Remarkably close male bonding exists among Costa Rican squirrel monkeys, Saimiri oerstedi. In this population males were philopatric, exhibited no male-male within-troop aggression, and only very slight evidence for a dominance hierarchy. Juvenile, young and full adult males were spatially clumped in same-age male cohorts. A less intense, secondary association was also shown between the young and full adult male cohorts. These associations were most dense over a spatial scale of 5 m or less, but can also be detected in a 5-10 m distance from a focal male. Males also cooperated in 1) sexual investigation of females during the mating season, 2) aggressive interactions with males of neighboring troops, and 3) valiant defense of infants and subadults from potential predators. In contrast, the spatial association among adult females never exceeded random expectations and only a transitory period of cooperation was observed among mothers during the birth season. Furthermore, there was little evidence of bonds between the sexes. Seasonal variation in affiliation patterns was best explained by fluctuations in food availability and, secondarily, reproductive activity. These results are surprising both from the perspective of the general pattern of male affiliation among primates and the often marked aggression and dominance relationships among males documented in captive and wild populations of South American squirrel monkeys. The ultimate explanation is suggested to be the disparate distribution of fruit resources exploited by South and Central American squirrel monkeys, mediated by effects on female affiliation and dispersal patterns.
Cognitive skills essential to dispersal remain a thorny, seldom-broached topic, especially among the putatively 'clever' primates. This essay, the final installment of a three-part monograph, considers the cognitive mechanisms underlying expression of three extremely distinctive species-specific dispersal outcomes within squirrel monkeys (genus Saimiri, Primates: Cebidae). Findings from two companion reports, which assess the costs and benefits structuring between-species differences (I. Divergent costs and benefits, Boinski et al., 2005a) and variation within-species (II. Within-species and local variation, Boinski et al., 2005b), provide the groundwork for my often speculative discussion. (1) In Costa Rica, female S. oerstedii do not form kin-based alliances. All females disperse prior to their first mating season and may disperse on numerous occasions throughout adulthood. Male S. oerstedii are philopatric and exhibit close social bonds with other natal males, particularly those from the same birth cohort. (2) Male dispersal and female philopatry, the prevalent pattern in most mammals, including primates, characterizes Peruvian S. boliviensis. Both sexes form life-long alliances with same-sex kin. After natal dispersal, male birth cohorts join all-male groups, from which they attempt to immigrate into mixed-sex troops. Female kin in a S. boliviensis troop form matrilines critical in within-group food competition. (3) All male and most female S. sciureus disperse from several to many occasions during their lifetime. In contrast to the other two species, male S. sciureus never exhibit stable alliances with other males, including probable kin. Similarly, female coalitions are transient, detectable only during periods of relative food abundance.What are the implications of this marked between-species disparity in dispersal outcomes for squirrel monkey cognition, and, by extension, the cognition of other social mammals? Two timely issues are addressed. First, squirrel monkeys exemplify the provocative parallels in the assessments required of individuals embedded within three circumstances usually treated separately: dispersal; coordination of group travel; and fission-fusion adjustments of group composition. Are arguments that dispersal is more or less cognitively demanding than either coordinated travel or fission-fusion social structures justified? Fundamentally, all three processes are reducible to frequency-dependent decision-making by individuals based upon concurrent social and ecological assessments across multiple dimensions, such as time, space, and participant number. Second, a common approach to identify the covariation of selective regimes and apparent cognitive abilities are taxonomically inclusive, multivariate parametric statistical models, which incorporate information on ecology, behaviour, morphology and phylogeny. However, such correlative analyses add little to what is arguably the major challenge in contemporary field investigations of animal behaviour: How can we distinguish complex, multivariate decision-making algorithms from simple 'rule of thumb' solutions? Must field workers await the findings of laboratory-based neuroethological and neuroanatomical investigations to improve understanding of what innate versus learned behaviour contributes to complex social and ecological decisions in group-living mammals, such as those incarnate in dispersal? My suggestion is that more research emphasis be given to detailed, longitudinal field observations of recognized individuals from infancy onwards. The resulting empirical data, although in most instances onerous to collect, will enable construction of a rich, multivariate, quantitative and qualitative longitudinal picture of individual development and changing contexts of experience. In turn, these descriptive data will afford a strong basis for rejecting or accepting predictions distinguishing experiential, socially learned and innate components of dispersal behaviour.
The white-faced capuchin, Cebus capucinus, employed a specialized vocalization, the trill, to coordinate troop movement at La Selva, an Atlantic wet-forest study site in Costa Rica. We analyse the contexts in which this intra-group vocalization was emitted, including responses elicited from other group members. A cumulative 26.6 hours of continuous samples and 3,314 spectrograms (including 1,295 trills) were analysed from a study troop with 16 focal subjects. These results generally corroboratc the conclusions of a comparable field study of white-faced capuchins at Santa Rosa, a Pacific coast dry-forest site in Costa Rica (BOINSKI, 1993, Amer. J. Primatol. 30, p. 85-100). At both sites, (I) trills were closely associated with the initiation of movement by a stationary troop in a specific direction. (2) Trills were emitted at a much higher rate in the leading edge of a travelling troop than in following positions. (3) Individuals often reinforced the efforts of other troop members to coordinate troop movement. (4) Lack of consensus among troop members over the travel route was evident. (5) In rare instances trills were employed in tactical maneuvers suggestive of intentionality and the ability to anticipate behavioural effects. Differences in the usage of trills at these two sites were also detected. (1) At La Selva all troop members, with the exception of infants, used trills in the coordination of troop movement, whereas at Santa Rosa marked age, sex and rank distinctions in the extent of participation were apparent. (2) Capuchins at Santa Rosa altered the trajectory of travelling troops with trills, even reversing directions, but not at La Selva. These disparities may follow from differences between the sites in the extent of visual and auditory contact typical among troop members, social structure, susceptibility to predation, and possible genetic variation.
Current theory frames animal dispersal as an outcome of potentially complex, multi-factorial interactions and tradeoffs that may vary across individual, sex, rank, age, social group, species, habitat and time. Empirical data relevant to a broad range of the potential costs and benefits incurred by dispersal are, not surprisingly, limited for many mammals and other vertebrates. Here we present the first report on dispersal in a wild population of the Neotropical primate Saimiri sciureus (Primates: Cebidae). Long-term observations (1998-2001) of this squirrel monkey represent part of a broader study of the forest community at Raleighvallen in the Central Suriname Nature Reserve. These new dispersal records for S. sciureus are combined with comparable information from congeners, S. boliviensis in Peru and S. oerstedii in Costa Rica. The resulting three-way compilation includes the ecological, social and mating context for each congener. Further enhancing the inherent phylogenetic control of a within-genus comparison, these data were collected with the explicit intent of joint analyses, and the study sites for these small, arboreal social mammals are three of the least disturbed extant Neotropical forests in the historical record.Saimiri appears to merit description as the genus with the most diverse set of species dispersal patterns yet documented among mammals. (1) S. sciureus of both sexes undertake dispersal on several to many occasions during their lifetime. Females and immatures commonly transfer between troops. The large portion of male S. sciureus spend their adult years as solitary or peripheral males. Few males attain secure residence in a mixed-sex troop, a prerequisite for mating success. (2) On attainment of sexual maturity, male S. boliviensis emigrate with their same-age cohort, first joining all-male bands, and eventually entering mixed-sex troops with this same natal male birth cohort. Natal female S. boliviensis are philopatric and form cohesive matrilines. Within-troop competition determines each matriline's priority of access to fruit resources. (3) In contrast to both S. sciureus and S. boliviensis, S. oerstedii males are philopatric and maintain tight affiliation with same age-cohort males. Natal female S. oerstedii emigrate as juveniles prior to their first mating season, and may undertake secondary dispersals in subsequent years.Squirrel monkeys represent a genus with realistic prospects of discriminating the costs and benefits germane to species-typical dispersal strategies. To this end, we collate 30 different causal parameters commonly invoked as influencing mammalian dispersal patterns. Each of these factors is assessed separately for possible influence on the empirically determined sex and species differences. We predict the possible consequences of direct and inclusive fitness interactions on dispersal outcomes for future testing with genetic data. Components of Saimiri selective regimes particularly salient to female dispersal strategies include food competition, foraging benefits provided by kin and inbreeding avoidance. Dispersal patterns among male Saimiri are constrained by mate competition and the consequent reproductive skew, in addition to enhanced predation risk during dispersal forays. Little evidence, however, suggests that relative to familiar landscapes, exploitation of novel ranging areas substantially increases foraging costs or predation risk for dispersing squirrel monkeys of either sex. We then compare the species-specific dispersal regimes initially identified with the univariate array of proposed costs and benefits to the tradeoffs predicted by a selection of contemporary multivariate dispersal models. The multivariate models did not, however, improve substantially upon the collective insights on cost-benefit regimes achieved with the univariate hypotheses. Conclusions regarding the selective regimes structuring dispersal among squirrel monkeys are best considered provisional until genetic data become available allowing tests of our inferences concerning kin relationships and population structure of the study populations.
Three long-term field studies, together with numerous supplementary sources of information, demonstrate that the Neotropical squirrel monkey, genus Saimiri (Primates: Cebidae) are distinguished among mammals by the wide divergence in dispersal patterns among congeners. Both sexes of Saimiri sciureus at Raleighvallen in the Central Suriname Nature Reserve undertake dispersal on several to many occasions during their lifetime. Male dispersal and female philopatry characterize S. boliviensis studies at Manu, Peru. Among S. oerstedii, studied at Corcovado and other locations in Costa Rica, females disperse and males are philopatric. This is the second in a series of three companion reports investigating patterns and processes relevant to dispersal in these three species of squirrel monkeys. The first report, I. Divergent costs and benefits (Boinski et al., 2005), predicts the direct and inclusive fitness costs and benefits structuring species, sex and individual dispersal strategies among squirrel monkeys. III. Cognition (Boinski, 2005), the final report comprising this monograph, considers the possible cognitive mechanisms underlying dispersal among squirrel monkeys and other taxa, and suggests useful strategies to collect and interpret additional data from laboratory and field contexts.Here we evaluate the sources and potential magnitude of variation in dispersal strategies within each squirrel monkey species. For all three congeners, local edaphic and anthropogenic regimes of habitat disturbance probably represent the major source of within-species variance in the density of wild populations. Squirrel monkey population density, all else being equal, positively increases with the intensity of habitat disturbance. New evidence suggests that in addition to edaphic and recent historical disturbance regimes, in some localities in the Neotropical lowlands anthropogenic disturbance caused by pre-Columbian Amerindians remains a potent factor enhancing squirrel monkey numbers.Squirrel monkeys are predicted to exhibit density-dependent behavioural responses. In turn, these responses are expected to modulate population-level dispersal outcomes in several predictable axes. Major between-site variation in dispersal strategies, however, is unlikely for either sex among S. oerstedii or S. boliviensis. Although all natal male S. sciureus almost certainly disperse before or at the time of sexual maturity, the proportion of females emigrating from a S. sciureus troop appears more variable, dependent on local levels of within-troop competition for food. In any year or season, those mature and immature female S. sciureus with high priority access to food resources are least likely to disperse.