difficult, in part because chemicals that cause avoidance of occupied
hosts (and uniform egg distributions) are hard to distinguish from a wider
array of chemicals that also happen to deter host acceptance in laboratory
choice tests.
Two examples illustrate potential problems. Although it is an external
leaf feeder rather than a seed or fruit parasite, the cabbage butterfly,
Pieris brassicae L., apparently avoids laying egg batches on leaves
that have already received eggs (Rothschild and Schoonhoven, 1977).
Novel alkaloids (miriamides) produced in the female’s accessory glands
are deposited with the eggs. Choice tests initially suggested that these
compounds caused reduced acceptance of occupied leaves (Blaakmeer
et al., 1994). Yet females also avoided leaves from which eggs are
removed, even though such leaves do not contain detectable amounts of
miriamides.
Subsequent experiments suggested that rejection of egg-bearing leaves
might actually be mediated by volatile, plant-derived compounds that
are induced by previous oviposition (Blaakmeeret al., 1994). Because
miriamides are not volatile, plant-derived compounds may better explain
the observation that females respond to egg-bearing leaves without
contacting them. The release of volatiles by infested plants is well known
to attract natural enemies of herbivorous insects, but volatiles may also
be used by the herbivores themselves to avoid competition (De Moraes
et al., 2001).
Similar confusion has surrounded attempts to identify the chemical
means of host discrimination byCallosobruchusbeetles. Early research
suggested that ether-soluble compounds (especially fatty acids) act as
bruchid marking pheromones (Oshimaet al., 1973; Sakaiet al., 1986).
These compounds can be obtained from washes of egg-laden glass
beads that serve as surrogate seeds. Yet washes of glass beads exposed
to males or virgin females also deterred oviposition in choice tests, and it
is difficult to imagine how these non-specific cues could mediate the
quantitative assessment of egg density described above (Messina and
Renwick, 1985).
Additional experiments suggested that Callosobruchus marking
pheromone is perceived through the maxillary palps. Ablation of these
mouth-parts eliminated uniform distributions of eggs; treated females
actually tended to aggregate their eggs (Fig. 4.1; Messinaet al., 1987).
However, palpectomized females were still able to avoid seeds coated
with putative marking pheromone in choice tests. Taken together, these
results suggested a weak correspondence between the sensory receptors
involved in the response to chemical extract and those needed for uniform
egg laying. It is certainly possible that multiple chemical cues are used for
detection of eggs on infested seeds (Credland and Wright, 1990), but
another complication is that fatty acids can both stimulate and deter
oviposition, depending on their concentrations (Parret al., 1998).
Identification of chemicals responsible for host discrimination will
probably be most successful if multiple behavioural assays (choice and70 F.J. Messina