Neonicotinoids and Bees: The Science and the Ban

In 1970, a chemist at Purdue University named Henry Feuer synthesized a compound called nithiazine. It killed insects by binding to nicotinic acetylcholine receptors in their nervous systems - the same mechanism as nicotine, the compound tobacco plants evolved specifically to kill insects. Nithiazine worked beautifully indoors. Outdoors, it disintegrated in sunlight. Commercially useless.

Fifteen years later, Shinzo Kagabu, working at Nihon Tokushu Noyaku Seizo in collaboration with Bayer, replaced nithiazine's chemical framework with an imidazolidine ring. The result - imidacloprid - was stable in sunlight, water-soluble enough to be absorbed by plant roots, and lethal to insects at concentrations measured in parts per billion. Bayer brought it to market in 1991. By 1999, it was the most widely used insecticide on Earth. Kagabu received the American Chemical Society's International Award for Research in Agrochemicals in 2010.

The neonicotinoid class - imidacloprid, clothianidin, thiamethoxam, acetamiprid, thiacloprid, dinotefuran, nitenpyram - now represents a $5.3 billion global market. The compounds share a defining characteristic: they are systemic. A seed coated with neonicotinoid grows into a plant where the insecticide is present in every tissue - roots, stems, leaves, pollen, and nectar. Any insect that feeds on any part of the plant ingests the compound. This includes the insects the chemical was designed to kill. It also includes the ones it wasn't.

The Dose That Doesn't Kill

The lethal dose of imidacloprid for a honey bee is 13 nanograms - 13 billionths of a gram. Clothianidin is worse: 3.68 nanograms. Both are classified as "highly toxic" to honey bees. But the lethal dose turned out to be the wrong number to focus on.

Bees foraging on treated crops encounter concentrations of 1 to 50 parts per billion in pollen and 1 to 8 parts per billion in nectar. These concentrations don't kill bees outright. They do something that, from a colony perspective, may be worse.

In March 2012, two papers landed in Science on the same day. Mickael Henry's team at INRA in Avignon tagged free-ranging honey bees with RFID microchips - the same technology used in pet identification - dosed them with sublethal thiamethoxam, and tracked their return to the hive. Treated bees were two to three times more likely to die while away from their nests. They weren't poisoned. They were lost. The navigation system that allows a bee to find a food source three miles away and return to one specific box in an apiary of hundreds was disrupted by a dose that, on paper, was too small to cause harm.

The same day, Penelope Whitehorn at the University of Stirling published data on bumblebees exposed to field-realistic imidacloprid levels. The colonies showed significantly reduced growth. The number of new queens produced - the only metric that matters for bumblebee population persistence - dropped by 85 percent.

Those two papers changed the conversation. The question stopped being "do neonicotinoids kill bees?" and became "what do sublethal doses do to colonies over time?"

The Sublethal Catalog

The answer, accumulated over the following decade, is: almost everything.

Learning and memory. Neonicotinoids bind to nicotinic acetylcholine receptors in the bee brain, causing overstimulation and oxidative stress. Exposed bees show reduced memory formation, altered waggle dance communication, and impaired foraging efficiency. The receptor they're disrupting is the same one responsible for learning and spatial navigation.

Immune suppression. Clothianidin exposure significantly decreases hemocyte proliferation - reducing the number of immune cells available to fight microbial invaders. Sublethal imidacloprid doses make honey bees more susceptible to Nosema, the gut parasite that already causes significant colony losses. The pesticide doesn't cause the disease. It removes the defense.

Queen fertility. Field-realistic neonicotinoid exposure during development severely compromises queen reproductive anatomy. Spermathecal-stored sperm quality drops. The number of queens that survive and successfully produce worker offspring declines by 34 percent. Ovariole hyperplasia - abnormal ovary development - appears in exposed queens. A queen that can't lay fertilized eggs effectively is a colony on a countdown.

Thermoregulation. All three major neonicotinoids - clothianidin, imidacloprid, thiamethoxam - decrease thermotolerance by more than 3 degrees Celsius. Chronic 16-week exposure leaves colonies unable to maintain thermal stability in cold temperatures, resulting in delayed brood development and lower pupation success. The mechanism: neonicotinoids suppress oxidative phosphorylation and lower ATP synthesis in mitochondria. The bees' furnace runs low on fuel. In the context of a winter cluster trying to maintain 93 degrees Fahrenheit in the center of the brood nest, 3 degrees is the margin between alive and dead.

In 2017, Ben Woodcock and colleagues published the largest field-scale assessment of neonicotinoid impacts in Science. Three countries - Hungary, Germany, the UK - three bee species, clothianidin and thiamethoxam on winter-sown oilseed rape versus untreated controls. In Hungary, negative effects persisted over winter: colonies were 24 percent smaller the following spring. Wild bee reproduction was negatively correlated with neonicotinoid residues across all sites.

The evidence was building a picture of a class of chemicals that didn't need to kill bees to destroy colonies. It just needed to make them slightly worse at everything - navigating, fighting disease, reproducing, staying warm - and let time do the rest.

95 Percent

The number that reframes the entire neonicotinoid debate is this: approximately 5 percent of the active ingredient in a treated seed is actually taken up by the crop plant. The remaining 95 percent disperses into soil, water, and the surrounding environment.

This isn't a design flaw. It's inherent to how systemic seed treatments work. The seed is coated. It's planted. Soil moisture dissolves the coating. The roots absorb some. The rest stays in the soil - where imidacloprid has a half-life of 1 to 3 years under aerobic conditions. Clothianidin's half-life ranges from 148 days to 7,000 days. Nineteen years. The chemical persists in fields that haven't been treated for over two years. A French survey found detectable imidacloprid in 62 of 67 soil samples, including fields where the compound hadn't been applied for two consecutive seasons.

Because neonicotinoids are water-soluble - that was the whole point, to let them travel through plant vascular systems - they leach into waterways. The USGS national reconnaissance found at least one neonicotinoid in 53 percent of US stream samples. In Midwestern streams near corn and soybean country, clothianidin showed up in 75 percent of samples. In coastal California, average neonicotinoid concentrations in streams were seven times greater than EPA ecosystem health guidelines.

The scale of application is staggering. At peak usage, approximately 95 percent of US corn acreage was planted with neonicotinoid-coated seeds - roughly 90 million acres. Seventy-six percent of soybean acres. Nearly all cotton. Add canola at 95 percent. The total: at least 150 million acres of US cropland, treated with a systemic insecticide where 95 percent of the active ingredient enters the surrounding ecosystem.

A 2019 study in Scientific Reports tested the economic premise underlying all of this and found that neonicotinoid seed treatments on soybeans provide "negligible benefits to US farmers." The compound that's in 53 percent of American streams doesn't measurably improve soybean yields.

The Regulatory Fracture

The EU acted first. In April 2013, 15 of 27 member states voted to restrict three neonicotinoids - imidacloprid, clothianidin, thiamethoxam - on crops attractive to bees, effective December 2013. In February 2018, the European Food Safety Authority concluded neonicotinoids posed a "high risk" to both domestic and wild bees. Two months later, EU member states agreed on a total outdoor ban. Only greenhouse use permitted. The ban took effect at the end of 2018.

France went further. On September 1, 2018, France became the first country to ban all five neonicotinoid compounds - including acetamiprid and thiacloprid, which the EU ban hadn't covered.

Bayer and Syngenta sued. In May 2018, the EU General Court threw out the challenge, ruling the European Commission had correctly applied its precautionary principle - giving "precedence to the requirements relating to the protection of public health, safety and the environment over economic interest." In 2021, Bayer lost its final appeal in the European Court of Justice.

Canada's Pest Management Regulatory Agency proposed phasing out the three main neonicotinoids by 2023. Then, in 2021, the PMRA reversed its decision. All three remain approved for agricultural use. A December 2024 investigation by Canada's National Observer found that Bayer "colluded with environmental and public health regulators in Canada to obstruct" the proposed ban.

The United States has not banned any neonicotinoid for agricultural use. The EPA finalized biological evaluations in 2023 showing the three main neonicotinoids are likely to adversely affect 67 to 79 percent of listed species. Draft risk assessments were released in July 2024. Proposed interim decisions with pollinator-protective measures were anticipated in 2025. The measures so far: label changes, application timing restrictions, cancellation of spray uses of imidacloprid on residential turf. The agricultural seed treatment market - the primary delivery mechanism, the one responsible for 150 million treated acres - remains untouched.

The same data. The same studies. The same bees dying in the same ways. Three regulatory outcomes: total outdoor ban, proposed-then-reversed ban, and no ban. The science isn't ambiguous. The politics are.

The CCD Question

Colony Collapse Disorder was first widely reported in 2006-2007: worker bees abandoning the hive, leaving the queen and brood behind. The neonicotinoid connection seems obvious in retrospect - a chemical that impairs navigation, suppresses immune function, and reduces colony thermoregulation could plausibly produce exactly those symptoms.

But France banned imidacloprid in 1999 and still experienced CCD-like losses. Many collapsing apiaries showed no trace of neonicotinoids. The Harvard studies by Chensheng Lu replicated CCD symptoms by feeding colonies low doses of imidacloprid and clothianidin - but the studies were criticized for small sample sizes and dosing methodology.

Scientific consensus has settled on a multiple-stressor model: neonicotinoids, Varroa destructor, Nosema, viruses, poor nutrition, habitat loss. No single factor alone. The colony losses of 55.1 percent in 2023-24 - 14.8 percentage points above the 13-year average, the highest since 2010-11 - aren't attributable to one cause. They're attributable to an environment where every stressor amplifies every other stressor, and neonicotinoids are part of that environment.

Dennis vanEngelsdorp at the University of Maryland, who co-developed the Bee Informed Partnership, found that "fungicides, which we didn't expect to harm insects, seem to have a sub-lethal effect on bee health." His work with colleagues showed sublethal imidacloprid makes bees more susceptible to Nosema. The compound doesn't need to be the cause. It just needs to be the amplifier.

The Replacement Problem

What replaced neonicotinoids in the EU? Mostly pyrethroids - older-generation insecticides used in 89 percent of cases. Pyrethroids have their own toxicity profiles and environmental concerns. Newer alternatives exist - chlorantraniliprole, cyantraniliprole, spinosad - and performed as well as neonicotinoids in trials. In 78 percent of cases, at least one non-chemical alternative could replace neonicotinoids: microorganisms, semiochemicals, surface coatings.

Then there's sulfoxaflor - marketed by Dow/Corteva as a neonicotinoid replacement. Like neonicotinoids, it acts as a competitive modulator of nicotinic acetylcholine receptors. Research in 2026 shows it "cannot be deemed a safe alternative," with diverse sublethal impacts on honey bees including metabolic, morphological, behavioral, and gene expression alterations. PAN Europe argues sulfoxaflor is functionally a neonicotinoid and should be classified as such. Both sulfoxaflor and flupyradifurone are up for EU approval renewal in 2026.

The replacement problem reveals the deeper issue. Neonicotinoids weren't adopted because they were the only option. They were adopted because they were convenient - coat the seed, plant the seed, forget about it. The 95 percent waste rate was invisible to farmers. The cost was externalized to waterways, soil organisms, and pollinators that contribute $18.9 billion annually to US agriculture - roughly four times the entire neonicotinoid market.

Dave Goulson's Numbers

Dave Goulson, professor at the University of Sussex, published two landmark reviews that synthesized the environmental case against neonicotinoids. The 2013 paper in the Journal of Applied Ecology established that current use was "likely impacting a broad range of non-target taxa including pollinators and soil/aquatic invertebrates." The 2018 review in Environmental Science and Pollution Research updated the evidence: environmental persistence and accumulation confirmed, water contamination widespread, non-target impacts ongoing.

Goulson's work made the case that neonicotinoids aren't just a bee problem. They're an ecosystem problem. When 95 percent of the applied compound enters soil and water, and the soil half-life is measured in years, and the chemical is water-soluble and leaches into every stream in agricultural country, the question stops being about bees and becomes about what happens to the base of the food web when the soil and water are chronically contaminated with a neurotoxin.

The bees are the indicator species. They're the ones we count. But the aquatic invertebrates, the soil organisms, the non-managed pollinators - the things nobody is tagging with RFID chips - are in the same water and the same soil.

150 Million Acres

More recent estimates suggest some decline from peak usage: approximately 58 percent of corn and 32 percent of soybean acreage used neonicotinoid-coated seeds in 2023, down from peak levels. Whether that represents a trend or a fluctuation depends on which market analyst you ask and which year you use as the baseline.

The $5.3 billion market continues. The 150 million treated acres (at peak) leave a chemical signature in soil that persists for years. The compounds are in 53 percent of US streams. The beekeepers lose 40 percent of their colonies annually on average, 55 percent in bad years. The EU banned the compounds outdoors. France banned all of them. Canada tried to ban them, then didn't. The United States reviewed them, evaluated them, studied them, drafted assessments about them, and treated no one's label change as an emergency.

Imidacloprid was synthesized in 1985. It entered the market in 1991. The first two Science papers documenting sublethal harm were published in 2012. The EU banned outdoor use in 2018. In 2026, the compounds are still on 58 percent of US corn seed and in 53 percent of US streams, and the EPA's proposed interim decisions are still proposed and still interim.

Thirty-four years from synthesis to the present. Twenty-one years since the first major alarm. The molecule that replaced nicotine - because nicotine wasn't stable enough to survive sunlight - turned out to survive everything else a little too well.