Legislative and forensic applications of melissopalynology, the study of pollen in honey

Abstract

Melissopalynology can provide scientific evidence about the authenticity of honey products, which is becoming a medical, economic and legal concern globally. Investigations can evaluate whether the pollen and spores (collectively termed palynomorphs) in a questioned honey sample are consistent with its purported geographical origin and botanical source(s). Legislation can therefore use such melissopalynological information to create benchmarks for authentic honey products and, by requiring explicit labelling of origins on honey products, can counteract honey fraud, protect consumers, and create premium economic products for small-scale honey production. If the authenticity of a honey product is challenged, melissopalynology must be able to provide a robust forensic service. This review evaluates how well melissopalynology qualifies as a forensic science that can be used to litigate under the legislation with which it is intimately associated. It meets criteria of scientific testability; professional peer review and publication; codified standard operating procedures for data collection, quantification and analysis; quantified error rates; and widespread acceptance within its scientific community, and it has been used in court cases. Its public databases are growing and proprietary databases need to be probatively auditable. Like other forensic sciences, melissopalynology is growing in technical sophistication but its ability to address specific relevant questions may depend on the contingencies of each case, e.g. the inadmissibility of illegally-obtained evidence. Melissopalynology is a science that satisfies the baselines of fitness for court, and this overview highlights areas where it can be refined so that its evidence is not merely admissible and legally defensible, but inherently robust forensic science.

Highlights

• •Honey fraud is illegal and thus has relevance to court and litigation.
• •Melissopalynology is a tool for honey fraud detection.
• •It has intimate links to food fraud legislation in several countries.
• •As a forensic science it satisfies criteria for admissible evidence.
• •Melissopalynological evidence has been used in criminal trials.

1. Introduction

Unlike science at large, which is an incremental investigation with provisional outcomes, courts need to reach effective judgements in constrained timeframes. Courts therefore need admissible, relevant, reliable, and competent evidence. Admissibility is the first hurdle for forensic evidence in court. Reviews of the fitness-for-court of forensic sciences in general have identified many opportunities for developing their rigor and robustness [1,2]. These critical perspectives have special value for the development of new fields of forensic science arising from technological innovations and novel juridical needs. Melissopalynology is an established science that has relevance for addressing escalating levels of international fraud involving honey, yet it remains underutilised as a forensic tool due to forensic melissopalynology being in its infancy. This overview examines how far forensic melissopalynology has come in terms of probative value, beginning with its relevance to legal matters.

Melissopalynology is relevant to detecting honey fraud and honey adulteration, which are growing medical, economic and legal concerns globally [[3], [4], [5], [6]], even affecting prestigious honey-producers’ competitions that were meant to showcase the honey market [7]. Litigation may involve, inter alia, adulteration, blending, geographical or botanical mislabelling, and sanctioned transhipping [6,8,9].

Honey fraud is the deliberate misrepresentation of a honey product through inaccurate labelling [9] usually to gain an economic advantage [10]. A honey label serves to inform consumers about the qualities of the product, which may include statements about the botanical origin (e.g. “Eucalyptus honey” or “Single origin honey” or “Wildflower honey”), geographical origin (“Product of South Africa” or “Local” or “Table Mountain honey”), or other desirable features (“Unfiltered” or “Raw” or “Organic” or “Pesticide free”), and misinformation is fraudulent. Melissopalynology cannot directly evaluate whether a honey is not organic but it may associate the honey under investigation with a harvest site that does not qualify as organic. In the case of blended honey, if a Canola honey from country A is mixed with a Canola honey from country B, the botanical origin would still be inferred as Canola honey after blending. However, the honey would not qualify for some geographical origin descriptions such as “Local” or “Product of country A” since these do not accurately reflect the properties of the product in its entirety, and this might be reflected in endemic secondary palynomorphs in the blend. Recent development in EU honey labelling regulations require that the percentage contribution of honey from each country in honey blends is provided [11]. Honey fraud may also occur through honey product mismanifesting and transhipping or “honey laundering”, where honey is moved via an indirect route to hide its geographical origin to circumvent restrictions or tariffs [8].

Honey adulteration increases the profit margin of fraudsters by increasing the volume of ‘honey’ to sell where the honey has been diluted with other materials. Adulteration of honey products with cheaper materials can both increase fraudsters' potential profit margins and undercut the profit of the pure honey industry to the point of making it unsustainable [4,12], particularly for small apiaries in less-developed countries.

Adulteration and fraud techniques have become increasingly sophisticated, making their detection commensurately more challenging. The chemical composition of authentic honey is currently distinct from that of other sweeteners [9,13], allowing forensic detection of adulteration. Concomitantly, increased understanding of natural chemical and melissopalynological variation in honeys has allowed the development of legislative benchmarks for identifying and labelling honey products as having specific geographical and botanical origins (See eAmbrosia, the EU register for honeys protected under geographical indications [14], similar to legislation relating to cheeses and beverages such as cheddar, gouda, port and champagne). Honey products with protected origins represent a novel value-enhancing marketing opportunity for small-scale honey producers, and melissopalynology provides a means to legislate this protected status. The botanical origin describes the nectar sources in a honey, typically characterized as either monofloral (e.g. “Manuka honey”), multifloral (“Wildflower honey”), or a vegetation community (e.g. “Fynbos honey”), and the geographical origin associates a honey with a particular region [15], be it a country (e.g. “South African” or “local”), a province (e.g. “Limpopo honey”), or topographical feature (e.g. “Table Mountain honey”).

Melissopalynology describes the profiles of pollen grains and spores (collectively termed palynomorphs) characteristically present in honey products, providing legislation with the foundation for the development of diagnostic honey benchmarks. It is thus relevant to inferring and authenticating a honey product's origin and purity (i.e. detection of blended honeys) (Fig. 1). It also cross-validates chemical benchmarks defined for the same goal [[16], [17], [18], [19], [20], [21], [22], [23]] and provides legislation with forensic evidence concerning the labelling of honey, especially in terms of botanical and geographical origins [9].

Fig. 1.

The place and roles of melissopalynology in forensic palynology.

While melissopalynology is relevant and widely used, it is currently under-evaluated as a forensic science. Understanding the caveats and risks of error for melissopalynology is important because if benchmarks are legislated too vaguely or permissively, fraudulent and adulterated honeys may remain in the market, while if they are too strict and do not accommodate the natural variation of authentic specialist products, sellers may be unfairly accused of fraud or unnecessarily excluded from international marketplaces.

Melissopalynology's relevance in formulating legislation, and therefore potentially in providing forensic evidence to courts, exposes it to rigorous standards as a forensic science [1,2]. Many jurisdictions have Rules of Evidence governing the admissibility of forensic evidence, including considerations of the relevance, reliability, accuracy, repeatability, error rates and general professional acceptance of the methods and techniques of the science [24,25]. The methods of general forensic palynology have been reviewed in the academic literature [[26], [27], [28], [29], [30]] and tested in court (Table 1), but their specific application in, and implications for, the subdiscipline of forensic melissopalynology are less well documented.

Table 1.

Some case reports where palynological evidence was used in an associative or investigative forensic analysis, grouped by the type of legal question being addressed. Cases specifically involving melissopalynology are set in bold print.

Legal focus	Associative	Investigative	Reference
LABEL FRAUD
Associated the absence of pollen in honey with product labelling fraud	X		[31]
Associated the absence of pollen in honey with product labelling fraud	X		[32]
Associated the absence of pollen in honey with product labelling fraud	X		[33]
Associated the absence of pollen in honey with product labelling fraud	X		[34]
BOTANICAL ORIGIN
Associated the pollen content of the honey with the botanical origin label claims	X		[35]
Associated the pollen content of the honey with the botanical origin label claims	X		[36]
Associated the pollen content of the honey with the botanical origin label claims	X		[37]
GEOGRAPHICAL ORIGIN
Identified fraudulent honeys marketed as authentic United States products	X	X	[38]
Identified hives that were illegally imported into a US state that was under quarantine for honeybee health concerns		X	[38]
Described the foraging behaviour of honeybees at an apiary site, indicating that the bees had not foraged from crops sprayed with pesticides	X		[38]
Associated the victim with the crime scene	X		[39]
Associated an offender with the last seen locality after escaping police pursuit	X		[40]
Along with other particulate evidence, associated an offender with a crime scene	X		[40]
Associated an offender with a crime scene	X		[40]
Associated sheep advertised at an auction with a farm where sheep had recently been stolen	X		[40]
Associated a suspect with a noose found at a crime scene	X		[40]
Inferred the probable geographical origin of an artefact as evidence of its authenticity		X	[40]
Associated the geographical origin of seized marijuana with the general area where the accused resided	X		[40]
Associated three subsamples of marijuana seized from individuals in respective cities with a single large consignment of marijuana. The pollen composition was also used to infer the likely geographical origin of the consignment	X	X	[40]
Associated the soil on an item in the possession of a suspect with the soil of an illegal marijuana plantation	X		[40]
Inferred the geographical crime scene where bags of soil were used to replace stolen machinery during a multinational shipment		X	[40]
Inferred the geographical crime scene where limestone rocks were used to replace stolen whisky in a shipment		X	[40]
Inferred that the purported origin of a product was inconsistent with the pollen profile recorded from the product	X		[40]
Inferred the season during which multiple victims recovered from a mass grave were murdered		X	[41]
Associated an offender with a crime scene	X		[42]
Supported victim's crime scene allegation	X		[43]
Inferred the geographical origin of a seized sample of marijuana		X	[38]
Inferred that the victim had been moved to the crime scene after their murder and proposed the geographical region where the victim most likely was murdered	X	X	[38]
Associated an offender to have been in proximity of the crime scene	X		[38]
Inferred the likelihood that particulate matter recovered from an airplane engine was responsible for a fatal accident		X	[38]
Inferred the geographical origin and subsequent distribution pattern of seized cocaine		X	[38]
Inferred the geographical origin of historical artefacts suspected to be smuggled illegally	X	X	[38]
Inferred the origin (place of residence) of an unidentified victim		X	[44]

In this paper we trace the origin of the need for melissopalynology as a source of forensic evidence and evaluate its theoretical and empirical ability to meet that need. Ultimately, this review aims to illustrate that forensic melissopalynology is well aligned with the needs of admissible forensic evidence.

2. Relevance: the nature of forensic palynology

2.1. General forensic palynology

The identification of pollen and spores is key to various fields of research, including paleopalynology, aeropalynology, entomopalynology, and melissopalynology (Fig. 1). Like the forensic applications of ballistics, genetics, fingerprint analysis, bite mark analysis, handwriting analysis, and several other investigative methods [1], forensic palynology provides probabilistic support to associate evidence with another item, person or scene, or infer the probabilistic origin of an item or person through pattern-matching analyses [26].

The general theory of palynology is that the morphological characteristics of palynomorphs are very diverse across the plant kingdom (Fig. 2), but reliably consistent within particular plant groups, generically termed taxa, allowing the construction of a qualitative or quantitative profile for a sample based on the variety of palynomorphs in it, and the inference of the biological identity of each type of palynomorph to varying degrees of taxonomic detail. Identification is achieved by matching unknown palynomorphs with identified specimens vouchered in reference collections [45], illustrated in pollen atlases [[46], [47], [48], [49]], or depicted in other academic literature [[50], [51], [52], [53], [54], [55]].

Fig. 2.

Five (1: Brassica sp. (Brassicaceae), 2: Apiaceae, 3: Searsia sp. (Anacardiaceae), 4: Proteaceae, 5: Scrophulariaceae) of the twenty-nine palynomorphs observed in a honey harvested from the Greater Cape Floristic Region, South Africa (© J Delport 2026). Expertise in recognizing the natural variation in the size, shape, and surface detail of pollen, such as is observable in this figure, is important for palynologists to infer accurate palynomorph identifications.

By similar pattern-matching logic, the suite of palynomorphs in a sample's profile can be used in two ways (Fig. 1), as investigative evidence to infer the sample's potential geographical, climatic, or botanical origin, or as associative evidence to associate or compare the sample with other samples [29].

Associative cases draw on Locard's Exchange Principle [56] that items and their environment exchange physical traces that allow investigators to link them, even if the traces are microscopic. Thus, forensic palynologists collect palynomorphs from relevant clothing, bodies, vehicles, etc. [30] and compare the profiles of these samples to those from items or people theorized to be associated with the crime, and to profiles from cross-validating control samples from the crime scene [29]. The palynomorph types in profiles must be differentiated consistently and expertly, and when taxonomic identification is also possible, it provides valuable additional evidence. Associative evidence is valuable for binary enquiries, such as whether the palynomorph profile of an exhibit of questioned origin is either “consistent” or “inconsistent” with a reference profile of known origin [29].

In investigative cases, forensic palynologists obtain palynomorph profiles from relevant items or persons and infer the most probable botanical community or locality from which they might originate. Interpreting investigative evidence requires more knowledge since there is no exhibit of known origin for comparison, and the palynologist must use taxonomic information drawn from the profile(s) of the exhibit of questioned origin and knowledge about those plant taxa to infer what plant community or geographical area is most likely to be associated with it.

Pattern-matching analyses in general provide forensically reliable individuation of questioned samples best when background rates of pattern occurrence are available to provide confidence in the uniqueness of a match [1,2]. This can be done in a frequentist probabilistic multivariate framework using, for instance, non-metric multidimensional scaling [57,58], or principal component analysis [[59], [60], [61]]. Alternatively, the probability of a match can be evaluated against such background rates in a Bayesian inference framework [[62], [63], [64], [65], [66], [67]] using e.g. Poisson Discriminant Function Analysis [59,61,[68], [69], [70], [71]], to show whether it is well beyond being coincidental or random, given what is known of palynomorphs in relevant localities and situations. In both cases, a database of scores to hundreds of background pollen profile samples representing the spectrum of variation in the relevant classification variables under investigation are used to provide context. To provide more confidence in the result, such databases should be large, unbiased and sufficiently public to be auditable. Palynology relies on herbarium collections and large databases containing images of identified pollen grains and spores, and associated metadata such as their geographical and seasonal origins. In some situations, the available background information may be too sparse, too biased in representation, or too imprecise to provide the relevant level of assurance about a match and investigators must collect additional cross-validating associative evidence to validate their inferences.

Case reports are available that attest to the positive probative performance of forensic palynology (Table 1). Although the forensic uptake of palynology has been slow, intermittent and tentative, palynological evidence has gradually become more used in investigations. The availability of databases is growing rapidly with the development of open-access on-line digital repositories, and the numbers of qualified expert witnesses is growing slowly. However, the still limited demand for, or adoption of, the field is driven by factors such as investigative and legal professionals’ low levels of awareness of the science and the scarcity of funding to train and sustain palynologists specialized in forensic investigations [38].

2.2. Forensic melissopalynology

The techniques of forensic palynology also apply to melissopalynological investigations [27,29,30]. However, melissopalynology differs from other branches of palynology in several ways. It focuses on honey products specifically, and it is intimately associated with (and, in fact, incorporated into) a specific sector of legislation. This legislation concerns honey-related commerce, sets labelling and pollen-related standards for products [14,72], and facilitates its relationship with forensic melissopalynology by requiring that pollen may not be filtered from honey products [11,73]. Melissopalynology draws forensic validity from general palynological sources and from additional specialized public and private database libraries of palynomorph profiles from honeys with known geographical and botanical origins. Finally, the reliability of its evidence can additionally be cross-validated with the physicochemical properties of the honey.

Like palynology generally, melissopalynology has investigative and associative applications (Fig. 1). In investigative melissopalynological analyses, individual diagnostic pollen grains and spores in profiles from samples of questioned honey products are matched against taxonomically identified images to create a qualitative ecological profile of the plant community around the source hive(s), leading to the inference of its geographical, seasonal and/or botanical origin. The pollen grains are not only from the flowers on which the bees foraged, but also from other vegetation that the bees interacted with, due to Locard's Exchange Principle. The more pollen types that are taxonomically identified, the more nuanced the inference of origins can be. This evidence can be compared to legislatively mandated information on the product's label as a test for honey fraud and honey laundering.

Associative melissopalynological analyses use peer-reviewed literature and public and private databases to compare, quantitatively or semi-quantitatively, the profile of a questioned honey product with profiles of known origin(s). When the botanical origin on a product's label is questioned, analysis must identify and rigorously quantify the relative abundances of key pollen types specified in legislation and compare these measurements to legislated standards [72]. When the geographical origin on the product's label is disputed, as many as possible of the profile's palynomorphs are identified to infer a qualitative ecological profile of the originating plant community that is matched probabilistically with databases of relevant known profiles to test the geographical claims on the label. Both approaches evaluate claims on the product's label as a test for compliance with legislation, particularly regarding trademarking and patenting.

3. Reliability: melissopalynology as a forensic tool

The pattern-matching forensic sciences have been used successfully in prosecutions in many cases in many countries, but also face a suite of criticisms applicable to all forensic techniques. These include whether the technique can produce reliable evidence, and the extent to which there has been critical evaluation of the expert witnesses’ interpretation of the evidence which may be influenced by factors such as bias, error, or the improper application of a technique [1,2,74]. These generic weaknesses have led to numerous instances of techniques such as DNA profiling, bite mark analysis, shoe print analysis, voice analysis, and hair analysis failing to serve justice in specific cases [74,75]. Forensic melissopalynology can enhance its evidentiary validity by attending to these criticisms, which provide part of a basis for evaluating forensic melissopalynology in relation to evidentiary standards for expert testimony.

In addition to such critiques, many jurisdictions have guidelines for evaluating the admissibility of expert opinion and/or forensic evidence. Examples of guidelines that provide criteria specifically for evaluating the reliability of scientific evidence are the United Kingdom's amended Criminal Practice Directions 2023 [25], the United States Federal Rules for evidence [76], and specifically the United States Federal Supreme Court's judgement in Daubert v. Merrell Dow Pharmaceuticals [24]. There are numerous congruences between these guidelines; for evaluating the admissibility of melissopalynological evidence, this manuscript will adopt the framework of the Daubert criteria because it has been discussed extensively in the forensic literature. The five criteria are as follows.

• (1)Can, and has, the theory or technique in question been tested?
• (2)Has it been subjected to peer review and publication?
• (3)What is its known or potential error rate?
• (4)Are standards for controlling its operation available and are these being maintained?
• (5)Has it attracted widespread acceptance within a relevant scientific community?

These factors are not a compliance checklist or a set of hurdles for expert testimony, but rather guidelines to weight a source of expert testimony in terms of the needs of court. They address criteria of relevance, reliability, and unambiguousness for forensic sciences. As such they provide a quality assurance framework for enhancing the forensic excellence of a scientific field [77]. The Daubert criteria thus provide a convenient framework for examining the forensic merits and level of probative development of melissopalynology.

3.1. Criterion 1: empirical testing

Reliability [78] is a key property for admissible evidence [24,25]. The Daubert criterion that a science can in principle be tested is underpinned by the Popperian philosophical position that for a subject to qualify as a science, its hypotheses should be falsifiable by testing [79]. However, the reliability of a theory is established as fact only after testing has demonstrated it consistently. By extension, a science's methods and techniques should be demonstrably inductively reliable through empirical testing.

The reliability of melissopalynological research can be evaluated based on how well palynomorph identification tools are developed and applied. Palynomorph identification tools, hereafter referred to as pollen tools, can take the form of literature (e.g. pollen atlases, postgraduate theses, systematic studies, and palynological studies) [[46], [47], [48], [49],51,54,55,80,81], digital databases such as the African Pollen Database, The Global Pollen Project, PalDat, PollenAtlas, and The Pollen-Wiki [45,[82], [83], [84], [85]], and (ideally public) physical reference slide collections held in institutions like herbaria and palynology laboratories. The latter resource is especially important as a primary source for the pollen photographs and descriptions archived in the literature and digital databases. Physical reference slides are preferably produced using pollen and spores collected directly from identified voucher specimens preserved in herbaria [86,87]. Benefits of this include assurance of the taxonomic identity of the palynomorph source, continuous preservation of the voucher material for re-examination, and opportunity for on-going curation after taxonomic revisions [86,87].

The reliability of pollen processing methods and the interpretation of pollen grains have been tested empirically. Methods for the extraction, processing, and analysis of melissopalynological samples have been tested, established and harmonized across institutions [15,88]. Harmonization allowed these methods to be subjected to tests in rings of laboratories that explicitly established their ability to provide consistent and reliable results [15].

Confounding influences on palynomorph morphology (e.g. processing methods, natural variation, or mutations) that may affect the accuracy of identifications have been and are still investigated. Several instances were documented where pollen morphology may be misinterpreted due to the observation or processing method used, or due to aberrant pollen structures [80]. Acetolysis has been shown empirically to facilitate differentiation between Brassica and Salix pollen grains, which are almost indistinguishable when not acetolysed [89]. Pollen grains of Pachycereus weberi show exceptional natural size variation that creates a risk of misidentification if pollen tools do not account for this variation [90]. Similarly, confounding deviations in the number of apertures can occur in pollen from several genera of Myrtaceae (e.g. gum trees) [[91], [92], [93], [94]]. As melissopalynologists familiarize themselves with these variations (e.g. Fig. 2), the accuracy and subsequent reliability of their inferences of palynomorph identity increases.

The reliability of palynomorph identifications and interpretation of melissopalynological data have been validated through published ring tests across different laboratories [15]. Identification reliability may further be improved if reference photographs of the palynomorph observed in the honey are available for audit by other palynologists to cross-validate the identification and interpretation of the data. Reference photographs of the palynomorphs observed in a honey would also serve as a record of the evidence used in a forensic investigation.

3.2. Criterion 2: peer review and publication

Peer review is part of the self-correcting character of science. The academic culture of science exposes new data and explanations to professional and public scrutiny and critique. Pre-publication review typically involves feedback from three independent experts recruited by the editor, sometimes in a double-blind format, and it can forestall publication of flawed works. Post-publication commentary from anyone can be published as a direct response or an aside in a related research or review article. Not every manuscript is reviewed before publication, and the progress of science may eventually expose latent flaws in peer-reviewed publications, so on-going post-publication peer review provides crucial quality assurance to science.

A rough estimate of the size of the pool of reviewers available in a field can be derived from the numbers of publications in that field, which can be gauged using public internet databases and search engines. In September 2025, a search using the terms “melissopalynology”, “melissopalynological”, “pollen analysis AND honey”, and “honey origin AND pollen” recovered ±17012 documents (excluding citations) from Google Scholar (which includes articles, books and book chapters) and ±1316 articles from Scopus. A recent review of melissopalynological research articles from Africa alone recovered 140 studies published in a variety of professional, peer-reviewed journals [95]. There is clearly a substantial number of peer-reviewed publications available on melissopalynology, and by implication there is a substantial population of post-publication reviewers.

Pre-publication review relies on gatekeeping by a very small number of opinions per manuscript, which poses a quality assurance concern for individual studies. Due to most scientific periodicals' policy of anonymous pre-publication peer review, it is almost impossible to assess the expertise of their reviewers directly, but editors typically match manuscripts to reviewers' expertise in the interests of their journal's reputation. Melissopalynological research does not (yet) produce studies at a sufficient rate to warrant its own specialist journal, but studies published in the palynological journals will fairly certainly have been reviewed by peers specialized in palynology, if not actually in melissopalynology. The quality of peer-review is thus an aspect of the quality of the journals themselves.

There are at least five peer-reviewed journals dedicated to palynology, each published and distributed internationally for over half a century (Table 2). Some of them are attached to professional palynological societies (Table 2). Melissopalynological research has also been published in international journals focused on food science [60,[100], [101], [102], [103], [104], [105], [106], [107]], exposing it to a broader critical professional audience. The publishers of many of these journals operate internationally and are generally regarded as reputable, adding to the perceived credibility of their articles, so the issue of predatory journals and publishers [[108], [109], [110]] and vanity publishing is far less of a concern than that of weak studies that pass pre-publication review in more academically lenient journals.

Table 2.

Alphabetical list of journals focused on publishing peer-reviewed palynological research, regularly including studies of melissopalynology.

Title	ISSN		Inception/Latest volume	URL	Academic affiliation	Current Publisher
	Print	On-line
Grana [Palynologica]; An International Journal of Palynology
	0017-3134	1651-2049	1954/64 (2025)	[96]	Scandinavian Palynological Collegium (CPS) and International Association for Aerobiology (IAA)	Taylor & Francis
Japanese Journal of Palynology
	0387-1851	2433-0272	1955?/71 (2025)	[97]	Palynological Society of Japan	Palynological Society of Japan
Journal of Palynology
	0022-3379	-	1965/59 (2023)	None found	None found	Today & Tomorrow's Printers and Publishers
Palynology
	0191-6122	1558-9188	1977/49 (2025)	[98]		Taylor & Francis
Review of Palaeobotany and Palynology
	0034-6667	1879-0615	1967/336 (2026)	[99]		Elsevier

Weaker scientific studies tend to be recognized for what they are and ignored to a correlated degree by subsequent authors. In fields with many active authors, the citation rates of articles are roughly correlated with their quality and/or general relevance, especially once self-citations are excluded. Some key articles in melissopalynology are particularly well cited; for instance, in January 2026 CrossRef [111] reported that Louveaux et al. [88] had received at least 1183 citations over 36 years and Von Der Ohe et al. [15] had garnered at least 518 citations over 20 years. Such publications set an upper bound on the citation rates that can be expected in a particular discipline, and the distribution of citations across the remaining articles is expected to follow a Pareto distribution that is affected by the quality, age, digital accessibility, and sometimes the language of each article. However, since citations rates are confounded measures of quality (especially with current trends in manuscript writing using artificial intelligence [AI] tools), each article should also be assessed directly, on its own merits [95].

Overall, melissopalynological research has clearly received extensive quality assurance through peer review, reputable publication and generally positive reception internationally over at least the last four decades. If the melissopalynologists and their publisher can resist emerging trends to abuse AI tools and prioritize self-interest over academic integrity [108], future melissopalynological publications should remain at their current court-worthy level.

3.3. Criterion 3: known or potential error rates

Qualifying and quantifying the error associated with a scientific method is important to calibrate the reliability of the evidence derived from that method. Here, the focus is on random error that may arise through the methods, data generation, or statistical analysis of melissopalynological investigations.

Analytical standard operating procedures (SOPs) specific to the methods of melissopalynology have been published [15,88] and widely applied. The authors of the SOPs quantified the repeatability and reproducibility of qualitative and quantitative melissopalynological analyses of pollen composition using interlaboratory ring trials consisting of seventeen and sixteen expert palynologists from fifteen laboratories respectively [15]. The precision, measured by the relative standard deviation of reproducibility (RSD_R) and relative standard deviation of repeatability (RSD_r), associated with quantitative analysis, which depends on the competent extraction of palynomorphs, was deemed acceptable for low, medium, and high pollen concentrations. This is as anticipated because the extraction process is a recipe-based, standardized laboratory protocol that, done competently, should deliver very consistent results.

The precision associated with qualitative analysis, which also depends on palynomorph extraction, data generation, and interpretation, was within an acceptable range too [15]. Again, data collection is fundamentally a recipe-based laboratory protocol that is designed to be reliable. While the interpretation of analyses is qualitative, it is based on quantitative thresholds [15] that can be applied in a rule-based way that leaves little room for interpretive error. Palynomorph misidentification was not calibrated for the 17 experts involved in the ring trials, but it could not have been concerning because it was a component of the overall precision. The differences between the reproducibility and repeatability of the qualitative analysis was described as “relatively small”, with the authors claiming that “… the human component is less important than usually considered …” [15].

A separate evaluation of palynologist's proficiency in quantifying aeropollen data found that, regardless of the amount of preparatory training they received, experienced palynologist had lower rates of error than inexperienced palynologists [112]. This finding probably applies to melissopalynology too because the methods of identification and quantification in aeropalynology and melissopalynology are similar, specifically in their need for the palynologist to distinguish palynomorphs from non-palynomorphs, to distinguish palynomorphs (consistently), to count occurrences of each palynomorph (correctly), and to identify each significant palynomorph (consistently and correctly). It is likely that the error associated with melissopalynological data collection will be ameliorated by experience too. The emphasis on experience may diminish as new technologies are introduced, such as automated image recognition for palynomorph identification and counting [[113], [114], [115]]. Such systems do not suffer from observer fatigue and can be designed to refer novel palynomorphs for human attention. Clearly, experienced palynologists will remain necessary to provide quality assurance to the outputs of such technology.

3.4. Criterion 4: standards

Practically every stage of the melissopalynological analysis can be matched against an established, formal professional standard. This includes standards for personnel, laboratories, protocols, interpretation and reporting for evidence.

At the personnel level, consulting melissopalynologists may be regulated by national professional and statutory bodies such as the South African Council for Natural Scientific Professions (SACNASP) [116]. These bodies may provide recognition of qualifications, require statutory registration, mandate and oversee professional development mechanisms, set ethical norms, and/or sanction misconduct. Certification of academic competence may occur through tertiary education institutions, particularly universities’ standards for doctoral programs, which may be nationally legislated, e.g. through the South African National Qualifications Framework (SAQA) [117]. Standards specific for evaluating the accuracy of individual melissopalynologists have not been proposed; ring tests [15] are a means of sustaining low error rates in the profession [[118], [119], [120]], but they are not generally practical for routine use in quality assurance. The peer review system of scientific publications also mediates an informal but potentially uneven standard.

General sample management protocols are regulated by jurisdictional chain-of-custody practices outside the laboratory [121,122] and informed by Standard Operating Procedures (SOPs) within the laboratory that are defined by bodies such as the International Organization for Standardization (ISO) [123] and ASTM International [124]. SOPs may be regulated by jurisdictional quality assurers such as the South African National Accreditation System (SANAS) [125]. Responsibility for these standards lies outside melissopalynology, providing independent quality assurance. Adoption of such standards provides confidence in the handling of melissopalynological samples and the running of laboratories, if not in melissopalynology itself.

Regarding the protocols of melissopalynology itself, standard methods for processing pollen evidence were originally not prioritized over other approaches [89], but they are now standardized, harmonized, widely accepted and applied in the interests of quality assurance and making melissopalynology relevant to its users [15,88]. In particular, the use of standardized laboratory methods facilitates comparisons of results across laboratories and studies. This has been key in developing melissopalynology as a forensic science with a definite user community.

These systems ensure that the evidence derived from melissopalynological sample management and processing is of a high standard. Similarly, the interpretation of this evidence requires its own standardisation criteria. Ideally, the quality of sample interpretation by human observers can be standardized through international on-line palynological databases [45,[82], [83], [84], [85]] of images of voucher specimens obtained directly from plants (and processed in various standard ways if these produce characteristic analytical effects). Practically, these databases are in their relative infancy. While they are developing, melissopalynologists must rely on matching evidence against their own pollen collections and illustrations in public documents such as pollen atlases [[46], [47], [48], [49]]. Pollen atlases and some digital databases generally provide the voucher details of the specimen(s) used in the study, and the associated material is archived in an accessible herbarium or palynomorph collection.

The basis of identification is generally not documented well in published palynological research, since reference is commonly made only to the pollen identification tools used for comparison, without providing accompanying voucher images of the palynomorphs found in the research samples. Automated pollen identification generally trains artificial intelligence software on bespoke private libraries of palynomorph images that have been identified by experts, relying on the experts’ expertise for quality assurance [114]. In either case, lack of access to the evidence impedes other palynologists from cross-validating the inferred taxonomic identity of the palynomorphs, which may cast doubt about the interpretation of palynomorph profiles submitted as evidence. The absence of documentation of the observed palynomorphs also limits the functionality of botanical and geographical palynomorph profiles that would be used as reference in associative and investigative investigations. Greater transparency of melissopalynological evidence to improve the reproducibility of palynomorph identifications and pollen profiles strongly merits more attention. Fortunately, there is no need to identify every palynomorph, provided that the key taxa mentioned on the label of a honey product are identified (correctly), and that each palynomorph is recognized consistently, especially in quantitative analyses of palynomorph profiles by both humans and software programs.

The interpretation of palynomorph profiles is less well standardized than the interpretation of the identity of individual pollen grains. For instance, the interpretive criteria for classifying honeys as monofloral or multifloral have not yet been standardized [72]. In the harmonized methods of melissopalynology, Von Der Ohe et al. [15] propose that in general, a honey may be classified as monofloral if one taxon dominates the pollen composition (>45%). However, the authors also list exceptions to this criterion that depend on whether a taxon is by definition typically under-represented or over-represented. Methods have been proposed for determining the appropriate benchmarks for honey types [20,126], but a relevant standard operating procedure has not yet been proposed. Thrasyvoulou et al. [72] reviewed the pollen benchmarks that countries have adopted for the classification of their monofloral types, and there are major differences between them. For example, the minimum requirement for recognizing Citrus honey in Greece is that 3% of the constituent palynomorphs must be Citrus pollen grains, whereas in Germany the requirement is 20%. At present, lawyers may particularly target this in legal challenges by questioning the validity of using pollen benchmarks that are higher or lower than those of other countries. However, this is currently an issue of legislation, not of melissopalynology.

As in other sciences, melissopalynology is also likely to evolve more sophisticated input on this issue. While modern approaches rely on interdisciplinary methods to design legally defensible pollen benchmarks [20], disagreements about the thresholds and which markers are suitable for classification may be debated. Thus, it would be beneficial to establish a scientific standard operating procedure with a list of required honey properties that should be used for determining appropriate minimum requirements [20,126], and this may even be implemented in software programs that can provide quantitative, probabilistic, multivariate classifications.

Standardization of how melissopalynological evidence is interpreted and reported in forensic investigations also warrants attention [127,128]. The descriptor (label) of the honey under investigation will determine the type of forensic question(s) asked, which in turn influences how the evidence is collected, evaluated and presented. Botanical origin investigations are measured against legislated thresholds, which would result in the pollen profile being interpreted as consistent or inconsistent with the descriptor, or inconclusive [26]. The binary nature of the question thus limits the scope of interpretation of the evidence. Geographical origin investigation requires more sophisticated analyses because the entire pollen profile is evaluated rather than a specific or dominant palynomorph as happens in botanical origin classification. The forensic question of geographical origin is also measured according to the descriptor but does not equate to a binary outcome like botanical origin inference. Rather, the pollen profile is evaluated using multivariate statistics [[57], [58], [59], [60], [61]] against reference profiles, and the weight of the discrepancies between the profiles, e.g. presence of anomalous palynomorphs, is also evaluated. The evidence should be reported as consistent or inconsistent with the descriptor, or as inconclusive, with detailed explanation about why the expert witness came to this interpretation.

3.5. Criterion 5: professional acceptance

Obviously, contentious science may provide equivocal evidence that has dubious relevance or reliability for court proceedings. Ideally, forensic evidence should not need to go on trial too. Fortunately, melissopalynology has already achieved acceptance by the scientific, legislative, commercial and judicial professions.

The scientific professions have a decades-long history of accepting palynology and melissopalynology as a technology in peer-reviewed research globally [58,95,102,[129], [130], [131], [132], [133], [134]], producing enough scientifically uncontentious outputs to support at least five international specialist journals (Table 2). The majority of articles are technical, but landmark theoretical studies began appearing decades ago [15,88,126], and melissopalynology is internationally recognized as a theoretical framework for inferring the botanical and geographical origin of honey [9,15,72,88,100,104,134].

That scientific recognition underlies the acceptance by the legislative profession in at least Croatia, Germany, Greece, Italy, and Serbia of melissopalynology as a basis for regulating standards for the classification and certification of honey products [72], and it is being integrated into EU legislation [11].

Both the scientific standing and the legislative relevance of melissopalynology as a source of evidence that can be taken to litigation have promoted its commercial acceptance, although few cases have been documented (Table 1). However, there is sufficient demand that melissopalynological analysis is offered as a professional commercial service by many industrial and research institutions (Table 3).

Table 3.

An illustrative list of examples of industrial and academic institutions that offer melissopalynological services. Institutions were identified by searching for keywords (“honey pollen testing”, “honey origin testing”, and “melissopalynology service”) and based on personal knowledge from direct engagements. This list does not constitute endorsement of the institutions as service providers.

Continent	Institution	Website
Africa
	Melimetric	NA.
Australasia
	Analytica Laboratories	[135]
Europe
	Ages	[136]
	Eurofins	[137]
	Food Forensics	[138]
	HoneyLab Teper & Waś	[139]
	Intertek	[140]
	Melissopal Pollenanalytik	[141]
	Mellifloral	[142]
	Niedersächsische Landesamt für Verbraucherschutz und Lebensmittelsicherheit Institut für Bienenkunde Celle	[143]
	Tentamus	[144]
	University of Worcester	[145]
North America
	The Penn State Honey and Pollen Diagnostics Laboratory	[146]
	Texas A&M University Palynology Research Laboratory	[147]

Finally, the existence of case law involving melissopalynology [38] (Table 1) suggests its evidentiary acceptance by the judicial profession too, at least in the originating jurisdictions. The relative scarcity of reported cases is due more to a general lack of awareness of the forensic relevance and probative potential of palynology in general, although some notable cases have been publicized (Table 1). The explicit inclusion of melissopalynological criteria in legislation guarantees that case law will accumulate as that legislation leads to prosecutions.

4. Conclusions

Melissopalynology is a branch of palynology with a few unique features (particularly its explicit incorporation into legislation regarding honey products) that suit it to specific forensic applications. It can thus draw forensic credibility from both the general theory of palynology as a pattern-matching science and from its specific use in setting standards and benchmarks in honey-related legislation, which rests on its direct operational relevance.

Melissopalynology is increasingly well established as a science, with a history as one of the earliest forms of forensic palynology [38]; a tested, standardized and harmonized suite of methods with measured error rates and known confounding conditions [15,88]; and a large, peer-reviewed, public literature of background data (e.g. Refs. [46,58,89,95,132,134] that is finding its way into synoptic databases that can fuel appraisals of melissopalynological evidence. It is pivotal though that these databases are accessible to the public to ensure transparency of the reference images and palynomorph profiles that palynologist refer to for their investigations. Melissopalynology is sufficiently accepted by relevant experts that it has been adopted by legislation as a means of gauging the authenticity of honey products [11,14,72,73]. It is in the process of integrating automated identification software that may further diminish its known error rates [113,114].

Such well-consolidated structure has, in turn, created the need for melissopalynology to provide forensic evidence to courts when the authenticity of a honey product is contested. Its history and findings have placed it well to serve this role, as illustrated by its good fit to quality assurance guidelines such as the Daubert criteria of the United States Supreme Court and the United Kingdom's amended Criminal Practice Directions 2023 [25].

As with most examinations, reaching a minimum threshold of qualification is a gateway to adequate performance, but achieving full marks is the most desirable goal for all stakeholders. Every science by its nature has scope for development, particularly through refining its quality assurance protocols to further identify and diminish error, increasing the robustness of its methods and interpretation in practice, and enhancing the reliability and trustworthiness of its practitioners and their publications. This includes further investment in research, embracing rigorous quality assurance processes (testing, controls, vouchering, documentation), illustration of voucher material in publications (even if only in their supplementary files), validating improved techniques (like those with higher rates of accuracy), recognizing competent practitioners and reviewing epistemological achievements so that melissopalynological evidence is not merely legally defensible, but inherently robust.

In summary, forensic melissopalynology satisfies all five of the Daubert criteria, each to a significant extent, and by implication meets the evidentiary requirements regarding forensic evidence of at least some other jurisdictions. It has definite opportunities for further development, like any forensic discipline. For instance, fingerprint analysis is routinely used with recognition of its known caveats [1,148,149]. Specifically, melissopalynology can standardise how its evidence is documented and archived to ensure transparency regarding palynomorph identification. Like all forensic evidence, melissopalynological evidence comes with context-dependent caveats, but it is not premature as a legislative, investigative and prosecutorial tool.

Data availability statement

All the data used to generate this review has been cited in this document and was accessed through the internet. No other datasets were generated.

Use of artificial intelligence (AI) tools

No AI tools were used in this work.

CRediT authorship contribution statement

Janaïs Delport: Conceptualization, Funding acquisition, Investigation, Visualization, Writing – original draft, Writing – review & editing. Martin Herrer Villet: Conceptualization, Investigation, Visualization, Writing – original draft, Writing – review & editing.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Janaïs Delport reports financial support was provided by National Research Foundation. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.