A common pesticide could hinder bumblebees’ ability to pollinate plants, says a new study — and that could be a big problem for both agriculture and the natural ecosystems that depend on the bees for survival.

A paper published Wednesday in the journal Nature tested the effects of a pesticide called thiamethoxam on the ability of bumblebees to pollinate apple trees. The study suggested that, at certain levels, the pesticide can have negative impacts on the bees’ pollination abilities, causing colonies to visit fewer flowers and return with less pollen, and resulting in apples with fewer seeds. These results could indicate poor fruit quality and a risk of decreased agricultural output down the line, fueling the fire in an ongoing debate over how certain pesticides affect the environment and whether their continued use should be permitted in the U.S.

The researchers used three groups of bees in their experiment. They exposed the first two groups to two different “field-realistic” levels of the pesticide, or amounts that are actually used by farmers to protect their crops — 2.4 parts per billion and 10 parts per billion. They didn’t expose the third group to any pesticides at all. Colonies exposed to the 10 parts per billion experienced the strongest effects.

“One of the important things about our work is that it highlights the importance of pollination services,” said lead author Dara Stanley, who was a researcher at Royal Holloway University of London at the time the study was conducted, and will soon be starting a position at the National University of Ireland Galway. According to Stanley, this paper is among the first studies to examine the way pesticides affect bumblebee pollination services, and not just bumblebee biology or physiology.

“Most of the studies in the past have focused on direct effects on the bees, both the adults and the larvae,” said Robbin Thorp, professor emeritus and bee expert at the University of California Davis, who was not involved in this paper. These can include effects on bee mortality or reproduction.

“This study now clearly demonstrates that in addition to effects on the bees, both direct effects and sublethal indirect effects, that these effects are influencing their ability to pollinate plants,” Thorp said. “And they used apple as an example of this, as an important crop.”

Thiamethoxam belongs to a class of pesticides known as neonicotinoids, which are common in U.S. agriculture. Their use has been widely debated in recent years, as research continues to emerge suggesting that the chemicals may have negative effects on pollinators, including bees — even as critics continue to dispute the findings. In the light of such research, the European Union restricted the use of a handful of neonicotinoid pesticides in recent years, including thiamethoxam. However, its use is still permitted in the U.S.

Thiamethoxam was developed by agribusiness Syngenta and is an active ingredient in several insecticide brands. The company says the insecticide has gone through myriad safety tests and that “scientific evidence clearly shows that bees and other pollinators can coexist safely with neonicotinoids.”

Peter Campbell, senior environmental risk assessor at Syngenta, further criticized the new study’s conclusions in a statement to the UK Science Media Centre, saying, “The conclusion reached in this study that thiamethoxam impairs pollination services provided by bumble bees to apple trees is not conclusive, it is premature, and only representative of a single experiment conducted under artificial conditions for the apple trees being pollinated and using unrealistically exposed bumble bees.” Campbell also made a number of other technical criticisms of the study. 

For one thing, he highlighted the fact that certain other aspects of the bees’ pollination services were not affected by the treatment. Although treated colonies were less active pollinators and were associated with fruits that contained fewer seeds, they did not seem to make a significant difference in whether apple trees produced fruit or how many fruits they produced.

But the authors maintain in the paper that it’s unclear whether pesticide-exposed colonies’ altered behavior may affect other plants differently. And they also noted that the reduced seed production observed in this study could be a serious problem for agriculture, indicating a lower fruit crop quality with the potential to reduce agricultural output down the road.

So far, most studies on neonicotinoids and pesticides have focused on honeybees, said Thorp. This is largely a reaction to widespread honeybees losses to colony collapse disorder, which causes colonies to suddenly abandon their hives. The phenomenon is still poorly understood, but is thought to be the result of a complex combination of factors, including habitat degradation, disease — and pesticides.  

But it’s important not to forget about bumblebees, Thorp cautions. While honeybees are generally considered a managed species in the U.S., “wild bees are kind of the under-appreciated resource that we have out there,” he said. “They’re extremely important in pollination of our native ecosystems, and many of them…are important contributors to crop pollination.”  

Furthermore, he added, if honeybees continue to decline, other pollinators — including the bumblebee — will become increasingly important to the survival of plants that rely on pollination for their reproduction.

So the study has some worrying implications for a valuable wild pollinator. But Stanley cautions that the study “only looked at one pollinator species and one crop species,” so it’s impossible to say for sure that the effects would be the same with other bees or other plants — or other pesticides, for that matter.

The study also doesn’t provide a clear explanation for why, exactly, thiamethoxam caused the bees’ pollination services to suffer. In a mysterious twist, the researchers found that when bumblebees were released into the field one by one — not as a whole colony — the bees exposed to pesticides were actually more active than the control group, visiting more flowers, spending longer periods of time foraging and switching between apple trees more frequently.

“If you’re having bees visiting more flowers and [being] more active, you would presume that they’re probably going to deliver better pollination services,” Stanley said. It was only when whole colonies were released at once that the bees’ collective pollination services suffered.

“When we opened up the colonies, less bees were actually coming out of the colony,” Stanley said. “The colony overall was less active in sending out foragers.” This phenomenon raises questions about the effects of pesticides on individual bee behavior versus their effects on colony behavior as a whole — and the reasons for the discrepancy are not completely clear. The study opens the door for future research on the mechanisms that actually affect bee behavior, Thorp said.

Even if the study can’t be generalized to all bees or all crops, it raises more questions in the ongoing debate over pesticide use in the U.S.

“I think it’s kind of a wake-up call to growers that they ought to be paying more attention to what they’re putting on their crops,” Thorp said. “Because it’s coming right out of their pocket as well if they’re damaging the ability of pollinators that they rely on to pollinate their crops.”