Correction: An earlier version of this article incorrectly said that the Center for Food Safety has sued the Environmental Protection Agency over its handling of clothianidin. The group has filed a petition with the EPA asking that the pesticide be banned but has not filed suit. This version has been updated.

New research has begun to unravel the mystery of why bees are disappearing in alarming numbers worldwide: Some of the pesticides most commonly used by farmers appear to be changing bee behavior in small but fatal ways.

Two new studies found that honeybees and bumblebees had trouble foraging for food and returning with it to their hives after exposure to the class of insecticides, which is widely used to protect grains, cotton, beans, vegetables and many other crops.

Ironically, the relatively new pesticides have been welcomed as an environmental plus because they are, by almost all accounts, less harmful to other wildlife than previous generations of pesticides.

Although the authors of the studies published Thursday in the journal Science do not conclude that the pesticides — called neonicotinoids — are the sole cause of the American and international decline in bees or the more immediate and worrisome phenomenon known as colony collapse disorder, they say that the omnipresent chemicals have a clearly harmful effect on beehives.

“People have asked me to sign petitions to ban or limit the use of the neonicotinoids for some time, but I never did because I really didn’t know if they were having a major impact on the bees,” said David Goulson of the University of Stirling in Britain, co-author of one of the studies.

“After seeing what we and the others found, I’m much more inclined to sign.”

Unlike older pesticides, the neonicotinoid pesticides are most often introduced directly into the seeds of crops planted by farmers and thus permeate the entire plant as it grows — including the pollen and nectar the bees feed on.

Spraying of the older pesticides could be halted when plants were flowering so bees and other pollinators would not be harmed. With the neonicotinoids, which kill pest insects by attacking their central nervous systems and are derived from the same nicotine found in tobacco, that kind of timing is not possible.

Failure to thrive

The initial changes in bee behavior found by the researchers may have been small, but the longer-term impact was large: Researchers found a sharp drop in the number of queen bumblebees produced, a decrease in the size and weight of beehives, and a demonstrated increase in the number of bees unable to find their way home. When the hives as a whole don’t thrive, then the individuals become more susceptible to disease and other threats.

The subject became a major focus of agricultural and environmental attention in 2006 when beekeepers reported massive losses in their beehives, an escalation of a decline seen for years.

The new research — one study of honeybees and the other of bumblebees — points to flaws in the way pesticides are evaluated by regulators, said the author of the study of a kind of honeybee widely used as a pollinator in agriculture.

“So far, they mostly require manufacturers to ensure that doses encountered on the field do not kill bees, but they basically ignore the consequences of doses that do not kill them but may cause behavioral difficulties,” said Mickael Henry of the French National Institute for Agricultural Research (INRA) in Avignon, France.

Environmental toxicologist David Fisher of Bayer CropScience, which makes some of the neonicotinoids, said the studies appear to be well done, but are inconsistent with earlier results.

Pesticides were an early suspect in the decline, but many other factors have been implicated as well — including a relatively new invasive mite that kills bees in their hives, the loss of open land for foraging, and the stresses on honeybee colonies caused by moving them from site to site for agricultural pollination.

“We know that these agents can kill bees at high dosages, but previous studies did not show that effect at the low doses found in fields,” Fisher said. “We have a definite problem with the health of our bee populations, but we don’t believe the research has shown neonicotinoid pesticides to be the reason why.”

He said the honeybee study in particular was problematic because it exposed the bees to concentrations of pesticide not found in farm environments.

‘Dramatic’ results

The research leader for bee issues at the USDA’s Agriculture Research Service in Beltsville, however, found the bumblebee study to be convincing and possibly able to alter the debate. Jeffrey Pettis said that by showing that the hive’s production of queens was significantly impaired — declining by 85 percent — by the kind of low dosages of neonicotinoids found in field environments, the authors provided important new information.

“The results were so dramatic that you just had to take notice,” Pettis said. “I think that line of research will now drive a lot more of the discussion.”

Bumblebee study co-author Gaulson also said the result pointed to the need for action.

“Bumblebees in the wild pollinate crops such as tomatoes, beans and cucumbers, as well as wildflowers,” he said. “This study shows that the use of neonicotinoid pesticides on flowering crops clearly poses a threat to their bumblebee health, and urgently needs to be reevaluated.”

Even before the release of Thursday’s papers, worry over the decline of the nation’s bee colonies had been high, with beekeepers and some environmentalists likening today’s dwindling populations to the disappearance of songbirds described in Rachel Carson’s 1962 book “Silent Spring.” In both cases, pesticide “collateral damage” to creatures doing no harm to crops galvanized opposition to the products, first DDT and now neonicotinoids (although the scientific case against the second is at this point less certain.)

Reflecting the heightened concern, beekeepers working with the Center for Food Safety filed a petition with the EPA in March over the agency’s handling of a particular drug in that class, clothianidin, asking that the pesticide be banned.

In the field

The new studies looked at the field experience of honeybees (most used in controlled pollination) and bumblebees (more often wild pollinators) after exposure to neonicotinoids. The honeybee research involved a novel technique to track the bees: Researchers in France glued tiny radio-frequency microchips to the insects’ thoraxes before the bees ventured out to feed.

Some beehives were exposed to the insecticide thiamethoxam (from the neonicotinoid class) and some were not. Those bees exposed to the pesticide were two to three times more likely to not return to their hive than the unexposed control bees. According to researcher Henry of the French agricultural research institute, the bees appear to get disoriented and lost.

In the bumblebee study, the bees were exposed to imidacloprid, the most commonly used neonicotionid, at levels comparable to what bees encounter in the wild. The colonies were placed in an enclosed field where bees fed under natural conditions for six weeks. The researchers weighed each bumblebee nest before and after the six weeks, and found that they were 8 to 12 percent smaller than the control nests at the end.

Researcher Goulson said that most queen bumblebees are produced in the largest nests, and that the sharp decline in queens may well have been caused by the sluggish growth of the pesticide-exposed nests as a whole.

Bee toxicity specialist May Berenbaum of the University of Illinois, who chaired a National Academy of Sciences study group on bee declines, said the studies were important in establishing that the neonicotinoids are affecting wild bees as well as those used in controlled pollinations.

“It’s easy to see the decline in honeybees kept for pollination, but much harder to see it in the wild bees with their nests underground and in trees,” she said. “Now we can say with more confidence that they are at risk as well.”