Japanese climbing fern (Lygodium japonicum) is an invasive vine that has been identified as an economic and ecological threat in forest ecosystems of the Southeast. In two separate studies, we examined the use of directed sprays of glyphosate, imazapyr, and metsulfuron-methyl herbicides, alone and in combination, for control of Japanese climbing fern and for impacts to associated vegetation in mixed bottomland hardwood–cypress forests at two first terrace sites of the Apalachicola River in north Florida. Two yr after treatment, various rates of glyphosate alone generally provided greater percent control (84 to 95%) than imazapyr alone (−25 to 86%) or metsulfuron-methyl alone (−5 to 53%). Combinations of 2% glyphosate and imazapyr and/or metsulfuron-methyl resulted in 81 to 97% control at 2 yr after treatment, with no significant differences among the combination treatments or 2% glyphosate alone. Change in percent cover of associated vegetation groups was not influenced by treatment except for one location where graminoid cover increased 1 yr after metsulfuron-methyl treatment but decreased after imazapyr or glyphosate application. Species richness of dominant understory vegetation remained nearly constant at both locations regardless of treatment. These data indicate that glyphosate alone or in combination with imazapyr or metsulfuron-methyl will provide effective, although not complete, control of Japanese climbing fern, and that directed herbicide sprays minimally impact nontarget vegetation.

Sulfur cinquefoil is native to Eurasia and has invaded meadows and grasslands of North America. It occurs in intermountain valleys of the Rocky Mountains, including a nature reserve in northwest Montana that supports a mosaic of mesic fescue and xeric needlegrass grassland communities. We did tests to determine germination requirements, followed the fate of mapped sulfur cinquefoil from 1998 to 2005 to determine vital rates, and used a matrix modeling framework to analyze population-level dynamics. Sulfur cinquefoil plants were highly fecund; large plants produced > 10,000 seeds that require light for germination. The survivorship curve of the 1999 cohort suggested low juvenile but high adult survival. Simulations indicated that a 10% decline in survival would decrease population size more than a 10% decline in recruitment. Annual recruitment, growth of nonreproductive plants, survival, and flowering frequency of sulfur cinquefoil were all higher in the three needlegrass sample plots compared with the one fescue plot, and projected equilibrium sulfur cinquefoil populations were approximately 12 times larger in needlegrass than fescue grasslands. Our results suggest that biological control agents that negatively affect survival, such as root borers, might be more effective at controlling established sulfur cinquefoil populations than those that diminish seed production. Our results also suggest that sulfur cinquefoil might have more potential to become dominant in xeric grasslands and demonstrate the need for doing multiple demographic studies over a variety of habitats because the same species will respond differently in different habitats.

We monitored colonization, spread, and growth of an exotic Old World climbing fern on tree islands in the interior marsh of Arthur R. Marshall Loxahatchee National Wildlife Refuge, Florida to investigate linkages between its colonization and growth and environmental factors. Number of the patches continuously increased over a 2-yr monitoring period, but mean patch area and height did not increase. Lower water depth was associated with larger number of patches whereas higher water depth was associated with higher mean patch height. Patch occurrence probability at each 3 m (9.8 ft) grid within tree islands also increased and indicated that patches spread from outer areas to interiors of tree islands over time. Patch occurrence probability was lower in sawgrass-dominated locations, which typically occur at the margin of tree islands, compared to locations dominated by woody plants. Patch occurrence probability was higher on west sides of tree islands corresponding to the side of the islands most impacted by a recent major hurricane. This study revealed that colonization and growth of Old World climbing fern are characterized by several spatial and temporal factors. A detection protocol using ground surveys that focus on the edges of tree islands could be an efficient way to identify early infestations on tree islands in the Refuge. Such a protocol also might be useful to identify new infestations on tree islands in other areas of the Everglades.

As Europeans colonized California, they introduced annual grasses from the Mediterranean Basin. These exotic annual grasses eventually invaded grasslands throughout the state, some of which were once dominated by native perennial grass species. Annual grasses differ from perennials in their phenology, longevity, rooting depth, litter chemistry, and interaction with the microbial community. As these traits may influence plant nitrogen (N) use, it is likely that the invasion by annual species resulted in changes in the availability and cycling of N in California grassland systems. We addressed the question of how invasive annual grasses influence rates of N cycling by measuring N pool sizes and rates of net and gross mineralization and nitrification, gross immobilization, and the denitrification potential of soils from experimentally planted annual and perennial-dominated grasslands. With an increase in annual grass cover, we saw increases in ammonium (NH4+) pool sizes and rates of N mineralization, nitrification, and denitrification in soils. These differences in N status suggest that N cycling in California grasslands was altered at sites where native perennial bunchgrasses were invaded by nonnative annual grasses. One consequence of annual grass invasion may be a legacy of NH4+-enriched soils that hinder the reestablishment of native perennial grass species.

Yellow starthistle (Centaurea solstitialis) is an exotic winter annual forb that is aggressively invasive and problematic in much of California, Idaho, Oregon, and Washington. Yellow starthistle control is particularly challenging in canyon rangelands where accessibility limits control options. Our objective was to evaluate the effects of late-season targeted goat grazing on yellow starthistle and nontarget grasses and forbs. A 3-yr grazing study was initiated in 2006 on a 380-ha (939 acres) canyon grassland site infested with yellow starthistle near White Bird, ID. Twenty-four paired plots were established, with each pair including a fenced subplot to exclude grazing and a similar-sized adjacent subplot that was grazed. Density of yellow starthistle plants and seedheads was assessed after grazing of each plot in all 3 yr and before grazing in the second and third years. Canopy cover of yellow starthistle, grasses, and forbs also was measured. Grazed subplots had 58% fewer yellow starthistle plants than the ungrazed controls after grazing was applied and 94% fewer seedheads after 3 yr of grazing. Cover of yellow starthistle did not differ between grazed and ungrazed subplots after grazing in 2006, whereas grazing decreased yellow starthistle cover in 2007 and 2008 by about 75%. Goat grazing had little impact on canopy cover of grasses and resident forbs, with the exception of after grazing in 2007 when there was less forb cover in grazed areas compared with ungrazed areas. Late-season (i.e., July to November) targeted goat grazing appears to be an effective way to reduce yellow starthistle plant densities at landscape scales, which creates a large window of opportunity for grazing treatment and flexibility for land and livestock managers.

Focus group methodology in a field demonstration setting was used to obtain qualitative data on the perceptions of family forest owners relating to treatment efficiency and feasibility of herbicide control methods. Interviews took place on sites where various strategic herbicide treatments were implemented for Chinese privet (Ligustrum sinense) control using the active ingredients glyphosate and metsulfuron. Forest owners expressed unease about the possibility for post-treatment privet reestablishment due to reseeding or other factors and opinions surfaced calling for selective chemicals or application methods that would spare non-target species. Furthermore, treatment cost effectiveness with regard to timber value, the possible need for expensive multiple treatments, cost-share incentives, and treatment guarantees from herbicide applicators were participant concerns. Environmental concerns surfaced about possible effects of both herbicide use and the invasion of privet on natural systems and an unexpected result was a strong feeling among the forest owners that focus groups are a powerful demonstration tool.

Scotch broom is an invasive leguminous shrub in California and other Pacific Northwest states, as well as New Zealand and Australia. It is highly competitive in forest and shrub communities and can significantly impact reestablishment of conifer forests. The objective of this experiment was to evaluate mechanical methods (Weed Wrench, lopping), several herbicides, and herbicide application techniques for control of Scotch broom in a premontane site in California. Three herbicides were evaluated (glyphosate, imazapyr, and triclopyr ester) for canopy reduction using foliar, drizzle, and basal bark treatments. All treatments were made in both fall and late spring. In addition, we conducted a cost analysis of the various herbicide treatments and application methods. Results indicate that both mechanical treatments were effective, but their optimum timing depended on soil moisture conditions. In addition, there were no significant differences among herbicides at both timings, among all rates, and for any application method. All herbicides provided effective control of Scotch broom. However, the cost analysis demonstrated that the drizzle application method with glyphosate was the most cost-effective treatment, due to low herbicide cost and reduced labor requirements. These results provide several options for Scotch broom control and give land managers considerable flexibility with timing, herbicide, and application technique in their management programs.

Garlic mustard, a biennial forb native to Europe, has invaded native ecosystems in forested regions in the United States. In anticipation of a biological control program being implemented in the United States for this plant, a garlic mustard monitoring program was initiated. The objective of this study was to characterize garlic mustard populations and the associated plant communities and their response to environmental conditions in Minnesota hardwood forest ecosystems. Additionally, we developed a baseline for long-term studies to determine future benefits and impacts of biological control agents on plant communities infested with garlic mustard, should they be released. To monitor garlic mustard populations, we used a nationally standardized protocol in which data were collected on garlic mustard population density and cover, garlic mustard plant heights and silique production, insect damage to garlic mustard, cover of the associated plant community, and litter cover. We also collected data on available photosynthetically active radiation in the understory. The results underscore the variability in garlic mustard population dynamics. At only 6 of 12 sites did garlic mustard densities follow the predicted two-point cycles due to their biennial life cycle, with the first- or second-year life stage dominating in any given year. Available light did not differ strongly among sites, but shading by adult plants is implicated in keeping the populations of first-year plants low. Sites with greater garlic mustard cover had lower native species richness and cover than sites with lower garlic mustard cover. Absent biological control agents, garlic mustard is currently experiencing very little herbivory in Minnesota with an average of 2% of leaf area removed by herbivores. Our work shows the importance of pre-release monitoring at multiple sites over multiple years to adequately characterize populations. Without control, garlic mustard will likely continue to have negative impacts on northern forests.

Parrotfeather is an invasive, aquatic plant in the United States that is native to South America. It has impaired the use of water bodies throughout the United States and is difficult to control, despite using a variety of management techniques. Our objectives were to examine the efficacy of subsurface applications of seven herbicides labeled for aquatic use and to compare those applications to herbicides that can also be applied to emergent foliage. A replicated mesocosm study was conducted in 378-L (100-gal) tanks beginning in August 2007 and repeated during the same period in 2008. The maximum and half-maximum labeled rates of copper chelate, diquat, endothall, fluridone, triclopyr, and carfentrazone-ethyl were applied to the water column in designated mesocosms. The maximum labeled rate for foliar applications of diquat, triclopyr, and 2,4-D were used to compare treatment methods. Six weeks after treatment (WAT), copper, endothall, fluridone, and carfentrazone-ethyl did not achieve 90% control; in fact, control was less than 50% for each herbicide, and therefore, the herbicides were not considered efficacious for controlling parrotfeather. Diquat at all rates and application methods resulted in 70 to 90% biomass reduction. Triclopyr, with both the highest aqueous concentration and foliar application, resulted in an 84 and 86%, respectively, reduction in biomass at 6 WAT. The foliar application of 2,4-D was the only herbicide and application method that resulted in ≥ 90% biomass reduction of parrotfeather. In these studies, regrowth occurred in all tanks regardless of herbicide or treatment method, indicating multiple applications would be necessary to provide longer-term plant control. Future research should identify possible herbicide combinations or timing of applications to maximize treatment efficacy.

Aminopyralid efficacy on Canada thistle (Cirsium arvense) and potential to injure native species was evaluated in a restored prairie at the Glacial Ridge Preserve managed by The Nature Conservancy in Polk County, MN. Canada thistle stem density was reduced from 17 to 0.1 stems m−2 10 mo after treatment (MAT) with aminopyralid applied in the fall at 120 g ha−1. Aminopyralid also altered the composition of both Canada thistle–infested and native plant communities. Aminopyralid controlled Canada thistle and removed or reduced several undesirable forb species from the restored prairie communities, such as absinth wormwood (Artemisia absinthium) and perennial sowthistle (Sonchus arvensis). A number of high seral forbs were also reduced or removed by aminopyralid, including maximilian sunflower (Helianthus maximiliani) and purple prairie clover (Dalea purpurea). Foliar cover of high seral forbs in the native plant community was reduced from 12.2 to 7% 22 MAT. The cover of high seral grass species, such as big bluestem (Andropogon gerardii) and Indiangrass (Sorghastrum nutans) increased after aminopyralid application in both the Canada thistle–infested and native plant communities and averaged 41.4% cover compared with only 19.4% before removal of Canada thistle. Species richness, evenness, and diversity were reduced after aminopyralid application in both Canada thistle–infested and native plant communities. However, the benefits of Canada thistle control, removal of undesirable species, and the increase in native grass cover should lead to an overall improvement in the long-term stability and composition of the restored prairie plant community, which likely outweigh the short-term effects of a Canada thistle control program.

An open-field test was conducted in southern France to assess the host-specificity of Ceratapion basicorne (Illiger), a candidate for biological control of yellow starthistle. Test plants were infested by naturally occurring populations of C. basicorne but were also exposed to sympatric herbivore species, including other Ceratapion spp. Insects from the test plants were collected directly into tubes of ethanol and were subsequently identified to species according to DNA sequence similarity with morphologically identified reference specimens. This integrated, morphological and molecular identification method was used in an effort to maximize the amount of data gained in the field bioassay and to minimize the number of taxonomist–hours necessary to complete the study. The results obtained showed that the French C. basicorne population only attacked yellow starthistle and cornflower, another known host of C. basicorne. Molecular phylogenetic analysis of the insects collected from all other nonhost plants rejected the possibility that any were C. basicorne.

Japanese barberry is listed as an invasive shrub in 20 states and four Canadian provinces. Control of Japanese barberry was evaluated using several two-step processes over 16 mo using a total of 1,100 clumps at six study areas. Initial treatments in spring (prescribed burning, mechanical mowing with a brush saw or rotary wood shredder) reduced the size of established barberry clumps. Follow-up treatments in midsummer to kill new ramets that developed from surviving root crowns were foliar applications of triclopyr or glyphosate, directed heating with a propane torch, and untreated controls. Mortality was defined as the absence of ramets from a root crown and not the mortality of individual ramets of a given clump. Clump mortality and size of new ramets did not differ among initial treatments. However, larger clumps had higher survival and larger sprouts than smaller clumps 16 mo after initial treatment. Effectiveness of follow-up treatments varied by clump size. Two follow-up treatments of directed heating using propane torches were as effective as herbicides for clumps that were initially smaller than 120 cm. For clumps with pretreatment sizes of 120 cm and larger, clump mortality following herbicide treatments (90%) and directed heating (65%) was greater than for clumps that had no follow-up treatments (35%). Although clump sizes did not differ between follow-up methods 1 yr after treatment, both follow-up treatments resulted in smaller clumps than untreated controls. Effective control of Japanese barberry can be achieved in a single growing season by integrating an early-season initial treatment (prescribed fire or mechanical) that kills the aboveground tissues with a midseason follow-up treatment such as directed heating or targeted herbicide application.

Broadleaf herbicides are commonly used to suppress exotic weeds with the intent of releasing native species from negative impacts of invasion. However, weed control measures can also have unintended consequences that should be considered along with intended effects. We conducted a controlled field experiment within bunchgrass communities of western Montana to examine if broadcast application of the broadleaf herbicide, picloram, may mitigate impacts of the exotic forb, spotted knapweed, on the dominant native grass, bluebunch wheatgrass, and forb, arrowleaf balsamroot. Local-scale relationships between native species and spotted knapweed cover served as a baseline for evaluating treatment effects at differing spotted knapweed invasion levels. To examine secondary invasion, we also measured treatment effects on the exotic grass, downy brome, relative to initial levels of spotted knapweed cover. Picloram application suppressed spotted knapweed cover by 70 to 80%. Treatment appeared to release cover and seed production of bluebunch wheatgrass, causing increases that varied positively with initial spotted knapweed cover. Bluebunch wheatgrass measures were elevated by as much as fourfold in treated vs. control plots, exceeding baseline levels in noninvaded plots. For arrowleaf balsamroot, negative effects of treatment prevailed, particularly where initial spotted knapweed cover was low. Arrowleaf balsamroot cover and fecundity variables were reduced by as much as 60% in treated vs. control plots, to levels typifying baseline conditions in highly invaded plots. In addition, treatment released downy brome, with cover increases from 2- to 20-fold. A controlled experiment selectively removing spotted knapweed showed similar release of downy brome. Our results show that picloram effects can depend on initial levels of weed invasion and may include substantial side effects, particularly when broadcast applications are used. Integrated approaches that include seeding of desirable species may be needed to enhance plant community resistance to secondary invaders and reinvasion by the target weed.

Invasion of nutrient-poor habitats might be related to the ability of a species to exploit nutrient-rich microsites. Recent research suggests fast-growing species might have a greater ability to allocate root biomass to nutrient-rich microsites (root foraging precision) than slow-growing species. We examined if differences in relative growth rate (RGR) between invasive and native species were related to differences in foraging precision. We hypothesized that invasive species would: (1) have greater foraging precision than native species but (2) greater foraging precision would come at a cost in terms of root nutrient uptake rate. Foraging precision was evaluated on plants growing in soils with uniform or patchy nutrient distribution. Plants were harvested at a common time and a common developmental stage to separate indirect effects of RGR on foraging. Nutrient uptake rate was examined by exposing plants to a low or high nitrogen pulse. Invasives foraged more precisely than natives but had lower nitrogen uptake rate. Although these results support the idea of a positive relationship between RGR and foraging precision, biomass production in heterogeneous soils showed no relationship to foraging precision. Instead, species with greater RGR produced more biomass and root length across all treatments, allowing greater nutrient capture in heterogeneous soils. Although these results do not exclude a role for proliferation in influencing invasion of nutrient-poor systems or the potential for heterogeneity to influence population processes, these results suggest other traits may have an overriding importance in determining invader success in these systems.

Many large-scale management programs directed toward the control of waterhyacinth rely on maintenance management with herbicides. Improving the implementation of these programs could be achieved through accurately detecting herbicide injury in order to evaluate efficacy. Mesocosm studies were conducted in the fall and summer of 2006 and 2007 at the R. R. Foil Plant Science Research Center, Mississippi State University, to detect and predict herbicide injury on waterhyacinth treated with four different rates of imazapyr and glyphosate. Herbicide rates corresponded to maximum recommended rates of 0.6 and 3.4 kg ae ha−1 (0.5 and 3 lb ac−1) for imazapyr and glyphosate, respectively, and three rates lower than recommended maximum. Injury was visually estimated using a phytotoxicity rating scale, and reflectance measurements were collected using a handheld hyperspectral sensor. Reflectance measurements were then transformed into a Landsat 5 Thematic Mapper (TM) simulated data set to obtain pixel values for each spectral band. Statistical analyses were performed to determine if a correlation existed between bands 1, 2, 3, 4, 5, and 7 and phytotoxicity ratings. Simulated data from Landsat 5 TM indicated that band 4 was the most useful band to detect and predict herbicide injury of waterhyacinth by glyphosate and imazapyr. The relationship was negative because pixel values of band 4 decreased when herbicide injury increased. At 2 wk after treatment, the relationship between band 4 and phytotoxicity was best (r 2 of 0.75 and 0.90 for glyphosate and imazapyr, respectively), which served to predict herbicide injury in the following weeks.

Meadow hawkweed is an invasive, perennial forb of upland forest openings, mountain meadows, permanent pastures, and abandoned farmlands in the Pacific Northwest. The objective of this study was to measure the plant community response, following meadow hawkweed control using selective herbicides and a single application of N–P–K fertilizer, across three levels of meadow hawkweed infestation in cleared forestlands that had been converted to pasture in northern Idaho. Clopyralid was applied to meadow hawkweed rosettes at a rate of 0.59l kg ae ha−1 (0.528 lb ac−1), and fertilizer (23–5–5, 1% Fe, 14% S) was broadcast following herbicide applications at two rates, 44 and 88 kg N ha−1. Foliar cover of meadow hawkweed, Idaho fescue, and other perennial grasses was measured in microplots at three levels of initial meadow hawkweed cover; < 25%, 40 to 60%, and > 75%. Clopyralid treatments resulted in excellent meadow hawkweed control 3 mo after treatment (MAT). Clopyralid alone appeared to shift the competitive balance in favor of perennial grasses, which increased 7, 44, and 65% above pretreatment levels in LOW, MED, and HIGH hawkweed cover plots, respectively, across fertilizer treatments. Meadow hawkweed cover remained < 5% in clopyralid plots at 52 MAT. Fertilizer effects were confined to interactions between Idaho fescue and other perennial grasses.

Noncropland such as levees, roadsides, field borders, fencerows, and wildlife areas are vulnerable to weed invasion. Many sites have undergone frequent human disturbance, such as manipulation from surrounding land uses, and lack competitive, desirable vegetation. This study addressed the importance of revegetation in an integrated weed management program including revegetation for noncrop areas. The study evaluated 14 cool-season perennial grasses (seven native species and eight introduced species) for their establishment, vigor, and ability to suppress weeds. It also evaluated the impact of herbicides on weed control and grass establishment. Treatments were applied at three noncrop sites in Northeast California that were heavily infested with weeds. Chemical weed control during the year of seeding and the following year was critical for perennial grass establishment. Weed cover was greater than 50% whereas average seeded grass cover was less than 6% in untreated plots at all sites 2 yr after seeding. In contrast, average seeded grass cover at all sites was 22 to 31% 2 yr after seeding for treatments where herbicide use resulted in wide-spectrum weed control and grass safety. Increasing perennial grass cover decreased total weed cover across perennial grass species 1and 2 yr after seeding. Individual grass species' cover differed among sites. Two introduced grasses (tall wheatgrass and crested wheatgrass) and three native grasses (western wheatgrass, bluebunch wheatgrass, and thickspike wheatgrass) showed broad adaptation and had > 20% cover at all sites 2 yr after seeding. In herbicide-treated plots, these grasses reduced total weed cover by 43 to 98% compared to unseeded plots 2 yr after seeding.

The extent and nature of spread of exotic plant species to and within Alaska by shipment of hay and straw was studied. The amounts of hay and straw imported into Alaska and the amounts and types of seed in imported and locally produced hay and straw was determined We purchased alfalfa hay, wheat straw, ryegrass straw, and timothy hay produced in Washington and Oregon (WA–OR) and locally produced straw and hay. The hay and straw were shaken over screens, and the remaining fines were mixed with sterile potting soil and incubated in the greenhouse. Forty-nine plant species were identified from hay and straw, nine of which are ranked as invasive in Alaska, including downy brome, foxtail barley, hare barley, narrowleaf hawksbeard, and quackgrass—a prohibited weed in Alaska. The number of seeds ranged from 0 to 6,205 seeds kg−1, with an average of 585 seeds kg−1, and the number of species ranged from 0 to 12, with an average of 4.2 species per bale. Crop seed comprised a large proportion of the germinated seeds in ryegrass straw, wheat straw, and timothy/brome hay (98, 78, and 62%, respectively), but was less prevalent (ranging from 0 to 38%) in the other three hay and straw crop types. Hay and straw from Alaska contained more total seeds and species than hay and straw from WA-OR, but the difference was not significant when only weed seed was used in the analysis. Alaska-grown timothy/brome hay contained significantly more total seed than alfalfa hay and wheat straw from WA–OR and Alaska-grown barley straw. The grower or distributor of the hay and straw also influenced the number of seeds and species among some crop types. Results of this study document that large numbers of alien plant species are transported by movements of hay and straw into and within Alaska.

One of the greatest threats to biodiversity is the invasion of ecological communities by alien plants. Management strategies for alien plants, however, rarely focus on specific biodiversity-conservation outcomes, and recovery actions rarely address the threat on a landscape scale. There are many reasons for these failures, including (1) limited knowledge of the native species at risk, (2) a disconnect between policy and management, (3) a disconnect between the fields of weed science and biodiversity conservation, (4) a dearth of data from management actions (for both threat abatement and recovery), and (5) the broadly held assumption that control of alien plants will by itself lead to a positive biodiversity response is often incorrect. Thus, alien plant management strategies with a conservation aim need to include planning processes and assessment measures to ensure that the aim is achieved. Here, I outline an example of the planning steps needed to ensure that the management of widespread alien plant species achieves the greatest possible conservation outcome, using the alien plant program for bitou bush (Chrysanthemoides monilifena) as a case study. In addition, I present an overview of the challenges faced and solutions developed during the transition from planning to management, along with some initial data to demonstrate the success and value of this investment in both planning and implementation during the past decade. Interim results from a series of sites have highlighted the success of the program and led to the program being acknowledged as an important ecological restoration project in Australasia. This success is highly dependent on extensive stakeholder involvement (across land tenures), dedicated coordination, and leadership. The planning process is now being adopted for other alien plants in Australia and can be modified to abate the threat from other alien organisms or other threats to biodiversity because the principles of each step are fundamentally similar.

English ivy is an exotic liana that invades forests by forming dense monocultures on forest floors and by climbing trees. In this study, we evaluated the effectiveness of high-intensity–short-duration goat browsing to control English ivy in the Willamette Valley of Oregon. Species composition and percent cover were determined in August 2006 before the first browsing treatment and before and after treatments in 2006 and 2007. A final assessment was made in July 2008. Browsing effects were evaluated by comparison of multiple means statistical methods. English ivy cover declined significantly in samples browsed once or twice compared with untreated samples. The difference between levels of browsing was also significant, indicating that repetition of browsing for a second year is effective. Species composition change was minimal, with the appearance of sword fern and removal of Himalayan blackberry from 2006 to 2008.