success. According to the resource defense hypothesis (Ost-
feld 1985; 1990), females defend their territories to provide
exclusive access to food. In contrast, the pup defense hy-
pothesis (Sherman 1980a, 1981b; Wolff 1993b; Wolff and
Peterson 1998) posits that the main function of female ter-
ritoriality is to protect vulnerable offspring from infantici-
dal conspecifics.
Considering these hypotheses with regard to the breed-
ing season, both hypotheses predict that territory defense
should be most pronounced during lactation. There are sev-
eral predictions, though, that should allow us to distinguish
between them.
- The resource defense hypothesis predicts that female
territorial behavior should begin to increase follow-
ing fertilization, whereas the pup defense hypothesis
predicts that it is not until after pups are born that fe-
male aggression to intruders should increase (Sherman
1981b). - According to the pup defense hypothesis, females
should only behave aggressively toward potential infan-
ticidal conspecifics (Wolff and Peterson 1998). In some
species infanticide is committed by male conspecifics
(e.g., house mice, Labov et al. 1985; deer mice, Wolff
and Circirello 1991); female conspecifics (e.g., meadow
voles, Webster et al. 1981; field voles, Microtus agres-
tis,Agrell 1995); or conspecifics of both sex (e.g., some
ground squirrel species, Sherman 1981b; Michener
1983a; Vestal 1991; white-footed mice, Wolff and Ci-
cirello 1991; bank voles,Clethrionomys glareolus,
Ylönen et al. 1997 and some species of marmots, Brody
and Melcher 1985; Blumstein 1997). Furthermore,
females of some species are much more likely to kill
the offspring of nonrelatives than relatives (e.g., Beld-
ing’s ground squirrels, Spermophilus beldingi,Sherman
1981b; Columbian ground squirrels, Stevens 1998).
In these species, females should be more aggressive to-
ward intruders that are nonkin. In contrast, the food
defense hypothesis predicts that females in all species
would be equally aggressive toward intruders regardless
of sex, and only very close kin would be allowed to
feed on the territory. - Home range size should decrease at the onset of lac-
tation according to the pup defense hypothesis, but
may increase according to the food defense hypothesis
(Boutin 1990; Wolff and Schauber 1996; Koskela et al.
1997).
Previous studies show that females do become more ag-
gressive immediately after parturition and during lacta-
tion (Svare and Gandelman 1976; Ayer and Whitsett 1980;
Sherman 1980a, 1981b; Ostermeyer 1983; Maestripieri
1992; Koskela et al. 1997), as predicted by both hypothe-
ses. Aggression toward conspecific females is seen in some
species in which females are the infanticidal sex (Madison
1980a; Koskela et al. 1997; Ylönen et al. 1997), as pre-
dicted by the pup defense hypothesis. Although intersexual
aggression is uncommon in mammals, females are aggres-
sive toward males (Agrell et al. 1998) in species in which
infanticide is committed by males (vom Saal 1984; Wolff,
J. O., 1985c; Mallory and Brooks 1978), or males and fe-
males (Wolff, J. O., 1985c).
Home range size in female bank voles decreased during
the time between mating and lactation, at the time when the
energetic demands of reproduction are increasing (Koskela
et al. 1997). These data are inconsistent with the prediction
of the food defense hypothesis but support the pup defense
hypothesis. In addition, supplemental food did not result in
decreased home ranges in field manipulations of prairie and
meadow voles populations when density is controlled, sug-
gesting that females may not be food limited as expected,
according to the resource defense hypothesis (Desy et al.
1990; Fortier and Tamarin 1998). Thus existing data are
more consistent with the pup defense hypothesis than with
the resource defense hypothesis.Territory bequeathal
After offspring are weaned they have three options: to dis-
perse from their natal nest, remain philopatric and share
space and resources with their mother (or parents), or re-
main at their natal nest after their mother/parents disperse.
In most species, offspring disperse around the time of wean-
ing (Anderson 1989), but in some, juveniles may remain
and inherit some or all of their natal territory after their
parents’ death (Anderson 1989). In a few species, adult fe-
males have been observed to disperse when offspring are
about weaning age (King 1955; Harris and Murie 1984;
Price and Boutin 1993). This post-breeding dispersal may
be a form of extended parental investment or parental fa-
cilitation (Brown 1987), in which parents allow their off-
spring to remain on their natal territory, where these off-
spring have access to necessary resources such as food and
a nest site (Myllymaki 1977).
The territory bequeathal hypothesis makes three critical
predictions:- Parents with weaned offspring are more likely to dis-
perse than those with no offspring around weaning age
(for example, Harris and Murie 1984; Price and Boutin
1993). - Offspring will remain within their natal home range af-
ter it is vacated by parents (Lambin 1997).
Reproductive Strategies in Female Rodents 53