (n.d.-b). Nutrition and health claims. Retrieved 29 August 2026 from https://www.blv.admin.ch/fr/allegations-nutritionnelles-et-de-sante
• Fernandes, K. E., Dong, A. Z., Levina, A., Cokcetin, N. N., Brooks, P., & Carter, D. A. (2024). Long-term stability and the physical and chemical factors predictive for antimicrobial activity in Australian honey. PLOS ONE, 19(5), e0303095. https://doi.org/10.1371/journal.pone.0303095
• Fortis, H. O., Ravikanti, S., Barrett, J. S., Lopez, E. M., Bampouras, T. M., Haworth, J. J., Areta, J. L., & Pugh, J. N. (2025). Could it bee? Honey ingestion induces comparable metabolic responses to traditional carbohydrate-based sports nutrition product during 3-hr steady-state cycling and subsequent exercise capacity test. International Journal of Sport Nutrition and Exercise Metabolism, 35(5), 424–432. https://doi.org/10.1123/ijsnem.2024-0244
• Frizzera, D., Del Fabbro, S., Ortis, G., Zanni, V., Bortolomeazzi, R., Nazzi, F., & Annoscia, D. (2020). Possible side effects of sugar supplementary nutrition on honey bee health. Apidologie, 51(4), 594–608. https://doi.org/10.1007/s13592-020-00745-6
• Gleiter, R. A., Horn, H., & Isengard, H.-D. (2006). Influence of type and state of crystallisation on the water activity of honey. Food Chemistry, 96(3), 441–445. https://doi.org/10.1016/j.foodchem.2005.03.051
• Gourdomichali, T., & Papakonstantinou, E. (2018). Short-term effects of six Greek honey varieties on glycemic response: A randomized clinical trial in healthy subjects. European Journal of Clinical Nutrition, 72, 1709–1716. https://doi.org/10.1038/s41430-018-0160-8
• Gregorc, A., Jurišić, S., & Sampson, B. (2020). Hydroxymethylfurfural affects caged honey bees (Apis mellifera carnica). Diversity, 12(1), 18. https://doi.org/10.3390/d12010018
• Guler, A., Kocaokutgen, H., Garipoğlu, A. V., Onder, H., Ekinci, D., & Bıyık, S. (2014). Detection of adulterated honey produced by honeybee (Apis mellifera L.) colonies fed with different levels of commercial industrial sugar (C3 and C4 plants) syrups by the carbon isotope ratio analysis. Food Chemistry, 155, 155–160. https://doi.org/10.1016/j.foodchem.2014.01.033
• Hawkins, J., de Vere, N., Griffith, A., Ford, C. R., Allainguillaume, J., Hegarty, M. J., Baillie, L., & Adams-Groom, B. (2015). Using DNA metabarcoding to identify the floral composition of honey: A new tool for investigating honey bee foraging preferences. PLOS ONE, 10(8), e0134735. https://doi.org/10.1371/journal.pone.0134735
• He, C., Liu, Y., Liu, H., Zheng, X., Shen, G., & Feng, J. (2020). Compositional identification and authentication of Chinese honeys by ¹H NMR combined with multivariate analysis. Food Research International, 130, 108936. https://doi.org/10.1016/j.foodres.2019.108936
• Helbling, A., Peter, C., Berchtold, E., Bogdanov, S., & Müller, U. (1992). Allergy to honey: Relation to pollen and honey bee allergy. Allergy, 47(1), 41–49. https://doi.org/10.1111/j.1398-9995.1992.tb02248.x
• Hills, S. P., Mitchell, P., Wells, C., & Russell, M. (2019). Honey supplementation and exercise: A systematic review. Nutrients, 11(7), 1586. https://doi.org/10.3390/nu11071586
• Idris, A., Abdalla, A., Zeitoun, M., & Ali, S. (2021). Influence of floral origin and storage conditions on physicochemical properties of Libyan honeys. European Journal of Nutrition & Food Safety, 13(8), 25–37. https://doi.org/10.9734/ejnfs/2021/v13i830442
• Ischayek, J. I., & Kern, M. (2006). US honeys varying in glucose and fructose content elicit similar glycemic indexes. Journal of the American Dietetic Association, 106(8), 1260–1262. https://doi.org/10.1016/j.jada.2006.05.003
• Jansen, S. A., Kleerekooper, I., Hofman, Z. L. M., Kappen, I. F. P. M., Stary-Weinzinger, A., & van der Heyden, M. A. G. (2012). Grayanotoxin poisoning: 'Mad honey disease' and beyond. Cardiovascular Toxicology, 12(3), 208–215. https://doi.org/10.1007/s12012-012-9162-2
• Joint Research Centre. (2023, 23 March). Food fraud: How genuine is your honey? European Commission. https://joint-research-centre.ec.europa.eu/jrc-news-and-updates/food-fraud-how-genuine-your-honey-2023-03-23_en
• Jull, A. B., Cullum, N., Dumville, J. C., Westby, M. J., Deshpande, S., & Walker, N. (2015). Honey as a topical treatment for wounds. Cochrane Database of Systematic Reviews, 2015(3), CD005083. https://doi.org/10.1002/14651858.CD005083.pub4
• Jull, A., Walker, N., Parag, V., Molan, P., Rodgers, A., & Honey as Adjuvant Leg Ulcer Therapy Trial Collaborators. (2008). Randomized clinical trial of honey-impregnated dressings for venous leg ulcers. British Journal of Surgery, 95(2), 175–182. https://doi.org/10.1002/bjs.6059
• Kanelis, D., Liolios, V., Tananaki, C., & Rodopoulou, M.-A. (2022). Determination of the carbohydrate profile and invertase activity of adulterated honeys after bee feeding. Applied Sciences, 12(7), 3661. https://doi.org/10.3390/app12073661
• Kast, C., & Ritter, R. (2014). Comment maîtriser la teneur en eau du miel ? Revue Suisse d'Apiculture, (8), 24–27.
• Kast, C., Dübecke, A., Kilchenmann, V., Bieri, K., Böhlen, M., Zoller, O., Beckh, G., & Lüllmann, C. (2014). Analysis of Swiss honeys for pyrrolizidine alkaloids. Journal of Apicultural Research, 53(1), 75–83. https://doi.org/10.3896/IBRA.1.53.1.07
• Kast, C., Kilchenmann, V., & Charrière, J.-D. (2021). Long-term monitoring of lipophilic acaricide residues in commercial Swiss beeswax. Pest Management Science, 77(9), 4026–4033. https://doi.org/10.1002/ps.6427
• Kędzierska-Matysek, M., Florek, M., Wolanciuk, A., & Skałecki, P. (2016). Effect of freezing and room temperatures storage for 18 months on quality of raw rapeseed honey (Brassica napus). Journal of Food Science and Technology, 53(9), 3349–3355. https://doi.org/10.1007/s13197-016-2313-x
• Kędzierska-Matysek, M., Teter, A., Daszkiewicz, T., Topyła, B., Skałecki, P., Domaradzki, P., & Florek, M. (2025). Effect of temperature of two-year storage of varietal honeys on 5-hydroxymethylfurfural content, diastase number, and CIE color coordinates. Agriculture, 15(6), 652. https://doi.org/10.3390/agriculture15060652
• Kiistala, R., Hannuksela, M., Mäkinen-Kiljunen, S., Niinimäki, A., & Haahtela, T. (1995). Honey allergy is rare in patients sensitive to pollens. Allergy, 50(10), 844–847. https://doi.org/10.1111/j.1398-9995.1995.tb05061.x
• Koepke, R., Sobel, J., & Arnon, S. S. (2008). Global occurrence of infant botulism, 1976-2006. Pediatrics, 122(1), e73-e82. https://doi.org/10.1542/peds.2007-1827
• Koo, Y. J., Pack, E. C., Lee, Y. J., Kim, H. S., Jang, D. Y., Lee, S. H., Kim, Y. S., Lim, K.-M., & Choi, D. W. (2020). Determination of toxic metal release from metallic kitchen utensils and their health risks. Food and Chemical Toxicology, 145, 111651. https://doi.org/10.1016/j.fct.2020.111651
• Koulis, G. A., Tsagkaris, A. S., Aalizadeh, R., Dasenaki, M. E., Panagopoulou, E. I., Drivelos, S., Halagarda, M., Georgiou, C. A., Proestos, C., & Thomaidis, N. S. (2021). Honey phenolic compound profiling and authenticity assessment using HRMS targeted and untargeted metabolomics. Molecules, 26(9), 2769. https://doi.org/10.3390/molecules26092769
• Kowalski, S. (2013). Changes of antioxidant activity and formation of 5-hydroxymethylfurfural in honey during thermal and microwave processing. Food Chemistry, 141(2), 1378–1382. https://doi.org/10.1016/j.foodchem.2013.04.025
• Kurt, A., Palabiyik, I., Gunes, R., Konar, N., & Toker, O. S. (2020). Determining honey adulteration by seeding method: An initial study with sunflower honey. Food Analytical Methods, 13, 952–961. https://doi.org/10.1007/s12161-020-01711-9
• Kwakman, P. H. S., te Velde, A. A., de Boer, L., Speijer, D., Vandenbroucke-Grauls, C. M. J. E., & Zaat, S. A. J. (2010). How honey kills bacteria. FASEB Journal, 24(7), 2576–2582. https://doi.org/10.1096/fj.09-150789
• Kwakman, P. H. S., te Velde, A. A., de Boer, L., Vandenbroucke-Grauls, C. M. J. E., & Zaat, S. A. J. (2011). Two major medicinal honeys have different mechanisms of bactericidal activity. PLOS ONE, 6(3), e17709. https://doi.org/10.1371/journal.pone.0017709
• LeBlanc, B. W., Eggleston, G., Sammataro, D., Cornett, C., Dufault, R., Deeby, T., & St. Cyr, E. (2009). Formation of hydroxymethylfurfural in domestic high-fructose corn syrup and its toxicity to the honey bee (Apis mellifera). Journal of Agricultural and Food Chemistry, 57(16), 7369–7376. https://doi.org/10.1021/jf9014526
• Lee, S. H., Jeong, H., Heo, M., Park, G. W., Lee, W.-Y., Mun, J.-E., Chang, M., & Kang, H.-S. (2025). Identification of 3-methoxytyramine as a specific biomarker for beet-sugar-fed honey: A two-year surveillance study in South Korea. Food Research International, 200, 115460. https://doi.org/10.1016/j.foodres.2024.115460
• Luca, L., Pauliuc, D., & Oroian, M. (2024). Honey microbiota, methods for determining the microbiological composition and the antimicrobial effect of honey-A review. Food Chemistry: X, 23, 101524. https://doi.org/10.1016/j.fochx.2024.101524
• Lucchetti, M. A., Glauser, G., Kilchenmann, V., Dübecke, A., Beckh, G., Praz, C., & Kast, C. (2016). Pyrrolizidine alkaloids from Echium vulgare in honey originate primarily from floral nectar. Journal of Agricultural and Food Chemistry, 64(25), 5267–5273. https://doi.org/10.1021/acs.jafc.6b02320
• Machado, A. M., Miguel, M. G., Vilas-Boas, M., & Figueiredo, A. C. (2020). Honey volatiles as a fingerprint for botanical origin-A review on their occurrence on monofloral honeys. Molecules, 25(2), 374. https://doi.org/10.3390/molecules25020374
• Majtán, J. (2024). In vitro testing of honey quality and biological functionality: Underestimated elements in the clinical testing of honey. Frontiers in Nutrition, 11, 1433786. https://doi.org/10.3389/fnut.2024.1433786
• Manickavasagam, G., Saaid, M., & Lim, V. (2024). Impact of prolonged storage on quality assessment properties and constituents of honey: A systematic review. Journal of Food Science, 89(2), 811–833. https://doi.org/10.1111/1750-3841.16921
• Moumeh, B., Garrido, D. M., Díaz, P., Peñaranda, I., & Linares, M. B. (2020). Chemical analysis and sensory evaluation of honey produced by honeybee colonies fed with different sugar pastes. Food Science & Nutrition, 8(11), 5823–5831. https://doi.org/10.1002/fsn3.1843
• Murcia-Morales, M., Heinzen, H., Parrilla-Vázquez, P., Gómez-Ramos, M. M., & Fernández-Alba, A. R. (2021). Presence and distribution of pesticides in apicultural products: A critical appraisal. TrAC Trends in Analytical Chemistry, 146, 116506. https://doi.org/10.1016/j.trac.2021.116506
• National Institute for Health and Care Excellence. (2019). Cough (acute): Antimicrobial prescribing (NICE guideline NG120). https://www.nice.org.uk/guidance/ng120
• New Zealand Ministry for Primary Industries. (n.d.). Tutin contamination of honey: Testing for maximum level. Retrieved 29 August 2026 from https://www.mpi.govt.nz/food-business/honey-bee-products-processing-requirements/managing-tutin-contamination-in-honey
• Nicolson, S. W. (2009). Water homeostasis in bees, with the emphasis on sociality. Journal of Experimental Biology, 212(3), 429–434. https://doi.org/10.1242/jeb.022343
• Nishimura, T., Muta, H., Hosaka, T., Ueda, M., Kishida, K., & Honey and Coughs Study Group of the Society of Ambulatory and General Paediatrics of Japan. (2022). Multicentre, randomised study found that honey had no pharmacological effect on nocturnal coughs and sleep quality at 1-5 years of age. Acta Paediatrica, 111(11), 2157–2164. https://doi.org/10.1111/apa.16509
• Norouzzadeh, M., Barazandeh, S., Hasan Rashedi, M., Jamshidi, S., Hatamifar, F., Maghsoomi, Z., Teymoori, F., & Malek, M. (2025). Dosage exploration of the effects of honey and its derivatives on cardiometabolic outcomes: An overview of systematic reviews and GRADE-assessed updated meta-analysis. Nutrition & Diabetes, 15, 48. https://doi.org/10.1038/s41387-025-00403-9
• Notbohm, H. L., Feuerbacher, J. F., Papendorf, F., Friese, N., Jacobs, M. W., Predel, H.-G., Zacher, J., Bloch, W., & Schumann, M. (2021). Metabolic, hormonal and performance effects of isomaltulose ingestion before prolonged aerobic exercise: A double-blind, randomised, cross-over trial. Journal of the International Society of Sports Nutrition, 18, 38. https://doi.org/10.1186/s12970-021-00439-z
• Nowak, A., & Nowak, I. (2023). Review of harmful chemical pollutants of environmental origin in honey and bee products. Critical Reviews in Food Science and Nutrition, 63(21), 5094–5116. https://doi.org/10.1080/10408398.2021.2012752
• Oberreiter, H., & Brodschneider, R. (2020). Austrian COLOSS Survey of Honey Bee Colony Winter Losses 2018/19 and analysis of hive management practices. Diversity, 12(3), 99. https://doi.org/10.3390/d12030099
• ODAlAn. FDHA Ordinance on foodstuffs of animal origin. See Federal Department of Home Affairs (2016b).
• ODAlOUs. Ordinance on foodstuffs and utility articles. See Swiss Federal Council (2016).
• Oduwole, O., Udoh, E. E., Oyo-Ita, A., & Meremikwu, M. M. (2018). Honey for acute cough in children. Cochrane Database of Systematic Reviews, 2018(4), CD007094. https://doi.org/10.1002/14651858.CD007094.pub5
• OIDAl. FDHA Ordinance concerning food information. See Federal Department of Home Affairs (2016a).
• Pasias, I., Raptopoulou, K., Makrigennis, G., Ntakoulas, D. D., Lembessis, D., Dimakis, V., Katsinas, R., & Proestos, C. (2022). Finding the optimum treatment procedure to delay honey crystallization without reducing its quality. Food Chemistry, 381, 132301. https://doi.org/10.1016/j.foodchem.2022.132301
• Persano Oddo, L., & Bogdanov, S. (2004). Determination of honey botanical origin: Problems and issues. Apidologie, 35(Suppl. 1), S2-S3. https://doi.org/10.1051/apido:2004044
• Persano Oddo, L., & Piro, R. (2004). Main European unifloral honeys: Descriptive sheets. Apidologie, 35(Suppl. 1), S38-S81. https://doi.org/10.1051/apido:2004049
• Peters, L. J. F., Majtán, J., Mossialos, D., Szweda, P., Mateescu, C., Ozturk, F., Wagener, F. A. D. T. G., & Cremers, N. A. J. (2025). Medical-grade honey: Its definition and refined standards. Journal of Wound Care, 34(6), 412–423. https://doi.org/10.12968/jowc.2024.0206
• Piepiórka-Stepuk, J., Sterczyńska, M., Stachnik, M., & Pawłowski, P. (2025). Effects of refrigerated storage on the physicochemical, color and rheological properties of selected honey. Agriculture, 15(14), 1476. https://doi.org/10.3390/agriculture15141476
• Pita-Calvo, C., & Vázquez, M. (2017). Differences between honeydew and blossom honeys: A review. Trends in Food Science & Technology, 59, 79–87. https://doi.org/10.1016/j.tifs.2016.11.015
• Polatidou, K., Nouska, C., Tananaki, C. I., Biliaderis, C. G., & Lazaridou, A. (2025). Physicochemical and rheological characteristics of monofloral honeys — Kinetics of creaming–crystallization. Foods, 14(10), 1835. https://doi.org/10.3390/foods14101835
• Postmes, T., van den Bogaard, A. E., & Hazen, M. (1995). The sterilization of honey with cobalt 60 gamma radiation: A study of honey spiked with spores of Clostridium botulinum and Bacillus subtilis. Experientia, 51(9-10), 986–989. https://doi.org/10.1007/BF01921753
• Přidal, A., Musila, J., & Svoboda, J. (2023). Condition and honey productivity of honeybee colonies depending on type of supplemental feed for overwintering. Animals, 13(3), 323. https://doi.org/10.3390/ani13030323
• Quinlan, G., Döke, M. A., Ortiz-Alvarado, Y., Rodriguez-Gomez, N., Koru, Y. B., & Underwood, R. M. (2023). Carbohydrate nutrition associated with health of overwintering honey bees. Journal of Insect Science, 23(6), 16. https://doi.org/10.1093/jisesa/iead084
• Raatz, S. K., Johnson, L. K., & Picklo, M. J. (2015). Consumption of honey, sucrose, and high-fructose corn syrup produces similar metabolic effects in glucose-tolerant and -intolerant individuals. The Journal of Nutrition, 145(10), 2265–2272. https://doi.org/10.3945/jn.115.218016
• Rajan, T. V., Tennen, H., Lindquist, R. L., Cohen, L., & Clive, J. (2002). Effect of ingestion of honey on symptoms of rhinoconjunctivitis. Annals of Allergy, Asthma & Immunology, 88(2), 198–203. https://doi.org/10.1016/S1081-1206(10)61996-5
• Roffet-Salque, M., Regert, M., Evershed, R. P., Outram, A. K., Cramp, L. J. E., Decavallas, O., Dunne, J., Gerbault, P., Mileto, S., Mirabaud, S., Pääkkönen, M., Smyth, J., Šoberl, L., Whelton, H. L., et al. (2015). Widespread exploitation of the honeybee by early Neolithic farmers. Nature, 527(7577), 226–230. https://doi.org/10.1038/nature15757
• Saarinen, K., Jantunen, J., & Haahtela, T. (2011). Birch pollen honey for birch pollen allergy-A randomized controlled pilot study. International Archives of Allergy and Immunology, 155(2), 160–166. https://doi.org/10.1159/000319821
• Sadeghi, F., Salehi, S., Kohanmoo, A., & Akhlaghi, M. (2019). Effect of natural honey on glycemic control and anthropometric measures of patients with type 2 diabetes: A randomized controlled crossover trial. International Journal of Preventive Medicine, 10, 3. https://doi.org/10.4103/ijpvm.IJPVM_109_18
• Sanhueza, J., & Fuentes, E. (2025). Assessment of role of glucose oxidase, flavonoids, copper and iron on the generation of hydrogen peroxide in honey. Food Research International, 202, 115532. https://doi.org/10.1016/j.foodres.2024.115532
• Šarić, G., Marković, K., Major, N., Krpan, M., Uršulin-Trstenjak, N., Hruškar, M., & Vahčić, N. (2012). Changes of antioxidant activity and phenolic content in acacia and multifloral honey during storage. Food Technology and Biotechnology, 50(4), 434–441.
• Scepankova, H., Pinto, C. A., Paula, V., Estevinho, L. M., & Saraiva, J. A. (2021). Conventional and emergent technologies for honey processing: A perspective on microbiological safety, bioactivity, and quality. Comprehensive Reviews in Food Science and Food Safety, 20(6), 5393–5420. https://doi.org/10.1111/1541-4337.12848
• Schiassi, M., Freitas, T. D., de Morais, L. C., Bueno, T. M., Queiroz, F., & Resende, J. V. de. (2025). Evaluation of honeys of different botanical origins stored under different conditions. Anais da Academia Brasileira de Ciências, 97(Suppl. 4), e20240658. https://doi.org/10.1590/0001-3765202420240658
• Schievano, E., Piana, L., & Tessari, M. (2023). Automatic NMR-based protocol for assessment of honey authenticity. Food Chemistry, 420, 136094. https://doi.org/10.1016/j.foodchem.2023.136094
• Schleiffer, M., & Speiser, B. (2022). Presence of pesticides in the environment, transition into organic food, and implications for quality assurance along the European organic food chain: A review. Environmental Pollution, 313, 120116. https://doi.org/10.1016/j.envpol.2022.120116
• Schneider, A., Horn, H., & Hammes, W. P. (2003). Zum Vorkommen osmophiler Hefen im Honig. Deutsche Lebensmittel-Rundschau, 99, 310–319.
• Seeburger, V. C., D'Alvise, P., Shaaban, B., Schweikert, K., Lohaus, G., Schroeder, A., & Hasselmann, M. (2020). The trisaccharide melezitose impacts honey bees and their intestinal microbiota. PLOS ONE, 15(4), e0230871. https://doi.org/10.1371/journal.pone.0230871
• Seraglio, S. K. T., Silva, B., Bergamo, G., Brugnerotto, P., Gonzaga, L. V., Fett, R., & Costa, A. C. O. (2019). An overview of physicochemical characteristics and health-promoting properties of honeydew honey. Food Research International, 119, 44–66. https://doi.org/10.1016/j.foodres.2019.01.028
• Serra Bonvehí, J., & Ventura Coll, F. (2003). Flavour index and aroma profiles of fresh and processed honeys. Journal of the Science of Food and Agriculture, 83, 275–282. https://doi.org/10.1002/jsfa.1308
• Snowdon, J. A., & Cliver, D. O. (1996). Microorganisms in honey. International Journal of Food Microbiology, 31(1-3), 1–26. https://doi.org/10.1016/0168-1605(96)00970-1
• Soares, S., Amaral, J. S., Oliveira, M. B. P. P., & Mafra, I. (2017). A comprehensive review on the main honey authentication issues: Production and origin. Comprehensive Reviews in Food Science and Food Safety, 16(5), 1072–1100. https://doi.org/10.1111/1541-4337.12278
• Soares, S., Pinto, D., Rodrigues, F., Alves, R. C., & Oliveira, M. B. P. P. (2017). Portuguese honeys from different geographical and botanical origins: A 4-year stability study regarding quality parameters and antioxidant activity. Molecules, 22(8), 1338. https://doi.org/10.3390/molecules22081338
• Solayman, M., Islam, M. A., Paul, S., Ali, Y., Khalil, M. I., Alam, N., & Gan, S. H. (2016). Physicochemical properties, minerals, trace elements, and heavy metals in honey of different origins: A comprehensive review. Comprehensive Reviews in Food Science and Food Safety, 15(1), 219–233. https://doi.org/10.1111/1541-4337.12182
• Stanojević, S. P., Milinčić, D. D., Smiljanić, N., Pešić, M. B., Nedić, N. M., Kolašinac, S., Dojčinović, B., Dajić-Stevanović, Z., & Kostić, A. Ž. (2024). Conventional vs. organically produced honey — Are there differences in physicochemical, nutritional and sensory characteristics? Foods, 13(22), 3573. https://doi.org/10.3390/foods13223573
• Starowicz, M., & Zieliński, H. (2019). How Maillard reaction influences sensorial properties (color, flavor and texture) of food products? Food Reviews International, 35(8), 707–725. https://doi.org/10.1080/87559129.2019.1600538
• Stellingwerff, T., & Cox, G. R. (2014). Systematic review: Carbohydrate supplementation on exercise performance or capacity of varying durations. Applied Physiology, Nutrition, and Metabolism, 39(9), 998–1011. https://doi.org/10.1139/apnm-2014-0027
• Subbiah, B., Blank, U. K. M., & Morison, K. R. (2020). A review, analysis and extension of water activity data of sugars and model honey solutions. Food Chemistry, 326, 126981. https://doi.org/10.1016/j.foodchem.2020.126981
• Subrahmanyam, M. (1991). Topical application of honey in treatment of burns. British Journal of Surgery, 78(4), 497–498. https://doi.org/10.1002/bjs.1800780435
• Subramanian, R., Hebbar, H. U., & Rastogi, N. K. (2007). Processing of honey: A review. International Journal of Food Properties, 10(1), 127–143. https://doi.org/10.1080/10942910600981708
• Swiss Federal Council. (2016). Ordinance on foodstuffs and utility articles (ODAlOUs; SR 817.02). Fedlex. Retrieved 29 August 2026 from https://www.fedlex.admin.ch/eli/cc/2017/63/fr
• Szczęsna, T., Waś, E., Semkiw, P., Skubida, P., Jaśkiewicz, K., & Witek, M. (2021). Changes in the physicochemical properties of starch syrups after processing by honeybees. Agriculture, 11(4), 335. https://doi.org/10.3390/agriculture11040335
• Słowik-Borowiec, M., Baryła, N., & Kubis, H. (2026). Determination of pesticides, antibiotics, PAHs, PCBs, and plasticizers in honeybees, honey, and other bee products-Modified QuEChERS method. Method review for 2018-2024. Food Chemistry, 504, 147864. https://doi.org/10.1016/j.foodchem.2026.147864
• Tananaki, C., Thrasyvoulou, A., & Menexes, G. (2005). Absorption of volatile compounds in honey from stored spices. Journal of Apicultural Research, 44(2), 71–77. https://doi.org/10.1080/00218839.2005.11101152
• Tang, Y., Chen, L., & Ran, X. (2024). Efficacy and safety of honey dressings in the management of chronic wounds: An updated systematic review and meta-analysis. Nutrients, 16(15), 2455. https://doi.org/10.3390/nu16152455
• Tappi, S., Glicerina, V., Ragni, L., Dettori, A., Romani, S., & Rocculi, P. (2021). Physical and structural properties of honey crystallized by static and dynamic processes. Journal of Food Engineering, 292, 110316. https://doi.org/10.1016/j.jfoodeng.2020.110316
• Tappi, S., Laghi, L., Dettori, A., Piana, L., Ragni, L., & Rocculi, P. (2019). Investigation of water state during induced crystallization of honey. Food Chemistry, 294, 260–266. https://doi.org/10.1016/j.foodchem.2019.05.047
• Tosi, E., Ciappini, M., Ré, E., & Lucero, H. (2002). Honey thermal treatment effects on hydroxymethylfurfural content. Food Chemistry, 77(1), 71–74. https://doi.org/10.1016/S0308-8146(01)00325-9
• Tosi, E., Martinet, R., Ortega, M., Lucero, H., & Ré, E. (2008). Honey diastase activity modified by heating. Food Chemistry, 106(3), 883–887. https://doi.org/10.1016/j.foodchem.2007.04.025
• Trifković, J., Andrić, F., Ristivojević, P. M., Guzelmeric, E., & Yeşilada, E. (2017). Analytical methods in tracing honey authenticity. Journal of AOAC International, 100(4), 827–839. https://doi.org/10.5740/jaoacint.17-0142
• Tsagkaris, A. S., Koulis, G. A., Danezis, G. P., Martakos, I. C., Dasenaki, M. E., Georgiou, C. A., & Thomaidis, N. S. (2021). Honey authenticity: Analytical techniques, state of the art and challenges. RSC Advances, 11(19), 11273–11294. https://doi.org/10.1039/D1RA00069A
• Tyowua, A. T., Echendu, A. M., Adejo, S. O., & Binks, B. P. (2022). Influence of particle wettability on foam formation in honey. Journal of Physics: Condensed Matter, 34(49), 494001. https://doi.org/10.1088/1361-648X/ac8f0b
• Vázquez, L., Verdú, A., Miquel, A., Burló, F., & Carbonell-Barrachina, Á. A. (2007). Changes in physico-chemical properties, hydroxymethylfurfural and volatile compounds during concentration of honey and sugars in Alicante and Jijona turrón. European Food Research and Technology, 225, 757–767. https://doi.org/10.1007/s00217-006-0479-6
• von der Ohe, W., Persano Oddo, L., Piana, M. L., Morlot, M., & Martin, P. (2004). Harmonized methods of melissopalynology. Apidologie, 35(Suppl. 1), S18-S25. https://doi.org/10.1051/apido:2004050
• Walker, M. J., Cowen, S., Gray, K., Hancock, P., & Burns, D. T. (2022a). Honey authenticity: The opacity of analytical reports-Part 1 defining the problem. npj Science of Food, 6, 11. https://doi.org/10.1038/s41538-022-00126-6
• Walker, M. J., Cowen, S., Gray, K., Hancock, P., & Burns, D. T. (2022b). Honey authenticity: The opacity of analytical reports-Part 2, forensic evaluative reporting as a potential solution. npj Science of Food, 6, 12. https://doi.org/10.1038/s41538-022-00127-5
• Wallner, K. (1999). Varroacides and their residues in bee products. Apidologie, 30(2-3), 235–248. https://doi.org/10.1051/apido:19990212
• Weston, R. J. (2000). The contribution of catalase and other natural products to the antibacterial activity of honey: A review. Food Chemistry, 71(2), 235–239. https://doi.org/10.1016/S0308-8146(00)00162-X
• White, J. W., Jr., & Subers, M. H. (1964). Studies on honey inhibine. 4. Destruction of the peroxide accumulation system by light. Journal of Food Science, 29(6), 819–828. https://doi.org/10.1111/j.1365-2621.1964.tb00455.x
• White, J. W., Jr., Subers, M. H., & Schepartz, A. I. (1963). The identification of inhibine, the antibacterial factor in honey, as hydrogen peroxide and its origin in a honey glucose-oxidase system. Biochimica et Biophysica Acta, 73(1), 57–70. https://doi.org/10.1016/0926-6569(63)90108-1
• Wise, S. K., Damask, C., Roland, L. T., Ebert, C., Levy, J. M., Lin, S., Luong, A., Rodriguez, K., Sedaghat, A. R., Toskala, E., Villwock, J., Abdullah, B., Akdis, C., Alt, J. A., Ansotegui, I. J., Azar, A., Baroody, F., Benninger, M. S., Bernstein, J.,… Zhang, L. (2023). International consensus statement on allergy and rhinology: Allergic rhinitis-2023. International Forum of Allergy & Rhinology, 13(4), 293–859. https://doi.org/10.1002/alr.23090
• World Health Organization. (2015). Guideline: Sugars intake for adults and children. World Health Organization. https://www.who.int/publications/i/item/9789241549028
• Wright, G. A., Nicolson, S. W., & Shafir, S. (2018). Nutritional physiology and ecology of honey bees. Annual Review of Entomology, 63, 327–344. https://doi.org/10.1146/annurev-ento-020117-043423
• Yalçın, G. (2021). Effects of thermal treatment, ultrasonication, and sunlight exposure on antioxidant properties of honey. Turkish Journal of Pharmaceutical Sciences, 18(6), 776–780. https://doi.org/10.4274/tjps.galenos.2021.53810
• Yao, L., Bhandari, B., Datta, N., Singanusong, R., & D'Arcy, B. (2003). Crystallisation and moisture sorption properties of selected Australian unifloral honeys. Journal of the Science of Food and Agriculture, 83(9), 884–888. https://doi.org/10.1002/jsfa.1421
• Yao, L., Dai, J., & Mei, S. (2026). Honey dressing for diabetic foot ulcers: A systematic review and meta-analysis of randomized controlled trials. Frontiers in Endocrinology, 17, 1759703. https://doi.org/10.3389/fendo.2026.1759703
• Yarom, N., Hovan, A., Bossi, P., Ariyawardana, A., Jensen, S. B., Gobbo, M., Saca-Hazboun, H., Kandwal, A., Majorana, A., Ottaviani, G., Pentenero, M., Nasr, N. M., Rouleau, T., Lucas, A. S., Treister, N. S., Zur, E., Ranna, V., Vaddi, A., Barasch, A.,… Elad, S. (2020). Systematic review of natural and miscellaneous agents for the management of oral mucositis in cancer patients and clinical practice guidelines-Part 2: Honey, herbal compounds, saliva stimulants, probiotics, and miscellaneous agents. Supportive Care in Cancer, 28(5), 2457–2472. https://doi.org/10.1007/s00520-019-05256-4
• Yiğit, Y., Yalçın, S., & Onbaşılar, E. E. (2024). Effects of different packaging types and storage periods on physicochemical and antioxidant properties of honeys. Foods, 13(22), 3594. https://doi.org/10.3390/foods13223594
• Zamora, M. C., & Chirife, J. (2006). Determination of water activity change due to crystallization in honeys from Argentina. Food Control, 17(1), 59–64. https://doi.org/10.1016/j.foodcont.2004.09.003
• Zhang, X.-H., Gu, H.-W., Liu, R.-J., Qing, X.-D., & Nie, J.-F. (2023). A comprehensive review of the current trends and recent advancements on the authenticity of honey. Food Chemistry: X, 19, 100850. https://doi.org/10.1016/j.fochx.2023.100850