The beetles are dead and gone, shriveled in the heat or eaten by ants, but otherwise the Owens Valley, Calif., research site looks the same as the last time scientist Tom Dudley saw it. Tinemaha Reservoir glimmers beyond a wall of brush. The sharp peaks of the Sierra Nevada decorate the skyline. Invasive Eurasian tamarisk, or saltcedar, trees still freckle the valley floor, their feathery branches casting long shadows across the native sagebrush in the early morning light. And Dudley's beige beetle cage, the last of a set, still envelops a single tamarisk, as if the five-foot cube of fabric could contain the plant and everything it stands for. Unzipping the mesh, Dudley steps inside with his screwdriver.

It's time to let go.

Dudley, a wild-haired, gray-bearded riparian ecologist from the University of California at Santa Barbara, is dismantling part of a more-than-decade-long research project on whether an Asian leaf beetle, Diorhabda elongata, can help rid the West of tamarisk. In June, partly in response to a lawsuit by the Center for Biological Diversity and the Maricopa Audubon Society, the U.S. Department of Agriculture's Animal and Plant Health Inspection Service (APHIS) indefinitely suspended its Diorhabda biocontrol program, including Dudley's research permits. It turns out that the insects -- a group of related species each about the size of a grain of rice -- may be threatening the endangered southwestern willow flycatcher, which nests in tamarisk in areas where the plant has displaced its preferred willow habitat.

The stakes holding down the cage are stuck in the ground, so with a rip and a clang Dudley uses a metal pole to pry the mesh away. He is frustrated with what he sees as putting the good of individual birds over the health of an ecosystem. "I'd prefer a system that was dominated by native species," he says. Finally the cage comes off, and he sets the plant free. "OK, bush, grow grow grow!" he says wryly.

Most Westerners have heard that tamarisks guzzle water, mess with soil salinity, push out native species, and destroy habitat. Land managers often crusade for its eradication. In the Owens Valley, the Inyo County Water Department spends around $300,000 a year fighting tamarisk with chainsaws and herbicides. The beetle, on the other hand, costs little to release. Under the right conditions, a few hundred insects can multiply across tens of thousands of acres in just three or four years. The larvae and adults strip tamarisk leaves summer after summer, gradually wearing groves into submission. Eventually, many trees die; the rest grow more modestly, their vigor and dominance reduced by the beetle's constant attacks.

Over the past decade, APHIS has permitted beetle releases in 18 states across the West and Midwest, and worked to establish nurseries of the insects in 10 of those. At many sites, the beetles failed to thrive, but at some they spread faster and farther than scientists had anticipated. Today, the insects swarm across the upper Colorado Plateau, and though no one tracks exact numbers, adding up local reports suggests they've defoliated tamarisk on a total of at least 11 million acres. By that measure, certainly, the beetle program has been a success.

But APHIS' recent decision makes it illegal to transport beetles across state lines. That means no new releases in states where the beetles aren't already established, such as Idaho and Washington –– much to the dismay of biocontrol supporters like Dudley. APHIS intended its moratorium to "raise awareness" and "encourage thought about intrastate movement" of the insects, and it has; in Montana and in the Owens Valley, for example, land managers have abandoned plans to use the beetles. Meanwhile, the insects are spreading deeper into flycatcher territory, and frustrated endangered species activists worry that APHIS' caution may have come too late to help the birds.

Yet despite all the furor, the agency's decisions –– both to approve the program in the first place and to suspend it now –– could well prove to be a net positive. The suspension may not close off all possibility for future biocontrol. And the march of the beetles is inspiring a new emphasis on proactive habitat restoration in flycatcher territory.

The current heart of the beetle saga is the Virgin River, a tamarisk-tinged Colorado River tributary that flows through Utah and Arizona before emptying into Nevada's Lake Mead. The Virgin hosts about 50 of the 1,300 willow-flycatcher nesting sites remaining in the Southwest, and connects, via the Colorado, to hundreds more. When APHIS began actively distributing leaf beetles in 2005, it agreed not to release any insects within 200 miles of the bird's known critical habitat. But insects have a way of ignoring boundaries.

According to the Center for Biological Diversity, APHIS employees allowed local officials in southern Utah to help themselves to some beetles from an approved nursery farther north in 2006. The Utah officials released the insects in the Virgin River watershed. Originally, the beetles couldn't survive so far south -- the  short  summer  days cued them to hibernate too early -- but they quickly adapted and began to thrive. Though scientists knew that might happen, they'd predicted it would take at least a decade. Instead, in 2008, the beetles crossed the invisible line into active flycatcher nesting territory, just a couple miles from the release area, and an unplanned, high-stakes experiment began.

The beetles strip the trees where flycatchers nest at exactly the wrong time of year, creating an ecological trap for the birds, according to biologists. "The southwestern willow flycatchers like it where it's dense and moist," explains Greg Beatty, a biologist with the U.S. Fish and Wildlife Service in Arizona. When the small brown birds arrive from their winter homes in Central America to build nests early in the summer, the tamarisk is attractive because it's still green. Then the beetles emerge, start munching and upend everything. Different humidity levels, more sunlight, and new predators stress the birds -- and in 2009, 13 out of 15 Virgin River nests in the affected area failed once the beetles arrived. (The average nest failure rate is more like 40-50 percent.)

This year, the beetles advanced another 20 miles down the Virgin. By next year, they could be at Lake Mead. Researchers still don't know how deep into the flycatcher's range the beetles will spread, nor exactly how they will affect the birds in the long term. Even the statistics for the current overlap are preliminary, though this year's results are encouraging: The birds learned from last year's nest failures and chose to nest in willows instead. In the long run, there's a chance the beetles could even help improve the bird's habitat, if they clear the way for a more diverse community of native plants. But that's a big if.

Because it takes many years for biocontrolled ecosystems to reach a new equilibrium, it's too early to know for sure what will happen once beetles transform a site. "Generally, whatever is on the edges is going to move in as tamarisk mortality occurs," says Tim Carlson, research and policy director of the nonprofit Tamarisk Coalition. Along the main stem of the Colorado, native willows are growing more vigorously now that the beetles have hit the tamarisk. But on the lower Humbolt River in Nevada, the invasive Russian knapweed that was already in the understory is thriving as the tamarisk dies out.

Beatty worries that, especially in the Southwest, the willows won't return. And it's not because tamarisk is inherently aggressive: Newer studies suggest that dams, water diversion and overgrazing have created conditions that favor the plant. "Killing tamarisk thinking your ecosystem problems are going to go away is kind of like wiping your nose thinking it's going to take care of the cold," he says. In fact, Beatty says, tamarisk may not be the evil invader everyone thought.

It turns out that tamarisk uses more water than native plants only under certain conditions. Removing it rarely results in more water for people or wildlife. Tamarisk doesn't necessarily make soils saltier. And in some cases, wildlife -- including the flycatcher -- does fine in tamarisk and would likely do worse if the plants were removed.
Much of this information isn't brand-new, but it seems to have reached a new threshold of attention: APHIS cited "changes in perception of saltcedar (water usage, wildlife habitat, etc.)" as part of the new information that led it to re-evaluate the beetle program. APHIS and Beatty both reference a 2009 paper tracing the historical "monstering" of the plant. If blanket hostility to the tree is hyperbolic and outdated, it may matter less that many land managers are no longer free to cultivate the leaf beetles to fight it.

Still, Beatty concedes that removing tamarisk makes sense in some cases, especially where it's ecologically, economically and practically feasible for natives to rebound. As Dudley points out, tamarisk burns fiercely and generally supports less biodiversity than native plant communities. And the plant narrows stream channels by holding sediment, thus speeding water and increasing flooding. (Removing it all can cause trouble, however, as the sediment dumps back into the system.) Carlson of the Tamarisk Coalition sees the issue in terms of time-scales: "Generally speaking, long-term, there's good consensus that native species are better than a monoculture of non-natives."

Ultimately, though, the scuffle over the beetle may lead to more comprehensive efforts to restore flycatcher habitat. In early September, the Tamarisk Coalition assembled scientists and agency representatives to talk about how to prepare the birds for the beetles' arrival. Fish and Wildlife's default practice of simply leaving tamarisk in flycatcher habitat no longer makes sense. Instead, the idea is to be proactive. On the Virgin River and other sites, researchers and restoration professionals could create patches of willows near flycatcher nest sites, so the birds won't have to go far if tamarisk declines. And in the long term, the vision includes grazing reduction, leaving more water in streams where feasible, and protecting patches of native plants to serve as seed sources. These tactics will help not just the flycatchers, but the ecosystem as a whole. But if they don't work, don't get funded, or don't happen fast enough, the flycatcher could still be in trouble.

Even so, Dudley now believes that "for the first time in years there's actually some kind of a progress towards doing positive work," his optimistic tone a marked change from his earlier frustration. He's eager to get at the root causes of habitat problems. "In a weird way, the presence of the beetles really brought a smoldering issue to a head."