a corridor and the behavior of wildlife species that may attempt to use the cor-
ridor (Merriam 1984). The general assumption underlying the value of land-
scape connectivity is that a fragmented landscape that is interconnected is
more likely to support viable faunal and floral populations and intact ecolog-
ical processes than a landscape that is made up of only isolated fragments
(Harris 1984; Bennett 1998).
This assumption is based on two theoretical concepts: island biogeography
theory (MacArthur and Wilson 1967) and metapopulation models (Levins
1970). Island biogeography theory proposed that the number of species con-
tained on an isolated community (such as an oceanic island or forest frag-
ment) is the result of a dynamic equilibrium between opposing forces of col-
onization and extinction. The theory predicts that with increasing isolation
between forest fragments, there will be a decreasing rate of immigration by
species unable to traverse the modified habitat. Numerous studies have docu-
mented that land bridge islands (i.e., islands artificially created by flooding a
piece of land) and forest fragments lose species after isolation, a phenomenon
called species relaxation (Diamond 1972; Terborgh 1974; see also Chapter 2,
this volume).
Rather than examining entire species assemblages, metapopulation models
consider the population of a single species, which occurs in spatially separate
subpopulations that are connected by dispersal (Levins 1970; Forman 1997).
Although there are various types of metapopulation models, there are two gen-
eral forms. The first identifies a major source population that disperses out-
ward to smaller sink populations. This refers to a situation in which small
habitat fragments are partly separated from a larger area of intact habitat. The
second is a population that is patchily distributed throughout the landscape
and connected by dispersal, as would occur when only small forest fragments
remain in a formerly forested landscape. Small populations in the models are
prone to local extinction, and the movement of individuals between patches
can both bolster dwindling populations via their genetic and demographic
contributions (called the rescue effect; Brown and Kodric-Brown 1977) and
result in the recolonization of local patches where the species has gone extinct.
In both island biogeography theory and metapopulation models, the
extinction rate of a species depends on the size and quality of the remnant
habitats, whereas recolonization depends on the level of landscape connectiv-
ity. Various landscape features may affect connectivity, such as the distance
between remnant patches, the type of surrounding matrix (modified habitats),
and the presence of corridors or small habitat patches that can function as
stepping stones.
Corridors were first proposed for conservation planning in 1975, based on
fragmentation and island studies (Diamond 1975; Wilson and Willis 1975).
Since that time many studies have demonstrated the major benefits of corri-
dors, which include facilitating wildlife movement, providing habitat, and
- Landscape Connectivity and Biological Corridors 51