Kudzu is an invasive perennial climbing vine characterized by fast growth rates and tolerance to control measures. Repeated applications with high rates of 2,4-D plus picloram provide effective kudzu control, but picloram use is not permitted in certain states due to groundwater pollution concerns. Studies were conducted in Alabama and Florida to compare kudzu control with aminocyclopyrachlor, a new herbicide, to control provided by aminopyralid, clopyralid, metsulfuron methyl, and picloram plus 2,4-D, which are common treatments for kudzu management. Two annual applications of the same herbicide treatment were evaluated for effects on kudzu cover, kudzu volume index, and cover of other vegetation. Aminocyclopyrachlor at 140 to 280 g ae ha−1 (2 to 4 oz ae ac−1) was as effective as the standard 4.48 kg ae ha−1 (4 lb ae ac−1) 2,4-D amine plus 1.2 kg ae ha−1 picloram for kudzu control. There were no differences in kudzu control among the three rates of aminocyclopyrachlor tested. Colonization by graminoids, forbs, and Rubus spp. at 2 yr was greatest for herbicides providing the best kudzu control: aminocyclopyrachlor, and 2,4-D plus picloram. Herbicide treatments were more effective in controlling kudzu at the Alabama location, but repeated annual applications for 2 yr did not completely eliminate kudzu with any treatment at either site.

Preinvasion baseline data on entire communities are absent for most taxa in most places, and this limits our ability to connect long-term ecological changes to particular invasive species or invasion events. We obtained data on forest understory composition from 94 stands in the 1950s and again the 2000s. We recorded within-stand frequency of occurrence for garlic mustard, European buckthorn, and Bell's honeysuckle and identified changes in native plant species density in 20, 1-m2 quadrats in invaded and noninvaded stands. All three invasive species were absent from all study sites 50 yr ago, yet at least one was present in 77.7% of the stands by the 2000s. All three species were present in 14.9% of the stands. Garlic mustard and European buckthorn were found at 47.9% of resurveyed sites, and Bell's honeysuckle was found in 40.4% of resurveyed sites. Native understory plant species density declined an average of 23.1% during the past 50 yr. Declines were not significantly different in stands with or without invasive plants. The absence of a measurable effect by invasive plant presence or frequency could be due to invasive plants being too few to have a measurable effect at the plot scale, species density being an insensitive response variable, time lags between invasions and effects, or regional factors like development pressure and fire suppression driving density declines in both invasives and native species.

Bridal creeper has become a serious environmental weed in southern Australia. Historically the invaded areas had low soil nutrient levels. However, our field surveys indicate that soils in bridal creeper–invaded areas have higher phosphorus and iron levels than soils in nearby native reference areas regardless of the proximity to agriculture or other disturbances. A glasshouse experiment was undertaken to determine the influence of increased nutrients on plants that co-occur with bridal creeper in order to (1) assess the impact of changed soil conditions and (2) predict the response of dominant species following the biological control of bridal creeper. The relative growth rate (RGR) of bridal creeper, two native shrubs (narrow-leaved thomasia [Thomasia angustifolia] and bluebell creeper [Billardiera heterophylla]), and an invasive exotic grass (annual veldt grass [Ehrharta longiflora]) were determined in three soil types: soil collected within a bridal creeper stand, soil collected from a nearby reference area, and a potting mix with nutrient levels higher than that recorded in the field. The plant species were chosen due to their association with bridal creeper. For example, the native species narrow-leaved thomasia was identified in a previous survey as the most abundant shrub at the invaded site where the soil was collected. The two other species, bluebell creeper and annual veldt grass, were identified from a previous seedbank trial as being abundant (in the seedbank) and able to readily germinate in invaded areas. When grown in either the bridal creeper–invaded soil or reference soil, bluebell creeper had significantly lower RGRs than narrow-leaved thomasia and annual veldt grass. However, as all these species showed increases in RGRs between reference soil and bridal creeper soil, this study indicates that for at least these three species the impact of increased nutrients may not be a barrier to the recovery of invaded areas following the control of bridal creeper.

We investigated the use of sheep for controlling the spread of purple loosestrife in a wet meadow in upstate New York from June to August 2008. Changes in the purple loosestrife population and vascular plant community structure were monitored as a function of the grazing of two ewes, “rotated” through four “experimental” paddocks at 2- to 3-d intervals. Comparative data were collected in “reference” paddocks from which sheep were excluded. Purple loosestrife was heavily grazed and most plants did not flower in experimental paddocks. Purple loosestrife cover declined by 40.7% in the experimental paddocks but did not change significantly in the reference paddocks. After grazing, species richness was 20% higher in experimental than reference paddocks.

Controlled burns and grazing are being tested to manage invasive grasses in the Prairie Pothole region of the Northern Great Plains. These practices, however, may inadvertently promote saltcedar infestations from seed by opening the vegetative canopy. Saltcedar seedling establishment was investigated in greenhouse experiments using intact soil cores from one summit and three footslope sites in eastern South Dakota. Establishment tests were conducted in soil cores collected from treatment and control plots immediately after spring fire treatment (postburn) and in cores that contained peak cool- or peak warm-season vegetation, with or without clipping (simulated grazing treatment), to simulate vegetation conditions typical of saltcedar seed-shed in northern regions. Cores were seeded with 100 saltcedar seeds and subirrigated to maintain high soil water conditions, characteristic of the environment near potholes during late spring/early summer. Seedlings were counted during the first 3 wk to estimate establishment and the height of five seedlings core−1 were measured weekly to estimate growth rates. Opening the canopy with fire or clipping increased saltcedar establishment. Cores taken immediately after fire treatment had two times more seedlings establish (38% vs. 19%) and greater average seedling growth rate (1.5 mm d−1 vs. 0.9 mm d−1) when compared with no-fire controls. Fire after seeding reduced seedling establishment to 5%, but did not affect growth rate. Saltcedar establishment in peak cool-season vegetation cores was 6% regardless of earlier fire treatment, whereas in peak warm-season vegetation, establishment ranged from 8% (no spring fire) to 17% (spring fire). If soils remain wet, invasion risk following spring fire may be greatest when warm-season grasses are flowering because this time coincides with northern saltcedar seed production. Areas adjacent to viable saltcedar seed sources should be managed to maximize canopy cover when seeds are released to limit further establishment. Fire after saltcedar seed deposition may control propagules and young seedlings.

Wild chervil is an invasive biennial or short-lived perennial weed introduced into North America that negatively impacts forage production and degrades habitat for native plant species. A 2-yr study using prebloom mowing followed by combinations of herbicide, tillage, and grass seeding was conducted in the Pacific Northwest to identify an effective integrated weed management strategy for this species. By 2 mo after herbicide treatment (MAHT), wild chervil control with glyphosate + ammonium sulfate (AMS) and clopyralid was 83 and 73%, respectively. Tillage with or without herbicide pretreatment resulted in 92 to 98% wild chervil control at 2 MAHT, whereas herbicide without tillage gave only 45% control across all treatments. Tillage with or without subsequent grass seeding reduced wild chervil density four-fold compared to herbicide alone at 9 MAHT. Herbicide + tillage + grass seeding resulted in similar wild chervil cover (1 to 5% cover) as herbicide + tillage (1 to 6% cover) without subsequent grass seeding. Wild chervil biomass at 1 yr after herbicide treatment (YAHT) was reduced to 487 kg ha−1 (439 lb ac−1) with herbicide + tillage compared to 4,256 kg ha−1 for herbicide treatment alone. Herbicide + tillage + grass seeding increased grass dry weight at 1 YAHT from 201 kg ha−1 for herbicide + tillage to 1,575 kg ha−1, compared to 351 kg ha−1 in herbicide-only plots.

Tamarisk, a shrub or small tree native to Eurasia, was introduced to North America in the early 1800s and is now naturalized throughout many riparian areas of the southwestern United States, where extensive research has been conducted. It is a more recent invader to the northern Great Plains, and fewer studies have been conducted on tamarisk ecology and management in this area. The objectives of this research were to investigate the overwintering potential of tamarisk seeds in Montana and the relationship between hydrologic conditions and historic tamarisk establishment. Emergence of seedlings from seeds stored for different time periods at a range of temperatures was evaluated in a greenhouse study. Emergence rates declined after a 7-d storage period, but storage time had no effect on subsequent emergence rates, and seeds stored at −14 C and 5 C had greater emergence rates than those stored at 20 C and 35 C. Patterns in tamarisk establishment were assessed through age and hydrologic data collected from a reservoir (Fort Peck), a regulated river (Bighorn), and an unregulated river (Yellowstone) in Montana. These data indicated that tamarisk establishment at the reservoir was closely related to historic water levels, whereas establishment on rivers was not related to flow. However, data from the rivers indicated that recruitment differed between regulated and unregulated rivers, with the regulated river having less recruitment after the period of initial colonization than the unregulated river. Our results show that tamarisk seeds have the ability to overwinter in Montana and can establish under a range of flow conditions, indicating potential recolonization of sites after tamarisk removal.

Rehabilitation of downy brome–infested shrublands is challenging once this invasive grass dominates native communities. The effectiveness of imazapic herbicide in reducing downy brome cover has been variable, and there is uncertainty about the impacts of imazapic on native species. We used a before-after-control-impact (BACI) field experiment and greenhouse studies to (1) determine if imazapic herbicide applied at 132 g ai ha−1 (8 oz/ac−1) and seeding with two native shrub species (Wyoming big sagebrush [Artemisia tridentata] and Mexican cliffrose [Purshia mexicana]) reduced downy brome cover and promoted shrub establishment, (2) assess potential effects of imazapic on nontarget plant species and plant community composition, and (3) determine if imazapic affected downy brome or seeded shrub species when applied at different developmental stages. Seeding shrubs, alone, or in combination with imazapic application, did not significantly increase shrub density, possibly because of droughty conditions. In the field, imazapic reduced downy brome cover by 20% and nontarget forb cover by 25% and altered plant community composition the first year after treatment. Imazapic was lethal to downy brome at all growth stages in the greenhouse and reduced shrub germination by 50 to 80%, but older shrub seedlings were more tolerant of the herbicide. We conclude that a one-time application of imazapic combined with seeding shrubs was only slightly effective in rehabilitating areas with high downy brome and thatch cover and resulted in short-term impacts to nontarget species. These results highlight the need to treat downy brome infestations before they become too large. Also, removing thatch prior to treating with imazapic, although likely lethal to the native shrubs we studied, could increase the effectiveness of imazapic.

Perennial pepperweed (Lepidium latifolium) is a potential threat to biodiversity and ecosystem function in the communities that it invades. The mechanism for its successful invasion of riparian and wetland environments includes reproduction via seed and root propagules and its ability to withstand long duration flooding and saline conditions once established. Controlling this species presents a number of challenges for land managers, including difficult property access, limited herbicide choice, and the varied success of weed control measures. In this study, we tested the efficacy of a nonchemical-modified tarp treatment, and compared posttreatment stem counts to herbicide treatments with Mow–glyphosate and chlorsulfuron in a wildland setting. We found that tarping applied in combination with a mow and till treatment was effective at controlling perennial pepperweed at levels similar to herbicide treatments with Mow–glyphosate and with Mow–chlorsulfuron. However, Mow–Till–Tarp treatment is extremely time consuming and has the potential to limit native plant community recovery.

As invasive species become increasingly abundant in forests, their presence may influence a number of key nutrient cycling processes. For example, Chinese privet has become well established in southeastern forests and continues to spread. Two studies, a multisite field investigation and a controlled approach on a single site, were conducted to examine the role of Chinese privet (Ligustrum sinense) on decomposition within riparian forests of the Georgia Piedmont. The field study also investigated the effects of privet presence on soil nitrogen (N) mineralization and microbial carbon and N immobilization. Both studies utilized a litterbag approach to examine how increasing proportions of privet in foliar litter influenced mass loss rates and nutrient dynamics. The field investigation included litterbags with representative proportions of the five dominant species from 16 sites. Litterbags in the controlled study were composed of specific levels of privet litter within bags (0, 10, 20, 30, 40, and 50% Chinese privet) as treatments. The litter quality of four native species was compared to Chinese privet in the controlled study. Both studies showed significant positive relationships between percentage of Chinese privet in litterbags and decomposition rates (2.6-fold rate increase with 30% privet in litterfall). Chinese privet leaf litter had lower lignin and cellulose concentrations, higher N concentrations, lower lignin : N ratios, and narrower C : N ratios than the native species. The positive relationship between mass loss rates and the proportion of Chinese privet in litter indicates that Chinese privet enhances decomposition rates as it becomes more abundant. During summer, N mineralization showed approximately a fivefold increase; during winter, microbial biomass N increased by approximately 30% on sites with the highest levels of privet in the understory. Consequently, C and N dynamics in Piedmont riparian forests were significantly influenced in direct proportion to the amount of privet present in the understory.

Quantification of interference with biological control agents can provide support for anecdotal claims of success or failure of agent establishment and efficacy. This study was initiated because of observed predation of cinnabar moth larvae by carpenter ants when releasing larvae for the control of tansy ragwort, an invasive plant in Montana. Biotic and abiotic factors were compared among three sites with historically variable moth population establishment. Two experiments were developed to (1) observe and document insect activity, predation, or disappearance on tansy ragwort stems either protected or accessible to ants; and (2) quantify the effects of ant exclusion on herbivory of tansy ragwort. Site comparisons indicated that ant colony density was highest at the driest of three sites, and, interestingly, no ant colonies were detected at the site with higher observed numbers of moth larvae and adults and lower densities of tansy ragwort. Available substrate (logs and stumps) for ant colonization did not differ between the three sites. In the ant exclusion experiments, a larger number of larvae were missing on plants accessible to ants (63%) compared with plants where ants were excluded (39%) after 36 h. Direct observation of predation of larvae by carpenter ants accounted for 9% of missing larvae on stems accessible to ants. Larvae were able to consume 81% of original flowers or buds on ant-excluded stems, compared with 18% consumption on ant-accessible stems, suggesting that ant predation could limit the efficacy of cinnabar moth larvae. These results provide one of many possible explanations for the anecdotal observations of large, persistent populations of cinnabar moths in moist areas. This work emphasizes the importance of post-release observation and monitoring to detect and, ideally, quantify factors to support anecdotal perceptions regarding the fate and subsequent efficacy of insect biological-control agents.

Coral ardisia (Ardisia crenata) has been present in Florida for more than 100 yr as an ornamental and has become invasive in hammocks of natural areas. This plant forms dense understory cover, often greater than 90%, which can suppress native plant recruitment and growth. Results from herbicide trials at two sites in Florida indicate that a single foliar treatment applied as a spot application of triclopyr amine, triclopyr ester, glyphosate, imazapic, dicamba, triclopyr amine + imazapic, or triclopyr ester + fluroxypyr reduced Ardisia crenata to less than 13% at 12 mo after treatment (MAT). A single treatment of imazapic (2.4 g ae L−1) or imazapic (2.4 g ae L−1) + triclopyr (10.8 g ae L−1) reduced cover of mature plants to less than 0.5% and seedlings to less than or equal to 4% at 12 MAT. Native plant cover was less than 5% prior to treatment indicating that dense infestations of Ardisia crenata may suppress native vegetation. In the dense infestations of Ardisia crenata observed in this study, nontarget damage was not a concern due to the rarity of native plants. However, applicators should use caution applying triclopyr and imazapic when small shrubs and trees are present in the treatment area. Additional follow-up treatments will be required for control of seedling and possible resprouts at 12 MAT.

Serpentine grasslands are relatively free of the invasive species that typify other California grasslands, but recently are threatened by the spread of barb goatgrass, a Eurasian annual that is unique in its tolerance of serpentine soils. We evaluated two graminicides (clethodim and fluazifop), one broad-spectrum herbicide (glyphosate), mowing, and hand pulling for their effectiveness in controlling barb goatgrass and for restoring native species cover and diversity. We also evaluated application timing for the graminicides: before goatgrass flowering (early season), at flower initiation (midseason), and at early seed development (late season). Glyphosate was applied at early seed development. The 10 treatment combinations were applied to 10 blocks of 4-m2 (43-ft2) experimental plots in an inner Coast Range serpentine grassland with high cover of barb goatgrass but few other nonnative species. After 2 yr of treatment, all treatments except glyphosate reduced goatgrass frequency. Hand pulling, fluazifop, and mowing were most effective at controlling goatgrass, reducing frequency by 60, 51, and 48%, respectively (compared to a 30% increase on control plots). Midseason applications of clethodim and fluazifop were more effective than those applied early or late. Hand pulling, fluazifop, and clethodim increased native forb frequency by 31, 46, and 74%, respectively, with the benefit of the graminicides decreasing with later applications. Native grasses were at least partially resistant to the graminicides. Cover of one sided blue grass (Poa secunda), the most widespread native grass, might have been reduced slightly by early-season applications but was increased by late-season applications. Cover of soft brome (Bromus hordeaceus), the only other widespread nonnative species in the study, was reduced by early-season applications of the graminicides. Clethodim and fluazifop show great potential to selectively remove barb goatgrass and other nonnative annual grasses in grasslands that are otherwise dominated by native grasses and forbs. Mowing is a less selective, but viable alternative.

Yellow starthistle is the most widespread broadleaf invasive plant in the western United States, and it is particularly prevalent in California. Prior to the registration of aminopyralid in 2005, the standard for chemical control of yellow starthistle was the herbicide clopyralid. We report on a compilation of several independent trials comparing the efficacy of aminopyralid and clopyralid on yellow starthistle. Treatments were applied at several rates and timings at 11 locations in four states between 2001 and 2007. Treatments were made pre-emergence and postemergence at the seedling and rosette stages of yellow starthistle. Results showed that aminopyralid, even at the low rate of 18 g ae ha−1, provided nearly complete control of yellow starthistle when treatments were made at the seedling stage. However, less consistent control (80 to 100%) resulted with applications made at the pre-emergence and rosette stages. At the seedling stage, aminopyralid is about four times more effective on yellow starthistle compared to clopyralid, based on the rate of acid equivalent. In the Central Valley of California, complete control was obtained at the lowest registered rate (53 g ae ha−1) when applications were made from December through February. At two locations we also evaluated control of the poisonous native plant coast fiddleneck. Although clopyralid does not adequately control coast fiddleneck, aminopyralid provided almost complete control when applied in the winter growing season. Applications of aminopyralid at the rosette stage resulted in a two-fold increase in annual forage grass biomass the following year. These results indicate that aminopyralid is a valuable tool for land managers and can play an important role in integrated management strategies for yellow starthistle and coast fiddleneck.

Smooth brome (Bromus inermis) is an introduced, cool-season perennial, sod-forming grass that has been shown to invade both native cool-and warm-season grasslands throughout North America. During the fall of 2005 through spring 2007, we implemented a smooth brome removal study at five sites in eastern South Dakota. Sites were selected to represent a range of soil and environmental conditions. Seven fall herbicide treatments, five spring herbicide treatments, an untreated plot that was planted with a native seed mix, and an untreated control that received no herbicide or seed addition were applied at each location in fall 2005/spring 2006 and fall 2006/spring 2007. Based upon first-year results, three fall herbicide treatments and two spring herbicide treatments were added in fall 2006/spring 2007. Sites were seeded with a native plant mix within 2 wk following spring herbicide treatment. Smooth brome cover in untreated plots ranged from 73 to 99% at the conclusion of the study. Smooth brome cover on herbicide-treated plots ranged from 0 to 84% on 2005/2006 plots and 0 to 98% on 2006/2007 plots after three growing seasons. Native plant response varied by site and treatment, possibly due to competition from exotic weeds. Although several herbicides show promise for control of smooth brome, future response of native plants will be important in determining the proper timing and herbicide combination.

Bushkiller, an aggressive perennial vine native to Southeast Asia, has invaded several sites in Alabama, North Carolina, Texas, Louisiana, and Mississippi. Bushkiller has only recently been discovered in North Carolina. The potential economic and environmental consequences associated with established exotic invasive perennial vines and the lack of published control measures for bushkiller prompted research to be conducted at North Carolina State University that may be used in an early-detection rapid-response program. Field and greenhouse studies were conducted to determine bushkiller response to selected foliar-applied herbicides. Field study 1 evaluated efficacy of glyphosate, triclopyr, triclopyr plus 2,4-D, triclopyr plus aminopyralid, and triclopyr plus glyphosate applied postemergence to bushkiller. No control was evident from any treatment at 10 mo after application. In a separate experiment, aminocyclopyrachlor, imazapyr, metsulfuron, sulfometuron, and sulfometuron plus metsulfuron were applied postemergence to bushkiller. Control with aminocyclopyrachlor, imazapyr, sulfometuron, and sulfometuron plus metsulfuron was 88 to 99% at 10 mo after application. Each treatment was also applied to bushkiller in a greenhouse trial. Aminocyclopyrachlor and triclopyr-containing treatments generally resulted in the greatest control, lowest dry weights, and shortest vine lengths among the treatments. These results indicate that several herbicides may be employed initially in an early-detection, rapid-response program for bushkiller. Additional research is needed to determine how effective these herbicides would be in multiple-season treatments that may be required at well established bushkiller infestation sites.

Vigorous stands of perennial grasses can effectively provide long-term control of many invasive plants on rangelands. However, in degraded conditions, successful reestablishment of perennial grasses can be compromised by invasive annual grasses, such as downy brome. Propoxycarbazone-sodium is a selective herbicide currently labeled for downy brome control in small grains, but its potential use on rangelands is unknown. Studies were conducted from 2004 through 2008 at three rangeland sites in Colorado and Nebraska to evaluate downy brome control and perennial grass injury with propoxycarbazone-sodium and imazapic. Propoxycarbazone-sodium provided satisfactory downy brome control with grass injury equal to or less than imazapic when rainfall followed the fall application. A second set of studies was conducted from 2007 to 2008 at Lingle, WY, and Scottsbluff, NE, to determine the plant-back interval and postemergence application response of seven perennial grass species to propoxycarbazone-sodium and imazapic. Grass tolerance to both herbicides was good when applied 90 and 120 d before planting (DBP). However, grass injury increased as plant-back interval decreased. The greatest impact on plant biomass was observed from herbicide applied at planting or after planting. Crested and intermediate wheatgrass (Agropyron cristatum and Thinopyrum intermedium) biomass production was not affected when herbicides were applied 90 or 120 DBP. Western wheatgrass (Pascopyrum smithii) and Russian wildrye (Psathyrostachys juncea) showed tolerance to imazapic applied before planting. Smooth brome (Bromus inermis), sheep fescue (Festuca ovina), and orchardgrass (Dactylis glomerata) showed the least amount of tolerance to propoxycarbazone-sodium and imazapic.

Land managers must set weed management priorities if limited resources are to be utilized effectively. Weed surveys and inventories assist land managers in this process by providing information regarding the identity, location, and relative abundance of weeds on their land. Although this information is vital, it can be challenging to select a method that provides the necessary data to meet management objectives while remaining accurate and cost effective. This paper critically evaluates four wildland weed mapping methods. These methods were defined as (1) paper-drawn, (2) buffered-point, (3) screen-drawn, and (4) perimeter-walked. Polygons were drawn by hand on topographic maps in the paper-drawn method. The other methods utilized handheld geographic positioning system (GPS) technology to digitally record infestations. Six experienced weed mappers independently recorded the location and size of eight sagebrush patches using each method. Time and accuracy were evaluated for each method based upon mapping time, distance walked, horizontal precision error, estimated size error, and shape error. The paper-drawn method was significantly less accurate than GPS-based methods at recording patch size and location. There was no significant difference in the accuracy of the buffered-point, screen-drawn, and perimeter-walked methods at reporting patch size and location. The need to cover land area quickly and efficiently favors the selection of the buffered-point or screen-drawn method because of time and distance factors. However, if patch shape is an important factor, the perimeter-walked or buffered-point methods may be the best choices of methods tested.

We evaluated herbicide treatments for control of umbrella dracaena, locally invasive in central Florida. Biannual treatments of imazapyr (3.2 and 6.4 g ai L−1), metsulfuron (0.08 and 0.16 g ai L−1), and imazapyr (0.8 g ai L−1) plus glyphosate (19.9 g ai L−1) reduced cover to < 3% at 12 mo, but no treatment eliminated the plant. In contrast, a single treatment on individual potted plants in greenhouse trials was highly effective with ≥ 98% necrosis/chlorosis at 6 mo and no regrowth at 9 mo posttreatment for most treatments. Management of umbrella dracaena will require more than two herbicide treatments with imazapyr, metsulfuron, glyphosate, or a combination of these herbicides if eradication is to be achieved. Because it is currently problematic and localized in one specific area in about 40 ha, it should be a high management priority before it spreads throughout Florida and possibly other states.

A total of 24 pre-and posttreatment plant frequency data sets were analyzed from 15 Wisconsin lakes treated with granular 2,4-D BEE herbicide for the control of Eurasian watermilfoil (Myriophyllum spicatum). Six data sets from four untreated control lakes were analyzed for comparison. The data sets included the results of line-transect aquatic plant surveys and point-intercept aquatic plant surveys. The results from these two survey methods were analyzed separately. Analysis of pre-and posttreatment changes in frequency of occurence for 46 species of aquatic plants indicated Eurasian watermilfoil was the only species to show significant declines in all the surveys. At application rates of 112 kg ha−1, Eurasian watermilfoil declined an average 65.9% among the line-transect surveys; and 58.0% among the point-intercept surveys. At application rates of 168 kg ha−1, Eurasian watermilfoil declined by 94.4% and 76.5% among line-transect and point-intercept surveys, respectively. Among the control lakes, Eurasian watermilfoil increased an average of 77% in year 1 and 24% in year 2. Northern watermilfoil (Myriophyllum sibiricum), a closely related native plant, underwent declines in frequency at the higher 2,4-D application rate (20.0%) but showed an increase (88.9%) at the lower rate among the line-transect surveys. Northern watermilfoil exhibited declines at both rates among the point-intercept surveys (48 and 50%, respectively); however, the plant also exhibited declines in the control lakes in year 2. Most other native aquatic plant species were unaffected or showed increases following treatment with 2,4-D BEE. The high degree of selectivity to Eurasian watermilfoil found in this survey of operational treatments with 2,4-D BEE suggests that this herbicide is an important tool for restoring plant communities that have been degraded by Eurasian watermilfoil.