be influenced by poor habitat quality and the lack of suit-
able dispersal sites (Hodgdon 1978; Brooks et al. 1980;
Smith 1997; Busher and Lyons 1999).
Thus a beaver colony is typically a family with as many
as three different related age groups of individuals present;
both species exhibit this type of social group. Variations
in retention of older, nonreproductive family members, as
well as summer wandering by family members, have also
been reported. Density and habitat quality appear to be the
proximate factors influencing the composition of a beaver
family.
Evidence of polygamous mating
Few studies have reported more than one reproductive fe-
male in a family group. Bergerud and Miller (1977) in New-
foundland found that three of twenty-seven (11.1%) groups
had more than one adult male and three of twenty-seven
(11.1%) groups had more than one adult female. However,
only one of twenty-seven (3.7%) groups had two lactating
females. Brooks et al. (1980) reported that 11.1% of the
groups studied in Massachusetts had more than one adult
female and 11.1% had more than one adult male, but no
group had more than one lactating female. Schulte (1993)
found 18.4% of his groups had additional adults, but also
reported no additional lactating females. However, in this
same population a few years later 8.57% (three of thirty-
five families) had two breeding females living with one male
(Sun 2003). Busher et al. (1983) reported multiple lactating
adult females in 87.5% (seven of eight groups over 2 years)
of the groups in a small isolated population of beavers in
Nevada. They also observed more movement among group
territories by adult males, adult females, and 2-year-olds,
but not by yearlings. This population had an extremely high
density, with close spacing between colonies, and there was
little if any additional available habitat into which matur-
ing beavers could disperse. Wheatley (1993), in Manitoba,
also found one case of two lactating females living within
the same lodge.
Thus while a few authors have reported more than one
lactating adult female in a family, these represent a very
small percentage of studies, and polygynous mating is far
from frequent in beavers, as suggested by Sun (2003). In
fact, throughout North America, while the group compo-
sition may range from single adults (or dispersing young
adults) to mated pairs, to families with one or two genera-
tions of young to other combinations of a single adult and
young, incidences of polygyny have rarely been observed.
Given the large latitudinal (and elevational) distribution
of beavers and the various habitat conditions (ranging from
large lakes and rivers to small ponds and streams) where
beavers are found, greater variety in the mating system could
be expected. However, the consistent social group observed
is the family (or extended family), with monogamous mat-
ing between adults. The situation appears to be the same for
the Eurasian beaver (Wilsson 1971; Rosell and Pedersen
1999; Sharpe and Rosell 2003).
Only under conditions of high density does the strictly
monogamous pair bond become compromised; and further
studies of beavers living in high-density populations are re-
quired to adequately evaluate the influence of density on
family composition and the pair bond.Duration of the pair bond
The general pattern in both species is for the pair bond
to exist for multiple years and that the adults are closely
associated during this time. Pair bond duration in eight bea-
ver pairs over a 19-year period in a California population
ranged from 1– 6 years, with a mean of 3.6 years (Taylor
1970). A 12-year study in Ohio found the average duration
of the bond to be 2.5 years for twenty-six pairs (Svendsen
1989). In California, a 3-year study reported that three of
four pairs remained together for 3 years (Busher 1980),
while a longer-term study recorded a mean duration for
thirteen pairs of 1.9 years (Woodard 1994). A study in New
York reported the average pair bond duration to be 4.3
years, with the longest duration 9 years (Sun 2003). Only
one case of desertion was documented in these studies (a fe-
male leaving a male), with most pair bonds dissolving due
to documented or presumed death of a mate. Since mating
generally occurs in winter, maintaining a pair bond would
ensure access to a mate, especially at latitudes or altitudes
where mobility is restricted by ice and snow.
The cost of reproduction provides support for the long-
term maintenance of the pair bond. For example, female
beavers (introduced Castor canadensis) in Finland that re-
produced in 2 consecutive years produced more offspring
(as measured by fetus production) than females who only
mated in 1 of the 2 years (Ruusila et al. 2000). Maintain-
ing the pair bond over a series of years should increase the
potential for mating, encourage reproduction in consecu-
tive years, and increase the lifetime reproductive success of
the female. Additional long-term studies of marked popu-
lations are needed, and could add to our understanding of
the pair bond duration and the conditions under which they
terminate.Social Organization within the FamilyOur understanding of beaver social behavior has been ad-
vanced by many recent studies (Tevis 1950; Schramm 1968;
Wilsson 1971; Hodgdon and Larson 1973; Hodgdon 1978;Social Organization and Monogamy in the Beaver 283