other three pups (fig. 26.1g). Hobbling on her injured leg,
she carries her pups one by one to a burrow about 25 me-
ters away, where the resident male accepts them.
The next day 9G returns to the site of her encounter, ex-
hibiting little evidence of impairment from her snakebite
as she tail flags and cautiously examines burrow entrances
(fig. 26.1h). But the snake has moved deep into 9G’s former
burrow system, and she does not enter and locate it there.
The snake resumes hunting 2 days later, continuing in that
area for the next 6 days, but captures no more pups among
this alerted kinship cluster.
The Variety of Links between Social
and Antipredator Systems
As this squirrel-snake episode illustrates, social and anti-
predator systems are linked in a variety of ways. We will
discuss several categories of such connections, including
shared components and linkages between means (e.g., so-
cial behavior) and ends (e.g., avoiding predation).
Shared components at the level of motor patterns
Many motor patterns are sufficiently broad in their ap-
plicability that they are obviously usable in both social
and antipredator contexts, including locomotion, biting,
pouncing, parrying, elongate investigatory postures, and
freezing in quadrupedal and bipedal postures. Other motor
patterns, however, require more extensive observations be-
fore their use in both contexts becomes evident; these in-
clude substrate throwing and alarm calling.
Substrate throwing with fore or hind paws is often used
in the context of burrow excavation and repair, which is
perhaps its original functional context. But the proximate
and ultimate processes that make behavioral systems op-
portunistic have added this activity to a defensive context,
often called defensive burying (Terlecki et al. 1979; De Boer
and Koolhaas 2003). Defensive burying is typically hy-
pothesized to serve an antipredator function (Dell’omo
et al. 1994), an idea inspired by observations that substrate
throwing is used by ground-dwelling sciurids to deal with
snakes — for example, by entombing them in a burrow
(Merriam 1901; Coss and Owings 1978; Halpin 1983;
Hersek 1990). But substrate throwing is used for additional
antipredator purposes, including harassing snakes above
ground (fig. 26.1f; Owings and Coss 1977; Halpin 1983),
as well as simply getting a snake’s attention (Hennessy and
Owings 1988). More broadly yet, substrate throwing is
also used in social contexts to harass and even entomb con-
specific interlopers (Armitage and Downhower 1970; Levy
1977; personal observations of wild Rattus norvegicus).
To our knowledge, no work has systematically explored
how the form of substrate throwing has been modified with
the acquisition of these social and antipredator functions.
For example, the most common way to move earth during
excavation is to kick it out behind the animal with the hind
paws. It is possible that substrate throwing with the fore-
paws, as California ground squirrels and rock squirrels
(Spermophilus variegatus) do, had its evolutionary origins
in burrow plugging as an antisnake defense below ground,
a context in which it might be important to continue mon-
itoring the snake by facing it. This hypothesis, if true, begs
the question of why black-tailed prairie dogs (Cynomys lu-
dovicianus) typically (but not always) use their hind paws
to kick substrate at snakes (Halpin 1983; Shier, personal
communication).
Alarm calling is another class of behavior that has proven
to be functionally diverse. Ground-dwelling sciurid alarm
calling provided some of the earliest and best evidence of
nepotistic behavior (Dunford 1977a; Sherman 1977; Hoog-
land 1983), and subsequently also yielded evidence that
some alarm calling is self-interested rather than nepotistic
(Sherman 1985; Owings et al. 1986). More recently, atten-
tion has turned to questions about what predator-related
information is available in these calls (Owings and Leger
1980; Slobodchikoff et al. 1991; Blumstein and Arnold
1995; Blumstein and Armitage 1997a; Blumstein 1999a).
But some species also often use these calls socially, a
rarely explored phenomenon (but see Smith et al. 1976,
1977; Leger et al. 1980; Owings and Leger 1980). At least
eight species of ground-dwelling sciurids use alarm calls in
sexual and agonistic contexts (Owings and Hennessy 1984).
Such social use of these calls is common in at least some
species, and might be expected to dilute the antipredator
function of these calls over evolutionary time. For example,
at least one tree squirrel species (Formosan squirrel, Cal-
losciurus erythraeus) uses the same type of call in both post-
copulatory and terrestrial-predator alarm contexts, as Cal-
ifornia ground squirrels do (Boellstorff et al. 1994); these
tree-squirrel calls do not differ structurally, and do evoke
the same suppression of conspecific movement during play-
backs (Tamura 1995).
Two questions are raised by the idea that the antipreda-
tor function of calls can be diluted by using the same call
type in social contexts. First, are conspecifics or the preda-
tor the primary targets of these antipredator calls? The di-
lution idea assumes conspecifics to be the primary targets.
If, however, the predator is the primary target (Owings and
Hennessy 1984; Hersek and Owings 1993), social use of
calling may not interfere with its antipredator function.
Second, how finely differentiated is the call system? In Cal-
ifornia ground squirrels, for example, detailed structural
analyses of the same call types in social and antipredatorSocial and Antipredator Systems: Intertwining Links in Multiple Time Frames 307