254 ISLAND THEORY AND CONSERVATION
Box 10.1 Fluctuating asymmetry in a fragmented forest system
Fluctuating asymmetry (FA) refers to differences
between the right and left sides in animal
anatomical characteristics that are usually
bilaterally symmetrical, and is a form of
developmental instability that may have genetic
and/or environmental causes. Whatever the
nature of the control, a high incidence of FA is
regarded as a sign of a population under
environmental stress. Recent work has suggested
that small, isolated populations in fragmented
habitats may exhibit a higher incidence of FA.
Anciães and Marini (2000) test this hypothesis for
wing and tarsus FA on 1236 mist-netted
passerine birds of 100 species captured from 7
fragments and 7 control areas in the Atlantic rain
forest area of Brazil. They recorded a higher
mean level of FA in both features in the forest
fragments than in the control areas, suggesting
that passerine birds in these forests exhibit
morphological alterations as a result of forest
fragmentation and habitat alteration. Possible
proximal causes might include increased pressures
from predators, competitors, parasites anddisease, and/or decreased food supplies, shelter
or nest site availability.
In general, terrestrial and understorey
insectivores showed greater relative FA changes
than other foraging guilds, in line with the general
sensitivity of this guild to habitat fragmentation. An
example is Dysithamnus mentalis, which forages in
the medium strata and may also follow army ants.
It has previously been found to be an area-sensitive
species, and indeed was observed in only two of
the fragments in the study, where it showed higher
levels of FA in the tarsus than in the control forests.
By contrast, the most abundant frugivores sampled
in the study (Chiroxiphia caudate, Manacus
manacus, and Ilicura militaris) were commonly
found in altered forest patches and seemed
unaffected by fragmentation in their incidence and
in terms of FA values.
This form of developmental instability may
provide a form of early warning system, indicating
stresses on vertebrate species in isolates prior to
more profound changes, such as rapid population
declines and species extinctions.ensure short-term fitness according to the calcula-
tions of I. A. Franklin in 1980 (reported in Shafer
1990). However, there will still be a loss of genetic
variation at such a population size, and Franklin
recommended 500 individuals in order to balance
the loss of variation by gains through mutation.
These commonly cited values appear to derive
respectively from data from animal breeders and
bristle number in Drosophila, and so should be
regarded as merely first approximations (Fiedler
and Jain 1992). Lacy (1992) noted that crudely
estimated MVPs for mammals based on data for
inbreeding effects from captive populations varied
over several orders of magnitude, and that it would
be premature to generalize on MVPs in the wild on
this basis.
Typically, only a proportion of the adultpopula-
tion participates in breeding; and it is these animals
that form the effective population size, which is
often substantially smaller than the total popula-
tion size (Shafer 1990). A study of grizzly bears in
the Yellowstone National Park showed that to pre-
vent inbreeding rates exceeding 1% required an
overall population size of at least 220 rather than 50
animals (Shafer 1990).
It is intuitively reasonable that the loss of a broad
genetic base is likely to increase the probability of
extinction. However, we learnt in an earlier chapter
that putting a population through a bottleneck of
just a few individuals may not be as damaging to
the genetic base as one might anticipate, providing
the population is allowed to expand again fairly
quickly. In illustration, the great Indian rhino was
reduced within the Chitwan National Park in
Nepal (one of its two main havens) to an effective
population of only 21–28 animals in the 1960s. In
this case the bottleneck was short. Rhinos have a
long generation time, and it has since recovered to