preferred the interior of such patches (Lidicker and Peter-
son 1999).
A third intrinsic variable affecting colonization poten-
tial is the nature of a species’ social system. Although often
underappreciated as a conservation issue, social behavior
has the potential for making a real difference (Wolff 1999).
Strong social bonds provide a centripetal force, inhibiting
dispersal (Laurance 1990; 1991), and they also may make it
difficult for potential immigrants to establish in a new habi-
tat patch, by resisting recruitment of strangers to the social
group. Such resistance may furthermore be biased as to sex
or age of the would-be immigrant. Dispersers searching for
a new home will also be preferentially drawn to inhabited
patches as opposed to uninhabited ones. Successful colo-
nization of empty patches may only be possible if a group
of individuals either travel together (as in humans) or nearly
simultaneously arrive at an empty patch. Sex-biased disper-
sal may make it less likely that a suitable social group can
be assembled in this way. Taken together, there is likely a
negative, albeit noisy, relationship between colonizing abil-
ity and increasing levels of social complexity. By contrast,
the beneficial effects of social living should improve success
for groups moving into uninhabited or marginally produc-
tive habitat fragments.
An extreme form of spatial structure is for species or
populations to inhabit islands or island-like habitat patches.
For rodents this means actual islands (surrounded by wa-
ter) or terrestrial “islands” such as isolated mountaintops
(Lomolino et al. 1989). The essence of island life is that
there is complete or almost complete isolation from immi-
gration. Therefore, to the extent that this is true, persist-
ence depends only on the probability of local extinction,
there being almost no chance of the population being re-
established (rescued) from the outside. It is not therefore
surprising that the extinction rate for species restricted to
islands is much higher than for those on similar-sized main-
land fragments, which maintain at least some level of con-
nectivity with other patches (MacPhee et al. 1989; Stead-
man 1995; Brown and Lomolino 1998; Lomolino and
Perault 2001). MacPhee and Flemming (1999) reported
that of forty-three well-documented rodent species extinc-
tions in the last 500 years, 78.3% have been from islands.
In the West Indies, fifty-two species of rodents are known to
have been present at the close of the Pleistocene, but only
thirteen (25%) remain today, and almost all of these are in
a precarious state (Woods 1989).
Life and Death of Small Populations
When species become of conservation concern, they gener-
ally are persisting as small populations or are rapidly mov-
ing in that direction. Hence conservation biologists must
pay attention to the behavior of small and persistently de-
clining populations. In an ultimate sense, small populations
are at a high risk of extinction simply because their numbers
are inherently close to zero, and any negative perturbations
are likely to send them to their demise. Therefore, if popu-
lations are small, conservationists must work to maintain as
many of them as possible so that inevitable demic extinc-
tions can be reversed by immigration from surviving demes.
Given this first principle, there remain many other proper-
ties of small population size that will be useful to know if
management procedures are to be as effective as possible.
Factors that affect population numbers are of two sorts:
deterministic and stochastic (Lidicker 1978, 2002). The
first are influences that are either predictably related to pop-
ulation density or are consistent or regular in their impacts
while being unrelated to population density. Such factors
may vary temporally and /or spatially, but once understood,
the effects are predictable. Secondly, there are stochastic
(chance) influences. These factors occur inconsistently, ir-
regularly, and independently of the status of the focal pop-
ulation. Hence their effects are unpredictable, both in their
frequency and magnitude. Both kinds of factors influence
all populations, but small populations are especially vulner-
able to the latter, and hence they are of special concern in
the present context. Stochastic influences are of three types:
(1) environmental— factors extrinsic to the subject popu-
lation that unpredictably impact its size and welfare, that is,
weather, food supplies, arrival of novel predators, competi-
tors, or parasites, and other catastrophes, (2) demographic
— chance variations in population age structure, sex ratio,
or reproductive output, any or all of which could affect
demographic behavior, and (3) genetic — random changes
(drift) in the genetic constitution of the deme, including loss
of advantageous alleles or other polymorphisms, disruption
of coadapted genetic complexes, and arrival of new muta-
tions, most of which are likely to be deleterious. Of special
note is that these various stochastic influences are not mu-
tually exclusive; various combinations are likely to be im-
portant in small populations, and their effects can be mul-
tiplicative (synergistic rather than additive).
Small populations are also influenced by deterministic
factors. On the positive side, density-regulating factors (Li-
dicker 1978, 2002) exert their strongest growth-stimulating
influence at low densities and hence encourage popula-
tion growth. On the other hand, density antiregulating
forces may also be present, and these have the effect of
making small populations decline more rapidly and dense
populations increase faster (the Allee Effect, inverse density
dependence). They are inherently destabilizing, and may
drive small populations to extinction. When antiregulating
forces combine with regulating ones, they produce an un-458 Chapter Thirty-Eight