Environmental Engineering Reference
In-Depth Information
The general steps involved in a modern classical biological control project have
been discussed in depth by several authors; see for example, Van Driesche and
Bellows (1996); these authors also cover taxa other than plants and arthropods
Two related topics have dominated in the literature for some time because of their
relevance to successful and safe biological control: the characteristics of an effi -
cacious agent and how to assess and manage ecological risks. Although several
agents have been introduced in most biological control projects, most successes
have come about through the action of just one or two agents. This aspect has
recently put further attention on the need for criteria for the selection of agents to
avoid introducing 'poor' agents that might increase the risks of non-target effects
(Sheppard and Raghu 2005).
Determining why some biological control projects are successful and yet many
more are failures has been an active area of research of population ecology since
the 1920s. Ideas and hypotheses have been put forward to try and explain the lack
of establishment of agents and also why established agents may or may not con-
trol the target (Hopper 1996). The general thrust of work on the latter has been
to examine the effects of various demographic parameters of natural enemies and
also the population response to the host's distribution on the equilibrium density
and population stability of the target (Hochberg 2002). The focus of this work
has primarily been on insect natural enemies that attack insect hosts. Numerous
mathematical models have been created and these have contributed much to the
theoretical basis of population ecology, but none of this has produced a useful pre-
dictive framework or guidelines that help with the selection of agents. Some have
suggested that such predictions will be almost impossible because of the inherent
variation in the systems involved. A basic problem has been that little experimental
work or observations have been done to examine the hypotheses and thus to accept
or reject these hypotheses, and this was highlighted in a number of reviews written
in the 1990s (e.g. Hopper 1996).
An important aspect in the process of the selection agents does dictate that
practitioners consider one particular ecological trait in agents: that of complete or,
in some circumstances, very narrow host specifi city. h is is because of the need to
minimize the risk to non-target organisms; this is one of the points discussed in
the next section.
h ere are now a number of more recent and important papers that do provide
ideas on how to improve biological control projects through careful and well-
planned research on the ecology and genetics of natural enemies. Some of these
papers have focussed on the topic of criteria for agent selection. For example, in
the case of invasive plants several authors have suggested that it is important to
identify the stages in the life cycle that herbivores or pathogens might have a major
impact and that ideally these studies should be done in the native and invaded
