Thesis (Ph.D.)--Mississippi State University. Department of Plant and Soil Sciences. / Title from title screen. Includes bibliographical references.

Thesis (M.S.)--Mississippi State University. Department of Plant and Soil Sciences. / Title from title screen. Includes bibliographical references.

Title from title page of PDF (University of Missouri--St. Louis, viewed March 10, 2010). Includes bibliographical references (p. 26-31).

Thesis (M.E.S.)--The Evergreen State College, 2006. / Title from title screen (viewed 3/11/2010). Includes bibliographical references (leaves 41-53).

Thesis (Ph. D.)--Ohio State University, 2009. / Title from first page of PDF file. Includes bibliographical references (p. 151-168).

Buffelgrass (Cenchrus ciliaris L.) is an invasive, perennial, C₄ bunchgrass. Buffelgrass has a rapid invasion rate, a tendency to displace native vegetation, and presents a fire risk to native plant communities, adjacent developed areas and their associated infrastructure. Mechanical control is impractical and unable to keep pace with regional spread. Chemical control has offered the most promise for successful and cost-effective management on a regional scale. The predominant herbicide used to control buffelgrass is glyphosate, which requires active vegetative growth when applied for optimum uptake and translocation to meristematic tissue. The timing and duration of active growth is difficult to predict. In this dissertation I addressed three related topics to improve effectiveness of buffelgrass management in the Sonoran Desert. First, I used digital time-lapse photography and weather data to predict the timing and length of future active growth based on day of year and antecedent weather at three sites in the Sonoran Desert near Tucson, AZ that were representative of habitats currently infested by buffelgrass in the region. I was able to correctly predict greenness above or below a threshold of herbicide susceptibility at 81 to 95% for a basin floor site and at 61-88% for slope sites. Second, I evaluated the effects of different rates of two herbicides (imazapic and clethodim), alone or in combination with different rates of glyphosate, for pre- (imazapic only) and postemergence control of buffelgrass. I found a minimum glyphosate application rate of 2.52 kg ae ha⁻¹ glyphosate consistently killed mature buffelgrass plants; clethodim at had no effect on mature buffelgrass; and imazapyr was successful in killing mature plants when applied during the dormant season at 0.56 kg ae ha⁻¹ and provided preemergence control. Imazapic severely damaged but did not kill mature buffelgrass plants at the maximum label rate of 0.21 kg ae ha⁻¹. Finally, we evaluated results from a helicopter broadcast herbicide application trial conducted in the Tucson Mountains. We demonstrated that most species and life forms were less affected than buffelgrass to glyphosate deposition rates achieved in the study, but asserted that procedural changes were necessary to achieve effective buffelgrass control.

Awareness of the spread of invasive plant species has grown, but quantitative measures of their impacts are lacking. This study analyses the impact of Himalayan blackberry (Rubus armeniacus) on breeding bird diversity finding a significant difference in bird diversity between “natural” and R. armeniacus-dominated understoreys. More bird species were noted in habitats with greater structural and compositional diversity. Simpson’s richness/evenness index was significantly different between habitat types for Stanley Park and Maplewood Flats (P<0.05) but not Jericho Park (P>0.05), likely due to lower overall bird diversity at Jericho Park and lack of overstorey trees at R. armeniacus thickets. When R. armeniacus is the dominant understorey shrub in a forested setting it has the greatest negative impact on breeding bird diversity.

According to Baker (1955), success of plant invasions can be attributed to autogamous breeding systems and generalist pollination systems. A test of Baker's rule was carried out on 19 invasive alien plant species in KwaZulu-Natal, South Africa. Natural levels of fruit set in these plants was high (median= 71.5% fruit set per plant). Breeding system experiments for 18 species showed that 17% of the species are allogamous, 72 % are autogamous and 11% are apomictic. This contrasts markedly with a general flora of 1472 species in which 65% are allogamous, only 14% are autogamous, and 21% are apomictic. Because the breeding systems of the alien species were largely autogamous, pollinators make only a small contribution to the overall reproductive success. Seventy one percent of the 14 alien species for which pollinators were obtained were pollinated by honeybees, which are super-generalists and ubiquitous in a generalist pollination systems. However, some species showed pre-adaptation to relatively specialized pollination systems, involving hawkmoths, butterflies and Xylocopa bees respectively. Thus, alien plants were not visited by a wide range of insects, but rather showed pre-adaption to one or more pre-existing guilds in the new habitat. Pollen limitation experiments showed no evidence that reproductive success in populations of the species is limited by pollen availability. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 2001.

Alien invasive vegetation threatens the functioning of natural ecosystems as well as their services, which sustain human welfare, both directly and indirectly. The proliferation of this vegetation in South Africa has been attributed to the ignorance of the social implications affiliated to their existence. Investment decisions are predominantly based on the marginal private costs of an activity, since individuals have not been made accountable for costs imposed on society in the past. Consequently, their marginal private costs always fall well short of the marginal social costs. In order to promote sustainable development as well as curtail widespread invasion by alien plants in South Africa, it is essential to close the gap between these costs and support research aimed at ascertaining monetary values for non-market goods, such as biodiversity. This study scrutinizes the economic viability of alien plant eradication in the Mhlatuze Catchment by comparing the costs and benefits associated with this eradication process. A series of formal in-depth interviews with major stakeholders residing in the upper reaches of the catchment were conducted in an attempt to evoke individuals' perceptions concerning alien vegetation together with the actual costs incurred in extirpating this vegetation. The contingent valuation method (CVM), which relies on surveys to elicit the maximum amount respondents would be willing to pay to obtain or retain some nonmarket good, was employed to glean monetary values for benefits such as biodiversity and augmented streamflow. The findings suggest that there is a positive net effect associated with alien vegetation extirpation in the Mhlatuze Catchment, since the social benefits outweigh the costs. Given the temporal and financial constraints that prevailed, it was only possible to examine some of the benefits affiliated to this process, such as increased timber harvesting, streamflow, biodiversity and reduced fire hazard. Consequently, this analysis represents a minimum estimate of the benefits, further enhancing the argument in favour of extirpation. The results lend support to calls for greater policy emphasis on, as well as funds for, the eradication of alien invasive plants. / Thesis (M.A.)-University of Natal, Pietermaritzburg, 2001.

Common mullein (Verbascum thapsus) is a non-native plant threatening the integrity of natural ecosystems throughout California. Mechanical, chemical, and biological methods have been utilized for control of common mullein. The objective of this study was to determine which of the following treatment cycles is most effective to reduce populations of common mullein at Lava Beds National Monument: annual (once per year), biennial (once every two years), or no treatment. Vegetation type, natural features, site characteristics, and human disturbance were factors considered when analyzing common mullein distribution and abundance. Forty-five pre-existing plots with populations of common mullein were visited during summer 2005. The interrelationship of common mullein with the surrounding high-desert plant communities was assessed by three vegetation sampling methods. Density, cover, and common mullein stages were measured by point-intercept transects, belt transects, and circular macroplots. Simple linear regressions were performed using common mullein density as the dependent variable with multiple independent variables such as shrub, rock, soil, and grass cover. Analysis of variance test showed no significant relationship between density of common mullein and the three treatment cycles (p > 0.05). However, trends observed imply that annual treatment reduces the population of large rosettes and reproductive stalks over no-treatment and biennial treatment. A bivariate correlation test found the circular macroplot to be similar to belt transects for measuring common mullein density (r = 0.84). / Department of Natural Resources and Environmental Management