Dw. Macdonald et O. Courtenay, ENDURING SOCIAL RELATIONSHIPS IN A POPULATION OF CRAB-EATING ZORROS, CERDOCYON THOUS, IN AMAZONIAN BRAZIL (CARNIVORA, CANIDAE), Journal of zoology, 239, 1996, pp. 329-355
1) Crab-eating zorros, Cerdocyon thous, in Amazonian Brazil weighed 5.
2 kg (S.D. 0.6, n = 19), exhibited no sexual dimorphism. lived in soci
al units of 2-5 adults of >1 year old, and occupied territories (restr
icted polygons) of 532 ha (range 48-1042 ha, n = 21). 2) The zorros we
re omnivorous, fruit being the most frequent food (occurring in 57% of
72 faeces in the dry season) comprising 26.4% by volume of undigested
faecal remains. Insects were frequently eaten (86%), and vertebrates
rarely (15.2%). 3) Zorros were territorial. The ranges of neighbours o
verlapped minimally and were very stable in configuration, both from m
onth to month and between dry and wet seasons. Members of each social
unit shared on average 88.3% of 200 x 200 m grid cells within their te
rritory. Members of male-female breeding pairs were within 100 m of ea
ch other for 63% of the activity period. their proximity being greates
t during the mating season (July-August). Parents travelled in close c
ompany with their adult-sized offspring, and stayed within 100 m for u
p to 93.3% of the activity period. 4) Territories differed in habitat
composition. Overall, zorros spent most time in wooded savanna (33.9%)
and scrub (30.5%). They did not utilize habitats in direct proportion
to their availability, using some more (e.g. scrub) and some less (e.
g. open savanna) than expected. Habitat preferences differed between w
et and dry seasons. with elevated habitats being favoured during the w
et season. 5) Individuals differed in their habitat utilization, and p
arents had different habitat usage to their yearling offspring; this d
ifference was exaggerated in the dry season when the yearlings used lo
w-lying habitats apparently disfavoured by the parents. Differences in
habitat utilization between group members were least in the wet seaso
n, when widespread flooding forced the shared use of higher ground. 6)
Each of two social units larger than two individuals comprised a pair
and their three adult-sized offspring (totalling five males and one f
emale). In three cases, a yearling daughter was present in the breedin
g season, but none of these bred. Non-breeding offspring were seen fre
quently in the whelping area and In close company with the breeding pa
ir's cubs. This is the first proof of group-living in South American z
orros. 7) Five dispersal events were monitored, and revealed a 'good n
eighbour' strategy in which newly formed pairs set up their territorie
s adjoining their natal territories. In each case individuals returned
intermittently to their original territory where they were in close a
nd amicable company with their parents. One male, having dispersed, re
turned home frequently without his mate to tend the next generation of
his parents' cubs. Of four dispersing males, two subsequently returne
d to their natal group following the deaths of their mates at least 3-
13 months after their initial dispersal, in one case after breeding el
sewhere.