have been suggested to explain this relaxation in
territoriality:
●Resource hypothesis. This hypothesis suggests
that territorial behaviour is primarily adjusted to
resource densities, and that resources on islands
are, because of the lack of competitors resulting
from isolation, more abundant than in the main-
land. More resources will result in reduced territory
sizes, increased territorial overlap, and changes in
the vigour of territorial defence on islands (Stamps
and Buechner 1985).
●Defence hypothesis. In addition to any effects of
resources, the buffer effect exerted by the ocean on
island weather will result in milder climates and
this, together with the absence of predators, will
result in the survival of more conspecific individuals
without territory. These ‘have-nots’ will intrude and
contend with those that own territories. This will ele-
vate the costs of defence for owners of insular terri-
tories, selecting for reduced territories, increased
territorial overlap, and acceptance of subordinates. If
defence costs became exaggerated, animals might
benefit by expending less energy in territory defence
and reallocating their resources into producing
fewer but more competitive young. Thus, this
hypothesis links the observed trend of small clutch
sizes of insular birds to territory size via the mecha-
nism of defence costs (Stamps and Buechner 1985).
As with many of the phenomena discussed in
this chapter, these ideas are interesting, but at
present we cannot conclude which of them might
hold greater explanatory power across different
islands.The island syndrome in rodents
Although the various evolutionary changes in eco-
logical niche in island populations have mostly
been considered separately so far, it will be appar-
ent to the reader that many of the changes evolve
together. A proper understanding of individual
changes requires models that incorporate several
such phenomena simultaneously. Adler and Levins
(1994) set out to construct such a model for island
rodents, synthesizing results from a range of empir-
ical studies, principally of mice and voles. They
noted that island populations of rodents tend to
evolve higher and more stable densities, better sur-
vival, increased body mass and reduced aggres-
siveness, reproductive output, and dispersal. It is
striking that island rodent populations of different
species and from disparate geographic areas, often
demonstrate similar sets of patterns. Adler and
Levins termed these collective island-mainland dif-
ferences the island syndrome. They summarize the
main traits and some of the more likely explana-
tions for them in tabular form (Table 7.3).192 ARRIVAL AND CHANGE
Table 7.3Short-term and long-term changes in island rodents and proposed explanations (from Adler and Levins 1994, Table 2)
Island trait Proposed explanation
Reduced dispersal Immediate constraint (short-term response) and natural selection
against dispersers (long-term response)
Reduced aggression Initially, reduced population turnover, greater familiarity with
neighbours, and kin recognition. Long-term directional selection
for reduced aggression
Crowding effect Isolation (‘fence effect’ resulting from reduced dispersal) and
reduced number of mortality agents such as predation, both of
which result in crowding of individuals and consequently higher
population densities
Greater individual body size Initially, a reaction norm as a response to higher density.
Long-term directional selection for increased body size in
response to increased interspecific competition
Lower reproductive output per individual Initially, a reaction norm as a response to increased density.
Long-term directional selection in response to decreased
mortality
Greater life expectancy (higher survival probabilities for individuals) Reduced number of mortality agents such as predation