How honey is tested

The laboratory panel behind every honey specification, method by method: what each test measures, what it costs, and what a clean result does not rule out.

Quality, testing and standardsReviewed 2026-11-02

The routine panel

These are the measurements written into the Codex standard and the EU directive. They are cheap, fast and widely available, and between them they establish whether a honey meets its specification. None of them is an authenticity test.

Routine compositional testing
Method	Measures	Typically establishes
Refractometry	Refractive index, converted to water content	Whether the honey will keep or ferment
HPLC or HPAEC	The individual sugars and their proportions	Sugar spectrum, sucrose limit, fructose-to-glucose ratio
Spectrophotometry	Hydroxymethylfurfural	Heating and storage history, freshness
Enzyme assay	Diastase number	Heating and storage history, read against HMF
Conductivity meter	Electrical conductivity of a solution	Blossom honey versus honeydew; supports floral claims
Titration	Free acidity, pH	Ripeness, early fermentation
Microscopy	Pollen types and their proportions	Botanical and geographic origin
Sensory panel	Aroma, flavour, defects, against reference descriptions	Whether a monofloral honey is recognisably that honey

The authenticity panel

Detecting added syrup is a different problem. The adulterant is sugar, the honey is sugar, and modern syrups are engineered specifically to survive the standard tests. What follows are the methods that actually work, in rough order of cost.

Stable carbon isotope ratio analysis (SCIRA or EA-IRMS)Plants fix carbon by different photosynthetic pathways, and cane and maize are C4 plants while nearly all nectar plants are C3. Comparing the isotope ratio of the honey with that of its own extracted protein reveals C4 sugar addition down to a few per cent. It is the long-standing workhorse — and it is blind to syrups made from C3 crops such as rice, beet and wheat.
Nuclear magnetic resonance profilingProduces a broad fingerprint of the sample and compares it against a large reference database of authentic honeys. It catches things no targeted test looks for, including syrups the analyst did not have in mind. Its power is entirely a function of the database behind it, which is commercially held and unevenly representative of the world's honeys.
Marker compound screening by LC-MSLooks for specific traces left by syrup manufacture — particular oligosaccharides, mannose, or processing residues. Effective against known syrup types and a step behind each new one.
Pollen analysisA honey with almost no pollen has very likely been filtered hard; since 14 June 2026 the EU treats honey filtered so as to remove pollen significantly as baker's honey, for cooking only. A pollen assemblage that contradicts the stated floral source or region is strong evidence of misdescription — read against what that honey type normally carries, because there is no single pollen percentage that makes every honey monofloral.

Why carbon isotope testing sees cane and maize syrup but not rice or beet syrup. Plants fix carbon by two main pathways, which leave different carbon isotope ratios: nearly all nectar plants are C3, as are rice, beet and wheat; cane and maize are C4. The method compares the honey with protein extracted from the same honey. Sugar from a C4 crop pulls the honey’s ratio away from its protein’s, and above about 7% that is detected. A C3 syrup leaves the two together, so it passes — which is why a clean result rules out one family of syrups, not all of them. The scale is drawn without values. Sources: Methods for verifying honey authenticity; AOAC Official Method 998.12 — C-4 plant sugars in honey: internal standard stable carbon isotope ratio method. Drawing: HoneyHQ (HEKNO Ltd), original.

The wider method landscape

The four methods above are the ones an enforcement laboratory is most likely to run. They are not the whole field. The list of analytical approaches applied to honey authenticity is long, and the useful thing to know about it is that the entries are at very different stages of maturity — some are official and harmonised, some are accredited routine work, and some are research that has not been standardised anywhere.

Analytical methods, the questions they address and how settled each is
Method	Questions it addresses	What it measures	What a result does	Status	Needs a reference database	Speed / complexity
Stable carbon isotope ratio analysis (SCIRA, EA-IRMS (AOAC 998.12))	Was sugar syrup added?	Isotope ratios	Detects a defined adulterant or marker	Official method	No. No external database: the honey's own protein is the reference.	Medium / Medium
Isotope analysis of individual sugars (LC-IRMS)	Was sugar syrup added?	Isotope ratios; Sugars and their ratios	Detects a defined adulterant or marker	Validated in peer-reviewed work	Yes. Ranges for the isotope differences between sugars in authentic honeys, against which a sample is read.	Medium / High
Sugar profiling (HPLC, GC)	Was sugar syrup added? Is it ripe and in sound condition?	Sugars and their ratios	Detects a defined adulterant or marker	Official method	No. Read against published limits and typical values, not a proprietary database.	Medium / Medium
Syrup marker screening (HPAEC-PAD, LC-HRMS (targeted))	Was sugar syrup added? Did sugar fed to the bees reach the honey?	Oligosaccharide and syrup markers	Screens samples for follow-up; Detects a defined adulterant or marker	Screening technique	Yes. Marker libraries for syrup types, and natural ranges for honey types.	Medium / High
Nuclear magnetic resonance profiling (1H-NMR)	Was sugar syrup added? Is it from the plant it names? Is it from the place it names?	Metabolite profile	Screens samples for follow-up; Compares a sample with reference honeys; Classifies a sample by a model	Screening technique	Yes. Authentic honeys covering the botanical sources, regions, seasons, years and processing the sample might come from. Often privately held.	High / High
Untargeted metabolomics (LC-HRMS (untargeted))	Is it from the plant it names? Is it from the place it names? Was sugar syrup added?	Metabolite profile	Classifies a sample by a model	Experimental or emerging	Yes. Authentic training samples for every class the model is asked to recognise.	Low / High
Infrared and Raman spectroscopy (NIR, MIR/FTIR, Raman)	Is it from the plant it names? Is it from the place it names? Was sugar syrup added?	Spectral fingerprint	Screens samples for follow-up; Classifies a sample by a model	Screening technique	Yes. A calibration set of authentic and, for adulteration, deliberately adulterated samples.	High / Low
Pollen analysis (Melissopalynology)	Is it from the plant it names? Is it from the place it names? Has its pollen been removed?	Pollen grains	Compares a sample with reference honeys; Detects a defined adulterant or marker	Validated between laboratories	Yes. Reference slides of pollen types, and the published descriptions of each honey type.	Low / Medium
DNA metabarcoding (NGS)	Is it from the plant it names? Is it from the place it names?	DNA	Classifies a sample by a model; Compares a sample with reference honeys	Experimental or emerging	Yes. Barcode sequence libraries covering the plants the honey might come from.	Low / High
Elemental and trace-element profiling (ICP-MS, ICP-OES)	Is it from the place it names? Is it from the plant it names?	Elements and trace elements	Classifies a sample by a model; Compares a sample with reference honeys	Single-study research	Yes. Authentic samples from each region, of the same honey types, analysed comparably.	Medium / High
HMF and enzyme activity (HMF, diastase, invertase)	Has it been overheated or stored too long?	Heating and storage markers; Enzyme activity	Detects a defined adulterant or marker	Official method	No. Read against published limits; the EU must set invertase-based criteria for heat treatment by 14 June 2029.	High / Low
Water content (Refractometry)	Is it ripe and in sound condition?	Water content	Detects a defined adulterant or marker	Official method	No. Read against published limits.	High / Low

Status is how settled the method is, not how good it is: an official method has been standardised for a defined question, while a screening technique flags samples for follow-up and a research method has not been validated for routine use. No column ranks the methods, because none is best for every question.

Checked 2026-10-31 against: Methods for verifying honey authenticity (European Commission, Joint Research Centre — Knowledge Centre for Food Fraud and Quality, 2026); AOAC Official Method 998.12 — C-4 plant sugars in honey: internal standard stable carbon isotope ratio method (AOAC International, Official Methods of Analysis, 1998); Frequency of honey adulteration (European Commission, Joint Research Centre — Knowledge Centre for Food Fraud and Quality, 2025); Standard for Honey (CXS 12-1981) (Codex Alimentarius Commission, FAO/WHO, 1981); Adulteration of honey (European Commission, Joint Research Centre — Knowledge Centre for Food Fraud and Quality, 2025); Detection of honey adulteration by liquid chromatography–high-resolution mass spectrometry: results from an EU coordinated action (Separations 12(2):47 (MDPI), 2025); EU coordinated action 'From the Hives' (honey 2021–2022) (European Commission, Directorate-General for Health and Food Safety, 2023); Framework for interrogation of honey authenticity databases (IHAD) (Government Chemist and Department for Environment, Food and Rural Affairs, 2025); The use of SPME-GC-MS, IR and Raman techniques for botanical and geographical authentication and detection of adulteration of honey (Foods 10(7): 1671 (MDPI), 2021); Harmonized methods of melissopalynology (Apidologie 35 (Suppl. 1): S18–S25, 2004); Using DNA metabarcoding to identify the floral composition of honey: a new tool for investigating honey bee foraging preferences (PLoS ONE 10(8): e0134735, 2015); Honey: inorganic composition as possible marker for botanical and geological assignment (Molecules 30(7): 1466 (MDPI), 2025); Directive (EU) 2024/1438 amending the 'breakfast directives', including honey (European Parliament and Council of the European Union, 2024).

What each method can and cannot prove

This is the table to read if you read only one thing on this page. The middle column is what a positive finding supports. The right-hand column is what the method does not establish on its own, however clean or however alarming its result.

What each test can actually prove
Method	Can support	Cannot alone establish	Depends on	Status
Stable carbon isotope ratio analysis	Addition of sugar from C4 plants such as cane and maize, at about 7% of the honey or more.	That no syrup was added: syrups from C3 plants such as rice, beet and wheat have carbon isotope ratios close to nectar's and are not detected. C4 addition below the method's level of detection. The honey's floral or geographical origin.	The protein extracted from the same honey, used as its internal standard. Syrups not engineered to match honey's isotope ratio — such syrups are on the market.	Official method
Isotope analysis of individual sugars	C4 sugar addition at lower levels than bulk analysis, by comparing the isotope ratios of the honey's individual sugars.	Addition of C3 syrups whose sugars carry nectar-like isotope ratios. Origin of any kind.	Chromatographic separation of the sugars before the isotope measurement. Reference ranges for the differences between sugars in genuine honey.	Validated in peer-reviewed work
Sugar profiling	Whether the sugars meet the compositional limits — fructose plus glucose, sucrose. A sugar balance unusual for the declared honey type.	Absence of syrup: syrups are now made to mimic honey's sugar composition. That an unusual profile is fraud — some honeys are naturally high in sucrose or low in fructose.	Published compositional limits and typical values for the honey type.	Official method
Syrup marker screening	The presence of markers of manufactured syrup — mannose, foreign oligo- and polysaccharides — that genuine honey is not expected to contain.	How much syrup was added: the markers say present or absent. Absence of syrup: a syrup without the markers sought passes. On its own, a regulatory finding: the Commission describes these methods as useful for suspicion and not yet validated.	A library of known syrup markers, which lags each new syrup. Knowledge of which oligosaccharides occur naturally in which honeys.	Screening technique
Nuclear magnetic resonance profiling	That a sample's profile is unusual compared with authentic honeys described the same way. Classification by type or origin, within the scope of the reference set.	Adulteration or a false origin on its own: an unusual profile has innocent causes, and a rare genuine honey can sit outside the reference set. Anything about honeys the reference set does not represent.	A reference database of authentic honeys — its scope, representativity, metadata and validation. The statistical model built on it, and whether that model was tested on samples it had not seen.	Screening technique
Untargeted metabolomics	Discovery of candidate markers for a honey type or region. Classification of samples by a model, in the study that built it.	A routine finding: most classifiers have been built and tested on one study's samples. Origin for honeys, regions or years outside the training samples.	The training samples, the instrument and the processing settings. External validation on independent samples, which many studies lack.	Experimental or emerging
Infrared and Raman spectroscopy	Rapid first-line sorting of samples by type or origin, for follow-up.	Adulteration or origin on its own: pure and adulterated honey give similar spectra, and the result is only as good as the model. A finding about a portable or one-off reading without laboratory confirmation.	A calibration model and the reference samples behind it. The instrument and its calibration, which do not transfer automatically between instruments.	Screening technique
Pollen analysis	Whether the pollen spectrum fits the declared floral source, read with sensory and physico-chemical results. That a pollen assemblage contradicts a declared region. That pollen has been largely removed.	A plant's share of the nectar from its pollen share: some plants are over-represented in honey pollen and others under-represented. The country of origin on its own: it shows which plants the bees met, and many grow in more than one country. Anything after heavy filtration has removed the pollen.	An analyst who knows the regional flora, and a reference collection of pollen types. Honey-type-specific criteria: there is no single pollen percentage that makes every honey monofloral.	Validated between laboratories
DNA metabarcoding	Which plant taxa are present, complementing microscopy. Detection of taxa that are hard to tell apart by pollen shape.	Each plant's share of the nectar: read counts reflect DNA extraction and amplification, not nectar contribution. Taxa present at low levels, which are less likely to be detected. Authenticity in general, or a substitute for pollen microscopy.	Sequence reference libraries for the region's flora. DNA that survived processing, filtration and storage.	Experimental or emerging
Elemental and trace-element profiling	That a sample's elemental profile fits, or does not fit, authentic samples from a claimed region.	That one element proves one country: elements reflect the plant as much as the soil, and regions overlap. Origin where no authentic samples from the region exist to compare against.	Authentic regional reference samples, which for much of the world do not exist. Freedom from contamination: tools and galvanised storage can add iron and zinc.	Single-study research
HMF and enzyme activity	Heating or long or warm storage, where HMF is high and enzyme activity low. Compliance with the compositional limits for HMF and diastase.	Adulteration: high HMF is a heating and storage indicator, not proof that a honey is fake. Origin of any kind.	The honey's age and storage temperature, and its natural enzyme level, which is low in some genuine honeys.	Official method
Water content	Whether the honey is within the water limit, and so how likely it is to ferment.	Authenticity: water describes ripeness and keeping, and some genuine honeys, heather among them, are naturally high. That a fermenting honey is fraudulent: fermentation is a condition, not evidence of fraud.	A calibrated refractometer and temperature correction.	Official method

Read the right-hand columns first. Every method either depends on a reference set or is blind to a kind of adulterant, and usually both, which is why authorities run several and assemble a conclusion from independent lines of evidence.

Several lines of evidence, not one test

Each question has its own usual combination: a first-line screen, a complementary method that sees what the first is blind to, and something to check the result against — published limits, a specification, a reference set or the paperwork. Independent lines that agree are stronger than one sophisticated result, and when they disagree, the disagreement is itself information.

Was sugar syrup added?
1. First line
  - Stable carbon isotope ratio analysis (official method)
  - Sugar profiling (official method)
  ↓
2. Complementary
  - Isotope analysis of individual sugars (validated in peer-reviewed work)
  - Syrup marker screening (screening technique)
  - Nuclear magnetic resonance profiling (screening technique)
  ↓
3. Checked against
  - Findings read together against internationally recognised validated methods, which EU authorities must use until the Commission adopts harmonised adulteration methods (due by 14 June 2028).
Did sugar fed to the bees reach the honey?
1. First line
  - Syrup marker screening (screening technique)
  ↓
2. Complementary
  - Sugar profiling (official method)
  - Stable carbon isotope ratio analysis (official method)
  ↓
3. Checked against
  - Apiary records of when feeding stopped and supers went on. Analysis alone rarely separates carry-over from feeding and syrup added after harvest.
Is it from the plant it names?
1. First line
  - Pollen analysis (validated between laboratories)
  ↓
2. Complementary
  - DNA metabarcoding (experimental or emerging)
  - Nuclear magnetic resonance profiling (screening technique)
  - Infrared and Raman spectroscopy (screening technique)
  ↓
3. Checked against
  - The description of that honey type — sensory, physico-chemical and microscopic characteristics together, as EU law requires for a floral name — or, for a protected name, its specification.
Is it from the place it names?
1. First line
  - Pollen analysis (validated between laboratories)
  ↓
2. Complementary
  - Elemental and trace-element profiling (single-study research)
  - Nuclear magnetic resonance profiling (screening technique)
  - Untargeted metabolomics (experimental or emerging)
  ↓
3. Checked against
  - Traceability documents back to the harvesting producer or importer and, for a protected name, certification against its specification. EU criteria for determining where honey was harvested are due by 14 June 2029.
Has it been overheated or stored too long?
1. First line
  - HMF and enzyme activity (official method)
  ↓
2. Complementary
  - No independent second method recorded
  ↓
3. Checked against
  - Compositional limits for HMF and diastase; EU invertase-based criteria for heat treatment are due by 14 June 2029.
Has its pollen been removed?
1. First line
  - Pollen analysis (validated between laboratories)
  ↓
2. Complementary
  - No independent second method recorded
  ↓
3. Checked against
  - EU criteria to verify that pollen has not been removed — pollen content, minimal pollen size, filter mesh size — are due by 14 June 2029.

From question to conclusion. For each question, the evidence usually gathered first, what sees what that misses, and what the result is checked against. Typical combinations, not prescribed procedures — and no question has a single test. Sources: Methods for verifying honey authenticity; Directive (EU) 2024/1438 amending the 'breakfast directives', including honey; Frequency of honey adulteration; Adulteration of honey; Harmonized methods of melissopalynology; Using DNA metabarcoding to identify the floral composition of honey: a new tool for investigating honey bee foraging preferences; Honey: inorganic composition as possible marker for botanical and geological assignment; Framework for interrogation of honey authenticity databases (IHAD); Standard for Honey (CXS 12-1981). Drawing: HoneyHQ (HEKNO Ltd), original.

From question to conclusion: the evidence typically combined
Question	First line	Complementary	Checked against	Why no single test
Was sugar syrup added?	Stable carbon isotope ratio analysis; Sugar profiling	Isotope analysis of individual sugars; Syrup marker screening; Nuclear magnetic resonance profiling	Findings read together against internationally recognised validated methods, which EU authorities must use until the Commission adopts harmonised adulteration methods (due by 14 June 2028).	Isotope analysis sees C4 sugar and not C3 syrups; markers see the syrups they know; profiles see what is unusual against a reference set. Each covers what the others miss.
Did sugar fed to the bees reach the honey?	Syrup marker screening	Sugar profiling; Stable carbon isotope ratio analysis	Apiary records of when feeding stopped and supers went on. Analysis alone rarely separates carry-over from feeding and syrup added after harvest.	Feed passes through the bees, so it can resemble either honey or syrup depending on what was fed and when.
Is it from the plant it names?	Pollen analysis	DNA metabarcoding; Nuclear magnetic resonance profiling; Infrared and Raman spectroscopy	The description of that honey type — sensory, physico-chemical and microscopic characteristics together, as EU law requires for a floral name — or, for a protected name, its specification.	Pollen shares mislead for over- and under-represented plants, DNA counts are not nectar shares, and profiles depend on the reference set.
Is it from the place it names?	Pollen analysis	Elemental and trace-element profiling; Nuclear magnetic resonance profiling; Untargeted metabolomics	Traceability documents back to the harvesting producer or importer and, for a protected name, certification against its specification. EU criteria for determining where honey was harvested are due by 14 June 2029.	Plants and soils cross borders, and every fingerprint method needs authentic samples from the region claimed — often the very samples that do not exist.
Has it been overheated or stored too long?	HMF and enzyme activity	No independent second method recorded	Compositional limits for HMF and diastase; EU invertase-based criteria for heat treatment are due by 14 June 2029.	HMF rises with heat and with age, and some genuine honeys have little enzyme to begin with, so the markers are read together and with the honey's history.
Has its pollen been removed?	Pollen analysis	No independent second method recorded	EU criteria to verify that pollen has not been removed — pollen content, minimal pollen size, filter mesh size — are due by 14 June 2029.	Honeys differ naturally in pollen content, so a low count is read against what that honey type normally carries.

These are typical combinations, not prescribed procedures: laboratories and authorities choose their own panels, and the EU has not yet adopted harmonised adulteration methods.

Checked 2026-10-31 against: Methods for verifying honey authenticity (European Commission, Joint Research Centre — Knowledge Centre for Food Fraud and Quality, 2026); Directive (EU) 2024/1438 amending the 'breakfast directives', including honey (European Parliament and Council of the European Union, 2024); Frequency of honey adulteration (European Commission, Joint Research Centre — Knowledge Centre for Food Fraud and Quality, 2025); Adulteration of honey (European Commission, Joint Research Centre — Knowledge Centre for Food Fraud and Quality, 2025); Harmonized methods of melissopalynology (Apidologie 35 (Suppl. 1): S18–S25, 2004); Using DNA metabarcoding to identify the floral composition of honey: a new tool for investigating honey bee foraging preferences (PLoS ONE 10(8): e0134735, 2015); Honey: inorganic composition as possible marker for botanical and geological assignment (Molecules 30(7): 1466 (MDPI), 2025); Framework for interrogation of honey authenticity databases (IHAD) (Government Chemist and Department for Environment, Food and Rural Affairs, 2025); Standard for Honey (CXS 12-1981) (Codex Alimentarius Commission, FAO/WHO, 1981).

What testing cannot settle

Whether a honey is good. Every test here measures composition and history. None measures whether anyone enjoys eating it.
Whether it is worth the price. Rarity, labour and provenance set price, and no instrument reads them.
Geographic origin, without a reference set. Isotope and trace element methods compare against authenticated regional samples, and where those do not exist the method has nothing to say.
Whether a beekeeper treated their bees well. There is no analytical marker for husbandry.

The claims on this page

Every substantive claim is placed on a tier, and the top two tiers must also state what they are not claiming.

Emerging research

Real studies exist and point somewhere, but they are small, in vitro, in animals, or not yet replicated. Direction is not the same as a result.

Emerging research

Carbon isotope testing detects cane and maize syrup in honey but not syrups made from rice, beet or wheat.

The method depends on the difference between C4 and C3 photosynthetic pathways. Sugar cane and maize are C4; rice, beet, wheat and nearly all nectar plants are C3, so a C3-derived syrup has an isotope ratio close to that of the honey itself. This limitation is well documented in the analytical literature and is the reason the field moved towards NMR profiling and marker screening.

What this does not claim: This is not a claim that carbon isotope testing is unreliable or should be abandoned. It remains a validated method for the adulterants it was designed to detect. The claim is about scope: a clean C4 result rules out one family of syrups and says nothing about the others.

Analytical approaches to honey authentication — Peer-reviewed food-analysis literature · Peer-reviewed article

Sources

HoneyHQ editorial synthesis — HEKNO Ltd · HoneyHQ synthesis
Analytical approaches to honey authentication — Peer-reviewed food-analysis literature · Peer-reviewed article
Standard for Honey (CXS 12-1981) — Codex Alimentarius Commission, FAO/WHO (1981) · Standards document Link
Council Directive 2001/110/EC relating to honey — Council of the European Union (2001) · Regulatory guidance Link
Methods for verifying honey authenticity — European Commission, Joint Research Centre — Knowledge Centre for Food Fraud and Quality (2026) · Regulatory guidance Link
AOAC Official Method 998.12 — C-4 plant sugars in honey: internal standard stable carbon isotope ratio method — AOAC International, Official Methods of Analysis (1998) · Standards document Link
Harmonized methods of melissopalynology — Apidologie 35 (Suppl. 1): S18–S25 (2004) · Peer-reviewed article Link
Directive (EU) 2024/1438 amending the 'breakfast directives', including honey — European Parliament and Council of the European Union (2024) · Regulatory guidance Link

How sources are selected and weighted is set out in the sources and evidence policy.

In short

Several methods, none of them complete. Carbon isotope analysis catches cane and maize syrups but misses rice and beet. NMR profiling compares a sample against a reference database and catches far more. Pollen analysis establishes floral and geographic origin. The adulterators adapt to each test as it becomes standard.
The routine panel — moisture, sugars, HMF, diastase, conductivity, acidity — describes a honey rather than authenticating it.
Authenticity testing is a separate and much more expensive set of methods.
No single test proves a honey is genuine; adulteration is inferred from a pattern across several.
Every method compares a sample against reference data, so its power depends on the reference set.
The methods differ in maturity: some are official and harmonised, others have been standardised nowhere.

This guide answers

What tests are actually run on honey, and what does each one show?

Tools that use this

Honey finder
Which honey should I buy for what I actually want to do with it?
Flavour selector
I like this taste — which honeys have it?
Honey comparison
How do these honeys actually differ, beyond colour?

How honey is tested

The routine panel

The authenticity panel

The wider method landscape

What each method can and cannot prove

Several lines of evidence, not one test

What testing cannot settle

The claims on this page

Emerging research

Carbon isotope testing detects cane and maize syrup in honey but not syrups made from rice, beet or wheat.

Sources

Read next

Tools that use this