all males disperse from their natal area at the end of their
juvenile season and adult males disperse between breeding
seasons (Lacey and Wieczorek 2004). Within a group of
C. sociabilis,all females reproduce and all individuals en-
gage in activities such as burrow excavation, foraging, and
alarm calling to predators (E. A. Lacey, unpublished data).
Although allonursing has been observed in the lab, it has
not yet been documented among free-living animals. In
short, this species exhibits none of the within-group repro-
ductive or behavioral specializations that characterize so-
cial bathyergids. Moreover, C. sociabilisis the only species
of ctenomyid for which quantitative evidence of group liv-
ing has been obtained; although anecdotal accounts of bur-
row sharing exist for several other species of tuco-tucos, the
majority of ctenomyids are apparently solitary (Reig et al.
1990; Lacey 2000; Lacey and Ebensperger, chap. 34 this
volume).
The final independently evolved example of subterra-
nean sociality is the cururo (Spalacopus cyanus). This spe-
cies has been studied less extensively than the colonial tuco-
tuco or social African mole-rats, but available data suggest
that a group consists of two to fifteen adults, including, in
at least some populations, multiple adult males (Reig 1970;
Lacey, Ebensperger, and Wieczorek, unpublished data). Na-
tal philopatry has been detected for both sexes, as has dis-
persal between groups by adult males. The relative fre-
quency of dispersal versus philopatry for each sex, however,
is unknown. Within a group, multiple females show evi-
dence of reproduction (Begall et al. 1999; Lacey, Ebensper-
ger, and Wieczorek, unpublished data), although the distri-
bution of direct fitness among group mates is unknown. As
in colonial tuco-tucos, all group members appear to engage
in burrow excavation and alarm calling, with no evidence
of behavioral specialization among individuals.
Comparing these taxa, it appears that colonial tuco-
tucos differ from social African mole-rats and cururos in
that groups of C. sociabilisnever contain more than a single
adult male. Thus while colonial tuco-tucos are female kin
groups that arise due to female-biased natal philopatry,
groups of social mole-rats and cururos contain multiple
adult males and females and arise due to philopatry by
both sexes. Only the social African mole-rats, however, are
known to be singular breeders in which direct fitness is re-
stricted to a small subset of colony members and numerous
nonbreeders provide alloparental care to the offspring of re-
productive animals.
Adaptive bases for subterranean sociality
Social African mole-rats, colonial tuco-tucos, and cururos
all form groups composed of multiple adults that live to-
gether in a single burrow system. The adaptive bases for
burrow sharing have been examined most thoroughly for
bathyergid mole-rats (Jarvis et al. 1994; Lacey and Sher-
man 1997; Bennett and Faulkes 2000; Faulkes and Bennett,
chap. 36 this volume). The ecological factors thought to fa-
vor sociality in these animals are summarized by the aridity
food-distribution hypothesis (AFDH), which proposes that
the energetic cost of burrowing through hard soils to locate
patchily distributed but locally abundant food resources is
the primary selective factor favoring group living. Specifi-
cally, in arid regions characterized by difficult-to-excavate
soils, the only time that it is energetically feasible for the an-
imals to expand their burrows is when the soil has been re-
cently softened by rain. Because the roots and tubers that
the animals feed on are patchily distributed, a single in-
dividual would be unable to tunnel far enough to locate a
new food supply before drying of the soil again renders the
cost of tunnel excavation prohibitive. By living together and
working cooperatively to excavate new tunnels, the animals
are able to locate sufficient food resources to survive until
the next rain.
Three lines of evidence have been used to evaluate the
AFDH. First, analyses of the geographic distribution of so-
cial bathyergids suggest that while group-living species can
occupy arid habitats, solitary taxa are limited to more me-
sic regions (Jarvis et al. 1994), as expected if sociality is
necessary in areas characterized by high energetic costs of
burrow excavation. Second, within species, comparative
studies of two populations of the common mole-rat (C. hot-
tentotus) have revealed that natal philopatry is more com-
mon in an arid than in a mesic habitat (Spinks et al. 2000a,
2000b). Third, long-term field studies of Damaraland mole-
rats indicate that larger groups are better able to survive
extended droughts, suggesting that sociality is an adaptive
response to harsh, arid conditions (Faulkes and Bennett,
chap. 36 this volume).
Although the AFDH appears to be consistent with be-
havioral and ecological patterns among African mole-rats,
it does not seem to explain sociality in colonial tuco-tucos.
Comparative studies of the habitats occupied by a popula-
tion of this species and a population of the syntopic but
solitary Patagonian tuco-tuco (C. haigi) revealed no signifi-
cant differences in rainfall or food resource distributions
(Lacey and Wieczorek 2003). Although significant differ-
ences in soil penetrability (a proxy for hardness) were de-
tected, it was the soils occupied by the solitary species that
were more difficult to excavate. This difference in soil con-
ditions appeared to result from interspecific differences
in habitat use. While the group-living C. sociabilisoccurs
primarily in patchily distributed areas of mesic grassland
known as mallínes, C. haigioccurs in these patches as well250 Chapter Twenty-One