50
MARINE MOLLUSCS AS HOSTS FOR SYMBIOSES
in decomposing meat. Thus, although A . manducator is not directly
attracted to its host, its selective preference for rotting meat, which is its
host’s habitat, indirectly aids in its selection of the preferred host.
Another example of host-preference among parasitic insects has to
do with the parasitic wasp Nerneritis. Thorpe and Jones (1937) have
demonstrated that if the eggs of Nemeritis are artificially introduced
into an unnatural host, the wax moth, Meliphora, instead of its natural
host, Ephestia, the resultant adults give a strong olfactometric response
to Meliphora when given a choice between it and a blank. These
workers, however, were not successful in attempts to demonstrate a
preference of conditioned Nemeritis for Meliphora over Ephestia. It is
nevertheless of interest to point out that wasps which develop in the
unnatural host become sufficiently conditioned so as to recognize the
new host. This model is useful in considering the origin of new hostsymbiont associations. It is conceivable that many such associations
arose when during a brief free developmental period in the life of the
symbiont it happened to encounter a potentially new host in some niche
different from that of its natural host. If the new host produces some
factor(s) related to, but not identical with, that with which the symbiont
has been associated in the natural host, it is conceivable that an effective
conditioning to the new host factor(s) may bring about a change in
host-preference, aided perhaps by a shortening or loss of the free stage,
and resulting in subsequent generations becoming genetically isolated
in or on their new host. From then on, the course of evolution of the
symbiont would be controlled by that of its new environment, the new
host (Davenport, 1955).
The mechanisms underlying the natural combination of new associations may not always be so simple; nevertheless, this hypothetical
model as presented by Davenport is worthy of serious consideration.
It is of interest to mention a t this point that Dethier (1954), in considering the feeding preferences of insects, has stated :
The crucial problem is whether a phenomenon like olfactory conditioning
can bring about sympatric splitting of a single population into two nonbreeding populations. It is clear that in the absence of effective isolating
mechanisms there would be free hybridization between individuals with
different . . . preferences so that the preferences fostered by olfactory conditioning could never become sorted out genetically. . . . Thus, in evaluating
the significance of olfactory conditioning as an effective factor in establishing
populations with new . . . preferences, a question of fundamental importance
arises, i.e. whether olfactory conditioning by itself can lead to any kind of
isolation . . . or whether the isolation must be interjected from another
quarter.
MARINE MOLLUSCS AS HOSTS FOR SYMBIOSES
in decomposing meat. Thus, although A . manducator is not directly
attracted to its host, its selective preference for rotting meat, which is its
host’s habitat, indirectly aids in its selection of the preferred host.
Another example of host-preference among parasitic insects has to
do with the parasitic wasp Nerneritis. Thorpe and Jones (1937) have
demonstrated that if the eggs of Nemeritis are artificially introduced
into an unnatural host, the wax moth, Meliphora, instead of its natural
host, Ephestia, the resultant adults give a strong olfactometric response
to Meliphora when given a choice between it and a blank. These
workers, however, were not successful in attempts to demonstrate a
preference of conditioned Nemeritis for Meliphora over Ephestia. It is
nevertheless of interest to point out that wasps which develop in the
unnatural host become sufficiently conditioned so as to recognize the
new host. This model is useful in considering the origin of new hostsymbiont associations. It is conceivable that many such associations
arose when during a brief free developmental period in the life of the
symbiont it happened to encounter a potentially new host in some niche
different from that of its natural host. If the new host produces some
factor(s) related to, but not identical with, that with which the symbiont
has been associated in the natural host, it is conceivable that an effective
conditioning to the new host factor(s) may bring about a change in
host-preference, aided perhaps by a shortening or loss of the free stage,
and resulting in subsequent generations becoming genetically isolated
in or on their new host. From then on, the course of evolution of the
symbiont would be controlled by that of its new environment, the new
host (Davenport, 1955).
The mechanisms underlying the natural combination of new associations may not always be so simple; nevertheless, this hypothetical
model as presented by Davenport is worthy of serious consideration.
It is of interest to mention a t this point that Dethier (1954), in considering the feeding preferences of insects, has stated :
The crucial problem is whether a phenomenon like olfactory conditioning
can bring about sympatric splitting of a single population into two nonbreeding populations. It is clear that in the absence of effective isolating
mechanisms there would be free hybridization between individuals with
different . . . preferences so that the preferences fostered by olfactory conditioning could never become sorted out genetically. . . . Thus, in evaluating
the significance of olfactory conditioning as an effective factor in establishing
populations with new . . . preferences, a question of fundamental importance
arises, i.e. whether olfactory conditioning by itself can lead to any kind of
isolation . . . or whether the isolation must be interjected from another
quarter.