(Anderson and May, 1991). Small changes in microhabitat choice might
mediate this process, but this is unlikely. There are easier ways to achieve
this, e.g. altering the reproductive apparatus. Among closely related
species ofTrichinellathat differ in their virulence, there is a significant
positive correlation between uterus size and the number of infective
stages produced (Sukhdeo and Meerovitch, 1980).
Much of the way we think about host/parasite interactions comes
from the gene-for-gene (GFG) hypothesis, which originated in studies of
plant pathogens (Flor, 1956). There are actually now two hypotheses
(variants) of GFG interactions, and geneticists are debating over them
because there is not enough evidence to unequivocally support either
(Newton and Andrivon, 1995). The original interpretation is that, for each
gene that conditions resistance in the host, there is a corresponding gene
that conditions virulence (pathogenicity) in the parasite (Vanderplank,
1991). The second and more popular interpretation is that the parasite
gene conditions avirulence (Kerr, 1987). The first implies a constant
escalation between the partners, while the second considers the fact that it
might not necessarily be advantageous for a pathogen to be virulent at the
expense of its host.
GFG is a simplistic explanation for how hosts and parasites interact
with each other, and it is now acknowledged that GFG interactions, if they
exist, are often complex polymorphisms, with large numbers of resistance
and virulence genes involved (Leonard, 1993). In animal-parasite studies,
there does not seem to be such an emphasis on GFG, but concepts such as
the ‘red queen’ and ‘arms race’ are modifications of GFG (Lively, 1996).
Parasitologists appear to be much more comfortable discussing an arms
race because it does not imply that there are specific rigid gene products
that control resistance and virulence.
In GFG scenarios, the immune response is thought to be the major
player. There are several examples of long-lived nematode infections
that are not actively controlled by host resistance and are slow to elicit
acquired responses, despite repeated exposure. For example, the human
hookwormNecator americanuscan live for 17 years in the host (Behnke,
1987) and the filarial nematode Onchocerca volvulus, which causes
tropical river blindness in humans, can live for 18 years (Plaiseret al.,
1991). It is not clear why there is no escalation by the host to kill the
parasite in these situations. It may be that host immune responses have
evolved to restrict infections within narrow tolerable bounds (narrow
niches), rather than totally eliminate infections. This may explain the
scarcity of sterilizing immunity among these worms (Behnkeet al., 1992).
Several theoretical and empirical studies now suggest that parasite
transmission and its need to produce infective stages at the cost of host
resources are directly coupled to the evolution of virulence in parasites
(Anderson and May, 1991; Ewald, 1994; Lenski and May, 1994). This
mechanism might clearly operate in microparasites that multiply within
their hosts, e.g. bacteria, viruses and protozoa (Bull, 1994; Read,
1994). However, intestinal nematodes (and other macroparasites) differ234 M.V.K. Sukhdeoet al.