Back in 2022 I wrote about a project in which I’d been involved with colleagues at the Natural History Museum, Imperial College London and several regional museums. We were using preserved bumblebees to ask a deceptively simple question: what can insects collected decades ago tell us about the environmental changes that their populations have experienced?
At the time I wrote that this opened up the possibility of looking at how the genomes of bee populations had changed through time. Well, we’ve now done it.
Published this week in Molecular Biology and Evolution, our new study focuses on the Moss Carder Bumblebee (Bombus muscorum), a beautiful species of flower-rich grasslands, heaths, marshes and coastal habitats. Once much more widespread across Britain, it is now largely concentrated in Scotland and the Scottish islands, with scattered populations surviving around the coasts of England and Wales.
Led by Victoria Mullin, the team generated whole-genome data from museum specimens collected between 1894 and 2019. The results provide a rather sobering picture of what has happened inside these bee populations as their geographical distribution has contracted.
Comparing bees collected before 1926 with those collected after 1975, we found about a 25% reduction in genome-wide genetic diversity. Looking at the full time series gives the same basic message: heterozygosity declined progressively through the 20th century, with no obvious sign of recovery.
That matters because losing a species is not the only way that biodiversity can decline. Long before extinction, populations can become smaller and more isolated and begin to lose genetic variation. That variation is part of the raw material that allows populations to respond to environmental change, disease and other challenges.
There was also a striking geographical pattern. In England and Wales, the amount of the genome occurring in long stretches of homozygosity increased by about 186% – in other words, almost threefold. These “runs of homozygosity” are consistent with populations becoming smaller, more fragmented and increasingly isolated from one another. Scotland showed a much weaker change, reflecting the fact that B. muscorum remains considerably more widespread there.
Importantly, however, the population structure is not simply “Scottish bees versus English bees”. The genomic data reveal a broad north-to-south gradient across Britain – what population geneticists call isolation by distance – rather than a set of sharply separated populations. That has implications for conservation because fragmentation may be breaking what was once a more continuously connected population into increasingly isolated regional groups.
There’s another interesting connection here with work that I published with colleagues more than a decade ago. In our 2014 Science paper we used historical records to show that Britain had lost 23 species of bees and flower-visiting wasps, with the fastest rate of extinction occurring from roughly the late 1920s to the late 1950s. We argued that large-scale changes in British agriculture, beginning just after the First World War, were an important part of that story.
The new study is looking at genetic change within a surviving species rather than the disappearance of whole species and it cannot pin that change on a single cause. But the parallels are difficult to ignore. The loss and fragmentation of flower-rich habitats associated with agricultural intensification is a likely contributor, alongside pesticides, climate change and other environmental pressures. The genomic erosion appears to have accumulated progressively across much of the same century in which Britain’s pollinator landscapes were being radically transformed.
One unexpected finding was also a useful warning about museum data. Of the 130 specimens originally labelled as Bombus muscorum, genetic analysis showed that 29 (almost a quarter) were actually two other, morphologically similar bumblebee species. Museum collections are extraordinarily valuable, but identifications cannot always simply be taken at face value, particularly for difficult groups. Genomics provides both a historical record and a way of checking the identity of the specimens supplying that record.
And that brings me back to what I think is one of the wider messages of this whole project: natural history collections are not dusty archives of dead organisms, they are biological time series.
A pinned bumblebee collected more than a century ago preserves information about where a species occurred, what environmental stresses it experienced, and, as we can now show, the genetic diversity of the population from which it came. Combined with modern genomic techniques, the millions of insects sitting in museum drawers around the world represent an extraordinary resource for understanding biological change.
The conservation message for the Moss Carder Bumblebee itself is fairly straightforward. Protecting the remaining populations is not enough if they become progressively smaller and more isolated. Restoring large areas of suitable, flower-rich habitat and reconnecting populations should help maintain population sizes and gene flow. More interventionist approaches such as translocation or assisted gene flow may eventually have a role, but would need careful assessment because of issues such as local genetic structure and disease transmission.
Four years ago, when I blogged about the first papers from this project, I finished with the idea that museum genomics might allow us to understand how insect populations had adapted, or failed to adapt, to a century of environmental change. We now have part of that answer. And it shows that beneath the visible contraction of a species’ range, another quieter form of biodiversity loss can be taking place: the erosion of diversity within its genome.
The paper is open access and you can download a copy here:
One of the things I love about working on the milkweed and dogbane family (Apocynaceae) is that it has an apparently inexhaustible capacity to produce strange and wonderful interactions with animals. Just when you think that the family’s pollination biology has yielded most of its surprises, along comes something like this.
And “fake figs” is not journalistic exaggeration, as you can see from the photos above, which are taken from the paper. The flowers of H. ficoides are globose, fleshy structures, about 1.5–2 cm across, with a small opening at the top. Superficially they look remarkably like the enclosed inflorescences (the syconia) that we normally call figs. Indeed, when the species was first described in 2019, its resemblance to a fig was obvious enough to inspire the name ficoides: fig-like. But looking like a fig is only the beginning of the story.
Fooling one of nature’s most specialised pollinators
The relationship between figs (Ficus) and their tiny agaonid wasps is one of the classic examples of an intimate plant–pollinator mutualism. Female fig wasps locate receptive figs using their scent, squeeze through the tiny opening or ostiole, and pollinate the enclosed flowers while attempting to reproduce inside. These relationships can be remarkably specific, in some cases representing rare examples of one-to-one plant-pollinator interactions.
So it was genuinely surprising to discover female Kradibia fig wasps entering flowers of Heterostemma ficoides. These wasps normally pollinate Ficus heterophylla, not an Apocynaceae, and the authors believe that nothing comparable, i.e., a fig wasp being fooled into visiting a completely unrelated flowering plant, has previously been documented. I have to concur, this relationship is (so far) unique.
The authors have neatly demonstrated that the deception of these wasps is not simply visual. The chemical bouquet produced by H. ficoides is closer to that of receptive Ficus heterophylla figs than it is to the flowers of its relative H. siamicum. More importantly, behavioural experiments showed that Kradibia wasps were significantly attracted to the scent of open H. ficoides flowers, though not to unopened flower buds. In other words, the plant seems to switch on its misleading olfactory advertisement at just the right time.
Once inside, however, the wasps discover that they have made a serious mistake. They wander around probing the inner surface, searching for egg-laying opportunities that simply aren’t there. Some die inside the flower, although more than 80% of observed individuals managed to get out again, and often went on to visit another Heterostemma flower.
In the process, pollen can be transferred between flowers, but as in other members of the Asclepiadoideae, Heterostemma does not dust its visitors with loose pollen grains. Instead, pollen is packaged into pollinia which become attached to the legs of the fig wasps.
So from the fig wasp’s perspective this is straightforward deception: the plant promises a breeding site that doesn’t exist, and the flowers get pollinated in return. But that is only half of the system.
The flies are not being fooled by the fake fig
The most important pollinator of H. ficoides appears to be a tiny phorid fly in the genus Megaselia, and its relationship with the plant is quite different. Female Megaselia enter the flowers and lay their eggs there. Mating may even take place on the flower beforehand. Their larvae then develop inside the fallen flowers, eventually completing their development in the decomposing floral tissue on the forest floor. For these flies the promise made by the flower is therefore an honest one: this really is a brood site.
This creates an ecological combination that is one of the most distinct examples of a “bimodal pollination system” that I’m aware of, in which the same flower operates simultaneously as a brood-site mutualism and as a brood-site deception, involving two completely distinct groups of insects.
We assembled usable pollination information for 567 species (a bit more than 10% of the family) and found an extraordinary diversity of interactions. About 73% of the species for which we had information were associated predominantly with a single broad pollinator group, whether bees and wasps, butterflies and moths, flies, beetles or birds. However, about 7% of the species had distinct bimodal pollination systems, such as bee + butterfly. Whether this is an unusually high proportion for such a family is currently unclear as there are few comparative statistics.
One of our broader conclusions from the study was that this interaction diversity was not randomly scattered across the Apocynaceae evolutionary tree. Pollination systems show strong phylogenetic structure and some transitions between pollinator types have occurred much more readily than others. The new Heterostemma study puts some wonderfully detailed natural history onto that much larger evolutionary picture.
Aroonrat and her colleagues suggest that Heterostemma ficoides may represent a shift away from the more open, nectar-producing flowers found in other parts of the genus. In section Oianthus, to which H. ficoides belongs, the flowers have become progressively more closed. In the extreme case represented here, the result is a globose chamber entered through a small opening, nectar production has apparently disappeared, and the plant has moved towards brood-site pollination by flies and deception of fig wasps.
Whether that evolutionary scenario survives a proper phylogeny and detailed pollination studies of the remaining species is an obvious question for the future. As the authors themselves acknowledge, both are currently lacking. But it is exactly the sort of transition that our 2019 study suggested we should be looking for.
And what does any of this have to do with eating Apocynaceae?
Quite a lot, in a roundabout way. Earlier this year my colleagues and I published another large synthesis of the family, this time asking a very different question: which Apocynaceae do people eat?
We documented 440 edible species, spread widely across the evolutionary tree of the family. That sounds like a lot, although statistically Apocynaceae are actually under-represented among the world’s food plants, probably in part because the family is so well endowed with toxic and pharmacologically active chemistry.
One point that emerged strongly from that work was that our knowledge of how those useful plants reproduce is remarkably incomplete. Many edible Apocynaceae are harvested for their fruits and seeds, which of course ultimately means that successful reproduction matters. Yet for most of those species we still know little or nothing about their pollinators.
That matters if we are serious about conserving wild food plants or developing some of them as future crops. You cannot fully understand the biology of a fruit-producing plant without understanding how the flowers get pollinated in the first place.
There are no Heterostemma species documented in our edible-plant study (though it appears that we missed one – see the comment below), so H. ficoides is not some promising new food species. But that isn’t my point. Rather, this new research is a reminder of just how much biological complexity can lie behind the production of an Apocynaceae seed. And we still know the pollination biology of only a fraction of this enormous family.
That, perhaps, is the broader lesson from all three papers. Large databases and family-wide syntheses are valuable because they reveal general patterns and, equally importantly, the gaps in our knowledge. But filling those gaps still depends on researchers spending time watching individual plants and individual insects, dissecting flowers, rearing larvae and asking exactly what the animals are doing. Sometimes what they find is stranger than anyone would have predicted.
Bedfordshire Natural History Society (BNHS) has invited me to give a talk on the theme of “Pollinator Conservation: Threats and Opportunities” at Maulden Village Hall (MK45 2DP) on Tuesday 15th September. It starts at 8pm and non-members are encouraged to attend.
Back in January I wrote a post entitled Pollination as a matter of national security, prompted by the UK Government’s rather extraordinary publication of this report: Global biodiversity loss, ecosystem collapse and national security: a national security assessment. My argument was that although pollination is obviously not a national security issue in the traditional sense of defending a country against hostile states or terrorist attacks, the loss of pollinators threatens food production, supply chains and the functioning of ecosystems upon which societies ultimately depend. In other words, it is a threat to national resilience, which we can define a country’s capacity to anticipate, withstand, adapt to, and recover from major shocks or long-term pressures while maintaining the essential functions of society, the economy, government, and critical infrastructure. I’ll come back to this national-level framing at the end.
I concluded that post by suggesting that the real question was not whether pollination could be regarded as a national security issue, but whether national security thinking had fully adapted to the biological foundations on which societies depend.
Since January, several reports, papers and policy developments have convinced me that this is becoming a much wider and more serious discussion. Pollination is only one component of it. What is emerging is a view of nature itself as part of the infrastructure that makes societies secure.
Nature as infrastructure
Perhaps the clearest expression of this idea comes from a paper by Bradley Cardinale, Emmett Duffy and Rod Schoonover entitled Nature’s role in national security, published this summer in the journal Nature-Based Solutions.
The authors consider five areas where ecological disruption can generate national security risks: food security, water scarcity, health security, protection from natural disasters, and environmental crime. In each case, environmental degradation can create or amplify economic and social pressures which, under some circumstances, translate into political instability and security problems.
What I particularly like about the paper is its use of the term “natural infrastructure”. Functioning ecosystems, the authors argue, operate rather like electricity grids, communications systems and transport networks. They provide services and, crucially, buffer societies against shocks.
This is an important shift in emphasis. We usually talk about the benefits or ecosystem services that nature provides. Both are perfectly useful concepts, but describing ecosystems as infrastructure changes how we think about their loss. A degraded wetland is then not simply a conservation problem. It may represent the deterioration of flood-control infrastructure. Similarly, a damaged watershed is a weakening of water-security infrastructure, whilst a decline in pollinators can be understood as damage to parts of our food-production infrastructure.
Seen in that way, restoring ecosystems begins to look less like discretionary environmental spending and more like preventive investment in resilience.
Rewilding as defence
An even more striking example appeared in April in The RUSI Journal, published by the Royal United Services Institute, one of the world’s oldest defence and security think tanks.
Sam Jelliman, Brian Schmidt and Alan Chandler propose the concept of “defensive rewilding”: strategically restoring forests, wetlands, peatlands and other ecosystems in locations where they could also impede, divert or slow an invading military force. This infographic from the paper outlines their argument quite persuasively:
At first sight this sounds rather eccentric, but the logic is straightforward. Military planners since at least the time of Alexander the Great have always understood that terrain matters. At the Battle of Issus in 333 BCE, for example, Alexander fought the much larger Persian army on a narrow strip of land between the sea and the mountains, restricting its ability to manoeuvre and helping to neutralise its numerical advantage. Wetlands, forests, rivers, hedgerows and boggy ground can also provide advantage by restricting visibility and movement and make the passage of vehicles and troops much more difficult. We have seen examples of this during the war in Ukraine, just as we did in numerous earlier conflicts.
What Jelliman and colleagues suggest is that this defensive function could sometimes be deliberately combined with ecological restoration. Rewetting peatlands or restoring floodplains near vulnerable borders might simultaneously enhance biodiversity, store carbon, reduce flooding and create terrain that is difficult for armoured vehicles to cross. A win-win for both national defence and this wider concept of national security.
Clearly this is a concept that requires much more testing, and ecological restoration should not be distorted into a military land-management programme. There would also be difficult questions about land ownership, farming and the ecological appropriateness of particular interventions. But that is almost beside the point. The fact that rewilding is now being seriously discussed in a defence journal as potential security infrastructure is itself significant.
It challenges the assumption that spending on defence and spending on nature necessarily compete with one another, though we’ve long known that some areas of the UK than are used by the military are also important wildlife sites, Salisbury Plain being an obvious example, the largest known expanse of unimproved chalk grassland in north-west Europe precisely because it is used for military purposes rather than being converted to agriculture.
Losing nature can make countries poorer and less financially secure
Another recent study extends the argument into the financial system. Matthew Agarwala and colleagues published a paper in Nature Ecology & Evolutionexamining how losses of ecosystem services might affect the creditworthiness of nations, which is the degree to which lenders and investors judge that a country’s government can reliably meet its debt obligations in full and on time. Their model incorporated scenarios involving the deterioration of three services – wild pollination, marine fisheries and tropical timber – into sovereign credit assessments for 23 countries containing about 5.5 billion people.
The numbers produced by their scenarios are striking. Under partial ecosystem collapse, additional annual sovereign interest payments across the countries studied could reach around US$162 billion. India and China alone could experience tens of billions of dollars in increased annual debt-servicing costs. Some countries could suffer very large reductions in GDP.
These are modelled scenarios rather than predictions, of course, and the ecological assumptions that sit beneath such economic modelling need careful scrutiny. But the basic mechanism is compelling.
Governments depend upon functioning economies to raise taxes, borrow money and respond to crises. If ecological degradation reduces agricultural output, fisheries production and other forms of economic activity, the fiscal position of a country deteriorates. Borrowing becomes more expensive precisely when governments may need to spend more money adapting to environmental change.
This summer’s record breaking drought and wildfires in the UK, and the promise of further disruption due to a mega El Nino event, make such adaptation all the more urgent. Nature loss can create a nasty feedback loop: ecological degradation weakens economies, weaker economies have less capacity to invest in resilience and restoration, and that in turn increases their vulnerability to further ecological degradation.
All of this moves biodiversity loss well beyond the conventional environmental policy box and there are positive signs that these ideas are becoming embedded in government.
The UK security machinery is beginning to respond
In June the UK Government established its first Climate Security Taskforce, bringing together experts from the security, military, academic and environmental communities. Significantly, its remit explicitly encompasses not only climate change but nature loss. This is important as many of us have been arguing for years that you cannot disentangle the two and they should not be treated as separate domains of environmental concern. The Government’s announcement stated that climate change and nature loss are drivers of instability, economic disruption and security risk, and linked the Taskforce to the broader approach set out in the UK’s National Security Strategy.
That matters because one of the perennial problems with biodiversity has been institutional fragmentation. Nature tends to be dealt with by environment ministries, while finance ministries think about economic stability, agriculture departments think about food production, health ministries think about disease, and defence and intelligence agencies think about security. But ecological risks do not respect departmental boundaries.
A collapsing ecosystem can simultaneously affect food prices, water availability, infectious disease risks, migration, trade, government finances and geopolitical stability. Treating each of these consequences separately misses the common ecological cause.
China, where I’ve been spending quite a bit of time in recent years, provides an interesting comparison with the UK. Whereas the explicit framing of biodiversity loss as a national-security risk is relatively new in Britain, China has for some years incorporated “ecological security” into its much broader concept of national security. Its first national-security white paper, published in 2025, explicitly listed ecological security alongside military, economic, food, resource, biological and other forms of security. This has been reinforced in 2026 by China’s new Five-Year Plans for ecological protection and the construction of a “Beautiful China”, both of which link biodiversity conservation, ecosystem restoration and resilience to national ecological security. Chinese researchers have meanwhile developed an extensive literature on “ecological security patterns”, identifying the habitats, corridors and ecosystem services required to maintain resilient landscapes. The Chinese use of the term is not identical to the emerging British idea of “nature security”, but the overlap is striking: both ultimately recognise that the security of a state rests partly upon the ecological systems that support it.
From an intelligence assessment to Parliament
The UK Government’s January assessment has continued to reverberate through Westminster. On 2 September, the House of Commons Library published a new briefing ahead of a parliamentary debate on the assessment. The briefing does not constitute a new security assessment, but usefully brings together the original findings with subsequent parliamentary scrutiny, stakeholder responses and government action, including the creation of the Climate Security Taskforce and a new three-year Nature Security R&D Programme.
The briefing rightly emphasises an important point about the original security assessment. It is based on a reasonable worst-case scenario. It is not predicting that these ecosystems will suddenly collapse tomorrow. National security planning routinely explores high-impact possibilities precisely because governments need to understand what would happen if severe events occurred.
On 3 September the subject was then debated in Westminster Hall, following a cross-party initiative led by Adrian Ramsay MP. Whatever ultimately comes from that debate, this represents an interesting political transition. In January, biodiversity loss appeared in a national security assessment, and eight months later it had become the subject of a dedicated parliamentary research briefing and debate. That does not mean that nature has suddenly moved to the centre of British security policy. Far from it. But it suggests that the language of nature security is beginning to acquire institutional traction.
And then there is food
As if to underline the point, today (4 September) the National Audit Office has published a report on the Resilience of the food supply chain to disruptions. This is not primarily a biodiversity report. It examines the resilience of the UK food system to a much wider range of threats, including extreme weather, energy disruption, cyber-attacks, pandemics and animal and plant diseases. But there is a direct connection with the nature-security debate.
Food is one of the UK’s 13 designated Critical National Infrastructure sectors. The NAO concludes that although the food system has proved resilient to previous disruptions, the ability of the sector to withstand catastrophic events has not been tested and that Defra’s actions have not yet been sufficient to ensure resilience to such events.
This brings us back to where I started in January with pollination. Food security is not simply about having enough farms. It depends on soils, water, climatic stability, pollinators, pest regulation, genetic diversity, functioning supply chains, energy, transport and international trade. Pull at enough of those threads simultaneously and resilience begins to unravel.
So where does this leave us?
What interests me most about these developments is that they come from very different intellectual and institutional directions. Ecologists are describing ecosystems as natural infrastructure. Economists are beginning to incorporate biodiversity into sovereign financial risk. Defence researchers are thinking about restored landscapes as strategic assets. Government has created a taskforce dealing with climate and nature security. Parliament is discussing ecosystem collapse as a national security problem. And the National Audit Office is warning that one of the country’s most fundamental pieces of critical infrastructure, its food system, needs greater resilience.
These are not separate conversations, they are pieces of the same emerging picture.
There is, of course, a danger in pushing the national-security framing too far. Nature deserves protection for many reasons that have nothing to do with its usefulness to states, economies or human societies. I would be very uncomfortable with biodiversity conservation being justified solely because it makes Britain, or any other country, more secure.
But that is not the argument I am making. Rather, I think we have systematically underestimated the extent to which national security is built upon ecological stability. Forests, soils, rivers, wetlands, oceans, pollinators and the millions of other species that make ecosystems function are not scenery surrounding the human economy, they are part of its physical foundations.
Earlier this year I argued that conserving pollinators could be regarded as a form of preventive security investment. The developments since then suggest that the same argument can be made much more broadly: Protecting and restoring nature is not an alternative to investing in national resilience. It is one of the ways in which we invest in national resilience.
That feels like an idea whose time has arrived.
There is, however, an obvious limitation to framing all of this in purely national terms. Modern states are embedded in a globalised economy in which food, energy, raw materials, finance, manufactured goods and ecological impacts cross borders continuously. A country may strengthen the resilience of its own landscapes and infrastructure yet remain highly vulnerable to ecosystem degradation, crop failures or political instability elsewhere. In that sense, “national resilience” is also dependent on international ecological resilience: security at home increasingly rests on the stability of environmental and economic systems far beyond national borders. Just as individual trees in a forest are dependent upon one another to create the conditions in which all can thrive, regardless of species, so the too are countries mutually dependent. The leaders of the world’s nations would do well to remember that.
That is not a complaint about the quantity of research being done. Quite the opposite: it is remarkable how much good work is now being published across the world. But the sheer volume creates a serious practical problem. How can any researcher keep track of what is relevant to their field, let alone read and properly absorb it?
For someone interested in pollination, the difficulty is compounded by the extraordinary breadth of the subject. Pollination is not confined to a single academic discipline, or even to a small cluster of them. It cuts across botany, zoology, ecology, evolution, conservation biology, agriculture, economics, food security, national security, palaeontology, biogeography, genetics, behaviour, climate science and many other areas.
A paper that changes how I think about pollination might appear in a specialist plant journal, an entomological publication, or journals covering conservation, agricultural economics, or palaeontology. It might concern the structure of ecological networks, the nutritional quality of crops, the evolution of flowers, pesticide regulation, the movements of migratory birds or the fossil record of insects. It may not even use the word “pollination” prominently in its title or abstract.
That breadth is one of the great attractions of a subject that has kept me fascinated for 35 years, and which I tried to capture in my book Pollinators & Pollination: Nature and Society. It is also what makes keeping up with it so challenging.
The limitations of conventional alerts
There are, of course, many ways to receive information about new research. Journals send tables of contents. Google Scholar provides alerts based on keywords or authors. ResearchGate regularly tells me that someone has cited one of my publications.
These services have their place, but they tend to provide a rather narrow window onto the literature.
For example, Google Scholar can alert researchers to newly indexed material matching a search query. This is useful, but it is neither a comprehensive record of everything published nor a carefully curated selection: broad searches produce noise, while narrow ones inevitably miss relevant work.
A citation alert from ResearchGate tells me about research connected to work that I have already published. That can be valuable, and occasionally flattering, but it inevitably looks backwards. It shows me the expanding wake of my own research rather than offering a broad view of where the subject is moving.
Keyword alerts have a different problem. They can generate large quantities of material with very little discrimination. A search for “pollination”, “pollinator” or “plant–pollinator interactions” will retrieve many relevant papers, but also conference notices, marginally related studies, duplicate records and work of highly variable importance.
More restrictive searches reduce the noise but risk excluding the unexpected paper that turns out to be especially interesting.
What I need is not simply a larger stream of titles. I want something closer to an informed research assistant: a system that could search widely, exercise some judgement, explain why particular items might matter, and alter its approach in response to my comments.
My current instruction is for ChatGPT to provide a shortlist every Friday morning of the most worthwhile new papers, preprints, reports and substantive analyses relating to plant–pollinator interactions, pollination networks, bird pollination and related broader biodiversity topics.
I have asked it to include no more than ten items, and fewer when the available material is weak. That final qualification is important: I do not need ten references merely to fill ten spaces, I’d rather receive four genuinely interesting papers than a padded list containing six that I will never read.
For each item, the system can provide the citation, a short account of the main finding and an explanation of why it may be relevant to my interests. It can also distinguish between peer-reviewed research, preprints, reports and other forms of analysis.
In that sense, the alert is already more useful than a conventional automated search. But the most important difference is that it is a conversation.
I can tell it that a particular paper was especially useful and ask it to look for more work of that kind. I can point out that another item was only marginally relevant. I can ask it to widen its search into palaeontology, ecological economics or agricultural policy, or to pay closer attention to a particular taxonomic group.
I can also tell it what not to do. When a major European pollinator-research white paper appeared, for example, the alert quite reasonably identified it as relevant. But I was one of its co-authors and did not need an artificial intelligence system to introduce it to me as a new discovery. I could therefore instruct the system to recognise my own publications and either omit them or flag them only when they were strategically relevant.
That adaptability is difficult to reproduce with conventional keyword alerts. The scheduled search is therefore not a fixed filter, it can be refined as my interests, projects and frustrations change.
From retrieval to assessment
The distinction between finding research and assessing it is also important.
A long list of newly published papers transfers the problem of selection from the search engine to the researcher. A useful alert should do more than retrieve documents. It should offer some preliminary judgement about novelty, relevance and significance.
Does a paper introduce a genuinely new idea, or does it repackage a familiar concept in new terminology? Is a striking conclusion supported by a strong study design? Does a paper matter because of its empirical results, its methods, its conceptual framework or its policy implications? Is it directly relevant to my work, or merely adjacent to it?
ChatGPT cannot answer such questions infallibly. Nor should its assessment be accepted without scrutiny. But it can help to triage the literature and identify which papers deserve closer attention.
This is particularly valuable outside one’s immediate specialism. I can usually make a rapid initial judgement about a field study of flower visitors or a paper on pollination networks. I may need more assistance in deciding whether a new economic analysis, remote-sensing method or palaeontological reconstruction is likely to be important.
The purpose is not to delegate scientific judgement, it is to direct that judgement more efficiently.
A few necessary cautions
There are obvious limitations.
An AI-generated research brief is only as good as the literature it can locate and the instructions it has been given. It may miss relevant work, misunderstand a paper, exaggerate novelty or place too much weight on a fashionable topic.
There is also a danger of creating an intellectual echo chamber. If I repeatedly tell the system what interests me, it may become increasingly good at confirming those interests while overlooking research that sits outside them. A useful alert needs some permission to surprise.
For that reason, I think the instructions should combine a clear core remit with an explicit request to include occasional unexpected items that could change how the field is viewed.
The summaries themselves must also remain starting points. Important papers need to be read, methods inspected and conclusions considered in relation to the evidence. An articulate paragraph generated by ChatGPT is not a substitute for engaging with the original study.
A better kind of alert
Used carefully, however, scheduled ChatGPT alerts offer something that conventional notification services do not: an iterative relationship between the researcher and the search process.
The system can learn that I am interested not only in pollination as an ecological interaction, but also in its agricultural, economic, evolutionary, historical and political dimensions. It can respond when I say that a particular area is becoming more important to a project. It can explain its selections, accept criticism and search differently the following week.
That makes it less like an electronic table of contents and more like a research assistant who arrives on Friday morning with a small pile of papers and a cup of coffee and says: “These are the ones I think you should look at, and this is why.”
Like any research assistant, it will sometimes be wrong. It may overlook something obvious, become overenthusiastic about a weak paper or misunderstand why a study matters. But it can be corrected. The conversation continues, and the next briefing can be better.
For a field as broad and rapidly expanding as pollination research, that may be the most useful feature of all.
I will finish with an example. Today’s alert arrived at the scheduled 08:00 (though I had to make my own coffee…) and included four items for me to look at; this was the fourth, which I’m reproducing verbatim:
The authors incorporate projected losses of wild pollination, tropical timber and marine fisheries into sovereign credit-risk models for 23 countries. Their scenarios suggest that partial ecosystem collapse could substantially increase national borrowing costs and expose a major omission in conventional credit ratings.
Why it matters: This is one of the clearer attempts to translate pollination loss from ecosystem-service valuation into a decision framework already used by governments and financial markets. For ecological-economics work, the methodological bridge—from spatial ecosystem-service scenarios through macroeconomic models to sovereign credit ratings—is more important than the headline monetary totals.
The ecological assumptions deserve scrutiny. “Wild pollination” is compressed into a tractable economic input, so uncertainty in crop dependence, substitution, adaptation and spatial mismatch may be obscured once propagated through the financial model.
Assessment:Strategically consequential interdisciplinary analysis, not new pollination biology.
The paper actually appeared on 4th June but I missed it, even though it cites this paper by Millard et al., in which I was involved, and presumably appeared on a ResearchGate citation alert last month. ChatGPT rightly spotted that it was something which would interest me and that’s what’s so fascinating about this approach to keeping up with the literature – this system is intelligent enough to “understand” my needs, even if it’s not conscious of precisely why it understands them.
As always, I’d be pleased to hear your views on this topic – feel free to comment below.
In my book Birds & Flowers, I included a chapter called “The curious case of Europe”. The point of that chapter was simple enough: compared with much of the rest of the world, Europe appears to be oddly deficient in bird pollination. There are no hummingbirds, no sunbirds, no honeyeaters, and very few native plants that obviously look as though they have evolved with birds as their main pollinators. And it Britain, it appears that bird pollination is totally absent.
But “appears” is doing a lot of work there.
For decades, European bird–flower interactions have tended to be treated as marginal curiosities: Blue Tits taking nectar from willow catkins, warblers dusted with pollen, finches messing about in blossom. Interesting natural history, certainly, but not necessarily pollination. The assumption has usually been that insects do the serious work, and birds are at best incidental visitors.
A new paper in Journal of Ecology called “Generalist passerine birds perform a functional role as pollinators in temperate Europe“, by Sandra Anderson, George Perry and Rose Thorogood challenges that assumption in a very useful way. Working at Wicken Fen in Cambridgeshire, they found that pollen transport by passerine birds was widespread. Most of the birds they sampled carried pollen, and several species — including Blue Tit, Blackcap, Chiffchaff, Wren, Redpoll and Bullfinch — regularly carried meaningful loads.
More importantly, they tested whether this mattered to the plants. By excluding birds from flowers while allowing insects access, they showed that fruit-set was reduced in several early-flowering woody plants, including Blackthorn (Prunus spinosa), Hawthorn (Crataegus monogyna) and Buckthorn (Rhamnus cathartica). In other words, the birds were not just getting dusty faces. They were contributing to plant reproduction.
I should say that I was one of the reviewers of this paper, so I have followed its development with particular interest. What I like about it is that it does not try to claim that Europe secretly has a hidden flora of classic “bird flowers”. These are not red tubular blossoms adapted to hummingbirds or sunbirds. They are familiar, open, pale, spring-flowering shrubs and trees. Nor are the birds specialised nectar-feeders. They are generalist passerines making use of seasonal resources.
That is precisely why the paper is interesting. It shifts the question from “does this look like bird pollination?” to “does bird visitation actually function as pollination?” That distinction matters. Pollination syndromes can be useful, but they can also blind us to interactions that do not fit the textbook categories.
The seasonal context is also important. These interactions peak early in spring, when willows, blackthorn and other woody plants are flowering, temperatures are still cool, insects may be unreliable, and birds are preparing to breed or arriving from migration. Under those conditions, even occasional bird visits could be valuable to plants needing pollen moved between individuals.
For me, this paper strengthens the argument I made in “The curious case of Europe”: Europe is not devoid of bird–flower interactions; rather, we have been looking for the wrong kind of bird pollination. Instead of obvious “ornithophilous” specialisation, we may have overlooked a more diffuse, opportunistic, generalist system involving common birds and common spring-flowering woody plants.
That may not be as spectacular as a hummingbird hovering at a tropical flower, but ecologically it is just as revealing.
NOTE FROM JEFF: This press release came out at the start of the week and I am involved in two of the EU-funded projects that are mentioned, Butterfly and ProPollSoil, and I’m a co-author of the White Paper.
For immediate release
University of Bergen Brussels office, Rue Guimard 1040, Brussels, Belgium, 22 June 2026
Eight EU-funded research consortia warn that Europe risks a crisis if it fails to halt pollinator loss. Their solution: a roadmap to reverse wild pollinator decline and protect managed bees.
A new White Paper from eight major EU-funded pollinator projects warns that the resilience of Europe’s vital societal functions and food security are at stake if the EU fails to halt and reverse wild pollinator declines, and to support managed pollinators. Behind the report is an interdisciplinary team of 135 leading researchers with expertise ranging from ecosystem ecology, pollinator ecology, ecological economics, social science, environmental history, behavioural psychology, political science, and environmental law. The report flags the EU’s siloed governance structures and resulting policy incoherence as the major barrier to pollinator restoration. It states that the EU and its Member States urgently need to act by making Pollinator Stewardship an explicit and measurable top-priority across policies on agriculture, environment, chemicals, research and innovation, trade, finance, planning, legislation, and education.
The report diagnoses the looming pollinator crisis as arising from a dysfunctional relationship between humans and nature. Seeing humans as separate from and superior to nature, and thinking of nature as an object for human use as a resource, reinforces institutional structures that exploit nature for short-term individual and material gain. This leads to unsustainable agricultural practices that risk jeopardising the resilience of the ecosystems on which humanity critically depends.
There is more at stake than food security, the report warns. Indeed, many of Europe’s economic supply chains and sectors depend on pollination of flowering plants. Think of medicinal plants, food supplements, biomass energy crops, biomaterials, textiles, fodder, cosmetics, decoration, art, culture, and tourism.
The report also highlights the low pollinator literacy of key societal actors whose daily actions can make the difference for pollinators. It advocates mandating ecoliteracy in the education of professionals in all key sectors that affect pollinators and their habitats.
According to the report’s lead author, Professor Jeroen van der Sluijs, many people whose actions affect pollinators and their habitat are already doing their best to help save the bees. But most lack the literacy to understand how their practices cascade into pollinator loss.
“Many farmers plant wildflower strips along their fields, but almost no one knows that some moths are more effective pollinators than honeybees. These little creatures of the night, clothed in velvet and moonlit dust, need host plants for their larvae, not only flowers. Host plants for pollinating hoverflies, beetles and moths are missing in most seed-mixtures for flower strips.”
Avoiding a scenario in which Europe is hit by a pollination crisis requires addressing the EU’s functioning and moving away from its siloed governance structures. This requires addressing fragmented responsibilities across sectors, top-down policy design, and weak coordination among administrations that currently hinders effective pollinator restoration. According to the authors, the conflict between short-term production goals and the need to maintain pollination services as a public good must be solved as soon as possible.
The report ends with a detailed roadmap of 15 urgent, evidence-informed recommendations for action that, when fully implemented, can reverse pollinator decline in Europe.
That is usually framed as failure: the experiment did not work, the results were inconclusive, the story was unclear. But science can also advance when we go back to old data and ask a different question. That is exactly what happened with a study of Field Scabious, Knautia arvensis, a familiar wildflower of meadows and grasslands.
Back in 2001, I carried out an experiment to test the function of the showy outer florets around the edge of the flower head. These enlarged “ray florets” look as though they ought to be important in attracting pollinators. The idea was simple: trim them in different ways and see whether pollinating insects responded – you can see the results of that trimming in the photo above.
The answer, at the time, was baffling. Bumblebees, hoverflies and butterflies (all of which are effective pollinators of this plant) did not seem to care very much. Visitation rates by the insects hardly changed, and neither did seed set. But seed weight did change, as did the amount of sugar being produced by the trimmed flower heads. At the time I could not make sense of this in relation to the question I had asked, so the data were archived and left alone.
But not forgotten.
Fast forward 25 years, and I was chatting about this data set with colleagues when I was at the Kunming Institute of Botany in China. That got me reading some more recent work about florivory -the damage done to flowers by animals – and I realised that this old experiment might make more sense if viewed from another angle. Perhaps I had not really been testing floral attraction at all. Perhaps, without intending to, I had carried out a simulated florivory experiment.
Seen in that light, the results became much more interesting.
Cutting the ray florets did not stop pollinators from visiting. Nor did it reduce the number of seeds produced. But it did change the plant’s internal economics. Nectar quality declined in the most heavily cut flowers, and the seeds that those flower heads produced were consistently lighter. In other words, the flowers still functioned well enough to get pollinated, but the plant appeared to invest less in rewards for pollinators and less in each of its offspring.
That matters because lighter seeds may have poorer prospects later in life, even if the plant initially appears to reproduce successfully. The damage did not cause total reproductive failure. Instead, it produced a subtler effect: hidden costs that only become visible when you look beyond simple seed counts.
Newly contexualised in this way, we wrote up this work, being completely honest about the history of the study, submitted it to the Journal of Pollination Ecology, where it received very positive reviews. It’s now been published and you can download a copy by following the link in the reference:
I like this study because it tells two stories at once. One is ecological: florivory may not always reduce pollinator visits, but it can still alter plant reproduction in potentially important ways. The other is about how science actually works. Not every good paper begins with a perfect hypothesis and a clean result. Sometimes the value lies in returning to an awkward, neglected dataset and realising that it was trying to tell you something different all along.
Science does not always go the way you plan. But occasionally that is when it becomes most revealing.
I have never published a paper that’s not been improved, to some degree, by peer review, and broadly the system works. But I do wonder if it’s sustainable in the long-term and whether in the future LLMs might actually be a more effective way of assessing manuscripts. I recognise that’s (currently) a controversial statement to make – but having recently run a few of my own manuscripts through ChatGPT and asked for its “opinion”, I can honestly say that the feedback has improved not just the writing but also the framing and focus of the work. It’s also picked up weaknesses and errors that I had otherwise missed.
That initiated an email conversation with Steve which resulted in me running a short experiment with ChatGPT model 5.5. I first loaded up the original manuscript that I’d submitted to a journal of this paper on pollinator effectiveness. I then asked ChatGPT to write a review of the manuscript as though it was a peer reviewer of the journal. Which it did – in some detail – in 28 seconds! If anyone is interested I can send them that review, but it’s the next bit that I think is especially interesting.
After ChatGPT had completed the review, I then uploaded the actual peer reviews I’d received from the journal, plus the editor’s comments, and asked it to summarise the degree to which its review agreed with those I had received.
Below I have copied the response, but it’s the take-home message, summarised in the graphic above, which I think is the most intriguing aspect: there was a 55% agreement with Reviewer 1, a 70% agreement with Reviewer 2, and a 65% overall agreement between ChatGPT’s assessment of the manuscript and that of the Editor.
That may not sound like a large overlap, but note that it’s the level of agreement between the two reviewers was only 65%! In my not inconsiderable experience (>130 peer reviewed papers and chapters, hundreds of peer reviews of my own) this is fairly typical – reviewers ALWAYS pick up on different strengths and weaknesses in a manuscript!
What particularly impressed me about ChatGPT’s response below, is that it went far beyond simply looking at the language being used, it provided a critique of how the study was conducted, how data were analysed, and the conclusions drawn. Which is just what we expect a human reviewer to do! Not only that, but it was more critical of the work than the reviewers and the editor.
So where does this leave us? Could Large Language Models (LLMs) like ChatGPT replace part of the academic peer-review process? Clearly there need to be guide rails and human agency involved in final decisions, but I think that, yes, they may well have a role to play in the not-too-distant future. As always, I’d be interested in your comments.
If you want to dive into this more deeply, here’s the verbatim account of what ChatGPT wrote (including its emboldened parts) when I asked how much agreement there was between its assessment and those of the other two reviewers and the editor:
There is a fair amount of overlap, but not a perfect match.
I would say my review was moderately similar in substance, but noticeably more critical in tone and in editorial recommendation. The strongest areas of convergence were on the paper’s core strengths and on one key statistical concern. The biggest mismatch is that the actual editor and reviewers were much more positive overall, and most of their requested changes are closer to clarification, presentation, and framing than to a deep challenge to the paper’s publishability.
The clearest points of agreement are these. First, I highlighted the manuscript’s strengths as the multi-year design, the focus on different functional groups, and the interesting role of Lepidoptera; that aligns very closely with the guest editor, who explicitly says that “the roles of lepidopterans, the inclusion of flight distance measures, and the multi-year study are particularly notable,” and with Reviewer 2, who lists the multi-year field study, multiple pollinator groups, and the unexpected importance of Lepidoptera as strengths. Second, I flagged the Poisson treatment of an ordinal flight-distance variable as statistically weak; Reviewer 1 independently makes essentially the same point and explicitly recommends ordinal regression instead. Third, I thought the discussion was somewhat over-extended and not always tightly aligned with the evidence; Reviewer 2 likewise suggests embedding the work better in the broader literature and moving some material out of the conclusion into the discussion.
There is also a partial overlap around the manuscript’s central interpretation. I argued that the paper sometimes pushes too hard from pollen deposition and coarse movement classes to larger claims about true generalisation, outcrossing value, and broader evolutionary or environmental buffering. The editor does not object to those broad claims in the same way; instead, the editor actually asks for the discussion to connect the results more explicitly to the theoretical statement that plants should generalise on equally effective pollinators under dynamic environments, and even invites discussion of anthropogenic disturbance and future interactions. In other words, where I wanted that interpretive arc toned down, the editor mostly wanted it better developed and better supported.
The main differences are quite important.
First, I was more sceptical about inference than the real reviewers were. I was worried that the manuscript moved too quickly from non-significant or coarse flight-distance patterns to statements about butterfly superiority, reduced geitonogamy, and broader adaptive significance. Reviewer 1 only objected narrowly to the modelling choice; neither reviewer makes my broader inferential critique, and the editor seems broadly comfortable with the interpretive direction so long as it is supported with more descriptive data and clearer exposition.
Second, I was more concerned about the coarse pooling of visitor taxa. I thought the manuscript should be more cautious because “butterflies,” “hoverflies,” “bumblebees,” and especially “Others” are broad bins. The actual comments do touch this, but in a softer, more taxonomic way: the editor asks whether there is a species or morphotype list, and Reviewer 2 asks for more detail on the butterfly taxa and justification for putting solitary bees into “Others.” That is related to my concern, but it is narrower and more presentational than the conceptual objection I raised.
Third, I did not anticipate how much of the real feedback would focus on descriptive content and figures. The editor repeatedly asks for more floral biology data, clearer results presentation, more absolute numbers, more behavioural quantification, clearer site description, and better figure design and consistency. Reviewer 2 similarly asks for effect sizes, better figures, units, and clearer display of the underlying data distribution. I mentioned presentational issues only in a general way; the actual reports are much more detailed and concrete here.
Fourth, I did not raise the absence of random effects, whereas Reviewer 2 does. That is probably the single most important substantive point that I missed. Reviewer 2 specifically asks why insect individual and/or observation day were not modelled as random effects.
Fifth, my editorial recommendation was harsher. I recommended major revision. The actual reports read much more like minor-to-moderate revision. The editor calls it “an excellent piece of work” and says it is well executed with robust methodology; Reviewer 2 says they have “little to criticise,” mostly editorial suggestions plus a few statistical and figure issues; Reviewer 1 has only a short list of points. That collective tone is clearly more encouraging than mine.
So, in condensed form:
Where I matched well: multi-year dataset as a major strength; Lepidoptera result as a notable contribution; the Poisson-for-ordinal issue; some concern that discussion/framing needed tightening.
Where I was too severe: I treated the interpretive overreach as a major obstacle, whereas the real reviewers mostly saw a strong paper needing clarification and enrichment rather than retrenchment.
What I underplayed or missed: requests for more descriptive natural history and floral biology data, much more detailed figure/presentation criticism, and the random-effects issue.
My bottom-line judgement is that my review was directionally right on some key scientific points, especially the statistical issue and the general strengths, but too pessimistic about the paper’s editorial position. If I had seen these reports first, I would have shifted from “major revision because the evidence-conclusion fit is shaky” to something more like “moderate revision: good paper, but it needs clearer descriptive support, better figure presentation, and tighter statistical justification. If revised along those lines, I think the manuscript could become a credible and interesting contribution”.
Here is a concise version of the editorial decision in journal-review language:
This manuscript addresses an interesting and worthwhile question using a useful multi-year dataset and a stronger-than-usual measure of pollinator performance. However, the current version over-interprets some of the results, particularly those concerning movement behaviour, outcrossing implications, and the broader adaptive significance of generalised pollination. The statistical treatment of flight distance is also not fully convincing. I therefore recommend major revision. The manuscript has clear potential, but its conclusions need to be more tightly aligned with what the data actually demonstrate.
One of the few positive things to come out of the COVID lockdowns was the unexpected opportunity to look much more closely at the nature right outside our doors. In 2020 I coordinated the Lockdown Gardens initiative, bringing together pollination ecologists from around the world to carry out standardised surveys of flower visitors in the gardens they could access during that strange and constrained period. That project generated an unusually rich global dataset: 67 gardens, almost 47,000 flower visits, and records from more than 650 pollinator species.
Since then, the dataset has started to yield some really interesting insights. A new paper led by Luis Perugini uses the Lockdown Gardens data to ask a deceptively simple question: what determines how specialised plant–pollinator interactions are in gardens? In other words, are garden flowers in some places visited by a narrow set of pollinators, while elsewhere they are more generalist?
The answers are not quite what we expected. Looking at 40 garden networks from four continents, we found that larger gardens support more plant species, and that suburban gardens tended to be richer in plant species than either rural or urban gardens. We also found that pollinator richness increased with plant richness and with precipitation. But when it came to the actual specialisation of interactions, climate and species richness did not seem to matter very much at all. Instead, variation in specialisation was mostly species-specific and showed no clear phylogenetic pattern.
That’s an important result, because it suggests that the factors that drive biodiversity in gardens are not necessarily the same factors that shape the ecological relationships within those gardens. Put more simply: having more species does not automatically mean having more specialised interactions.
For me, this is a nice example of how a project born out of a global crisis can continue to produce useful science. The Lockdown Gardens surveys began as an improvised response to an extraordinary moment, but the data are now helping us to understand how gardens function as ecological systems, and how they might better support pollinators in an increasingly human-dominated world.
Here’s the full reference with a link to the paper, which is open access:
