This work examined the importance of structural complexity of habitat, availability of prey, and competition with ants as factors influencing the abundance and community composition of arboreal spiders in western Oregon. In 1993, I compared the spider communities of several host-tree species which have different branch structure. I also assessed the importance of several habitat variables as predictors of spider abundance and diversity on and among individual tree species. The greatest abundance and species richness of spiders per 1-m-long branch tips were found on structurally more complex tree species, including Douglas-fir, Pseudotsuga menziesii (Mirbel) Franco and noble fir, Abies procera Rehder. Spider densities, species richness and diversity positively correlated with the amount of foliage, branch twigs and prey densities on individual tree species. The amount of branch twigs alone explained almost 70% of the variation in the total spider abundance across five tree species. In 1994, I experimentally tested the importance of needle density and branching complexity of Douglas-fir branches on the abundance and community structure of spiders and their potential prey organisms. This was accomplished by either removing needles, by thinning branches or by tying branches. Tying branches resulted in a significant increase in the abundance of spiders and their prey. Densities of spiders and their prey were reduced by removal of needles and thinning. The spider community of needle-sparse branches was dominated by orb weavers (Araneidae), whereas tied branches were preferably colonized by sheet-web weavers (Linyphiidae and Micryphantidae), and nocturnal hunting spiders (Anyphaeilidae and Clubionidae). Spider species richness and diversity increased in structurally more complex habitats. In 1994 and 1995, I excluded foraging Camponotus spp. ants from canopies of sapling Douglas-fir. Biomass of potential prey organisms, dominated by Psocoptera, increased significantly by 1.9 to 2.4-fold on the foliage following ant exclusion. Hunting spiders, dominated by the Salticidae, increased significantly by 1.5 to 1.8-fold in trees without ants in the late summer. The exclusion of ants did not affect the abundance of web-building spiders. Documented aggressive behavior of aphid-tending ants suggests interference competition between hunting spiders and ants. / Graduation date: 1996

published_or_final_version / Ecology and Biodiversity / Master / Master of Philosophy

Female Misumenoides formosipes crab spiders, unlike their highly mobile, short-lived male counterparts, are relatively sedentary predators that forage long after the males have died off. Associated with their foraging ability is a phenomenon rare amongst animals: the capacity for a reversible change in body color. This color change ability, exhibited by several species of crab spiders, has historically been interpreted as an adaptation providing enhanced crypsis during movement between hunting substrates (inflorescences). Misumenoides formosipes females were relocated onto matched and mismatched substrates in the field to assess their propensity for color change, the rate at which it occurs, and any impact on foraging success. Yellow females transferred to white inflorescences were the only category that did not remain in their new location. White females changed to yellow over a 9 day period. We found conflicting evidence as to whether or not foraging success was enhanced for females on matched backgrounds. / Department of Biology

Stegodyphus mimosarum Pavesi,1883 and S. dumicola Pocock, 1898 are two species of philopatric, inbred, permanently communal, non-territorial spiders that co-occur in parts of South Africa. The patchiness of colony distribution, limited dispersal capabilities and the observation of periodic, but rare mass dispersal events raised interest in factors influencing dispersal. The aim of this project was firstly, to determine which factors influence the spiders' readiness to leave a colony (two laboratory experiments), and secondly, to map nest dispersion in Weenen Nature Reserve, Kwa-Zulu Natal, and to use this to explain nest distribution. The first experiment assessed whether group size and variance in access to resources influenced the decision to disperse. Four colony sizes (8, 16,32 and 64) of S. mimosarum were established under a proportional feeding regime. I expected more spiders to leave larger colonies due to intra-group competition. However, there was no significant increase in the number of spiders leaving with increasing group size. Significantly more spiders left a colony during spring and when spiders were large. In the second experiment, I assessed whether the mean amount of food available, in liberally fed or starved colonies influenced the decision to disperse. Five colonies were fed daily on an abundance of prey items and five were starved. I expected more spiders to leave the starved colonies. However, a significant number of spiders left colonies where food was abundant. During a field survey nests were tagged within 40 plots of 50 m radius, and randomly. Retreat dimensions, height above ground, nest position, nearest neighbouring nests, and species were recorded. Nest status was tracked over six months to three years. I confirmed that nest height above ground was significantly higher for S. mimosarum. The two species differed in retreat volume and nearest neighbour distances. Most nests occurred on the northern aspect of trees. Few nests survived beyond three years, although many new nests were established. Access to resources influenced the decision to disperse. However, only well-fed (larger) spiders had the resources to relocate. Patchy nest distribution could be a consequence of nest site selection, short distance dispersal by budding and bridging, and long distance dispersal by ballooning. / Thesis (M.Sc.)-University of Natal, Durban, 2002.

The objective of this thesis was to understand the spatial patterns and processes responsible for canopy and understorey spider (Arachnida: Araneae) diversity at multiple spatial scales in north-temperate hardwood forests. I sampled tree trunks (sticky traps) and foliage (beating) of sugar maple and American beech tree canopies and their understorey saplings in old growth forests near Montreal, Quebec. Results show the composition of canopy and understorey assemblages differed significantly, and so did sugar maple and American beech canopy assemblages. Each stratum was also dominated by different species. The rank-abundance distribution of species from each habitat wsa also verticaly stratified because it fit different distribution models. Different factors likely structure assemblages in both habitats, particularly since the canopy is a less stable environment. Spiders from canopy and understorey foliage were tested in a laboratory for their propensity to balloon. General linear models indicated that small sized web-building spiders of the RTA and Orbicularia clades have the highest propensity to balloon. Small bodied species initiated ballooning regardless of the habitat they were collected in or their developmental stage. My results support the mixed evolutionarily stable strategy theory and indicate the absence of risk-spreading in the dispersal strategy of canopy spiders. My last chapter focused on dispersal capacity and diversity patterns of spiders at multiple spatial scales. Analyses of the species diversity of limited and high dispersal capacity species subsets through nested-multivariate ANOVA, additive diversity partitioning, and species-abundance distribution curves all point towards species-sorting processes as the main driver of local community spider diversity at the tree and stand spatial scales. Mass-effects and patch-dynamic processes drive site and regional scale diversity patterns. This thesis demonstrates that spiders provide good models to test many biological hypotheses. The research chapters of this thesis test hypotheses on the vertical stratification of forest spider diversity, the evolution of local dispersal adaptations, and the importance of dispersal capacity on species diversity patterns through a metacommunity framework.

No description available.