A critical question in invasion biology involves the relative importance of propagule rain and community invasibility. For plant invasions, invasibility is often related to disturbance, but few studies of forest invaders have simultaneously investigated both canopy and ground-level disturbance. We investigated the relative importance of seed rain, canopy disturbance, and soil disturbance in a mature forest in Maryland on the recruitment of four invasive species: wine raspberry (Rubus phoenicolasius Maxim.), Japanese barberry (Berberis thunbergii DC), multiflora rose (Rosa multiflora Thunb.), and Japanese stiltgrass [Microstegium vimineum (Trin.) A. Camus]. Using complete censuses of a 9-ha plot at two points in time (2011-12 and 2014), we mapped new recruits, and related their locations to canopy and soil disturbance, as well as to a seed rain index based on locations of reproducing plants and seed-dispersal kernels. We found that propagule rain, as measured by the seed rain index, was a significant predictor of recruitment for B. thunbergii, R. phoenicolasius, and M. vimineum. For R. multiflora, seed sources were not located, precluding assessment of propagule rain, but recruitment was linked to canopy disturbance, as was recruitment of M. vimineum. However, because reproduction of R. phoenicolasius and, in some years, of B. thunbergii is higher in treefall gaps, these gaps experience higher propagule rain, with the result that recruitment is indirectly associated with these gaps. Ground-layer disturbance was an important predictor of recruitment only for B. thunbergii. Our findings reveal that the importance of propagule rain is the most consistent driver of recruitment, but canopy or ground-layer disturbance promotes recruitment of some invasive plant species.

Cogongrass [Imperata cylindrica (L.) Beauv.] is an invasive grass in the southeastern United States, and its impacts strongly affect the region, especially Florida. Herbicide strategies have been limited to glyphosate in natural areas and imazapyr in managed pine forests and non-crop areas where its soil residual activity is of less concern. This lack of options has raised concern for herbicide resistance, which has never been documented for I. cylindrica. Land managers have also reported variable I. cylindrica control, especially with glyphosate. To determine whether herbicide resistance was a possible explanation, we examined glyphosate response of I. cylindrica from 12 Florida populations. We also tested aminocyclopyrachlor with and without glyphosate and flumioxazin with glyphosate. Results indicated that herbicide performance was similar across I. cylindrica populations: glyphosate and aminocyclopyrachlor reduced I. cylindrica biomass by 78% and 76%, respectively, and the combined tank mix reduced cogongrass biomass by 91%. Flumioxazin tank mixed with glyphosate did not improve control compared with glyphosate alone. There were no differences in pretreatment I. cylindrica shoot height, with the exception of two panhandle populations that were shorter. Subsequent harvests indicated few differences in shoot and root plus rhizome weights among untreated controls for almost all populations, with the exception of one of the initially shorter panhandle populations. Our findings indicate that variability in glyphosate efficacy, as suggested by managers, is unlikely due to any conferred resistance. Other abiotic factors such as drought and shade and applicator factors such as carrier water quality should be examined to better understand this issue. Additional studies examining non-target impacts of aminocyclopyrachlor should be conducted to determine its potential fit into I. cylindrica management.

Invasive smooth cordgrass (Spartina alterniflora Loisel) eradication is important for the health of many coastal ecosystems. An integrated regime of continuous submergence after clear mowing, with three interval levels between mowing and submergence (5, 10, and 15 d) and three submergence depths (20, 30, and 50 cm), was implemented in cofferdams enclosing invader populations along a Chinese coast. In July of the following year, after the roots of mowed S. alterniflora had been submerged for 12 mo, some ramets grew under the regime with an interval of 15 d and the regime with a submergence depth of 20 cm, but no ramets occurred under the regimes with submergence depths of 30 or 50 cm and intervals of 5 or 10 d. Four crab species were documented: Helice tridens tientsinensis Rathbun, Sesarma dehaani H. Milne-Edwards, Ocypode stimpsoni Ortmann, and Chiromantes haematocheir de Haan. Biomass and abundance values of crab species in the cofferdams were similar to those in the mudflats but different from those in S. alterniflora populations. Thus, the treatment of submergence after mowing, which was implemented in the cofferdams, can control S. alterniflora and provide a mudflat-like habitat that promotes crab recovery if this treatment uses the proper combination of submergence depth and interval between mowing and submergence.

We distributed a 16-question survey concerning noxious weed abundance, impacts, and management to livestock producers grazing on privately owned or leased grazing lands in Montana. The noxious weeds most commonly reported as being present on respondents' grazing units were Canada thistle [Cirsium arvense (L.) Scop.] (64% of grazing units) and leafy spurge (Euphorbia esula L.) (45% of grazing units), and these species also reportedly caused the greatest reductions in livestock forage. Houndstongue (Cynoglossum officinale L.) was more prevalent than either spotted knapweed (Centaurea stoebe L.) or diffuse knapweed (Centaurea diffusa Lam.) (39% vs. 32% and 10%, respectively, of grazing units), but collectively C. stoebe and C. diffusa were reported to cause greater forage reductions than C. officinale. The top three strategies used to manage noxious weeds were chemical control, grazing, and biological control. Combining survey responses with forage-loss models derived from field data for C. stoebe and E. esula, we estimated the combined cost of noxious weed management and forage losses on privately owned rangeland to be $3.54 ha^-1 yr^-1, or $7,243 annually for an average size grazing unit (i.e., 2,046 ha [5,055 ac]). Our estimates of economic losses are lower than many estimates from previous studies, possibly because we focused only on direct costs related to private grazing land, while other studies often consider indirect impacts. Nonetheless, our estimates are substantial; for example, our estimated loss equates to 24% of the average per-hectare lease rate for Montana grazing land.

Hoary alyssum [Berteroa incana (L.) DC.] is a nonnative invasive forb that is noxious in California, Idaho, Michigan, Montana, Oregon, Washington, Wyoming, Alberta, British Columbia, and Saskatchewan. Managing B. incana is difficult, because it has an extended flowering period, during which plants simultaneously flower and produce seeds. Consequently, poorly timed herbicide applications may kill B. incana flowers but not prevent viable seed production. We examined how different herbicide management practices used by invasive plant managers affected B. incana seed production and viability the year of application as well as population density 1 yr after application. Professional invasive plant managers sprayed B. incana with various herbicides as part of their current management practices at six sites in southwestern Montana in summer 2016. We collected B. incana plants at 4 wk postapplication for seed biology analyses. Across the six sites, nonsprayed B. incana produced 5 to 1,855 seeds plant ^-1 and averaged 429 seeds plant ^-1 Seed production was reduced by 64% to 99% with 7 of the 11 herbicide applications. Berteroa incana seed viability in nonsprayed areas averaged 53% and ranged from 36% to 73% across the sites. Nine of the 10 herbicide applications used by invasive plant managers reduced seed viability 49% to 100%. Few of the herbicide management practices reduced B. incana's population density the following growing season, suggesting that managers should expect reoccurring infestations at least 1 yr after application. Our results show that invasive plant managers can reduce B. incana viable seed production even when spraying plants that have flowered and formed seed pods. However, sites may need to be monitored for additional years to treat reoccurring infestations.