"Ecological risk was severely underestimated," say Svata Louda of the University of Nebraska in Lincoln and Charles O'Brien of Florida A & M University in Tallahassee in the June issue of Conservation Biology.
The weevil (Larinus planus) is from Eurasia and is being released in the western U.S to control Canada thistle, which despite its name is also from Europe. Canada thistle is an aggressive weed and may threaten large areas of range and crop land. The weevil damages thistles in two ways: the adults eat the leaves and the larvae eat -- and so destroy -- developing flowers and seeds. Since a 1990 study suggested that the weevil preferred Canada thistle to native ones, the weevil has been widely released in the U.S., notably in western national parks, forests and monuments. However, Louda and O'Brien reanalyzed the 1990 results and found that the weevil fed equally on Canada and native thistles in laboratory tests.
While doing another study, Louda and O'Brien unexpectedly found that the weevil also feeds on a native thistle in the wild: Tracy's thistle, an relatively uncommon species found only in western Colorado and eastern Utah. In 1992 and 1993 the U.S. Forest Service released the weevil on the edge of Gunnison National Forest, which is near Almont, Colorado. In 1999 the researchers collected 30 Tracy's thistle flower heads from a roadside stand near the weevil release site. In 2000 the researchers double-checked their surprising find by collecting 185 Tracy's thistle flower heads from the same stand and 166 from another stand that was further away, as well as 375 Canada thistle flower heads from three nearby stands.
Louda and O'Brien found that the weevil fed extensively on Tracy's thistle. More than 75% of the flower heads either contained weevil larvae or had signs of larval damage. Worse, the weevil reduced seed set in infested flower heads by 98%: infested flower heads produced about one viable seed each while undamaged flower heads produced about 45 seeds each. Overall, the weevil reduced the seed production of the Tracy's thistle stands by two-thirds.
Moreover, the researchers found that the weevil had little effect on the non-native thistle it was supposed to control. In fact, there was no evidence of weevil feeding in any of the three stands of Canada thistle studied. This is striking because Tracy's thistle is sparse while Canada thistle is relatively common in the study area.
Louda and O'Brien call for re-evaluating the release of non-native insects to control non-native weeds in natural areas. "Current practices involving such introductions of exotic insects into nature reserves and national parks rely on incomplete assessments of ecological risk," they say.
CONTACT:
For PDFs of papers, contact Robin Meadows: robin@nasw.org
Conservation Biology
Journal