Jersey). Using binoculars and a 60-power telescope, stu-
dents and I observe marked individuals from 4-meter-high
towers.
Because female prairie dogs rear their offspring in sep-
arate nursery burrows, maternity is usually easy to estab-
lish (Rayor 1988; Hoogland 1997, 1999, 2001). By sur-
rounding nursery burrows with livetraps on the day after
juveniles first appear aboveground, when they are about
5.5 weeks old, I capture, eartag, and mark all littermate sib-
lings before they mix with juveniles from other litters.
Juvenile prairie dogs are (1) individuals that have not yet
emerged from the natal burrow, or (2) individuals that first
emerged from the natal burrow 8 months ago (Hoogland
1995). Adults (including yearlings) are individuals that first
emerged from the natal burrow 8 months ago.
For each species of prairie dog that I have studied, almost
all my information has come from a single colony. Conse-
quently, I have been able to collect detailed, long-term data
from individuals of known age, kinship, and reproductive
success. I must assume, however, that information from a
single colony is representative of the entire species.
Experiments with stuffed badgers
In response to predators, black-tailed, Gunnison’s, and
Utah prairie dogs run to burrow mounds and then some-
times give an alarm call (King 1955; Waring 1970; Fitzger-
ald and Lechleitner 1974; Wright-Smith 1978). Predators
strike quickly at prairie dog colonies, however. A prairie
falcon (Falco mexicanus), for example, can capture a young
prairie dog and fly away in 15 seconds, and a coyote (Ca-
nis latrans) can run away with an adult or juvenile victim
30 seconds after charging into a colony. Accurate iden-
tification of alarm callers during such attacks is almost im-
possible. To investigate alarm calling, I use stuffed speci-
mens (N4) of a natural predator, the American badger
(Taxidea taxus), mounted in a lifelike pose and attached to
a Plexiglas sled (Hoogland 1995). Using fishing wire wound
around a garden hose reel, one person pulls the badger
across a family’s home territory, and another person rec-
ords whether or not each aboveground member of the fam-
ily gives an alarm call. We perform all experimental runs in
the 4 weeks after juveniles first emerge from their natal bur-
rows. The assumption is that prairie dogs respond to the
stuffed badger as if it were real. Attacks by live badgers in-
dicate that this assumption is valid for all three species.
Specifically, as in our experiments, some individuals call in
response to live badgers, others remain silent, and calling is
more likely for individuals with nearby kin (Hoogland
1995, unpublished data).
Alpine and hoary marmots (M. marmotaand M. cali-
gata) and several species of ground squirrels have at least
two alarm calls for different types of predators (e.g., aerial
versus terrestrial), or for different levels of danger from
the same predator (Balph and Balph 1966; Melchior 1971;
Taulman 1977; Davis 1984a; Hofer and Ingold 1984; Sher-
man 1985). Black-tailed, Gunnison’s, and Utah prairie dogs,
on the other hand, resemble Olympic marmots (M. olym-
pus;Barash 1973, 1989) and Columbian and thirteen-lined
ground squirrels (S. columbianusand S. tridecemlineatus;
Betts 1976; Matocha 1977; Schwagmeyer 1980) by having
only one distinct alarm call. Subtle variation in prairie dog
alarm calls, however, might indicate different types of pred-
ators or different levels of urgency (Slobodchikoff et al.
1986, 1991). For all three species, for example, individuals
seem to call at a faster rate (i.e., more barks per minute)
when danger is imminent (King 1955; Waring 1970), but
I have not quantitatively investigated this possibility. Nor
have I investigated why some individuals give short alarm
calls (5 seconds) whereas other individuals give longer
alarm calls (10 minutes).
The stuffed badger allows me to expose each prairie dog
to the same level of (simulated) danger, and also enables
me to obtain large sample sizes. For black-tailed prairie
dogs, results are from 4,000 responses of 323 individu-
als (Hoogland 1995); for Gunnison’s prairie dogs, results
are from 287 responses from 88 individuals (Hoogland
1996b); and for Utah prairie dogs, results are from 1,500
responses from 242 individuals. After five to ten experi-
mental runs, I calculate the percentage of times that each in-
dividual gave an alarm call, and then use a single data point
(i.e., the percentage) for each individual for all statistical
analyses. Prairie dogs call as often for late experimental runs
as for earlier runs (Hoogland 1983, 1995), and therefore
do not seem to habituate to the stuffed badger.
For all three species, calling by juveniles is rarer than
calling by adults, and for this chapter I only consider call-
ing by adults. Calling among adults does not vary system-
atically with age for black-tailed prairie dogs (Hoogland
1995), but I have not investigated calling versus age for
Gunnison’s and Utah prairie dogs.Observations of mating behaviors
Like females of other ground-dwelling squirrels (Hanken
and Sherman 1981; Sherman 1989; Boellstorff et al. 1994;
Murie 1996), females of black-tailed, Gunnison’s, and Utah
prairie dogs are sexually receptive for several hours on only
1 day of each year, usually in the afternoon. To detect cop-
ulations, students and I watch marked individuals from
dawn until dusk during the period of 3 –5 weeks when mat-
ings occur (Hoogland 1995, 1998a, 1998b, 2001). Most440 Chapter Thirty-Seven