Flea beetles (Aphthona spp.) were released in the Little Missouri National Grasslands (LMNG) in western North Dakota in 1999 to control leafy spurge. The changes in leafy spurge density and soil seedbank composition were evaluated on two ecological sites 10 yr (2009) after Aphthona spp. release to monitor the effectiveness of the insects on weed control and the associated changes in plant communities. In 2009, leafy spurge stem density averaged 2 and 9 stems m−2 (0.19 to 0.84 ft−2) in the loamy overflow and loamy sites, respectively, compared with 110 and 78 stems m−2, respectively, in 1999. Leafy spurge constituted nearly 67% of the loamy overflow seedbank in 1999, compared with 2% in 2009. In the loamy seedbank, the weed represented nearly 70% in 1999, compared with approximately 15% in 2009. As leafy spurge abundance was reduced, native species richness and seed count increased 10 yr after Aphthona spp. release. Late seral species represented 17% of the loamy overflow seedbank in 2009, an increase from 5% in 1999. However, Kentucky bluegrass, a nontarget weedy species, increased more than 250% in the loamy overflow seedbank. Late seral species were 38% of the loamy seedbank in 2009, compared with 13% in 1999. The number of native species increased from 31 in 1999 to 39 in 2009 in the loamy overflow seedbank, but only changed from 32 to 34 species in the loamy site during the same period. The reestablishment of native species has been slow, but seedbank analysis indicates the number and type of species found before the leafy spurge infestation have increased. Planting native species in selected areas may have reduced the lag time in these species return to the seedbank and reduced invasion from other nondesirable species, such as Kentucky bluegrass.

Hardwood forests in eastern North America are being colonized by multiple nonnative plant and animal species. Colonization rates can be affected by stand structure and distance from edge. We sampled earthworm densities and understory plant species cover in transects located in paired old- and second-growth forests in Indiana. Two 100-m transects were established within each forest stand during late April to early May in each year. One transect was placed parallel to and within 5 m of a south- or west-facing edge. The second transect was placed parallel to the first. but at no less than 100 m from any edge. Nonnative earthworms and plants were found in forest edge and interior regardless of structural stage (second-growth vs. old-growth). The number of native plant species decreased linearly as the densities of adult Lumbricus and Aporrectodea earthworms and the percent cover of multiflora rose (an invasive plant species) increased. Densities of L. terrestris and Aporrectodea earthworms and percent cover of multiflora rose cumulatively explained 39% of the variation in the number of native plant species found in transects across the state. However, multivariate analyses suggested that the species composition of Indiana understory plant communities was affected more by geography than by earthworm densities. Our results suggest that nonnative earthworms and plants are ubiquitous in Indiana hardwood forests and that they may reduce the number of native plant species.

Invasion of closed canopy forests by shade-tolerant alien plants has the potential to modify species composition, stand structure, ecosystem function, and long-term forest development patterns. Ligustrum sinense is a shade-tolerant alien shrub that has invaded bottomland forests throughout the southeastern United States. This species has received comparatively little attention in the literature despite its potential to drastically alter invaded sites. The overarching goal of our study was to document the relationships between Ligustrum sinense invasion and woody plant biodiversity and development patterns in an intact southeastern U.S. bottomland forest. The forest was dominated by Quercus nigra and Liquidambar styraciflua. Ligustrum sinense ranked fifth in basal area contribution, occurred in 97% of our plots, and represented 95% of all understory stems. Spearman's rho for dominance (based on basal area of stems > 5 cm diameter at breast height [dbh]) of L. sinense and woody plant species richness for each plot revealed a significant negative relationship (rs = −0.69, P < 0.01). A similar relationship was revealed between L. sinense density and woody plant species diversity (rs = −0.78, P < 0.01) and evenness (rs = −0.82, P < 0.01). Spearman's rho for L. sinense density and native understory stem density (individuals ≥ 1 m height, < 5 cm dbh) also revealed a significant negative association (rs = −0.48, P < 0.01). Under the current disturbance regime and without active management, we projected the forest would shift to support a stronger component of L. sinense and that structure would transition from tree to shrub dominance for sites within the forest.

We examined relationships between land disturbance and the extent and abundance of exotic buffelgrass (Pennisetum ciliare) at the interface of cultivated pastures and native desert lands in Sonora, Mexico. Plot and transect surveys of lands inside and outside pasture fences and general linear mixed models revealed complex relationships among buffelgrass, native vegetation, distance from pasture fences, and three categories of land disturbance (undisturbed, moderate, and severe). Results illustrate that buffelgrass invasion is extensive in lands surrounding pastures, and that buffelgrass abundance declines steeply with distance from pasture fences. The role of disturbance is weak but significant in its interaction with distance from the fence. Buffelgrass is more successful at colonizing severely disturbed lands than native vegetation, and its decline in abundance on severely disturbed lands is relatively more gradual than on other disturbance regimes, so landscapes where severely disturbed lands are interspersed with buffelgrass pastures could become centers of extensive buffelgrass invasion. In light of its potential to transform the Sonoran Desert, buffelgrass outside pastures warrants attention in a region-wide control scheme, as well as in future research, which ideally would involve remote sensing.