However, Hare (1992, 1994) suggested that this discrimi-
nation ability is transitory and that older juveniles fail to dis-
tinguish between littermates and nonlittermates, although
the results of his behavioral tests with temporarily captive
juveniles are difficult to interpret because of the large range
in ages tested (e.g., 9 – 61 days after natal emergence). In
many rodent species, individuals’ social worlds expand as
young leave their natal environment and begin to encoun-
ter distantly-related and unrelated conspecifics (Nunes,
chap. 13 this volume). We are unaware of any longitudinal
rodent kin-recognition studies that have assessed an indi-
vidual’s discrimination abilities at various ages, before and
after different kinds of social experience. Such studies would
be valuable because they would help address developmen-
tal questions like whether experience at a particular age
(e.g., just prior to weaning or dispersal, when litters begin
to dissolve) has a long-term impact on kin recognition.
Rodent Kin Recognition —Empirical Evidence
In this section, we describe and analyze what is known
about rodent kin-recognition abilities at the proximate and
ultimate levels of analysis. Because this literature is vast (see
reviews in Blaustein et al. 1987; Dewsbury 1988; Halpin
1991; Mateo 2003, 2004), we concentrate on a few well-
studied species.
Behavioral ecologists in North America have made de-
tailed studies of social behavior in three genera of ground-
dwelling squirrels (Cynomys, Marmota,andSpermophilus).
These animals are diurnal, readily observable due to their
size and preference for open habitats and because individ-
uals can be uniquely marked for long-term identification
(Murie and Michener 1984). Field studies in the 1970s (e.g.,
Michener 1973a; Barash 1974c; Dunford 1977b; Sherman
1977) found that social interactions often varied with kin-
ship in several ground-dwelling squirrels, and Michener’s
(1983a) review of sociality in this group emphasized that
social organization was based on genetic relatedness be-
tween females, with mother-offspring groups as a common
social unit (reviewed in Hare and Murie, chap. 29 this vol-
ume). Accordingly, various investigators (e.g., Sheppard
and Yoshida 1971; Michener 1974; Davis 1982b) began
to investigate kin-recognition abilities, especially in the ge-
nus Spermophilus,hereafter “ground squirrels” (reviewed
by Schwagmeyer 1988a; Mateo 2003; Holmes 2004). We
use “littermate” rather than “sibling” in our discussion of
ground squirrel kin recognition because multiple paternity
is common in Spermophilus(see references in Lacey and
Wieszorek 2001; Solomon and Keane, chap. 4, this vol-
ume) so that litters routinely comprise full- and maternal
half-siblings.Kin recognition in Belding’s ground squirrels
The kin-recognition abilities of Belding’s ground squirrels
(fig. 19.2) have received considerable empirical attention
after field studies (Sherman 1977, 1980a, 1981a, 1981b,
1985) showed that females displayed nepotistic behavior in
the context of antipredator warning calls (Sherman 1977,
1985) and cooperative defense of unweaned young by re-
lated females to prevent infanticide (Sherman 1980a, 1981a,
1981b). Cross-fostering procedures and a combination of
laboratory and field experiments revealed that (1) year-
lings discriminate between familiar littermates (young born
in the same litter) and unfamiliar nonlittermates, (2) female
yearlings discriminate between unfamiliar (reared apart)
female littermates and unfamiliar female nonlittermates,
and (3) mothers discriminate between their own familiar
young and alien young borne and reared by other moth-
ers (fig. 19.3). These results suggested that two recognition
mechanisms, prior association and phenotype matching,
are both involved in these instances of differential treatment
(Holmes and Sherman 1982, 1983). We examine these
mechanisms in detail because they also mediate kin recog-
nition in several other species (see the following).Prior association
In this recognition mechanism, an individual learns the
phenotypic traits of conspecifics with which it interacts di-
rectly (e.g., nestmates), stores them in its memory, and later
distinguishes between familiar and unfamiliar conspecifics.
If early rearing environments comprise only kin (e.g., only
full siblings share a nest) or if individuals have opportuni-220 Chapter Nineteen
Figure 19.2 Three juvenile Belding’s ground squirrels that have come above
ground for the first time from their underground natal nest. Juveniles can dis-
criminate between their littermates and nonlittermates, based primarily on olfac-
tory cues, and they can also recognize some classes of unfamiliar kin when they
are first encountered. Photo by J. M. Mateo.