354
P. 8. MEADOWS AND J. I. CAMPBELL
mutants of sedentary marine invertebrates whose larvae displayed
unusual gregarious or territorial patterns of behaviour.
If a species colonizes a new type of habitat by a process not
dependent on genetic change, for example by new habitats suddenly
becoming available or because of population pressure, one may ask
whether over a number of generations the new habitat can change the
preferences of the species to such an extent that it prefers the new to the
old habitat and can survive more successfully in it.
There are no studies relating to this topic on aquatic invertebrates.
However, several investigations on terrestrial animals have been
designed to answer these questions and can now be considered. They
fall into two categories. I n the first, a species is reared for a number
of generations in or on a particular habitat and then offered a choice
of this habitat with another. Sladden and Hewer (1938) have demonstrated an alteration in the food preferences of the stick insect
Carausius (Dixippus) mrosus (Br.) for privet and ivy leaves in this way
and four other examples are quoted by Dethier (1954, p. 43). On the
other hand Morimoto (1939) and Wood (1963) in analogous experiments
on other insects were unable to show any changes in preferences. I n the
second category, animals are bred for a number of generations and at
each generation are given a choice of two habitats, A and B, the
subsequent generation always being bred from only those animals that
chose A. At the end of the experiment the population is given a choice
of A and B on the assumption that it will prefer A to a greater extent
than would an unselected population. If two populations are run in
parallel, then one is selected for the A choice and the other for the
B choice; at the end of the experiment both populations are offered
the choice of A and B in the hope that the population selected for A
will prefer A to B and the population selected for B will prefer B to A.
There are furfher permutations of this type of experiment, but these
need not concern us here. Successful experiments falling into this
category have been performed by Eloff (1936), Wilkes (1942), Dobzhansky and Spassky (1962), Connelly (1966) and Ogden (1970), on
the preferences of insects for water, temperature, gravity, and on
activity respectively. In at least one instance the changed preferences
selected for over a number of generations have been correlated with
genetic changes in the population (Dobzhansky and Spassky, 1962,
1967; Ehrman et ab., 1965; Dobzhansky et al., 1969), and it is genetic
changes of this sort which may in some cases prelude the origin of new
species.
Any part that habitat selection might play in the initial steps of
speciation can clearly occur only during sympatric speciation, that is,
P. 8. MEADOWS AND J. I. CAMPBELL
mutants of sedentary marine invertebrates whose larvae displayed
unusual gregarious or territorial patterns of behaviour.
If a species colonizes a new type of habitat by a process not
dependent on genetic change, for example by new habitats suddenly
becoming available or because of population pressure, one may ask
whether over a number of generations the new habitat can change the
preferences of the species to such an extent that it prefers the new to the
old habitat and can survive more successfully in it.
There are no studies relating to this topic on aquatic invertebrates.
However, several investigations on terrestrial animals have been
designed to answer these questions and can now be considered. They
fall into two categories. I n the first, a species is reared for a number
of generations in or on a particular habitat and then offered a choice
of this habitat with another. Sladden and Hewer (1938) have demonstrated an alteration in the food preferences of the stick insect
Carausius (Dixippus) mrosus (Br.) for privet and ivy leaves in this way
and four other examples are quoted by Dethier (1954, p. 43). On the
other hand Morimoto (1939) and Wood (1963) in analogous experiments
on other insects were unable to show any changes in preferences. I n the
second category, animals are bred for a number of generations and at
each generation are given a choice of two habitats, A and B, the
subsequent generation always being bred from only those animals that
chose A. At the end of the experiment the population is given a choice
of A and B on the assumption that it will prefer A to a greater extent
than would an unselected population. If two populations are run in
parallel, then one is selected for the A choice and the other for the
B choice; at the end of the experiment both populations are offered
the choice of A and B in the hope that the population selected for A
will prefer A to B and the population selected for B will prefer B to A.
There are furfher permutations of this type of experiment, but these
need not concern us here. Successful experiments falling into this
category have been performed by Eloff (1936), Wilkes (1942), Dobzhansky and Spassky (1962), Connelly (1966) and Ogden (1970), on
the preferences of insects for water, temperature, gravity, and on
activity respectively. In at least one instance the changed preferences
selected for over a number of generations have been correlated with
genetic changes in the population (Dobzhansky and Spassky, 1962,
1967; Ehrman et ab., 1965; Dobzhansky et al., 1969), and it is genetic
changes of this sort which may in some cases prelude the origin of new
species.
Any part that habitat selection might play in the initial steps of
speciation can clearly occur only during sympatric speciation, that is,