2007 SOUTHEASTERN NATURALIST 6(3):535–550 Effects of Ligustrum sinense Lour. (Chinese Privet) on Abundance and Diversity of Songbirds and Native Plants in a Southeastern Nature Preserve Joshua Wilcox1 and Christopher W. Beck1,* Abstract - Invasive plant species can have substantial negative impacts on native flora and fauna. We investigated the effects of the invasive shrub Ligustrum sinense (Chinese privet) on the abundance and diversity of songbirds in a southeastern forest during summer, fall, and winter. We sampled 15-m x 15-m plots assigned to one of three privet density categories (low, n = 5; medium, n = 4; high, n = 5). In addition, we sampled all flora in each plot. Bird abundance and species richness varied only during the winter, both increasing in high privet density. In general, the behaviors and types of birds did not differ among privet-density categories. In contrast, abundance and richness of native plants were reduced in high privet-density plots. Our results suggest that removal of privet would improve native plant communities, while having no substantial impact on songbird populations. Introduction Nonindigenous species invasions are a major environmental threat both to native habitats and the species that live in them. Many species of exotic plants and animals are able to outcompete native species for resources through superior reproductive potential, quick growth, alleopathic qualities, and a number of other survival mechanisms (Simberloff et al. 1997). These types of invasives tend to dominate their new environment, pushing out the native species. They have been known to cause mass extinctions and habitat destruction, as well as to cost billions of dollars annually in control measures (Simberloff et al. 1997). The effects of invasive plant species may extend beyond their effects on native plants. Several recent studies have shown that specific bird species are negatively impacted by invasive plants. For example, infestation of grasslands by Euphorbia esula L. (leafy spurge) led to a reduction in the density of Ammodramus savannarum Gmelin (Grasshopper Sparrows) and Passerculus sandwichensis Gmelin (Savannah Sparrows)(Scheiman et al. 2003). In addition, Calcarius ornatus Townsend (Chestnut-collared Longspurs) have reduced reproductive success in monocultures of the invasive Agropyron cristatum (L.) Gaertn. (crested wheatgrass) as compared to in native prairie (Lloyd and Martin 2005). Entire bird assemblages can be negatively impacted by invasive plant species as well. Invasion of wetlands by Lythrum salicaria L. (purple loosestrife) has resulted in a decline in 1Department of Biology, Emory University, 1510 Clifton Road, Atlanta, GA 30322. *Corresponding author - christopher.beck@emory.edu. 536 Southeastern Naturalist Vol. 6, No. 3 wetland bird species (Blossey et al. 2001). Similarly, Australian wetlands invaded by Urochloa mutica (Forssk.) T.Q. Nguyen (para grass) have lower numbers of birds than habitats without para grass (Ferdinands et al. 2005). Although invasive plants often have negative effects on bird populations, in some cases, invasive plants have negligible or positive effects on birds. For instance, neither Dolichonyx oryzivorus L. (Bobolink) nor Sturnella neglecta Audubon (Western Meadowlark) is influenced by the density of leafy spurge in grasslands (Scheiman et al. 2003). In addition, a study of the effect of invasive plant species on the native bird community in the Muddy River drainage of the Mohave Desert found no effect on species richness of birds, as long as the structural diversity in the plant community remained unchanged (Fleishman et al. 2003). Furthermore, Blossey et al. (2001) found that generalist bird species readily use purple loosestrife, although it leads to the exclusion of wetland bird species. Four species of privet have been classified as invasive: Ligustrum vulgare L. (common privet), native to Europe; and Chinese privet, Ligustrum japonicum Thunb. (Japanese privet), and Ligustrum lucidum Ait. f. (glossy privet), native to east Asia. Chinese privet poses the biggest threat to the native habitats of Georgia (Ward 2002). It is the most invasive of the four species and the most suited to the Georgia climate. Introduced in the mid-1850s, Chinese privet has naturalized itself in 19 states east of the Rockies, ranging from Massachusetts to Texas, and as far south as Florida (Ward 2002). In natural areas, Chinese privet may be able to outcompete the native understory species of a habitat through characteristics common of many species of privet, such as high fruit production, rapid seed dispersal, high germination rates, clonal reproduction, ability to capture sunlight, and shade tolerance (Debussche and Isenmann 1994, Lavergne et al. 1999, Matlack 2002, Morris et al. 2002). Preferring moist, well-drained soils with large amounts of sunlight, Chinese privet is commonly found in lowland floodplains, but has been known to populate widely varied habitats and can tolerate drought-like conditions and low soil fertility. It will often form large monospecific stands, completely dominating the area (Langland and Craddock Burks 1998). These stands can have dramatic effects on the abiotic environment because the privet plants are able to catch sediment in their root structures during floods, elevating the soil and making the habitat more suitable for their reproduction (Louisiana Invasive Plants Database 2001). Although Chinese privet is known to be invasive, its effects on native species of flora and fauna are not well-known, with a few exceptions. For example, Merriam and Feil (2002) noted a decrease in the abundance of herbaceous plants, but not small trees, in a mixed-hardwood forest that had been invaded by privet. In contrast, Stromayer et al. (1998) found that Odocoileus virginianus Boddaert (white-tailed deer) depend on privet for 2007 J. Wilcox and C.W. Beck 537 browse. Data on the effects of privet on native flora and fauna will allow us to determine the importance of privet-control efforts in natural areas. The main objective of our study was to examine how the density of privet affects the abundance and species richness of songbirds. Therefore, we recorded the number and species of songbirds that occurred in plots that differed in privet density. We sampled in the summer, fall, and winter to determine whether the effect of privet on abundance and diversity of birds was consistent across seasons for two reasons. First, different assemblages of birds can be found in the nature preserve during each season. Second, the phenology of privet might influence which birds are attracted to plots with high densities of privet. In Georgia, privet is evergreen, flowers between May and June, and ripens fruit between September and November (Radford et al. 1968). In addition to sampling songbirds, we also determined the abundance and diversity of native flora in each plot, because the effects of privet on songbirds might be mediated through its impact on native understory plant species. Anecdotal observations of birders suggest that songbirds are often found in privet, because its dense cover provides nesting habitat in the summer and protection for fall migrants. Miller and Miller (2005) suggest that privet has a moderate value as cover for birds. In addition, in the winter, privet berries might be an important food source for some species of songbirds, although privet is considered to have a low value as a food source (Miller and Miller 2005). Therefore, we predicted that we would see a greater number of birds in high privet-density plots. However, areas with high privet density were predicted to exhibit a general decline in native plant biodiversity, as was found by Merriam and Fiel (2002) in a similar forest. Field-site Description Clyde Shepherd Nature Preserve is a 28-acre property located in Decatur, GA, inside metropolitan Atlanta. The Preserve contains a 4-acre beaver pond, wetlands, secondary growth pine groves, and wooded uplands. Much of the land is a floodplain for South Peachtree Creek, which borders the Preserve. Walking paths run through portions of the Preserve for public use. Various plant and animal species can be found in the park including beavers, raccoons, foxes, turtles, fish, snakes, frogs, rabbits, and a large variety of tree and plant species. To date, 168 bird species have been observed in the park, which has been recognized by the Atlanta Audubon Society as a wildlife sanctuary. The Preserve is a prime example of secondary growth in a southeastern piedmont floodplain forest. Much of the floodplain area is surrounded by dense stands of privet that dominate the understory. The Preserve contains three distinct forest areas. The floodplain area is generally at a low elevation in the Preserve. These areas are poorly drained, often with standing 538 Southeastern Naturalist Vol. 6, No. 3 water several cm above the soil. They are dominated by Liriodendron tulipifera L. (tulip poplars), Acer negundo L. (boxelder maples), and Acer rubrum L. (red maples). The pine groves appeared physically similar to the floodplain areas, but drier and dominated by young Pinus taeda L. (loblolly pine). The uplands forest areas are at a higher elevation within the Preserve, are drier than both the floodplain forests and the pine groves, and are dominated by Fagus grandifolia Ehrh. (American beech) and various Quercus (oak) species. Methods Privet plots We chose fourteen 15-m x 15-m plots, each with a 30-m buffer from any other plot. The plot size was chosen due to the small size of the nature preserve and the lack of large plots with consistent densities of privet. Plots were selected based on observed privet density in three categories: high, medium, and low privet density. Five plots with the maximum density of privet available were selected for the “high” category (90.6 ± 4.6 % cover); five plots containing very little (i.e., privet seedlings) to no privet were selected for the “low” category (0.4 ± 0.6 % cover); and four plots of intermediate privet density, ranging between maximum privet density and minimum privet density, were chosen to represent “medium” privet density (13.6 ± 1.5 % cover). The actual density of privet differed significantly across the three categories (Kruskal-Wallis: H = 11.7, P = 0.003) and between each category (Mann Whitney U tests: P < 0.015 for all pairwise comparisons). Bird sampling For the summer analysis, birds in each plot were surveyed once a week over the months of June–August 2003 using a 20-minute stationary survey point count, following Blondel et al. (1981). Sorace et al. (2000) note that 10-minute counts are sufficient to detect approximately 80% of birds, but longer counts ensure that rare species are counted. Longer counts may increase the probability of counting the same individual more than once. However, our plots were sufficiently small that individual birds could be tracked within a plot, thus avoiding overcounting. We conducted sampling using sight and sound between one-half and three and one-half hours after sunrise. Plots sampled were rotated randomly for days of the week and time slots in each day. Each plot was surveyed a total of 10 times. No surveys were conducted within 30-minutes of any rainfall or during other abnormally adverse weather conditions. Since understory birds are the only birds that would be affected by the shrubby nature of privet, only birds occurring within 15 m of the ground were counted. In addition, only birds detected within the plot boundaries were counted. We collected data during each survey including number, species, and behavior of the birds observed (e.g., Anderson 1981, Dawson 1981). We 2007 J. Wilcox and C.W. Beck 539 recorded the behaviors of birds to determine how birds were using the plots. The following categories of bird behavior were utilized (following Ekert 1999): fly through (a bird that flies through the plot and performs no other behavior before leaving the plot or the termination of the observation period), perch (a bird remaining stationary on a perch for a time), call (a bird making a vocal call while in the plot), sing (a bird making a vocal song while in the plot), forage (a bird foraging for or eating any type of food while in the plot), and nest (a bird constructing a nest or visiting an existing nest inside the plot). Fall and winter sampling were conducted in essentially the same manner, with some slight exceptions. Two plots from each privet-density category were chosen randomly for each sample day to ensure that single days were not skewing the data for a particular privet density. So that all six plots could be surveyed in a single day in order to detect transient migrants, the observation length was reduced to 15 min, which is still sufficiently long to observe greater than 90% of birds (Sorace et al. 2000). The order in which plots were sampled in a given day was randomized. These new methods were employed for the last four days of the fall sampling (five total samples between September 15 and September 27, 2003, based on the arrival of fall migrants) and all of the winter sampling (seven samples between December 27, 2003 and January 13, 2004). Plant sampling All plants occurring inside each plot were surveyed between August 4 and August 17, 2003. We recorded species identity, number, condition, and dbh (for trees over 2 m tall). We determined cover of privet and other comparable understory trees and shrubs using the crown-diameter method (Mueller-Dombois and Ellenberg 1974). Because of the tendency of Chinese privet to grow in large monospecific stands with numerous individual trunks and overlapping branches, the crown-diameter method was modified to assess the combined canopy size of all privet individuals occurring in the plots. Statistical analyses In all three seasons, bird abundance and species-richness data were not normally distributed and could not be transformed to normality. As a result, we used nonparametric statistics for all analyses. For all three seasons, we examined the effect of privet density on bird abundance and species richness in two ways. First, we compared bird abundance and species richness across plots in different privet-density categories using Kruskal-Wallis tests. If there was a significant effect of privet-density category, we carried out pairwise comparisons of privet-density categories with Mann-Whitney U tests with 􀁟 = 0.05/3 = 0.017 to control the experiment-wise error rate. Second, we used Spearman rank correlations to examine the relationship between actual privet density and bird 540 Southeastern Naturalist Vol. 6, No. 3 abundance or species richness. During the summer, each plot was surveyed every week. Therefore, we could examine changes in bird abundance and species richness across the season. Individual plots were ranked based on bird abundance or species richness across all weeks, with plots with the lowest abundance or richness having the lowest rank. Then, the effect of plot, time, and the interaction between the two on ranks were determined using a repeated-measures analysis of variance (Conover 1999). In addition to effects on bird abundance and species richness, we examined the effect of privet density on behavior (e.g., foraging, singing) by comparing the frequency of behaviors in the three privet-density categories with a chi-square test. Finally, we categorized birds into groups based on primary and secondary diet (e.g., frugivores, insectivores), foraging behavior (e.g, ground gleaners, foliage gleaners), and primary nesting habitat (e.g., ground nesters, shrub nesters) based on Ehrlich et al. (1988), because certain types of birds, such as shrub nesters in the summer or frugivores in the winter, might be more likely to use high privet-density areas. We considered both primary and secondary diet to account for changes in the diet of species across seasons. We compared the frequency of the different foraging or habitat groups in the three privet-density categories with chi-square tests. To determine the effects of privet density on native plants, we divided plant species into trees (over 2 m tall), shrubs, and ground cover. Small trees (< 2 m tall) were classified as shrubs and grouped together as a single type. Although this reduced species-richness values, it should not bias comparisons among privet-density categories. For each group, we compared the number of individuals and the number of species or types across plots in different privet-density categories using Kruskal-Wallis tests. If there was a significant effect of privet-density category, we carried out pair-wise comparisons of privet-density categories with Mann-Whitney U tests with 􀁟 = 0.5/3 = 0.017 to control the experiment-wise error rate. We also calculated Spearman rank correlations between actual privet density and number of individuals or number of species for each group. Results Bird sampling Cardinalis cardinalis L. (Northern Cardinals), Thryothorus ludovicianus Latham (Carolina Wrens), Cyanocitta cristata L. (Blue Jays), and Dumetella carolinensis L. (Gray Catbirds) were the most abundant species of songbirds. However, the number of individuals and species of birds varied across season and privet-density category (Table 1). Yet, during the summer, privet density did not significantly affect the number of birds (Kruskal-Wallis H = 0.39, df = 2, P = 0.83; Fig. 1a) and the number of species of birds (Kruskal-Wallis H = 2.15, df = 2, P = 0.34; Fig. 1b) recorded 2007 J. Wilcox and C.W. Beck 541 Table 1. Identified bird species by season and privet density. Summer Fall Winter Species Low Medium High Low Medium High Low Medium High American Crow (Corvus brachyrhynchos Brehm) 6 2 American Robin (Turdus migratorius L.) 1 31 37 Blue Jay (Cyanocitta cristata L.) 12 6 8 3 1 3 2 2 Blue-Gray Gnatcatcher (Polioptila caerulea L.) 1 Brown-headed Cowbird (Molothrus ater Boddaert) 2 1 Brown Thrasher (Toxostoma rufum L.) 2 1 1 Carolina Chickadee (Poecile carolinensis Audubon) 1 3 2 1 1 Carolina Wren (Thryothorus ludovicianus Latham) 15 10 39 4 2 3 1 4 11 Cedar Waxwing (Bombycilla cedrorum Vieillot) 1 9 2 4 6 Eastern Towhee (Pipilo erythrophthalmus L.) 3 3 10 1 3 1 Gray Catbird (Dumetella carolinensis L.) 16 1 5 3 1 2 2 Great Crested Flycatcher (Myiarchus crinitus L.) 1 1 Hairy Woodpecker (Picoides villosus L.) 1 House Wren (Troglodytes aedon Vieillot) 1 Mourning Dove (Zenaida macroura L.) 1 3 Northern Cardinal (Cardinalis cardinalis L.) 11 42 42 2 9 8 3 8 8 Red-headed Woodpecker (Melanerpes erythrocephalus L.) 2 2 2 1 1 Tufted Titmouse (Baeolophus bicolor L.) 1 2 1 White-breasted Nuthatch (Sitta carolinensis Latham) 1 White-eyed Vireo (Vireo griseus Boddaert) 1 Wood Thrush (Hylocichla mustelina Gmelin) 8 1 7 3 1 542 Southeastern Naturalist Vol. 6, No. 3 on plots. In addition, neither was significantly correlated with proportion of privet cover within plots (number of birds: r = -0.26, P = 0.37; number of Figure 1. The effect of privet-density category on the abundance and species richness of birds by season. Data are means ± 1 SE. In all cases except abundance during the winter, abundance and species richness was not significantly affected by privetdensity category. For abundance in the winter, bars with the same letter are not significantly different based on a Mann-Whitney U test, after significance level was adjusted for multiple comparisons. 2007 J. Wilcox and C.W. Beck 543 species: r = -0.23, P = 0.43). When we considered abundance and species richness among plot types across sample dates, we also found no significant effects of privet-density category, sample date, or the interaction between the two (Table 2). Similar to the summer, we found no significant differences among privet-density categories in the number of birds (Kruskal-Wallis H = 1.27, df = 2, P = 0.53; Fig. 1c) and the number of species of birds (Kruskal-Wallis H = 3.31, df = 2, P = 0.19; Fig. 1d) in the fall. Again, the proportion of privet cover within plots was not significantly correlated with the number of birds (r = 0.27, P = 0.34) or the number of species (r = 0.43, P = 0.12). In contrast to both the summer and the fall, in the winter, significantly fewer birds were found on plots in the low privet-density category as compared to plots in the high privet-density category (Kruskal-Wallis H = 6.33, df = 2, P = 0.04; Fig. 1e). However, the number of birds did not differ between plots in the medium privet-density category as compared to plots in either of the other privet-density categories (Fig. 1e). Yet, the number of birds was positively correlated with the proportion of privet cover in a plot (r = 0.61, P = 0.02). The number of species also was positively correlated with the proportion of privet cover in a plot (r = 0.59, P = 0.03). However, the number of species did not differ among the privet-density categories (Kruskal-Wallis H = 5.27, df = 2, P = 0.07; Fig 1f). The types of behaviors observed varied significantly among birds in the different privet-density categories during the summer (􀁲2 = 19.6, df = 10, P = 0.03), but not in the fall (􀁲2 = 14.1, df = 10, P = 0.17) or winter (􀁲2 = 7.2, df = 8, P = 0.51) (Table 3). The significant effect of privet-density category on Table 3. Frequency of different behaviors by season and privet-density category. Summer Fall Winter Behavior Low Medium High Low Medium High Low Medium High Call50 50 56 75 7 14 13 9 12 22 Fly10 10 19 28 1 5 3 4 4 5 Forage 2 3 11 0 0 3 1 5 16 Nest 0 0 1 0 0 1 0 0 0 Perch 77 70 116 12 10 21 12 23 37 Sing 24 17 13 3 5 1 2 3 3 Table 2. Effect of privet-density category and sample date on abundance and species richness of birds. Number of birds Species richness Source df MS F P MS F P Privet category 2 12,744.4 2.6 0.12 9967.7 2.8 0.10 Error 11 4831.4 Time 9 1537.5 1.5 0.16 1678.2 1.4 0.19 Time x Privet category 18 1313.4 1.3 0.23 972.3 0.82 0.68 Error (Time) 99 1035.8 1189.3 544 Southeastern Naturalist Vol. 6, No. 3 behavior seems to be due to birds singing more often in low privet-density plots and less often in high privet-density plots (Table 3). When birds were grouped by primary diet, the frequency of different diet types varied among privet-density categories during the summer (􀁲2 = 29.8, df = 6, P < 0.001) and the winter (􀁲2 = 12.6, df = 4, P = 0.01), but not during the fall (􀁲2 = 5.65, df = 6, P = 0.06) (Table 4). In both the summer and winter, a greater number of omnivores were found in the low privet-density plots than expected (Table 4). In addition, in the summer, we observed a greater number of birds that eat berries (primarily Bombycilla cedrorum Vieillot [Cedar Waxwings]) in the medium privet-density plots (Tables 1 and 4). The results for secondary diet were qualitatively similar to those for primary diet, except that the frequency of different diet types varied among privet-density categories in all seasons (P < 0.05 in all cases; Table 5). As with the primary diet, a greater number of omnivores were found in the low privet-density plots in the winter than expected (Table 5). However, in contrast to the primary diet, bird species whose secondary diet is fruits were Table 6. Frequency of birds grouped by foraging type for each season and privet-density category. Groupings based on Ehrlich et al. (1988). Summer Fall Winter Foraging type Low Medium High Low Medium High Low Medium High Bark glean 2 2 4 0 0 1 0 0 0 Foliage glean 2 13 5 0 1 0 5 5 6 General glean 47 21 67 8 4 8 0 0 0 Ground glean 31 48 51 5 9 9 14 18 29 Hawk 1 0 1 0 0 0 1 0 0 Table 5. Frequency of birds grouped by secondary diet for each season and privet-density category. Groupings based on Ehrlich et al. (1988). Summer Fall Winter Diet Low Medium High Low Medium High Low Medium High Fruits 26 5 14 3 1 2 5 3 8 Omnivores 22 8 11 4 0 2 6 2 2 Seeds 19 50 59 2 11 11 3 10 8 Small vertebrates 15 10 39 4 2 3 1 4 11 Insects 1 11 5 0 0 0 0 4 6 Table 4. Frequency of birds grouped by primary diet for each season and privet-density category. Groupings based on Ehrlich et al. (1988). Summer Fall Winter Diet Low Medium High Low Medium High Low Medium High Berries 1 9 2 0 0 0 0 4 6 Insects 59 67 112 9 14 16 9 17 27 Omnivores 22 8 11 4 0 2 6 2 2 Seeds 1 0 3 0 0 0 0 0 0 2007 J. Wilcox and C.W. Beck 545 found in greater densities than expected in low privet-density plots during the summer. In the fall, fewer seed-eaters were found in low privet-density plots than expected. The frequency of different foraging types also varied among the privetdensity categories during the summer (􀁲2 = 30.2, df = 8, P < 0.001), but not in the fall (􀁲2 = 6.25, df = 6, P = 0.40) or winter (􀁲2 = 7.2, df = 4, P = 0.13); (Table 6). During the summer, a greater number of foliage gleaners and a lower number of general gleaners than expected were observed in medium privet-density plots (Table 6). As with diet type, the high number of foliage gleaners in the medium privet-density plots was due to the number of Cedar Waxwings that we observed in those plots (Table 1). When we grouped birds by primary nest location, the frequency of birds in each nest-location category did not differ among privet-density categories in any season (summer: 􀁲2 = 8.57, df = 4, P = 0.07; fall: 􀁲2 = 5.6, df = 4, P = 0.23; winter: 􀁲2 = 3.5, df = 4, P = 0.47; Table 7). Plant sampling The effects of privet density on the abundance and species richness of native plants were similar for trees, shrubs, and ground covers (Table 8, Fig. 2). The abundance and species richness of native plants were higher in low and medium privet-density plots as compared to high privet-density plots (Fig. 2). The one exception was the species richness of trees, which did not differ significantly among privet-density categories (Table 8, Fig. 2). When we examined the correlation between proportion of privet cover in a plot and abundance and species richness of native plant species, we found significant negative correlations for all comparisons (r < - 0.56, P < 0.35), except for species richness of trees (r = -0.21, P = 0.48). Table 8. Results of Kruskal-Wallis tests for the effect of privet-density category on native plant abundance and species richness. Variable Kruskal-Wallis H df P Number of trees 9.05 2 0.01 Number of tree species 4.76 2 0.09 Number of shrubs 9.23 2 0.01 Number of shrub types 10.4 2 0.006 Number of ground-cover plants 6.11 2 0.05 Number of ground-cover species 7.08 2 0.03 Table 7. Frequency of birds grouped by primary nest location for each season and privet-density category. Groupings based on Ehrlich et al. (1988). Summer Fall Winter Nest location Low Medium High Low Medium High Low Medium High Ground 3 3 10 0 1 3 0 1 0 Shrub 29 44 48 6 10 10 5 8 8 Tree 51 84 70 7 3 5 10 14 27 546 Southeastern Naturalist Vol. 6, No. 3 Discussion The abundance and species richness of songbirds were not significantly affected by the density of privet, contrary to what we had expected. The Figure 2. The effect of privet-density category on the abundance and richness of native plants by plant type. Data are means ± 1 SE. See Table 7 for results of statistical analysis. Bars with the same letter are not significantly different based on a Mann- Whitney U test, after significance level was adjusted for multiple comparisons. 2007 J. Wilcox and C.W. Beck 547 dense cover in high privet-density plots might have made it more difficult to detect birds in those plots (Richards 1981), thus reducing our estimates of abundance and species richness. However, given the small size of our plots, it seems unlikely that detection differences biased our results. In contrast to the general pattern of no effect of privet density on abundance and species richness of songbirds, during the winter, we found a greater number of birds on high privet-density plots, as compared to low privet-density plots (Fig. 1e), and both abundance and species richness were positively correlated with actual privet density. During the winter, birds might have been found in high privet-density plots more often because they were foraging on privet berries. However, birds were not any more likely to be observed foraging in high privet-density plots than in low privet-density plots (Table 3). In addition, birds found in high privet-density plots during the winter were not predominantly frugivores (Table 4). In addition to effects of privet density on abundance and species richness of birds in the winter, the behaviors observed and the types of birds observed varied among plot types, although most often during the summer rather than the fall or winter (Tables 3–7). The effect of season is most likely due to the low number of birds observed during the fall and winter. As a result, detecting a significant effect of privet-density category on behavior or bird type was more difficult in these seasons. However, the significant effects found during the summer are instructive. Birds in low privet-density plots were more likely to be singing than birds in other plots (Table 3). This suggests that areas without dense underbrush might be used by birds for attracting mates. In addition, although most species used all plot types, certain species preferred particular plot types more than others (Table 1). In particular, cedar waxwings frequented medium privet-density plots more often that the other two plot types during the summer, which led to a significant effect of privet density on foraging type and primary diet type during the summer (Tables 4–5). In contrast to the general lack of an effect of privet density on songbirds, privet substantially impacted the abundance and diversity of most types of flora (Fig. 2). However, since we only sampled plants in August, our results do not account for winter annuals or spring ephemerals. Yet, for plants found in the summer, we found significantly reduced densities of trees, shrubs, and ground covers in high privet-density plots. In addition, richness of shrubs and ground covers were reduced in high privet-density plots. These results are in agreement with those of Merriam and Feil (2002), who also found a decrease in both abundance and species richness of native plants under privet. How Chinese privet leads to a reduction in abundance and richness of native plants is unknown. However, glossy privet, a related species of privet, has been shown to increase mortality of saplings of native trees in Argentina (Lichstein et al. 2004). In some high and medium privet-density plots, two other invasive plant species—Hedera helix L. (English ivy) and Pueraria 548 Southeastern Naturalist Vol. 6, No. 3 montana (Lour.) Merr. (kudzu)—were abundant. Although these species may have contributed to the low abundance and richness of native plants in these plots, we found no discernable differences in the native plant community in plots with ivy and kudzu as compared to those without these species. The co-occurrence of privet, English ivy, and kudzu does suggest that certain ecological factors may make some forested areas more prone to invasion by exotic species. Taken together, the results of our bird and plant sampling suggest several things about the control of privet in southeastern piedmont forests. Songbirds are not relying on low privet-density plots for habitat at our site. Therefore, removal of privet is not necessary to maintain bird populations. However, birds are not using high privet-density plots to the exclusion of other areas, even during the winter when privet berries might represent an important food source. Thus, removal of privet should not adversely affect bird populations. Yet, removal of privet would prevent birds from being active dispersers of privet seeds, thereby slowing its spread (Debussche and Isenmann 1994, Gosper et al. 2005). In addition, native plants clearly would benefit from the removal of privet. In fact, Merriam and Feil (2002) showed that the abundance and diversity of native plant species increased after privet removal in a mixed-hardwood forest in North Carolina. Thus, the ecological benefits of removing privet seem to outweigh any costs. Acknowledgments We thank Andy Davis for comments during the design and implementation of this study as well as comments on the manuscript, and Dave Butler for providing access to the Clyde Shepherd Nature Preserve. Literature Cited Anderson, S. 1981. Correlating habitat variables and birds. Studies in Avian Biology 6:538–542. Blondel, J., C. Ferry, B. Frochot. 1981. Point counts with unlimited distance. Studies in Avian Biology 6:414–420. Blossey, B., L.C. Skinner, J. Taylor. 2001. Impact and management of purple loosestrife (Lythrun salicaria) in North America. Biodiversity and Conservation 10:1787–1807. Conover, W.J. 1999. Practical Nonparametric Statistics, 3rd Edition. Wiley, New York, NY. 584 pp. Dawson, D. 1981. Experimental design when counting birds. Studies in Avian Biology 6:392–398. Debussche, M., and P. Isenmann. 1994. Bird-dispersed seed rain and seedling establishment in patchy Mediterranean vegetation. Oikos 69:414–426. Ekert, P. 1999. Winter use of large-leafed privet Ligustrum ludicum. Proceedings of the Linnean Society of New South Wales 121:29–38. 2007 J. Wilcox and C.W. Beck 549 Elton, C. 1958. The Ecology of Invasions by Animals and Plants. Methuen, London, UK. 181 pp. Ehrlich, P.R., D.S. Dobkin, and D. Wheye. 1988. The Birder’s Handbook. Simon and Shuster, New York, NY. 785 pp. Ferdinands, K., K. Beggs, and P. Whitehead. 2005. Biodiversity and invasive grass species: Multiple-use or monoculture. Wildlife Research 32:447–457. Fleishman, E., N. McDonal, R. MacNally, D.D. Murphy, J. Walters, and T. Floyd. 2003. Effects of floristics, physiognomy, and non-native vegetation on riparian bird communities in a Mojave Desert watershed. Journal of Animal Ecology 72:484–490. Gosper C.R., C.D. Stansbury, and G. Vivian-Smith. 2005. Seed dispersal of fleshyfruited invasive plants by birds: Contributing factors and management options. Diversity and Distributions 11:549–558. Langland, K.A., and K. Craddock Burks. 1998. Identification and Biology of Nonnative Plants in Florida’s Natural Areas. University of Florida, Gainesville, FL. 165 pp. Lavergne, C., J.-C. Rameau, and J. Figier. 1999. The invasive woody weed Ligustrum robustum subsp. Walkeri threatens native forests on La Réunion. Biological Invasions 1:377–392. Lichstein, J.W., H.R. Grau, and R. Aragón. 2004. Recruitment limitation in secondary forests dominated by an exotic tree. Journal of Vegetation Science 15:721–728. Louisiana Invasive Plants Database. 2001. Species: Ligustrum sinense Lour. Available online at http://www.lsuagcenter.com/invasive/chineseprivet.asp. Accessed June 9, 2003. Lloyd, J.D., and T.E. Martin. 2005. Reproductive success of Chestnut-collared Longspurs in native and exotic grassland. Condor 107:363–374. Matlack, G.R. 2002. Exotic plant species in Mississippi, USA: Critical issues in management and research. Natural Areas Journal 22:241–247. Merriam, R.W., and E. Feil. 2002. The potential impact of an introduced shrub on native plant diversity and forest regeneration. Biological Invasions 4:369–373. Miller, J.H., and K.V. Miller. 2005. Forest Plants of the Southeast and their Wildlife Uses, 2nd Edition. University of Georgia Press, Athens, GA. 454 pp. Morris, L.L., J.L. Walck, and S.N. Hidayati. 2002. Growth and reproduction of the invasive Ligustrum sinense and native Forestiera (Oleaceae): Implications for the invasion and persistence of a nonnative shrub. International Journal of Plant Sciences 163:1001–1010. Mueller-Dombois, D.M., and H. Ellenberg. 1974. Aims and Methods of Vegetation Ecology. Wiley, New York, NY. 547 pp. Radford, A.E., H.E. Ahles, and C.R. Bell. 1968. Manual of the Vascular Flora of the Carolinas. The University of North Carolina Press, Chapel Hill, NC. 1183 pp. Richards, D.G. 1981. Environmental acoustics and censuses of singing birds. Studies in Avian Biology 6:297–300. Scheiman, D.M., E.K. Bollinger, and D.H. Johnson. 2003. Effects of leafy spurge infestation on grassland birds. Journal of Wildlife Management 67:115–121. Simberloff, D., D. Schmitz, and T. Brown. 1997. Strangers in Paradise: Impact and Management of Nonindigenous Species in Florida. Island Press, Washington, DC. 467 pp. 550 Southeastern Naturalist Vol. 6, No. 3 Sorace, A., M. Gustin, E. Calvario, L. Ianniello, S. Sarrocco, and C. Carere. 2000. Assessing bird communities by point counts: Repeated sessions and their duration. Acta Ornithologica 35:197–202. Stromayer, K.A.K., R.J. Warren, A.S. Johnson, P.E. Hale, C.L. Rogers, C.L. Tucker. 1998. Chinese privet and the feeding ecology of white-tailed deer: The role of an exotic plant. Journal of Wildlife Management 62:1321–1329. Ward, R.W. 2002. Extent and dispersal rates of Chinese privet (Ligustrum sinense) invasion on the Upper Oconee River floodplain, North Georgia. Southeastern Geographer 42:29–48.