polygyny. Defense polygyny occurs if males are territorial
and /or defend access to females during the breeding sea-
son, whereas lack of any form of defense is considered non-
defense polygyny (Dobson 1984). Defense polygyny has
also been called male territoriality, female-defense polyg-
yny, harem-defense polygyny, resource-defense polygyny,
and male-defense polygyny (Schwagmeyer 1990). Armitage
(1986b) and Schwagmeyer (1990) argue that these distinc-
tions do not reflect the real differences in reproductive be-
havior, and suggest that female defense polygyny is the
appropriate term. Non-defense polygyny may include dom-
inance hierarchies or competitive searching (i.e., scramble
competition). Whether defense or non-defense polygyny
occurs is determined by the costs of defense. The economi-
cal defensibility of mating territories is affected by (1) pop-
ulation density, (2) the spatial distribution of females, and
(3) the temporal distribution of receptive females (Emlen
and Oring 1977).
As density increases, so does the number of competitors,
making territory defense uneconomical (“the density-male
hypothesis,” Dobson 1984). This hypothesis predicts that,
for species that are not widely dispersed, non-defense po-
lygyny will occur if density is high during the mating sea-
son, and defense polygyny will occur when densities are
low (Dobson 1984). Dobson (1984) tested this hypothesis
in polygynous ground-dwelling squirrels. Using only the lo-
cal density of males during the mating season, the hypoth-
esis was not supported, but when local density of both sexes
was examined, the predictions of the hypothesis were sup-
ported. Dobson concluded that the cost of monopolizing fe-
males was affected not only by an increase in the number of
male intruders, but also by the cost of having to sequester
more females.
As discussed previously, mating patterns can be flexible,
and species with male territoriality may respond to envi-
ronmental changes when territoriality becomes economi-
cally expensive. Males defend territories at low densities
and form dominance hierarchies at high densities in species
such as mouse mice (Mus musculus;Davis 1958), white-
footed mice, and deer mice (Wolff 1989). Grey-sided voles
can switch to territoriality if the density of competitors is
low (Ims 1987b).
Spatially clumped females are more economical to de-
fend than females that are spatially dispersed, leading to
territoriality. In ground-dwelling squirrels, which live in rel-
atively open areas, female clumping and low densities are
characteristic of male territoriality (Dobson 1984). How-
ever, female clumping does not result in territoriality in
voles, as habitat structure may make territoriality and de-
tection of intruders difficult for many small mammals (Ims
1988).
Temporal clumping, as well as spatial clumping, affects
male mating strategies. The timing or degree of synchrony
of sexual receptivity by females has a profound effect on
the operational sex ratio (OSR), the “ratio of fertilizable
females to sexually active males at any given time” (Emlen
and Oring 1977, 216). As female receptivity becomes more
asynchronous, the OSR becomes more biased toward males,
which should intensify male mating competition (Clutton-
Brock and Parker 1992) and change male mating strategies.
If the OSR is highly skewed toward males, then territorial-
ity would be uneconomical, as any territorial male would
have to defend against many intruders, and non-defense
polygyny would be predicted. Three factors would increase
this skew and make territorial defense uneconomical: a long
mating season, short and asynchronous periods of receptiv-
ity by females, and an actual skew in the population sex ra-
tio (Dobson 1984). In Richardson’s ground squirrels, which
typically use female defense polygyny, 50% of males in one
population became nonterritorial when a late snowstorm
increased asynchrony of receptive females and skewed the
OSR toward males (Davis and Murie 1985).
The OSR hypothesis has been supported in many spe-
cies of animals (Kvarnemo and Ahnesjo 1996); however, in
rodents the support has been mixed. Randall et al. (2002)
found that shifts in the OSR toward a male bias did inten-
sify male-male competition for mates in giant kangaroo
rats. Ims (1987a, 1988) proposed that the spacing system of
male microtines can be predicted from the temporal distri-
bution of females, in that males should have overlapping
home ranges when females are asynchronous and exclusive
home ranges when females are synchronous, which would
fit the predictions of Emlen and Oring (1977). However,
Agrell et al (1996) found just the opposite in male field
voles (Microtus agrestis).
The value of the OSR as a predictor of the intensity of
male competition in ground squirrels has also been ques-
tioned (Dobson 1984; Schwagmeyer and Wootner 1985;
Michener and McLean 1996). Michener and McLean
(1996) found more overt conflict during low OSR in Rich-
ardson’s ground squirrels. Similarly, Schwagmeyer and
Wootner (1985) determined that the density of female
thirteen-lined ground squirrels (S. tridecemlineatus) affected
male-male competition more than the OSR. Dobson (1984)
tested the relationship between OSR and the type of mating
system in ground-dwelling squirrels and rejected it as an ex-
planation of male mating strategies. However, his calcula-
tion of OSR was based on the average number of receptive
mates over a breeding season, which may be an inappropri-
ate measure (see Michener and McLean 1996 for a dis-
cussion). A better measure of OSR would be the local OSR
on a daily basis, which would more closely reflect the
changes in intensity of competition among males (Water-
man 1998).34 Chapter Three