ranging search) after contact with different species of host were observed
and may be due to differences in host preference or suitability (Glazer and
Lewis, 2000). Cues that trigger penetration and acceptance behaviour are
not well understood, but infective stages are known to respond to a
number of host-associated materials, such as faeces (Grewalet al., 1993a),
gut contents (Grewalet al., 1993b) and haemolymph (Khlibsuwanet al.,
1992). For example, contact with host faeces increased head-thrusting
behaviour, which may be involved in penetrating through the insect’s
digestive system (Grewalet al., 1993a). The functional benefit of these
behavioural responses, however, has not been tested.
There is evidence that entomopathogenic nematodes can evaluate
host quality and make infection decisions about which hosts to infect.
This is most apparent in the host-acceptance decision associated with
infecting an infected versus an uninfected insect. For steinernematid
nematodes, there appears to be a selective advantage to infecting a host
that is already infected. The risk of immediate mortality is reduced
due to degradation of the host immune response and the probability
of encountering a parasite of the opposite sex is increased. Over time
postinfection, the fitness advantages of infecting a host that is already
infected will probably diminish. In addition, there is an Allee effect on
the number of infective juveniles produced and a decline in infective
juvenile size as a function of increasing number of founders (Selvanet al.,
1993b).
Several species of entomopathogenic nematode infective juveniles are
more strongly attracted to infected hosts 24 h after infection than to
uninfected hosts (Lewis and Gaugler, 1994; Grewalet al., 1997). Infective
juveniles also distinguished between conspecific infections and hetero-
specific infections, but there was considerable variation in this ability
among species (Grewalet al., 1997). For example,S. feltiaewas not
attracted to hosts infected by any of the tested heterospecifics, but
S. glaseriwas attracted to all infected hosts except those infected by
S. riobrave. In some cases, infective juveniles were actually repelled by
the heterospecific-infected hosts. J.F. Campbellet al. (unpublished data)
found that, when given a choice between a host infected by conspecifics
24 h earlier and an uninfected host,S. feltiaepreferentially infected the
previously infected hosts, but showed no preference for hosts infected by
conspecifics 2, 4 or 8 h prior (Fig. 2.2). In contrast, Glazer (1997) found
that 9 h after injecting nematodes into an insect’s haemocoel, the number
of infective juveniles that would infect that insect was reduced. Exposure
ofS. carpocapsae,S. riobraveandS. feltiaeinfective juveniles to a surface
previously exposed to an infected host suppressed their infectivity when
exposed to an uninfected host, and this suppression was removed by
washing the infective juveniles (Glazer, 1997).
It has been widely reported that only a small proportion of infective
juveniles, typically less than 40%, infect a host when exposed under
laboratory conditions. The explanation most commonly invoked was
that this was due to differences in internal state, i.e. not all individuals28 J.F. Campbell and E.E. Lewis