their foraging behavior. In a study on voles exposed to three
nights of predators, the voles shifted from their normally
polyphasic activity pattern toward a decidedly nocturnal
activity pattern (J. Eccard et al. 2001). In this study, the
main predator, the least weasel, is primarily diurnal (Sun-
dell et al. 2000). In both studies by Kotler et al. (1988) and
by J. Eccard et al. (submitted ms.), the predators were free-
ranging, actively hunting, and posing a real and imminent
threat to the rodents.
An important issue in experimental tests of predation
risk concerns mimicking increased risk. Risk has most often
been simulated by either olfactory or visual cues, and more
rarely by the introduction of actual predators capable of
preying upon the experimental animals. It may be that per-
sistent indirect cues of predation risk such as microhabi-
tats and moon phases have fixed, long-term effects, whereas
some short-lived, direct cues of predation risk may simply
cause spiked responses because the animal only responds to
a perception of imminent peril.
Youth and predation
A common perception among vole ecologists is that young,
inexperienced voles and males are more vulnerable to avian
predation (e.g., Pearson and Pearson 1947; Lagerström and
Häkkinen 1978; Halle 1988; Mappes et al. 1993), whereas
female voles may be the preferred target of mammalian
predators because of scents associated with estrus (Cushing
1985). Both generalizations are not fully supported by re-
cent prey selection studies. Koivunen et al. (1996) identified
a “doomed surplus” of animals in which young field voles,
regardless of sex, were more prone to owl predation than
older individuals (but see Koivunen et al. 1998). While in-
experience may contribute to vulnerability in young voles,
their patterns of habitat use and mobility enforced by social
circumstances may play a larger role. Increased mobility
increases individual vulnerability to avian predation (Nor-
rdahl and Korpimäki 1995, 1998). In response to experi-
mental reductions in small mustelid predators, all voles
increased their mobility (Norrdahl and Korpimäki 1998).
Furthermore, in accord with Cushing’s (1985) prediction,
small mustelids preferentially prey upon female voles.
Desert rodents such as the kangaroo rat differ from most
voles by having smaller litter sizes, fewer litters per year,
and greater investment in young. Whereas young voles may
become independent at less than one-half final adult mass,
young kangaroo rats do not become independent until they
are around two-thirds or three-quarters adult mass. This
greater investment in young may be an antipredator strat-
egy that insures higher chances of survival of the young.
Patterns of dispersal in kangaroo rats may further reduce
predation risk to the young. For example, bannertail kan-
garoo rats maintain elaborate multi-entranced burrow sys-
tems that appear as mounds (ca. 3 m across and 0.5 m high)
across the landscape. Mothers may either disperse them-
selves, leaving their present mound to an offspring, or pro-
tect an unoccupied mound that is then bequeathed to an
offspring (Randall 1993). In the more nomadic Merriam’s
kangaroo rat, adults rather than juveniles frequently dis-
perse to new home ranges. In both species, females may be
indirectly bequeathing safety to their offspring.
Juvenile desert rodents purportedly face greater risks of
predation than adults for two possible reasons (however,
empirical proof of this conjecture is lacking at present). First,
juveniles may be less experienced in detecting and evad-
ing predators. In fact, juvenile eastern chipmunks (Tamias
striatus) appear highly susceptible to avian predation (e.g.,
Cooper’s Hawk, Accipiter cooperii;Bosakowski and Smith
1992). Second, juveniles, even more than adults, may bias
their foraging toward high-food, high-risk microhabitats be-
cause of their lack of foraging experience, less efficient for-
aging, or greater need for energy.
The open landscapes of deserts may be uniquely risky
to rodents faced with avian and mammalian predators. Fe-
male kangaroo rats may reduce predation risk to their off-
spring by (1) postponing independence, (2) bequeathing
burrows of known location, and (3) bequeathing a feeding
range with a known landscape of fear.Odors and olfactory signals
Odors emitted by voles (and presumably other taxa of ro-
dents) vary with gender, age, and social status. These odors
influence social interactions and communication. Ylönen
et al. (2003) studied the effect of sex- and age-specific odors
on the ability of weasels to hunt reproductive females, ma-
ture males, and nonreproductive females and males. They
found no demonstrable olfactory biases in the weasels’ hunt-
ing success — to a hunting weasel, lunch is lunch.
Olfactory signs of a prey individual are but one part in
the hunting sequence of a predator (Ylönen et al. 2003).
While directing a predator, like a mustelid, to sites with
abundant prey, olfactory signatures may play less of a role
than acoustic signals in facilitating a successful attack on
a vole. Movement by males and young may provide visual
and acoustic cues, while females attached to nest sites may
consequently leave more olfactory signals for predators (Pu-
senius and Ostfeld 2000).
Female deer mice (Peromyscus maniculatus) frequently
carry newborn pups to new nest sites (Sharpe and Mil-
lar 1980). This relocation may serve to deter predators by
reducing cues that might otherwise accumulate. Also as aFear and the Foraging, Breeding, and Sociality of Rodents 333