The samurai wasp (Trissolcus japonicus) arrived by accident in the United States before scientists were ready to release it.


Read our COVID-19 research and news.

Scientists spent years on a plan to import this wasp to kill stinkbugs. Then it showed up on its own

BRIDGETON, NEW JERSEY—In a peach orchard down a rural road here, an uninvited guest has run amok. The brown marmorated stinkbug (Halyomorpha halys) has been gorging on the unripe fruit. The bugs have sunk their needle-sharp stylets into the peaches, creating wounds that ooze a clear, sugary goo; form corky brown blemishes; and leave the trees more vulnerable to infection.

In this orchard, managed by the Rutgers University Agricultural Research and Extension Center, the mottled, shield-shaped stinkbug is a research subject. In surrounding farms and homes, however, it's a despised invasive pest known for its indiscriminate appetite, its tendency to escape cold weather by crowding into homes—sometimes by the thousands—and the pungent, cilantrolike odor it releases when crushed. (Exterminators often recommend that homeowners vacuum up the insects instead.) Native to Asia, the bug was first spotted in the United States in 1998; it has since reached 43 states and Washington, D.C., attacking fruit trees, corn, soybeans, berries, tomatoes, and other crops. Statistics are scarce, but an industry group estimates that Mid-Atlantic apple growers alone lost $37 million to stinkbug damage in 2010.

In the peach orchard, however, another surprise invader also is on the march—and it may prove to be the stinkbug's nemesis.

Like many invasive species, the brown marmorated stinkbug has no major enemies in its new home to keep its population in check. So in 2005, entomologist Kim Hoelmer and his team at the U.S. Department of Agriculture's (USDA's) Agricultural Research Service (ARS) in Newark, Delaware, turned to a strategy known as classical biological control: They traveled to Asia to find natural enemies of the stinkbug that they might release in the United States.

Fanning out to agricultural fields and botanical gardens, the team searched for the bug's tiny clusters of barrel-shaped eggs. They checked whether any had been invaded by parasitoid wasps, which inject their own eggs into the stinkbug's, leaving larvae that eat the developing bugs before chewing their way out. By far the most pervasive parasite they found was the samurai wasp (Trissolcus japonicus), which, despite its fearsome name, is stingerless and smaller than a sesame seed. The ARS team imported several strains of the wasp to a quarantined facility in Newark and began painstaking tests to decide whether it was a good biocontrol candidate.

Then in 2014, Hoelmer got an unexpected phone call. Elijah Talamas, a taxonomist at the Florida Department of Agriculture and Consumer Services in Gainesville, had been helping another ARS team identify native wasps parasitizing stinkbug eggs in Maryland. Talamas, an expert on Trissolcus species, had recognized that some were samurai wasps.

"It was stunning news," Hoelmer recalls. He had spent years studying the wasp in the lab to make sure that, if released, it would do its job without harming native species. But the insect was already here. Genetic tests confirmed that the wasps in Maryland hadn't escaped from any of his quarantined strains. Somehow, they had immigrated on their own.

The brown marmorated stinkbug is a versatile pest that attacks valuable crops and plagues homeowners.


Over the decades, a variety of uninvited biocontrol candidates have popped up on new continents, including a fungus that kills forest-stripping gypsy moths and a beetle that devours allergy-inducing ragweed. "The examples definitely are piling up," says Donald Weber, an ARS entomologist in College Park, Maryland, whose team found the first U.S. samurai wasps. "We've had this mindset that natural enemies would be less likely to establish" than invasive pests, he says. But sometimes, "It might be fairly easy."

Those unexpected arrivals can unsettle scientists and regulators. Rules aimed at carefully controlling insect releases can seem nonsensical when the species in question is already happily spreading on its own. And the arrival of the samurai wasp—a promising biocontrol agent against a high-profile pest, with a formal proposal for release already in the works—has prompted a fresh look at some U.S. regulations.

But unplanned introductions also free researchers from some of the usual constraints, allowing them to explore key questions about a biocontrol agent's impact in field experiments rather than just the lab. The team at the peach orchard, for example, is one of about a dozen U.S. groups now releasing the samurai wasp into fields and orchards to see whether it will be an ally in fighting the exotic stinkbugs—or yet another problematic invader.

Classical biological control has logged some undeniable successes, such the release of the South American wasp Apoanagyrus lopezi in Africa in the 1980s to control the cassava mealybug (Phenacoccus manihoti). That project preserved a staple crop and saved an estimated 20 million lives, earning its architect, Swiss entomologist Hans Rudolf Herren, the 1995 World Food Prize. But many of the best-known biocontrol efforts are the historical disasters: the mongooses unleashed for rat control in Hawaii in 1883 that devastated native birds and turtles, and the cane toads sent to Australia in 1935 that failed to control sugarcane-destroying beetles but—because the toads themselves are poisonous—killed native reptiles, frogs, birds, and mammals that ate the toads.

As the field matured, many nations began to strictly regulate the release of biocontrol agents—which can include insects, fungi, and bacteria—and required studies to predict potential "nontarget effects." As Weber puts it, "People are a lot more responsible now than when they were running around releasing mongooses." In the United States, researchers must submit a proposal to USDA's Animal and Plant Health Inspection Service (APHIS). That proposal must include data from lab experiments gauging whether their candidate is likely to eat or parasitize species other than the targeted pest. Three groups then vet the evidence of its safety: a scientific review panel with representatives from Canada and Mexico, an APHIS official, and sometimes the U.S. Fish and Wildlife Service. The process can take years.

A samurai wasp emerges from the brown marmorated stinkbug egg in which it hatched. Scientists hope the wasp will reduce populations of the pest in U.S. crops.


But organisms have a way of sidestepping bureaucracy. Over and over, potentially beneficial species have popped up uninvited, likely reaching new continents by the same shadowy routes of international trade and travel that spread pests. Entomologist Paul DeBach of the University of California, Riverside, in a 1971 essay, called this phenomenon fortuitous biological control. And it can be a boon. Recently, the North American leaf beetle (Ophraella communa), which feeds on the invasive ragweed Ambrosia artemisiifolia, sneaked into Europe, potentially saving many people from pollen-induced misery. And when an exotic fungus—Entomophaga maimaiga, which kills the gypsy moth in its caterpillar stage—started to spread around New England in the late 1980s, "it was very exciting," says entomologist Ann Hajek of Cornell University. Her team had tried without success to establish the fungus, but once it arrived, she spent years making sure it wasn't harming local caterpillar species. "We were lucky," she says: The natives were mostly unaffected.

Unplanned arrivals can bring heartache, however. Entomologist Tim Haye of the Centre for Agriculture and Bioscience International in Delémont, Switzerland, and his collaborators spent a decade developing plans to release the European parasitoid wasp Trichomalus perfectus to control the invasive weevil Ceutorhynchus obstrictus, a canola plant pest, in Canadian prairies. Then in 2009, the wasp appeared on its own, in Quebec in Canada. "I was very disappointed," Haye says.

Moreover, such invaders don't always prove fortuitous. A team led by ARS entomologist Keith Hopper investigated the parasitoid wasp Aphelinus certus as a biocontrol agent against the soybean aphid (Aphis glycines), but lab tests revealed it had a broad range of aphid hosts, including some native ones. "I would never have brought that thing in," says ecologist George Heimpel of the University of Minnesota in St. Paul, a collaborator on that project. But in 2005, the wasp showed up in Pennsylvania. Now that it's established, Heimpel's group is studying whether native aphid populations are in danger. "A lot of people don't care about native aphids," he says. "I am one of those that do."

Yet such unplanned introductions also hand researchers an opportunity, creating what Haye calls a gigantic field trial for testing predictions about the control agent's effects. For the samurai wasp, researchers can now get permission from state regulators to launch experiments that involve breeding and releasing accidentally introduced wasp strains in the wild—work that wouldn't be allowed with intentionally imported insects, which must stay quarantined until federal regulators are convinced they're safe.

Ancient enemies reunited

Since the samurai wasp's first U.S. appearance in 2014, surveys have turned up at least three genetically distinct populations in areas affected by the brown marmorated stinkbug.

Agricultural and nuisance problems Nuisance problems only Stinkbugs detected First field recovery of samurai wasps Severe agricultural and nuisance problems No data 2015 2014 2016 2017 2018

One night last May, 3600 samurai wasps streamed from mesh cages into the stinkbug-infested New Jersey orchard. A team led by Rutgers entomologist Anne Nielsen and entomologist Kevin Rice of the University of Missouri in Columbia had strung up yellow sticky cards baited with stinkbug egg clusters among the peaches and along the adjacent forest edge. They planned to wait a few days, collect the cards, and count the wasps to see whether they had ventured into the orchard to pursue the peach-destroying bugs.

The insects were descendants of wasps that Nielsen first discovered in a nearby New Jersey orchard in 2017—the first find in a U.S. agricultural crop. Since 2014, the samurai wasp has turned up in 10 states and Washington, D.C. Researchers have identified at least three genetically distinct strains, suggesting multiple introductions. Maybe wasp embryos were hiding in stinkbug eggs on plants aboard a cargo ship. An adult wasp may even have hitched a ride with an unsuspecting airline passenger. (While awaiting a flight from New York City to Russia, Talamas once watched a different parasitoid wasp species, native to the United States, land on a page of his book. "All it had to do was fly down the walkway … next stop: Russia." He trapped the hitchhiker in his contact lens case and, on arrival, preserved it in vodka.)

Now that the wasps are in the United States, research questions abound, Nielsen says. In their native range, they parasitize up to 90% of brown marmorated eggs. But will their behavior be different here? Where will they congregate and forage? Will they dramatically reduce stinkbug populations? Could farmers support the wasp by adjusting their practices—for example, not spraying pesticides where the insects are most concentrated? The chance to probe basic questions about a little-studied exotic species, Rice says, is "fabulously exciting."

For U.S. regulators, however, the wasp's unexpected arrival poses an conundrum. "This is a good chance for us to codify policy and decide, ‘How are we going to handle these circumstances?’" says Robert Pfannenstiel, an APHIS entomologist in Riverdale, Maryland, who reviews release applications. "Will we allow changes from our policies and processes that are already in place, or not?"

Brown marmorated stinkbugs have done major damage to apples and other fruit trees in the mid-Atlantic United States.

Matt Rourke/AP Photo

Studies so far suggest the samurai wasp is a promising biocontrol agent. Although in laboratory tests it has parasitized some eggs laid by native species, it has shown a strong preference for brown marmorated stinkbug eggs. Scientists can release the accidental strains in states where they've already been discovered, but for now they can't spread the wasps indiscriminately. APHIS prohibits moving exotic species that haven't been formally cleared for release into new states. (Nielsen and Rice, for example, couldn't legally perform their same experiment if they drove 4 hours north to Connecticut, where samurai wasps haven't been found—so far.)

The wasp needs to go through regulatory review, just like any other candidate, Pfannenstiel says. "The danger, in one case, of saying, ‘Oh, we can tell it's not a risk,’ and then releasing it [is that] there's pressure to do that repeatedly, and start making judgment calls rather than determinations" based on data. Field studies of the accidental strains could speed the evaluation and help the wasp's chances of approval—or reveal new reasons not to release it, he says. "I go into these evaluations with no preconceptions."

Hoelmer's team at ARS is still preparing a petition to APHIS to release one deliberately imported strain, which he hopes could serve as a backup if the accidentally introduced strains spread slowly. He also intends to include the accidentally introduced strains in his release petition, because their biology is so similar to that of the strain he has studied extensively. He plans to submit his petition by the end of this year, and hopes for a decision next year. For now, he says, growers and researchers in states where the wasp hasn't been detected will simply "have to wait until it crosses the border."

The caution can seem excessive, given the wasp's steady spread. "Should we really wait?" asks Haye, who also believes the wasp is a good biocontrol agent. "Or should we speed up the spread and prevent damage—and help farmers—knowing that it's not 100% specific and that eradicating it is not an option anyways?"

Those are just some of the questions likely to come up at a special symposium on the implications of accidentally introduced biocontrol candidates, set for the annual meeting of the Entomological Society of America in Vancouver, Canada, in November. Weber, co-organizing the symposium, hopes that regulators weighing the benefits and risks of a biocontrol proposal will factor in the potential for surprises: "The longer the regulatory process, the more likely the species will accidentally show up," he says. "There might be a fairy tale world where you can regulate everything that comes in," but in reality, "that horse leaves the barn on a regular basis.

Rutgers University entomologist Anne Nielsen and graduate student Nick Avila survey a yellow sticky card for samurai wasps to track their dispersal in the peach orchard.

Dean Polk

The researchers at work in the Rutgers orchard aren't yet endorsing the samurai wasp as a biocontrol agent. First, they'd like more evidence that it won't harm native species. "They're invasive," Rice says. "They're not in a different bucket" from the stinkbugs.

The restrictions on spreading the wasp "can be frustrating, and it can seem arbitrary, but the regulations are there for a reason," Nielsen adds. Still, she says, the wasp "is likely our best hope of controlling the brown marmorated." Examining the sticky cards this summer, her team found a roughly equal distribution of wasps in the peaches and the nearby woods. That finding suggests the wasps are perfectly happy foraging for stinkbug eggs among the fruit, which bodes well for the wasps' ability to control the pests. The team plans to run a similar experiment soon to see how the wasps spread into another crop, soybeans.

Meanwhile, researchers in California have sent Talamas another surprise: a new, accidentally introduced Trissolcus parasitoid wasp, this one native to India and Pakistan, which emerged from the egg of another exotic stinkbug pest, Bagrada hilaris. "I think that these introductions are happening constantly," Talamas says, but come to light only when taxonomists bother to take a close look. He'll publish the new finding later this month in the Journal of Hymenoptera Research.

Such arrivals are "humbling," Weber says—a reminder of the limits that humans face in shaping their environment. "We have less control over things than we think."