Filtern
Volltext vorhanden
- ja (2) (entfernen)
Gehört zur Bibliographie
- ja (2)
Dokumenttyp
- Dissertation (2) (entfernen)
Sprache
- Englisch (2) (entfernen)
Schlagworte
- Regenwald (2) (entfernen)
Institut
- Theodor-Boveri-Institut für Biowissenschaften (2) (entfernen)
Large parts of the tropical lowland rain forests of Sabah (Malaysia) were transformed into secondary forests due to heavy logging. Additionally the remaining forest remnants are isolated from each other by large scale oil palm plantations. Biodiversity patterns and responses of the community of leaf litter ants were studied in anthropogenically disturbed habitats and primary forests of different size. In logged over forests, only 70 per cent of the species of a primary forest were present even 25 years after timber extraction. The ant communities were thinned and could be described by a lower species density producing lower species numbers and a different community composition. The similarity in species number and community composition between logged over forests of different degrees of disturbance was explained by source-sink dynamics within a heterogeneous forest matrix. Rain forest fragments displayed even higher reductions in species density, numbers and diversity due to a more pronounced thinning effect. Even forest isolates exceeding 4 000 ha in size did not support more than 50 per cent of the species of the leaf litter ant community of a contiguous primary rain forest. Additionally, an increase in tramp species was recorded with decreasing size of the forest fragments, leading to a very different community composition. Regarding the leaf litter ant community, the remaining rain forest fragments of Sabah are effectively isolated by a barrier of oil palm plantation, now stretching all over the lowlands of the east coast. Only 13 species, which belonged to the forest ant community in highly disturbed areas were collected in these plantations. Some of the 10 other species of the highly reduced ground-dwelling ant community in the plantations are known as invasive tramp species, forming large exclusive territories. Correlative evidence and a field experiment implied, that leaf litter humidity, volume and temperature affect the distribution and community composition of forest leaf litter ant species. The smaller primary forests and the most disturbed logged over forests in this study revealed higher temperatures and lower humidity levels and a reduction in leaf litter volume compared to a large primary forest or forests affected by a lower impact of timber harvesting. If the pattern for leaf litter ants is confirmed for other taxa, the implications for any efficient management design aiming to preserve the majority of the biodiversity of the country are tremendous and current concepts need rethinking.
Leonia cymosa (Violaceae) is a small tree from the under story of the Amazonian rain forest. I investigated the seed dispersal ecology of L. cymosa in plots of old growth terra firme forest located within the Cuyabeno Faunistic Reserve in north-eastern Ecuador. This species offered good conditions to examine the variation of traits of individual trees and the way they are linked with fruit removal from each tree. With this study I aimed to address the question whether frugivores exert selection pressures on fruits and the fruiting regime of fleshy fruited plants. The mean height of a fruiting L. cymosa was 6.6 m (range: 2 - 12.6 m). The median tree density was 11.8 trees per hectare. Trees grew in clusters consisting of different numbers of trees of different heights. L. cymosa flowered two times a year, in late February to March and in October. The respective fruiting seasons occurred in August/September and between March and May. The fruit pulp of L. cymosa contained the sugars fructose, glucose, and sucrose, the total soluble sugar being the first important nutritional compound of the fruit pulp. The second important compound was proteins. No lipids were found in the fruit pulp. The variation of nutritional quality of the fruits was high within trees. Nonetheless, significant differences were found among trees in all nutrient constituents studied. The maximum of ripe fruits produced per season by a single tree was 427. Median productivity of the trees was 45 ripe fruits throughout the fruiting season in 1999 (n=57) and 36 ripe fruits in 2000 (n=92). The maximum standing crop of fruits in a tree was 324 fruits (counted in 2000). Black mantle tamarins, Saguinus nigricollis (Callitrichidae), and squirrel monkeys, Saimiri sciureus (Cebidae), and possibly an unknown nocturnal frugivore consumed the fruits of L. cymosa at my study site. Green-rumped acouchis (Myoprocta pratti, Dasyproctidae) consumed fallen fruits and seeds underneath the trees. Black mantle tamarins and squirrel monkeys differed widely in their effectiveness as seed dispersers. Black mantle tamarins swallowed the seeds together with the fruit pulp and defecated intact seeds far away from the mother tree. Squirrel monkeys opened the fruits to suck and gnaw on the fruit pulp, and then dropped seeds to the forest floor below the tree crowns. Each of my study plots fell into the core home range of one group each of S. nigricollis and S. sciureus. Thus, the frugivore assemblage is small and disperser availability is limited for the individual tree of L. cymosa. In a sample of 6 trees of comparable and high fruit crop size, the total of ripe fruits removed from a tree throughout the whole fruiting season by the reliable seed disperser S. nigricollis was neither significantly correlated with the content of any of the nutrients measured in the fruit pulp (fructose, glucose, sucrose, total protein; pulp does not contain lipids), nor with total metabolisable energy, seed to pulp weight ratio, or water content of the fruit pulp. Feeding preferences for single sugars determined by other laboratory studies were not confirmed by this field study. The reliable seed disperser S. nigricollis does not seem to exert selective pressure on the nutrient content of the fruits of L. cymosa. Seasonal fruit crop size was the main predictor of all aspects of fruit removal by the effective disperser of L. cymosa, Saguinus nigricollis, as well as by the non-disperser, Saimiri sciureus. Trees with larger seasonal fruit crop size had a higher probability to have fruits removed by the disperser than those with small seasonal fruit crop sizes. They also had a higher number of fruits removed by the seed disperser. However, the proportion of fruits removed by the disperser decreased with increasing seasonal fruit crop size. In contrast, probability of fruit removal, the number of fruits removed, and the proportion of fruits removed by the non-disperser increased with increasing seasonal fruit crop sizes. The observed differences between disperser and non-disperser are due to differences in feeding capacity, group size and foraging behavior. Tamarins were less likely to harvest Leonia trees that were not or less completely covered by surrounding vegetation. This probably reflects a behavior to avoid predation by forest raptors. At high con-specific fruit abundance in the neighborhood, the proportion of fruits removed by tamarins was reduced. This suggests competition of trees for the disperser. My study revealed selection of the disperser on seasonal fruit crop size of L. cymosa. My results are consistent with the “fruit crop size hypothesis”. FCSH appears to constitute a valid framework also in the monkey-dispersed L. cymosa. My findings also show that factors beyond the tree’s control influenced fruit removal from Leonia trees. Disperser-mediated selection may be constrained (yet not impeded) by neighborhood conditions.