Eradicating a population — removing every last individual — is a seductive but highly controversial idea. Some question the value of attempting eradication of well-established, widespread populations (e.g., Dahlsten, 1986; Rejmanek in Borenstein, 1999). Probably some of the skepticism arises from several highly visible, costly failures (Newsom, 1978). Perhaps the worst was the attempt to eradicate the imported fire ant (

Solenopsis invicta

) from the southeastern U.S. (Davidson and Stone, 1989). This was such a catastrophe in terms of collateral damage (impacts on nontargets, including humans) and expense (over $200 million) that E.O. Wilson has termed it the

Vietnam of entomology

(Brody, 1975). Part of the controversy surrounding eradication can be traced to the fact that the term is used

colloquially and imprecisely to mean partial removal of a pest species and control at some lower, acceptable density. This usage is common among politicians (e.g., Chiles, 1996) but also among scientists (e.g., Langland and Sutton, 1992). Thus, a program not really designed to eliminate every individual of a population can be said to have failed even if it confers substantial control. Of course, if a management program aiming at complete elimination — an eradication campaign — uses the same methods that would have been used had the goal been to lower densities to an acceptable level, even a failed campaign can be useful. No harm need be done even if the species is not eradicated (Simberloff, 1997), and substantial control can be achieved, as in the current campaign to eradicate

Spartina

spp. from New Zealand (Nicholls, 1998). However, if different means are used in an eradication attempt than would have been used for

maintenance control, a real problem can arise (Dahlsten, 1986). Consider, for example, a major premise of biological control: the maintenance of both a pest and its natural enemy at low levels, with a homeostatic relationship between them such that an increase in pest density triggers a rapid increase in that of the natural enemy. An eradication project that attempted to kill every pest individual but failed could nonetheless leave pest densities so low that natural enemy populations are extinguished, and subsequent increase of pest densities would be unimpeded.