beans, egg distributions of cowpea-selected females were slightly less
uniform than those of mung bean-selected females. On cowpea seeds,
cowpea-selected females were much ‘sloppier’ than mung bean-selected
females in how they spread their eggs among seeds (Fig. 4.4).
By showing a decreased tendency to avoid occupied hosts, females
from cowpea-selected lines converged towards the poorer host discrimi-
nation observed in cowpea-adapted populations. Selection for strong
avoidance of occupied hosts may have been relaxed in the cowpea-
selected lines because of the simultaneous evolution of less competitive
larvae. The relatively quick decline in averageUscores on cowpea (≤ 40
generations) also suggests that there may be a cost associated with a high
degree of host discrimination, although the nature of this cost remains
unclear (Messina, 1993). Taken together, the results of the selection exper-
iment demonstrate how a host trait (size) can predictably modify both a
juvenile trait (competitive ability) and an adult trait (host discrimination)
in the parasite.Effects of Parasite Enemies
Juvenile stages of seed parasites are of course susceptible to their
own natural enemies as well as to competition or cannibalism. Several
mechanisms have been identified by which natural enemies can modify
the tendency for seed-parasite females to avoid occupied hosts. Risk of
mortality from natural enemies also provides a separate explanation for
why females may reject a potential host (or lay few eggs) when the current
density on the host is well below the single-host maximum. Bruchid
beetles have been useful subjects for addressing relationships between
egg-laying behaviour by seed parasites and mortality risk from natural
enemies. Eggs and larvae of most seed beetles are attacked by at least one
species of parasitoid wasp.
Natural enemies could reduce advantages of host discrimination if
cues that allow seed parasites to detect previous infestation are used by
natural enemies to find parasite eggs or larvae. In a fruit parasite, marking
pheromone deposited after oviposition acted as a kairomone to attract a
specialist parasitoid (Prokopy and Webster, 1978). Use of marking phero-
mone as a kairomone causes a trade-off between minimizing competition
and minimizing predation risk, and can therefore maintain different
levels of host discrimination (and genetic variation) within a population
(Roitberg and Lalonde, 1991; Hoffmeister and Roitberg, 1998).
A similar trade-off would occur if predation risk were inversely
density-dependent, i.e. if single eggs or eggs in small clutches suffer
greater mortality from enemies than do eggs in large clutches. Inverse
density dependence could arise, for example, if a parasitoid tends to
attack only one or two seed-parasite eggs before leaving the seed. Mitchell
(1977) suggested thatMimosestesseed beetles often lay a stack of two eggs
rather than a single egg because the presence of the top egg diminishesHost Discrimination by Seed Parasites 79