Discussion
SLOTH POPULATION GROWTH RATES AND THE
INFLUENCE OF IMMIGRATION
The analysis herein is, to the best of our knowledge, the
first comprehensive demographic assessment of any vertebrate
species occurring in a shade-grown agro-ecosystem.
Point estimates of k indicated that population growth was
positive for C. hoffmanni when the matrix model was
parameterized with estimates of both true and local adult
survival rates. Indeed, these estimates suggest the local
population is increasing by 3–6% annually and that individuals
initially marked in the agro-ecosystem were
increasing 5–8% annually (i.e. more than replacing themselves).
Nevertheless, lower 95% confidence limits for k
fell slightly below 10, and we could not rule out the
possibility that the population was in decline. However,
we suspect that this is a locally stable population because
C. hoffmanni densities have been high in the study area
since 2004, and we have no evidence indicating that
C. hoffmanni has declined during this period (Vaughan
et al. 2007, this study). Moreover, only a modest amount
of immigration (005–007, depending on AFB) was
needed to produce estimates of k with lower 95%
confidence intervals that did not overlap 10.
In contrast, point estimates of k for B. variegatus indicated
c. 6% annual decline and upper 95% confidence
limits did not overlap 10. Negative population growth
was the case for both estimators of k since true and local
survival were equivalent in the absence of dispersing individuals.
Thus, the local population of B. variegatus did
not appear to be replacing itself and was expected to
decline in the absence of immigration. Adult survival and
reproductive estimates were very similar between species,
and the difference in population growth was largely due
to lower juvenile and subadult survival in B. variegatus.
While we only documented the cause of mortality for a
small number of juveniles and subadults (Table 2), young
B. variegatus may be more susceptible to predation given
this species’ smaller size and more docile behaviour
when handled (M. Z. Peery & J. N. Pauli, personal
observation).
While population modelling indicated that B. variegatus
would decline in the absence of immigration, genetic parentage
analyses suggested that significant immigration
occurred into our study site. Indeed, half of the subadults
that we sampled did not have a father in the sample,
whereas paternity was assigned to all sampled juveniles.
Because all juveniles were assigned paternity, we believe
that we sampled a high proportion of potential fathers
and that subadults without paternity assignments represented
immigrants. It could be argued that subadults were
less likely to have a father in the sample because they
were born 1–2 years earlier than juveniles and their
fathers had a greater probability of dying before being
sampled (as opposed to occurring outside of study area
boundaries). However, adult male survival was high
(0901) for B. variegatus, and we consider it unlikely that
many of the subadults for whom paternity was not
assigned were fathered by a deceased resident male. Estimating
immigration rates for B. variegatus based on the
proportion of inferred immigrants is challenging, but if
one assumes that the subadult phase lasts 1 year
(AFB = 2), a crude estimate of annual immigration can
be derived by dividing the sampled number of non-juvenile
immigrants by the sampled number of non-juveniles.
For B. variegatus, this calculation yields an annual immigration
estimate of 0089 (four of 45), in which case
k = 0981. Thus, our results indicate that, while the local
B. variegatus population may not be replacing itself,
immigration may contribute to maintaining an approximately
stable population, or one that is declining only
slowly.
ENVIRONMENTAL FACTORS INFLUENCING LIFEHISTORY
TRAITS
Our observation of the timing of breeding in C. hoffmanni
and B. variegatus is consistent with previous work indicating
that C. hoffmanni reproduces essentially throughout
the year, whereas B. variegatus has a distinct, albeit
lengthy (c. 6 month) breeding season (Taube et al. 2001).
Despite the ability to breed each year, the mean proportion
of breeders was similar for the two species, perhaps
because fewer adult female C. hoffmanni were nonbreeders.
We also note that the proportion of breeders varied
little among years in B. variegatus, an observation possibly
explained by the relatively constant environmental
conditions among years (compared to, for example,
temperate regions).