Natural communities are structured by a complex of direct and indirect species interactions. It is well recognized theoretically that if these interactions are weak, the entire community is more likely to persist. Several mechanisms can weaken a predator-prey interaction. I studied interference among conspecific predators and heterogeneity at the prey level. Incorporating these mechanisms into realistic functional responses is required for accurate model predictions at the community level. However, controversy remains on which dependencies need to be included. Using laboratory microcosms, I was able to demonstrate moderate predator-dependence in my model system. This effect was present even at low predator densities and after accounting for prey depletion. In separate experiments, I experimentally compared the functional responses of a gape-limited predator feeding on its prey in the absence or presence of species outside the predator's diet (non-prey). I demonstrated that both density and diversity of non-prey species can also substantially reduce the strength of predator-prey interactions. I further tested this non-prey effect on a long time-scale, where I compared the population dynamics of predator and prey alone to the dynamics of predator and prey when a non-prey species was present. Prey and predators had both gone extinct at the end of the experiment for all replicates containing only predator and prey. However, in the treatment that included non-prey, all species persisted in 4 out of the 5 replicates until the last day of the experiment. Prey species also spent significantly more time above the detection limit in the treatment with non-prey. In addition, I studied how inducible anti-predator defenses affect an aquatic food web with intraguild predation. I detected substantial clonal variation in the expression of a morphological inducible defense and the long-term experiment showed that the intraguild prey with greater ability to increase their body width as a response to predation also have longer persistence times. These results show that predator interference and prey heterogeneity may be important factors that increase persistence of predator and prey as predicted by theory. Their incorporation into simple food web models can improve our ability to reliably predict community dynamics.

Chipmunk antipredator calls were examined during the summers of 1990 and 1991. The structural characteristics of the three calls, chipping, chucking and the trill, were obtained through taped recordings and sonagraphic analysis. Behavioural observations indicated that chucking by choruses of individuals occurs in the presence of aerial predators, chipping by choruses of individuals occurs in the presence of terrestrial predators and the trill is given by single individuals when fleeing from predators. / Experiments were carried out to determine the function of the trill and chipping. Demographic and contextual effects indicate that the trill is in part a call which functions to warn kin but may also indicate to conspecifics that the caller has escaped into a refuge. An experiment with a tethered cat concluded that chipping is likely to function to deter predators from hunting in the area. This is done through vocal mobbing by several individuals whose home ranges overlap.

In this study, predation efficiency of three guppy ( Poecilia reticulata) size classes was evaluated at various vegetation densities in a series of circular aquaria containing plastic plants closely resembling Ceratophyllum demersum. The effect of vegetation density was most apparent among large fish where predation efficiency greatly diminished from 12.35 mosquito larvae (Aedes aegypti) to 4.68 as vegetation densities rose from 3 to 19 plants/700 cm2. Over that same range of increasing vegetation densities, predation among small fish remained unchanged, whereas among fish of intermediate size predation declined significantly but less precipitously than for large individuals. When presented with a choice between second and fourth mosquito larvae, small fish preferred to prey on second instars, whereas large fish preferred fourth instar larvae. Fish of intermediate size did not show a statistically significant preference. The functional response of fish to increasing prey densities, as measured over a 12-h period, was of Type III.

Forest tent caterpillar (Malacosoma disstria Hübner; FTC), a major defoliator of aspen trees, occupies both aspen and mixedwood forest stands in Alberta’s boreal forest. Forest stand composition could influence the spatial pattern of FTC outbreaks if mortality from natural enemies differs between stand types. I conducted field experiments to determine whether predator- or parasitoid-caused mortality of FTC differed between aspen and mixedwood forest stands and developed a spatial population model to determine the effects of variation in generalist predation on the spread of an FTC outbreak, including the effects of potential predator-caused Allee effects. Generalist predation on FTC was higher in aspen stands than in mixedwood stands, and the spatial model suggests that these observed differences may be sufficiently large to impact FTC outbreak spread rates. Forest stand composition may contribute to the spatial pattern of FTC outbreaks through variation in the impacts of predators on FTC populations. / Ecology

Predation is a frequently studied subject, but estimating crushing predation in mollusk communities is challenging. Shells record successful attacks, but it is not always possible to identify attacks on an individual basis. Repair scar frequency is a common proxy for crushing mortality, but shell repair does not directly measure mortality, so results are ambiguous. Borrowing a technique from Vermeij (1982), crushing mortality frequencies were estimated in a recent shell assemblage. Because crushing damage can be confused with taphonomy, a taphonomic baseline was established: the cause of death of drilled shells is known, so additional damage is postmortem. The frequencies of several damage types were tallied for drilled shells to estimate a taphonomic baseline for the assemblage. The same frequencies were calculated for undrilled shells (cause of death unknown). In many cases, undrilled shells had significantly higher frequencies than drilled shells. The differences in damage frequencies likely are caused by crushing predation.

Thesis (Ph. D.)--Illinois State University, 2007. / Title from title page screen, viewed on February 11, 2008. Dissertation Committee: Steven A. Juliano (chair), Diane L. Byers, L. Philip Lounibos, Charles F. Thompson, William L. Perry. Includes bibliographical references (leaves 152-163) and abstract. Also available in print.

Diss. (sammanfattning) Umeå : Sveriges lantbruksuniversitet, 2005. / Härtill 5 uppsatser.

Thesis (Ph.D.)--University of Adelaide, Dept. of Applied Mathematics, 2000? / Erratum pages inserted onto front end papers. Bibliography: leaves 226-244.

Thesis (Ph.D.) - James Cook University, 2004. / Typescript (photocopy) Bibliography: leaves 127-154.

Theses (Ph.D.)--University of New Hampshire (Dept. of Zoology), 2007. / Includes bibliographical references. Also available online.