in host acceptance and what determines patch-leaving decisions. Since
the audience is a general one, we have chosen not to extensively review
the abundance of literature that is available on (some of) these subjects but
instead illustrate key points by specific examples with key references. At
the end of our chapter we shall discuss the future of a behavioural ecology
approach and its value for understanding the ecology and evolution of
insect parasitoids.Patch Finding
Parasitoids generally do not emerge in the vicinity of a population of
insects that can serve as hosts for their offspring. Hence, emerging wasps
have to search for new sites. This first step in the searching process is
patch finding – finding the location of a site where suitable hosts are
likely to occur (Fig. 3.1). Parasitoids cannot simply detect new hosts
as bees detect new flowers. Since hosts are under severe selection to
avoid being found and devoured, they are difficult-to-detect objects. This
low detectability of hosts constrains the use of direct host cues, and
parasitoids have to deal with ‘surrogate’ indirect environmental cues.
Parasitoids primarily focus on olfactory cues to locate potential host sites,
although there is growing evidence for the use of visual cues (Wäckers and
Lewis, 1999). The food of the host is a primary source of information
during patch finding. Host-food chemicals can attract parasitoids even in
the absence of the host itself. Parasitoids of flies, for example, are attracted
to odours from the decaying ephemeral materials their hosts are feeding
from: carrion, rotting plants and fermenting fruits, fungi and so on. To
find their first host patch, parasitoids rely on innate responses to a library
of cues that have been shown to be of importance to the parasitoid over
evolutionary time (Vetet al., 1990, 1995). We use the term innate in the
sense of ‘unlearned’, without suggesting an instinct–learning dichotomy.
The ‘innate part’ of a response is shown by naïve individuals, without
the animal having had apparent experience with the stimuli concerned.
However, as soon as a first host is found, the parasitoid’s task becomes
easier. They learn through experience which environmental cues reliably
predict host presence and which do not, where we define learning by the
following criteria: (i) behaviour can change in a repeatable way through
experience; and (ii) learned responses can be forgotten (wane) or disap-
pear as a consequence of another experience (see Vetet al. (1995) for a dis-
cussion on the definitions of learning, on the adaptive value of learning
and on variation in how learning is expressed). During the last 20 years,
many examples of learning have been reported for more than 25 species
(for reviews, see Turlingset al., 1993; Vetet al., 1995). In often classic
‘Pavlov-like’ experiments, parasitoids are typically given experience with
hosts (the unconditioned stimulus) in the presence or absence of a con-
ditioned stimulus and their response to this conditioned stimulus is sub-
sequently tested in single- or dual-choice experiments in olfactometers,42 L.E.M. Vetet al.