Understanding spatial and temporal diversity and distribution patterns of species along with the drivers of these patterns has long been a theme of community ecology. Primates, a very species-rich taxonomic group, able to occupy various habitat types, are known for their broad behavioral repertoire and flexibility. This, in turn, allows them to adapt quickly to different ecological conditions. Therefore, they represent an ideal group for examining aspects of community ecology such as species diversity and co-occurrence, distribution patterns, and the ecological determinants of such factors. This dissertation investigates the ways in which members of a multispecies primate community inhabiting a mosaic landscape comprised of flooded and non-flooded forests in northeastern Peru (western Amazonia) distributes themselves across time and space. The main objectives of this study are threefold: 1) to categorize, confirm, and differentiate previously identified habitat types across a mosaic landscape; 2) to examine the diversity and distribution patterns (spatial and temporal) of a large multispecies primate community, and 3) to combine the habitat data with the primate data in order to determine which habitats are occupied by which species and to suggest ways in which the primates share and utilize the landscape throughout an annual cycle. However, special attention is paid to one particular species, the red uakari (Cacajao calvus ucayalii), because of its unusually large home and day ranges and its fluid social system known to change in size and composition daily and even hourly. One of the least studied primates to date, red uakaris, are investigated in order to determine whether or not their behavior varies across habitat types, seasons, and when other primates are present as they navigate a mosaic landscape. By identifying the ways in which uakaris modify their behaviors as they traverse multiple habitats throughout the year, determining both habitat-typical behaviors and seasonal behaviors exhibited by uakaris becomes possible. Moreover, shedding light on the community structure and habitat requirements of one of the least known primates has conservation implications. Research was conducted at the Tahuayo River Amazon Research Center (TRARC), located in northeastern Peru. Systematic data collection on primates and the environment occurred between September 2012 and February 2014, except April 2013. Sampling methods for primates consisted of two parts—a combination of line transect and reconnaissance (recce) surveys in order to determine encounter rates for each species across habitat types and 10-minute interval scan sampling during uakari follows in order to determine the effect that environmental and social factors have on their behavior as they traverse multiple habitats. Environmental sampling occurred in thirty plots established throughout the various habitat types. All trees within the test plots with a diameter breast height (DBH) > 10 cm were marked and the following parameters were recorded: DBH, height, and taxonomic classification. Additionally, plots were monitored monthly to record flooding data in order to determine variation in flooding patterns across habitat types. Plots allowed for the determination of which floristic variables and flood patterns are suitable to differentiate the habitat types described at the TRARC. Results indicate flood duration, average tree height, and (Importance Value Indices [IVIs] at the family, genus, and species levels) are suitable measures for defining and differentiating the five previously identified habitat types at the TRARC. Analysis of the entire primate community showed that the occurrence of species and patterns of distribution across a mosaic landscape vary throughout the year. Distribution patterns are more dependent on forest structure (habitat type) than on rainfall seasonality. While a few primates showed preferences for particular habitats (e.g., Lagothrix) or for certain seasons (e.g. Saimiri), the majority of primates demonstrated more generalized modes of ranging and foraging, with relatively equal encounter rates in all habitats across both wet and dry seasons. Results of the uakari data revealed that habitat and season had an effect on their behavior. There were meaningful differences in the behavioral categories of vocalizing, resting, moving, feeding, infant clinging, and being in polyspecific associations (PSAs) between flooded and non-flooded habitat types. Seasonal differences were seen for vocalizing, traveling, resting, and time spent in PSAs. Although results revealed that red uakaris spent the majority of the time alone (71.86%), when they were in PSAs with one other primate species, squirrel monkeys (Saimiri sciureus) were the most common monkeys. When uakaris were in PSAs with two other species, squirrel monkeys were always present, but woolly monkeys (Lagothrix lagotricha poeppigii) were the second most frequent species. In general, uakaris demonstrated behavioral differences across habitats and seasons. However, some behavioral consistencies across seasons were evident (e.g., moving, feeding, infant clinging), suggesting ecological flexibility in the species. Results of this dissertation provide a basis for understanding ecological parameters best suited for characterizing and differentiating habitat types in upper Amazonia and describe the diversity and distribution patterns of a multispecies community of primates occurring across a mosaic landscape, reinforcing the view that New World primates are largely ecological generalists within forest environments. Understanding the spatio-temporal relationships between species and their environments can aid in predictions of species occurrence/abundance and contribute to better management strategies and conservation prioritization.

As pessoas do senso comum repetem os equívocos dos primeiros exploradores europeus quanto ao número de animais visíveis na Amazônia. Induzidas no passado pela grandeza do próprio bioma e incertezas do que existia além das matas ciliares, ou modernamente por programas televisivos, pressupõem a existência de grandes concentrações de animais selvagens neste bioma, semelhantes às encontradas nas planícies africanas ou no pantanal mato-grossense. No entanto, reza à lenda, que a Amazônia é ciumenta com seus bichos. Vê-los exige sacrifício e paciência do pesquisador. No presente estudo foi necessário percorrer 1.600 km, distribuídos em 17 meses entre 2005 e 2006, oitocentos quilômetros na Floresta Nacional do Tapajós (FNT) e 800 km na Floresta Nacional de Saracá-Taquera (FNS-T), ambas no estado do Pará (Brasil), para reunir detecções sobre os primatas nelas contidas que permitissem um robusto tratamento estatístico. Mesmo assim, para algumas espécies naturalmente raras, como por exemplo, Pithecia p. chrysocephala da região do Rio Trombetas, ou Ateles marginatus na região do Rio Tapajós, a freqüência de detecção ao redor de 10, para ambas as espécies, foi muito baixa para permitir inferências seguras. Considerando 40 detecções como o ideal, seria necessário então quadruplicar o esforço amostral, o que por limitações de tempo e recursos é inviável. Outras espécies descritas e observadas por residentes na região do Rio Trombetas, como Saguinus midas e Saimiri sciureus, sequer foram detectadas na floresta de terra firme. Provavelmente não suportando a competição com os primatas simpátricos do interior das florestas intactas foram compelidas a ocuparem e especializarem-se na exploração de florestas periféricas, onde seus alimentos preferidos, os insetos e frutos menores são mais abundantes. Por meio do programa SAS, foi calculado a ANOVA para testar a hipótese nula, de que não existiam diferenças populacionais entre as duas comunidades de primatas, uma vez que ambas ocorrem dentro de um mesmo bioma: floresta de terra firme do oeste da Amazônia e não estão a mais de 240 km uma da outra. Esta hipótese foi rejeitada em favor da hipótese alternativa de que as comunidades diferem entre si. Os 397 grupos de primatas detectados foram desigualmente distribuídos (130 x 267 em favor da FNS-T). Ou dito de outra forma: 1 grupo detectado a cada 6 km de caminhada no Tapajós contra 1 grupo detectado a cada 3 km no Trombetas. Os testes de Tukey e G de Sokal e Hohlf foram importantes auxiliares para facilitar o entendimento de como as duas comunidades de primatas diferiram, respondiam e se adaptavam a estrutura das respectivas florestas. Das 200 árvores amostradas em cada uma das flonas, a FNS-T apresentou 92 espécies diferentes, contra 74 na FNT. Assim o índice de Shannon e Wiener foi mais alto na região do Rio Trombetas do que na do Rio Tapajós (6,17 x 5,74 respectivamente). A cobertura do dossel também foi maior na região do Rio Trombetas quando comparado com a do Rio Tapajós (96% na FNS-T contra 88% na FNT). Estes dois índices reforçam-se mutuamente e sua interpretação sugere que a FNT vem sofrendo maiores perturbações do que a FNS-T. Considerando a colonização mais antiga e maior da população de entorno da Cuiabá-Santarém (BR-163) e o maior número de residentes dentro da FNT (10.500 pessoas), do que dentro da FNS-T (2.500 pessoas), é provável que o número tão discrepante de grupos de primatas em favor da FNS-T, reflita a maior pressão antrópica sofrida pela comunidade de primatas do Rio Tapajós. Interpretamos que as diferenças antrópicas entre as flonas (embora um epifenômeno), foram mais importantes na discrepante abundância de primatas do que as especificidades locais (na estrutura da floresta e oferta de recursos alimentares). (Figura 94, apêndice). / Common sense people have repeated the mistakes of early European explorers as to the number of observable animals in the Amazon region. In the past based, on the greatness of the biome itself and uncertainties as to what existed beyond the riparian forests, or nowadays, through television shows, they assume the existence of great concentrations of wild animals in such biome, likewise those found in African plains or in the Mato Grosso pantanal. However, the Amazon region is said to be \'jealous\' about its animals. Observing them requires the researcher\'s sacrifice and patience. At this study, one had to traverse 1,600 km in 17 months between 2005 and 2006, eight hundred kilometers within the Tapajós National Forest (FNT) and 800 km within the Saracá-Taquera National Forest (FNS-T), both in the state of Pará (Brazil), to gather detections of the primates therein allowing a solid statistical treatment. Even so, for some naturally rare species, such as Pithecia p. chrysocephala of the Trombetas river region, or Ateles marginatus, at the Tapajós river region, the frequency of detection nearing 10 for both species was too low for accurate inferences. Considering 40 detections as the ideal number, the sampling effort would be four-fold, which is not viable due to time and resource limitations. Other species described and observed for the Trombetas river region, as Saguinus midas and Saimiri sciureus, were not even detected at the upland forest. Probably succumbing to the competition with sympatric primates of inner intact forests, they were compelled to occupying and specializing in exploring peripheral forests, where their favorite foods - insects and small fruit - are abundant. By means of SAS software, the ANOVA was calculated to test the null hypothesis, that there were no population differences between both primate communities, since both occur within the same biome - upland forest of western Amazon and are no more than 240 km apart. This hypothesis was rejected in favor of an alternative hypothesis that the communities differ from one another. The 397 primate groups detected were unequally distributed (130 x 267 favoring the FNS-T). In other words, 1 group detected every 6 km of trekking at the Tapajós region versus 1 group detected every 3 km at the Trombetas region. The Tukey\'s test and G-test of Sokal and Hohlf were important tools in facilitating the understanding of how the two primate communities differed, responded and adapted to the structure of the respective forests. Out of 200 trees sampled in each forest, the FNS-T presented 92 different species, versus 74 at the FNT. This way, the Shannon-Wiener index was higher at the Trobetas river region than at the Tapajós river region (6.17 x 5.74 respectively). Also the canopy was higher at the Trombetas river region in comparison with that of the Tapajós river region (96% at the FNS-T versus 88% at the FNT). Both indexes are mutually reinforced, suggesting that the FNT has undergone more disturbances than the FNS-T. Considering the older and larger colonization of the population neighboring the Cuiabá-Santarém highway (BR-163) and the larger number of residents within the FNT (10,500 people), than within the FNS-T (2,500 people), such discrepant number of primate groups in the FNS-T is likely to reflect the higher anthropic pressure suffered by the primate community of the Tapajós river region. The anthropic differences between the forests, although an epiphenomenon, are interpreted as more important at the discrepant abundance of primates than the local specificities (on the forest structures and food resource supply).

As pessoas do senso comum repetem os equívocos dos primeiros exploradores europeus quanto ao número de animais visíveis na Amazônia. Induzidas no passado pela grandeza do próprio bioma e incertezas do que existia além das matas ciliares, ou modernamente por programas televisivos, pressupõem a existência de grandes concentrações de animais selvagens neste bioma, semelhantes às encontradas nas planícies africanas ou no pantanal mato-grossense. No entanto, reza à lenda, que a Amazônia é ciumenta com seus bichos. Vê-los exige sacrifício e paciência do pesquisador. No presente estudo foi necessário percorrer 1.600 km, distribuídos em 17 meses entre 2005 e 2006, oitocentos quilômetros na Floresta Nacional do Tapajós (FNT) e 800 km na Floresta Nacional de Saracá-Taquera (FNS-T), ambas no estado do Pará (Brasil), para reunir detecções sobre os primatas nelas contidas que permitissem um robusto tratamento estatístico. Mesmo assim, para algumas espécies naturalmente raras, como por exemplo, Pithecia p. chrysocephala da região do Rio Trombetas, ou Ateles marginatus na região do Rio Tapajós, a freqüência de detecção ao redor de 10, para ambas as espécies, foi muito baixa para permitir inferências seguras. Considerando 40 detecções como o ideal, seria necessário então quadruplicar o esforço amostral, o que por limitações de tempo e recursos é inviável. Outras espécies descritas e observadas por residentes na região do Rio Trombetas, como Saguinus midas e Saimiri sciureus, sequer foram detectadas na floresta de terra firme. Provavelmente não suportando a competição com os primatas simpátricos do interior das florestas intactas foram compelidas a ocuparem e especializarem-se na exploração de florestas periféricas, onde seus alimentos preferidos, os insetos e frutos menores são mais abundantes. Por meio do programa SAS, foi calculado a ANOVA para testar a hipótese nula, de que não existiam diferenças populacionais entre as duas comunidades de primatas, uma vez que ambas ocorrem dentro de um mesmo bioma: floresta de terra firme do oeste da Amazônia e não estão a mais de 240 km uma da outra. Esta hipótese foi rejeitada em favor da hipótese alternativa de que as comunidades diferem entre si. Os 397 grupos de primatas detectados foram desigualmente distribuídos (130 x 267 em favor da FNS-T). Ou dito de outra forma: 1 grupo detectado a cada 6 km de caminhada no Tapajós contra 1 grupo detectado a cada 3 km no Trombetas. Os testes de Tukey e G de Sokal e Hohlf foram importantes auxiliares para facilitar o entendimento de como as duas comunidades de primatas diferiram, respondiam e se adaptavam a estrutura das respectivas florestas. Das 200 árvores amostradas em cada uma das flonas, a FNS-T apresentou 92 espécies diferentes, contra 74 na FNT. Assim o índice de Shannon e Wiener foi mais alto na região do Rio Trombetas do que na do Rio Tapajós (6,17 x 5,74 respectivamente). A cobertura do dossel também foi maior na região do Rio Trombetas quando comparado com a do Rio Tapajós (96% na FNS-T contra 88% na FNT). Estes dois índices reforçam-se mutuamente e sua interpretação sugere que a FNT vem sofrendo maiores perturbações do que a FNS-T. Considerando a colonização mais antiga e maior da população de entorno da Cuiabá-Santarém (BR-163) e o maior número de residentes dentro da FNT (10.500 pessoas), do que dentro da FNS-T (2.500 pessoas), é provável que o número tão discrepante de grupos de primatas em favor da FNS-T, reflita a maior pressão antrópica sofrida pela comunidade de primatas do Rio Tapajós. Interpretamos que as diferenças antrópicas entre as flonas (embora um epifenômeno), foram mais importantes na discrepante abundância de primatas do que as especificidades locais (na estrutura da floresta e oferta de recursos alimentares). (Figura 94, apêndice). / Common sense people have repeated the mistakes of early European explorers as to the number of observable animals in the Amazon region. In the past based, on the greatness of the biome itself and uncertainties as to what existed beyond the riparian forests, or nowadays, through television shows, they assume the existence of great concentrations of wild animals in such biome, likewise those found in African plains or in the Mato Grosso pantanal. However, the Amazon region is said to be \'jealous\' about its animals. Observing them requires the researcher\'s sacrifice and patience. At this study, one had to traverse 1,600 km in 17 months between 2005 and 2006, eight hundred kilometers within the Tapajós National Forest (FNT) and 800 km within the Saracá-Taquera National Forest (FNS-T), both in the state of Pará (Brazil), to gather detections of the primates therein allowing a solid statistical treatment. Even so, for some naturally rare species, such as Pithecia p. chrysocephala of the Trombetas river region, or Ateles marginatus, at the Tapajós river region, the frequency of detection nearing 10 for both species was too low for accurate inferences. Considering 40 detections as the ideal number, the sampling effort would be four-fold, which is not viable due to time and resource limitations. Other species described and observed for the Trombetas river region, as Saguinus midas and Saimiri sciureus, were not even detected at the upland forest. Probably succumbing to the competition with sympatric primates of inner intact forests, they were compelled to occupying and specializing in exploring peripheral forests, where their favorite foods - insects and small fruit - are abundant. By means of SAS software, the ANOVA was calculated to test the null hypothesis, that there were no population differences between both primate communities, since both occur within the same biome - upland forest of western Amazon and are no more than 240 km apart. This hypothesis was rejected in favor of an alternative hypothesis that the communities differ from one another. The 397 primate groups detected were unequally distributed (130 x 267 favoring the FNS-T). In other words, 1 group detected every 6 km of trekking at the Tapajós region versus 1 group detected every 3 km at the Trombetas region. The Tukey\'s test and G-test of Sokal and Hohlf were important tools in facilitating the understanding of how the two primate communities differed, responded and adapted to the structure of the respective forests. Out of 200 trees sampled in each forest, the FNS-T presented 92 different species, versus 74 at the FNT. This way, the Shannon-Wiener index was higher at the Trobetas river region than at the Tapajós river region (6.17 x 5.74 respectively). Also the canopy was higher at the Trombetas river region in comparison with that of the Tapajós river region (96% at the FNS-T versus 88% at the FNT). Both indexes are mutually reinforced, suggesting that the FNT has undergone more disturbances than the FNS-T. Considering the older and larger colonization of the population neighboring the Cuiabá-Santarém highway (BR-163) and the larger number of residents within the FNT (10,500 people), than within the FNS-T (2,500 people), such discrepant number of primate groups in the FNS-T is likely to reflect the higher anthropic pressure suffered by the primate community of the Tapajós river region. The anthropic differences between the forests, although an epiphenomenon, are interpreted as more important at the discrepant abundance of primates than the local specificities (on the forest structures and food resource supply).