The samango monkey (Cercopithecus mitis erythrarchus) is the southern most representative of the polytypic mitis group. The samango is also the only truly arboreal guenon to have radiated as far as 30°S. At southern latitudes a greater seasonality of climate and an attendant seasonal shift in food availability is expected to restrict the foraging strategy of the arboreal guenon. In the absence of arboreal congenerics and few frugivorous bird and bat species, the samango experiences a level of competitive release at Cape Vidal not found in other equatorially located mitis populations. In this thesis I examine and contrast the diet and feeding behaviour of the mitis species group. In this way I illustrate the consequences of seasonality of food abundance and competitive release on the foraging strategy of the samango, and provide an explanation for the unique distribution of the samango monkey as the only arboreal guenon in southern Africa. Despite seasonality in climate and abundance of food resources, my data show that, in general, samango monkeys at Cape Vidal are not food limited. For this reason Cape Vidal samangos have large troop sizes (25+), use small home ranges (15 ha) and have the highest density (2.02 ind/ha) of any C. mitis population researched to date. Fruit forms an important part of the diet year-round and therefore, energy and carbohydrate are abundant. There is very little intra-group aggression for food, although interindividual distances are greatest when feeding. There are age-sex differences in the diet, and adult males eat more fruit while females eat more leaves than other age-sex classes. The most important aspect of the feeding strategy of the samango is concerned with obtaining adequate protein in the diet, and throughout the range of the mitis group, populations differ most in feeding strategies used to secure protein-rich foods, such as young leaves, flowers and invertebrates. In this respect seasonal nutrient (protein) availability, rather than seasonality of food abundance per se, is the most limiting component of the forest environment. Unlike equatorial populations of mitis that derive most of their protein from insects, samangos are unable to adopt a similar strategy. Insects were available to the monkeys, and then only in low numbers, in the wet summer season at Cape Vidal. Samango monkeys, therefore, make greater use of a wider variety of plant items for protein acquisition. During the wet summer months, insects, flowers and young leaves are used by samangos to obtain sufficient protein for important reproductive activities, such as lactation, and initial growth of the infant. During the drier winter months only mature leaves and small quantities of young leaves are available as protein resources, consequently samangos use more mature leaf in the diet than other C. mitis populations. Adaptations of the gut and specialized gut microflora permit this high degree of folivory in the diet (Bruorton and Perrin 1988) of the samango monkey. This is a characteristic that does not appear to be shared with other arboreal guenons. In so far as protein is essential for reproduction, folivory in c. mitis has been important in permitting this arboreal guenon species-group to radiate into southern latitudes where protein is more seasonally available. The ability of samango monkeys to eat large quantities of leaves at anyone time, accounts for their unique position as the only arboreal guenon species in southern Africa. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1990.

The understanding of behavioral patterns in different species is an important part of the proper management and conservation of wild populations of animals. This study aims to contribute to the understanding of behavioral patterns in the samango monkey (Cercopithecus albogularis erythrarcus) of northern South Africa. Using the scan- sampling procedure, the behaviors of an isolated population of free-ranging samango monkeys in the Soutpansberg, Limpopo Province, were recorded during 16 days in the summer of 2010. The day was divided into the intervals: morning, midday and afternoon, and the behaviors social, resting, movement, and feeding were recorded and analyzed. The results showed a behavioral pattern in which the relative frequency of occurrence of social behaviors and movement were significantly different depending of the time of day, whereas the behaviors resting and feeding were not. During midday, social behaviors increased, while movement decreased. The groups’ degree of arboreality was also recorded and analyzed. The group spent significantly more time on the ground during midday compared to morning and afternoon. The amount of time this group spent on the ground is not entirely consistent with what has been described in the literature, where the samango has been described as strictly arboreal. A longer study including more environmental parameters, and using focal animal sampling together with the scan sampling method would be valuable for the further understanding of the behavior of the samango monkey.

PhD (Zoology) / Department of Zoology / A general introduction about the need for studying forest dwelling primate populations in fragmented landscapes and more specifically in the Soutpansberg is given in Chapter 1. Forests affected by fragmentation are at risk of losing primate populations over the long term. In addition, although the impact of fragmentation on primate populations has been studied in many places in Africa, Asia and South America there is no consensus of how the different primate species react to forest disturbance and fragmentation. This study aims to investigate the impacts of natural and anthropogenic forest fragmentation on the Soutpansberg samango monkey population including their distribution, genetics, and phylogeography, identifying threats and mitigation measures. To date, no detailed population level research has been undertaken on samango monkey populations in the far north of their South African distribution. By collecting a diverse data set, in addition to existing ecological data, this study generates conservation and management recommendations suited specifically to the study area and the study species and provides baseline data for future monitoring. Chapter 2 details how through creating an accurate forest distribution map and collecting samango monkey distribution records through surveys we were able to investigate how both natural and anthropogenic forest fragmentation influences the distribution of samango monkeys in the Soutpansberg. We explored forest patch occupancy and connectivity, determined the degree and nature of matrix utilisation and identified possible threats to forests and samango monkeys between the contrasting landscapes of the eastern and western Soutpansberg. Here we found that samango monkeys largely occupied forest patches <100 ha in size and that the Soutpansberg has very few forest patches > 100 ha available. We showed that samango monkeys used all components of the surrounding matrix and that lone or bachelor group males used the matrix more extensively than groups. We found that paved roads pose a major threat, however not a barrier, to samango monkey populations when navigating the matrix. Forest connectivity was found to be influenced by the distance between patches and possible corridors or stepping stones of isolated forest patches connecting them. Patterns found in this chapter contrasted between the eastern and western landscapes requiring different adaptive strategies from the samango monkeys and different conservation approaches from practitioners. In this chapter we conclude that samango monkeys, having evolved in fragmented landscapes, are comparatively tolerant and adaptable to a human- transformed matrix. In Chapter 3 we focused on the effects that natural and anthropogenic habitat fragmentation in the Soutpansberg has on the genetic diversity and structure of the samango monkey population in the mountain range. Here we used microsatellite and mitochondrial DNA data from faecal and tissue samples collected from four local samango monkey populations across the Soutpansberg. We also included samples from an outlying population on the escarpment south of the mountain, the most likely historical migration route into the Soutpansberg. Our analyses showed that the Soutpansberg population is divided across the mountain and that genetic diversity within the Soutpansberg decreases with increasing distance to the escarpment population and from east to west. We found a lack of contemporary gene flow suggesting that the Soutpansberg population is completely isolated from the closest source population in the escarpment and that populations within the mountain range are isolated from each other. Based on our results we suggest that extensive anthropogenic changes to the landscape in the eastern Soutpansberg and distance between high canopy forest patches in the western Soutpansberg appear to have reduced the ability of samango monkeys to disperse between sampling areas. Overall, we found that natural and anthropogenic fragmentation and geographical distance are potential drivers for the observed population genetic differentiation and that the matrix surrounding forests and its suitability for samango utilisation plays a role at the local scale. In this chapter we conclude that the degree of samango monkey population subdivision and the apparent lack of contemporary migration between populations raises concerns about the long-term viability of populations across the mountain range. Having identified road fatalities as the main direct anthropogenic threat samango monkeys are exposed to when utilizing the matrix, we further investigated this issue in more detail in Chapter 4. Here we aimed to get a better understanding of where roadkills most likely occur and how to mitigate these through using suitable canopy overpasses. As road fatalities threaten primate populations globally, we used the samango monkey (Cercopithecus albogularis) as a model species to test the suitability of two different canopy bridge designs through field experimentation and behavioural data collection for arboreal guenon roadkill mitigation. Analysis of actual roadkill data collected in the study area provided insights into the nature of high-risk localities. We showed that canopy overpasses are a viable intervention for mitigating arboreal guenon road fatalities, reducing the probability that monkeys will cross a road on the ground. Samango monkeys clearly preferred a pole bridge over a rope ladder design and canopy bridges were preferred to trees and the ground when the tree canopy was open. Pole bridges were also used by other non-guenon primates and non-primate species. We showed that although samango road kills were not predictable in time (no seasonality), adult female and immature fatalities were predictable in space, restricted to bisected riparian zones and roads close to intact forests. We further found that adult male road fatalities can be expected in seemingly unsuitable habitat areas. This chapter highlights the importance of the correct interpretation of spatial, temporal and demographic data on road fatalities and how experimental research prior to installing crossing structures could increase mitigation impact. In Chapter 5 we investigate the regional phylogeography of samango monkeys, providing a deeper understanding into their evolutionary history in repeatedly fragmented forest habitat due to paleoclimatic fluctuations. Here we used mtDNA and microsatellite data obtained from tissue samples from a coastal population (Vamizi Island) in Mozambique and compared this to existing data from South Africa. The additional analysis of Mozambique animals allowed us to further test the number and timing of radiation events of Cercopithecus monkeys in southern Africa. In this chapter we propose the occurrence of a single, north-south radiation event during the midPleistocene along the Afromontane forest belt and that after the Last Glacial Maximum, samango populations re-radiated into (re)established coastal forests on a more local scale. Taking the findings from all chapters together we provide overall conclusions and conservation and management recommendations in Chapter 6. Here we also include interview data to give insights into public perceptions of samango monkeys and possible human-primate conflicts as we consider this an important aspect of conservation planning. In this final chapter we conclude that although samango monkeys appear comparatively adaptable to changes in the matrix surrounding their forest habitat, direct threats and forest patch isolation by distance may pose risks to populations in future. We advocate that conservation and management strategies aimed at the longterm persistence of the samango monkey populations and their forest habitat across the Soutpansberg should be formulated separately for the eastern and western parts of the mountain range as differing landscape variables pose different threats to forest and samango monkeys, thus requiring separate conservation and management approaches. We further recommend explicit mention of distinct management requirements for maternal groups and lone or bachelor group males to achieve a holistic conservation approach for samango monkey populations on the landscape scale. Findings of our study are not only relevant for samango monkey populations in the Soutpansberg but also for populations elsewhere in South Africa and southern Africa generally where forests and samango monkeys share the same paleohistory and current anthropogenic threats. / NRF