more effectively isolated than real islands that hap-
pen to be located within a few hundred metres of a
mainland. Many isolates are actually sufficiently
close to one another that their populations are in
effect linked: they form metapopulations, the
second theoretical element that we will consider.
When we have added these tools to our armoury,
we can then consider the whole-system implica-
tions of fragmentation and how ‘island’ approaches
have contributed to their understanding.
Over the past few decades there has been a shift
in ecology from what Pickett et al. (1992) term the
balance of natureparadigm, which sees the world
as a self-regulating, balanced system, towards the
flux of natureparadigm, which Pickett et al. (1992)
believe provides a more realistic basis for conserva-
tion planning and management. This framework
does not deny the possibility of equilibrium, but
recognizes equilibrium as merely one special state.
The flux of nature viewpoint stresses process and
context, the role of episodic events, and the open-
ness of ecological systems.
Conventionally, the ‘island theory’ input to con-
servation debates has been based on the general
assumption that a ‘natural’ area is in a state of bal-
ance, or equilibrium. Humans then intervene to
remove a large proportion of the area from its natu-
ral state, leaving behind newly isolated fragments,
which are thus cast out of equilibrium. Subsequently,
these fragments must shed species as the system
moves to a new equilibrium, determined by the area
and isolation of each fragment. This way of thinking
is entirely consonant with the balance of nature
paradigm.
By contrast, in the present volume we have
stressed the importance of considering physical
environmental factors in island biogeography, and
have shown that ecological responses to environ-
mental forcing factors are often played out over too
long a time frame for a tightly specified dynamic
equilibrium to be reached. Moreover, relatively few
areas of the planet, if any, are pristine, untouched
by human hand, so most areas that we may be frag-
menting already contain the imprint of past land
uses (e.g. cultivation and abandonment, hunting)
(e.g. Willis et al. 2004). Hence, if the starting
assumptions concerning the prior equilibrial state
of the system are flawed, some of the theoretical
island ecological effects may, in practice, be fairly
weak. These points are addressed in the present
chapter in the applied, conservation setting. Hence
we will consider the physical effects of fragmenta-
tion as well as the biotic, and we will examine what
sorts of factors in practice cause species to be lost
from fragments.10.3 Minimum viable populations and minimum viable areas
How many individuals are needed?
What is the minimum viable population(MVP)?
By this we mean the minimum size that will ensure
the survival of that population unit, not just in the
short term, but in the long term. It is often defined
more formally in terms such as the population size
that provides 95% probability of persistence for 100
or for 1000 years. Attempts to calculate the viability
of single populations, i.e. whether the population
exceeds the minimum requirement, are termed
population viability analyses (PVA). Important
elements that have to be considered in PVA include
demographic stochasticity, species traits, genetic
erosion in small populations, and extreme events
such as hurricanes or fires.
Demographic stochasticity of initially small pop-
ulations can lead to losses from a series of isolates
without a need to invoke any common factor such
as predation or loss of fitness. However, where
small populations persist for a reasonable length of
time, they may also lose genetic variability as they
pass through bottlenecks. They may then lack the
genetic flexibility to cope with either the normal
fluctuations of environment or an altered environ-
ment, and they may also accumulate deleterious
genes. In short they lose fitness(Box 10.1). It is
generally held as axiomatic that an increase in
inbreedingin small populations reduces fitness in
animals, although unambiguous demonstrations of
the effect in natural populations are relatively
scarce (Madsen et al. 1996). The basic rule of con-
servation genetics is that the maximum tolerable
rate of inbreeding is 1% per generation. This trans-
lates into an effective population size of 50 toMINIMUM VIABLE POPULATIONS AND MINIMUM VIABLE AREAS 253