2012 SOUTHEASTERN NATURALIST 11(4):575–588 Effects of Garlic Mustard Invasion on Arthropod Diets as Revealed through Stable-Isotope Analyses Pieter A.P. deHart1,* and Sarah E. Strand1,2 Abstract - Alliaria petiolata (Garlic Mustard) is an invasive plant species which displaces native communities by lowering levels of mycorrhizal fungi essential to native plant nutrient acquisition. Consequently, the diets of arthropods using these native plants as a primary food source may be altered. To assess the magnitude of this disruption, stable- isotope analyses of carbon, nitrogen, oxygen, and hydrogen were used to trophically differentiate the diets of arthropods in Garlic Mustard-invaded areas. In invaded areas, arthropods were depleted in δ13C and enriched in δ15N relative to arthropods in uninvaded areas, suggesting a change in trophic position among generalist predators. Slight trophic repositioning was observed in all 4 isotopes, indicating interactions of 3 primary predators throughout the study area. Most observable shifts are likely due to predators either altering prey source or traveling further to acquire nutrients. Introduction Since its introduction to the United States in 1868, Alliaria petiolata M. Bieb (Garlic Mustard) has become one of the most abundant invasive herbaceous species in the Northeast (Roberts and Anderson 2001, Yates and Murphy 2008). Garlic Mustard is a biennial herbaceous plant, spending its first year in lowgrowing juvenile rosette form and its second year as fast-growing reproductive adults. Abiotically, the presence of newly introduced Garlic Mustard to forest understory systems significantly lowers leaf-litter density and increases soil pH (Rodgers et al. 2007). The phenotypic plasticity of Garlic Mustard aids in out-competing native species for light, water, and additional nutrient sources (Myers and Anderson 2002). Garlic Mustard accomplishes this by decreasing the amount of both arbuscular and ectomycorrhizal fungi that interacts with the native plants’ roots, consequently lowering the nutrient levels that the native plants acquire from the environment (Stinson et al. 2006, Wolfe et al. 2008). Forests dominated by Pinus strobus L. (Eastern White Pine) are particularly sensitive to such an invasion, suffering from inhibited seedling establishment and biodiversity of native plants and altered biogeochemical cycling (Wolfe et al. 2008). These changes to soil biogeochemistry and plant communities could also affect the abundance and diversity of consumers. Indeed, several studies (Bultman and Dewitt 2008, Rudgers and Clay 2008) demonstrate that the presence of Garlic Mustard in Eastern White Pine forests correlates with a reduction in the abundance and dominance of native plants and seedlings, which could also affect consumer feeding behavior. Native consumer species found in this system 1Department of Biology, Virginia Military Institute, Lexington, VA 24450. 2Current address - Department of Earth and Enviornment, Florida International University, Miami, FL 33199. *Corresponding author - dehartpa@vmi.edu. 576 Southeastern Naturalist Vol. 11, No. 4 include springtails (order Collembola), harvestmen (order Opiliones), ants (family Formicidae), wolf spiders (family Lycosidae), cabbage loopers (family Noctuidae), and aphids (family Aphididae) (Keeler and Chew 2008, Stobbs and Schagen 1987, Strand 2010). Although feeding behavior is very difficult to observe directly (Martinez et al. 1999), it can be inferred by following the flow of elements through food webs. Estep and Dabrowski (1980), Ostrom et al. (1997), Peterson and Fry (1987), and Post (2002) have demonstrated the value of using stable-isotope analysis (SIA) as a means of examining energy flow and food-web dynamics. SIA is used to describe the unique ratio of heavy to light isotopes for carbon (13C/12C, expressed as δ13C), nitrogen (15N/14N or δ15N), oxygen (18O/16O or δ18O), and hydrogen (2H/ H or δD, deuterium) in an organism’s tissues (Grey and Jones 2001). These isotopic “signatures” are useful in reconstructing food webs because body tissues retain an isotopic signature similar to the food sources consumed by the animal (Hobson 1999). There are some changes across trophic levels, however. As carbon and nitrogen isotopes are incorporated into tissues, they tend to undergo trophic enrichment, or “fractionation”. Carbon and nitrogen isotopes are enriched at roughly 1‰ and 3.4‰ per trophic level, respectively (DeNiro and Epstein 1978, 1981a, 1981b; Tieszen 1978; Tieszen et al. 1983). While δ13C and 15N are the most commonly used isotopes in studies investigating trophic positioning, including some focused on arthropod prey consumption and trophic positioning of generalist predators in forest ecosystems (Post 2002, Wise et al. 2006, Yates and Murphy 2008), δ18O and δD can provide further insight into water and food sources (e.g., Bowen et al. 2005). Ideally, all 4 of these elements should be used to pinpoint the trophic location of each species in a study (deHart 2006). In this study, we use SIA to determine whether invasion by Garlic Mustard alters food-web dynamics. We compare the isotopic ratios of C, N, O, and D in arthropods in an un-invaded and an invaded system, and describe the dietary and life-history shifts that correlate with the invasion of this species. These shifts could be due to direct utilization of Garlic Mustard as the primary food source, a lower abundance of preferred prey, or a shift to food resources with different isotopic signatures. This study should provide a better understanding of the degree to which arthropod foraging ecology and physiology is affected by an invasive plant species, and show how SIA can be used as an effective method in studying complex questions in arthropod biology. Methods Field-site description We selected field plots in a mixed temperate forest with or without Garlic Mustard present, characterized each plot, and both observed and collected the arthropods present for SIA. The field plots were located in a forest dominated by Eastern White Pine in Lexington, VA (37°47'N, 79°25'W). Six 5-m2 field plots were randomly selected to match equal abundance and age of mature plant assemblage. Three of the plots were invaded with Garlic Mustard (indicated 2012 P.A.P. deHart and S.E. Strand 577 as GM; ≥50% coverage of Garlic Mustard), while the other 3 were uninvaded (UI; ≈0% coverage of Garlic Mustard). Each of the individual plots exhibited a similar understory plant assemblage, tree coverage, and slope gradient. The native plants found in these plots were Parthenocissus quinquefolia (L.) Planch. (Virginia Creeper), Acer negundo L. (Box Elder), Ailanthus altissima (Mill.) Swingle (Tree of Heaven), Lindera benzoin L. (Spice Bush), Fraxinus americana L. (White or American Ash), and Platanus occidentalis L. (American Sycamore). Both GM and UI plots had a combined average tree circumference of 64.3 cm and averaged 8 trees per plot. Sampling protocol Sampling was done over 2 summer field seasons, with the first extending from June through July 2010, and the second ranging from the end of May through July 2011. Sampling was performed in 2-hour increments with consistency on each sampling date. One 1-m2 sampling quadrat within each plot was chosen randomly on each sampling date. Opportunistic sampling was used to sample epigeal arthropods in all of the plots. Sampling quadrats were observed by 2 adjacently positioned researchers to ensure efficient sampling coverage of the quadrat. Both researchers searched the leaf litter and vegetation for two 5-minute periods. All non-flying arthropods were collected and stored in glass scintillation vials for freeze-drying and subsequent isotopic analyses. Opportunistic sampling is an efficient and simple procedure for obtaining arthropods (Strand 2010), that did not involve capturing or storing the collected arthropods using any chemicals that may artificially alter their isotopic signature. Procedures such as pitfall traps and sticky traps were avoided, due to the potentially isotopically-influencing capture and preservation methods involved in each. The primary arthropods collected reflect those found in previous studies in this area (Strand 2010), and include springtails (order Collembola), harvestmen (order Opiliones), ants (family Formicidae), wolf spiders (family Lycosidae), cabbage loopers (family Noctuidae), and aphids (family Aphididae). Given the large array of species defined as harvestmen, and to accurately reflect the potential trophic positioning of these species, all harvestmen collected were differentiated into large (>0.009 g) and small (<0.009 g) size classes using an analytical balance. In GM plots, the densities of adult Garlic Mustard plants in the delineated square were counted. The abundance of Garlic Mustard was estimated by each of the researchers as percent cover using the Braun Blanquet method (Wikum 1978). These percentages were averaged to calculate the mean percent coverage at each site. To confirm the role of Garlic Mustard on the abiotic properties of each region, we measured the light intensity, soil moisture, and soil pH at each sample plot. Light intensity was measured in μmol/m-2s at 5 random points at each site using a Field Scout light sensor (Spectrum Technologies, Inc., Plainfield, IL). Soil moisture measurements were taken at each plot twice/year and calculated by using 30.0 g of wet soil from 5 random points within each plot, as detailed in the gravimetric method by Gardner (1986). Soil pH was calculated by modifying a 578 Southeastern Naturalist Vol. 11, No. 4 method using 4.0 g of soil and 6 mL of distilled water from 5 random points in each of the plots in both the invaded and uninvaded plots (McLean et al. 1982). Alerding et al. (2011) provided values for these abiotic properties at our sampling plots for the 2010 field season. Stable-isotope analyses All arthropod and plant samples were sub-sampled and prepared at the VMI Conservation Biology Laboratory prior to being sent to the UC Davis Department of Plant Sciences Stable-Isotope Facility for isotope analyses. In order to obtain values for δ13C and δ15N, 1 mg ± 0.2 mg for insect material and between 2–3 mg for plant material were sub-sampled into tin capsules (Costech 5 x 9 mm) using a Sartorius CPA2P microbalance. Sub-sampling for δD and δ18O analysis required 0.075 mg and 0.400 mg of plant and insect material, respectively, utilizing silver capsules (Costech 3 x 5 mm) in the same fashion. Sub-sampling techniques varied according to organism size. Larger arthropods were ground and homogenized using a Wig-L-Bug grinding mill. Bulk samples of aphids and ants were used to ensure sufficient material for analysis. For siliques, a section of the specimens was cut and placed in the cups. Leaves were crushed, and small pieces were obtained with forceps and placed in the capsules. All capsules were filled using clean microspatulas or forceps, folded and sealed using clean forceps, and stored in a Dry-Keeper upright desiccator cabinet to ensure sample stability for shipping and analysis. At UC Davis, the combined δ13C and δ15N was analyzed using a PDZ Europa ANCA-GSL elemental analyzer interfaced to a PDZ Europa 20-20 isotope-ratio mass spectrometer. δ18O and δD values were analyzed separately using a Heckatech HT oxygen analyzer interfaced to a PDZ Europa 20-20 isotope-ratio mass spectrometer. Stable isotope ratios were expressed as: δ18O, δD, δ15N, or δ13C = [(Rsample/Rstandard)-1] x 1000, where Rsample/Rstandard are the ratios of 18O/16O, 2H/ H, 13C/12C, 15N/14N. Data are expressed using delta notation (δ) in parts per thousand (‰) with the reference material for δ13C as Vienna Pee Dee Belemnite (V-PDB) and for δ15N as atmospheric air (At-air) (National Institute of Standards and Technology, Gaithersburg, MD). Measurement precision was estimated at ±0.24, ±0.05, ±0.4, and ±1.05 for δ15N, δ13C, δD, and δ18O, respectively. Statistical analyses Results in the text and figures are presented as means ± SD. Variations between GM and UI plots were analyzed by performing two-way ANOVA, with treatment and year as main effects. After ANOVA, we used Tukey comparisons to test for differences in 15N and 13C between arthropod groups in each plot type. Estimates of the relative contributions of potential dietary sources in arthropods were established using both a simple linear mixing model and the concentration dependence model as displayed in the programs IsoConc (version 1.01) and IsoSource (version 1.3.1) using standard discrimination values of 1‰ (δ13C) and 3.4 ‰ (δ15N) (Phillips and Gregg 2003, Phillips and Koch 2002). 2012 P.A.P. deHart and S.E. Strand 579 Results A total of 393 arthropods was collected, with all organismal groups (harvestmen, ants, wolf spiders, cabbage loopers, and aphids; Fig. 1) found in similar frequency distributions in both 2010 (n = 292) and 2011 (n = 101). The frequency distribution of small harvestmen was significantly lower in both GM and UI plots in 2011 than 2010 (ANOVA: P = 0.05). The average light intensity, leaf-litter depth, soil pH, and soil moisture followed similar trends between the 2 sampling seasons (Table 1). Over the 2-year study, we found a significant difference between the treatment plots, with leaf-litter depth lower and soil pH greater in GM plots than UI plots (ANOVA: P = 0.05; Table 1). GM plots had significantly lower Figure 1. Total number of arthropods found in both Garlic Mustard-invaded (shaded columns) and un-invaded (un-shaded columns) plots over the 2-year study period. Table 1. Comparison of abiotic factors (X + 1 SD) from the two collections in the summer of 2010 and 2011. *Data presented for 2010 is from Alerding et al. (2011), which is temporally and spatially concomitant with our study. Abiotic characteristics Plot 2010* 2011 Light intensity (μmol/m-2s) GM 45.9 (17.2) 165.50 (54.60) UI 126.8 (34.3) 164.20 (27.50) Soil moisture (%) GM 30.2 (0.2) 24.78 (0.99) UI 51.0 (10.8) 22.69 (1.59) Soil pH GM 5.5 (0.1) 5.64 (0.15) UI 4.2 (0.1) 5.09 (0.13) Litter density (mm) GM 27.4 (2.9) 26.40 (3.52) UI 34.2 (3.2) 41.00 (3.43) 580 Southeastern Naturalist Vol. 11, No. 4 mean light intensity and soil moisture than UI plots in 2010, but did not differ in 2011 (ANOVA: P = 0.05, Table 1). δ 13C and δ15N isotopic values varied widely between collected organisms in the 2 plot treatments (Table 2). δ13C and δ15N signatures varied among predators when comparing GM and UI plots (Fig. 2), and were significantly lower for small opiliones than the other predator groups in both GM and UI plots (ANOVA: P = 0.05). In GM plots, wolf spiders are enriched in 15N compared to ants and large harvestmen (Tukey’s HSD: P < 0.05; Fig. 2). Large harvestmen and ants have similar isotopic enrichment in both δ13C and δ15N (Tukey’s HSD: P > 0.05), and therefore Isosource and Isoconc were used to evaluate the estimated overall contribution of prey for each individual (Table 3, Fig. 3). Both groups show shifts in the portion of their dietary dependency from springtails to cabbage loopers in GM areas. The contribution of large harvestmen and ants to the diet of wolf spiders increased when comparing UI to GM plots, and combined comprised ≈61% of the dietary input to wolf spiders in GM plots. While ants and large harvestmen fall well within the range of potential prey sources, wolf spiders are trophically enriched relative to these predators (Tukey’s HSD: P < 0.05; Fig. 3). δD values varied widely, ranging from -98.1 to -49.4‰, throughout all organismal groups (Table 2) and insignificantly within each organismal group between Table 2. Isotopic signatures (‰, mean ± 1 SD) and concentration of carbon and nitrogen calculated for each organism type. Organism Plot n δ13C (SD) [C] δ 15N (SD) [N] δ 18O (SD) δD (SD) Wolf spiders GM 2 -27.09 45.80 5.99 11.00 42.5 -62.9 UI 2 -26.37 47.65 5.04 11.63 42.5 -49.4 Ants GM 8 -27.59 (0.68) 43.06 4.65 (0.99) 5.10 42.0 (1.88) -91.8 (15.19) UI 8 -26.59 (0.68) 41.62 5.19 (0.98) 9.08 39.5 (3.98) -72.8 (14.95) Large harvestmen GM 8 -27.12 (0.38) 46.01 4.31 (0.59) 8.54 42.4(3.61) -76.1 (16.79) UI 8 -26.49 (0.62) 47.65 4.21 (1.31) 7.39 40.8(2.52) -59.0 (15.05) Small harvestmen GM 2 -26.62 45.37 2.13 10.88 40.6 N/A UI 2 -24.85 47.07 2.86 11.60 31.2 N/A Aphids GM 5 -31.87 (0.95) 48.88 -3.02 (1.88) 6.50 25.1 -151.7 Cabbage loopers GM 6 -32.77 (1.89) 41.17 -1.01 (1.55) 10.40 32.5 (2.37) -75.6 (4.23) Springtails UI 2 -25.15 52.98 -0.66 6.90 37.6 -54.1 GM Silique GM 6 -30.98 (1.08) 43.57 -2.17 (1.53) 2.35 32.7 (1.57) -42.5 (12.27) GM Leaf GM 6 -33.55 (0.33) 43.00 -0.75 (1.31) 3.74 28.5 (4.24) -75.0 (8.97) 2012 P.A.P. deHart and S.E. Strand 581 Figure 2. Mean δ15N and δ13C (± SD) for arthropods examined in Rockbridge County, VA over the sampling dates found in both Garlic Mustard-invaded (shaded) and un-invaded (unshaded) plots. Table 3. Relative contribution (± 1 SD) of arthropods to the diets of the primary predatory arthropods identified in this study, large harvestmen and ants, as estimated by a concentration-dependent, dual-isotope linear mixing model IsoSource (Phillips and Gregg 2003) using trophic-level fractionation factors of 1‰ for δ 13C and 3.4 ‰ for δ 15N. Relative contribution of prey (% of diet) Wolf Large Small Cabbage Predator Plot spiders Ants harvestmen harvestmen Aphids loopers Springtails Wolf spiders GM - 26 ± 0.18 35 ± 0.16 9 ± 0.07 9 ± 0.07 18 ± 0.06 4 ± 0.03 UI - 18 ± 0.11 20 ± 0.13 22 ± 0.14 12 ± 0.07 12 ± 0.09 16 ± 0.09 Ants GM 20 ± 0.11 - 23 ± 0.15 16 ± 0.12 16 ± 0.09 17 ± 0.09 8 ± 0.06 UI 21 ± 0.13 - 23 ± 0.15 17 ± 0.11 13 ± 0.07 14 ± 0.08 19 ± 0.08 Large harvestmen GM 11 ± 0.08 13 ± 0.09 - 23 ± 0.17 16 ± 0.09 22 ± 0.08 13 ± 0.09 UI 11 ± 0.08 11 ± 0.08 - 23 ± 0.11 17 ± 0.09 13 ± 0.08 25 ± 0.11 582 Southeastern Naturalist Vol. 11, No. 4 UI and GM regions (ANOVA: P > 0.05; Fig. 4). The dry weight of small harvestmen was insufficient to obtain measurable δD values. The δ18O values for the selected arthropods ranged from 39.5 to 42.5‰ (Table 1) and varied insignificantly within each organismal group between UI and GM regions (ANOVA: P > 0.05; Fig. 4). Discussion Because Garlic Mustard can severely alter the primary producer properties in forest ecosystems, we hypothesized that its introduction would alter the structure and function of the primary and secondary consumers throughout the community, as well. Previous studies have shown that such invasive species can alter abiotic and biotic components in various ecosystems through nutrient limitation and decreasing taxon diversity (Bultman and Dewitt 2008, Rudgers and Clay 2008, Stinson et al. 2006, Wolfe et al. 2008). In this work, we have shown that Garlic Mustard invasion can also have trophically cascading effects up to the generalist predator level, shifting the relationships of the arthropods found in these areas. Plant diversity was relatively constant between GM and UI plots, but plots with Garlic Mustard had significantly lower amounts of leaf litter and significantly Figure 3. Distribution of isotopic value means of potential food sources and the mean value (± 1 SD) of the predators large harvestmen (circle), ants (diamond), and wolf spiders (square) found in Garlic Mustard plots. Concentration-dependent mixing model using the prey A) aphids, B) small harvestmen, and C) cabbage loopers. Values were estimated by a concentration-dependent, dual-isotope linear mixing model IsoConc (Phillips and Koch 2002). Values reflect trophic-level fractionation of 1‰ for δ13C and 3.4‰ for δ15N for all organisms. 2012 P.A.P. deHart and S.E. Strand 583 higher pH than uninvaded plots, confirming the results of Rodgers et al. (2007). Previous research suggests that invasive species would decrease arthropod diversity (Rudgers and Clay 2008); however, the areas examined in our study showed little to no impact of Garlic Mustard on overall diversity, as all taxa were found in both GM and UI plots. The changes in light intensity for all plots between 2010 (Alerding et al. 2011) and 2011 were most likely due to variations in forest canopy rather than the presence of Garlic Mustard. Due to the inavailability of further forest canopy characteristics in 2010, we were unable to provide further analysis of forest canopy characteristics, which may be an influence on leaf-litter density and pH. While previous studies have targeted the trophic structure of arthropods in forest understory systems (Bennett and Hobson 2009), this study is to our knowledge the first to use stable isotopes to examine the trophic impact of a Garlic Mustard invasion specifically, and thus provides a valuable baseline of isotopic values for these target species. Depleted δ13C values of predators observed in GM plots could be due to a depletion of the δ13C of prey being transferred trophically, without a change in prey type. The presence of Garlic Mustard yielded a slight enrichment in δ15N for most organisms, which indicates that the low measured nitrogen concentration of Garlic Figure 4. Mean δ18O and δD (± 1 SD) for arthropods examined in Rockbridge County, VA over the sampling dates found in both Garlic Mustard invaded (shaded) and un-invaded (unshaded) plots. 584 Southeastern Naturalist Vol. 11, No. 4 Mustard may be minimizing its trophic signature throughout the food web. Alternatively, the presence of Garlic Mustard in these invaded areas may cause a physical disruption in the natural diets of the arthropods due to prey displacement. As has been suggested for other organisms feeding at a similar trophic level, such a displacement may force these secondary consumers to search further and longer for the same prey, or to shift to lower-quality prey sources (McNabb et al. 2001, Oelberman and Scheu 2002, Wise et al. 2006). The trophic relationships of the generalist predators suggested in previous research were noticeably different in the current study. Small harvestmen are shown to be isotopically depleted relative to the other 3 predator groups, confirming that our a priori size classification was indeed reflective of trophic level. Large harvestmen previously served as the most trophically enriched generalist predator in uninvaded areas (Strand 2010), but further isotopic analyses suggest that ants and wolf spiders actually play a larger role in ecosystem processing in the forest understory. This trend is more consistent with patterns observed in other regions (Oelbermann and Scheu 2002). Ant predation is typically difficult to track because of their broad diet, but research has shown that they typically eat organisms within their guild (Oelbermann and Scheu 2002). The stable isotope analyses reveal that both ants and wolf spiders occupy a similar and overlapping ecological role in uninvaded systems, feeding at a higher trophic level than large harvestmen. In particular, the δ13C and δ15N signatures of the 3 taxa overlapped in the 3 plots. One of the most noticeable differences revealed by SIA was that wolf spiders were significantly enriched in 15N compared to ants and large harvestmen in the GM plots, suggesting they had shifted to a higher trophic level. This shift is consistent with the role wolf spiders were shown to serve in other forest understory systems, where both intraguild predation and cannibalism have been documented (Rypstra and Samu 2005). The results of our linear mixing model indicate a widespread distribution of all potential prey components to the 3 predator groups in both GM and UI regions, but also revealed significant changes in feeding preference and trophic positioning in response to the presence of Garlic Mustard. Ants and large harvestmen had subtle responses to the presence of Garlic Mustard, appropriate for their classification as “generalist predators”. As expected, both groups shifted a portion of their dietary dependency from springtails to cabbage loopers in GM areas. Wolf spiders showed a more dramatic change in trophic positioning in response to the presence of Garlic Mustard. In uninvaded plots, wolf spiders had a diet typical of a generalist predator and similar to that of harvestment and ants. In the GM plots, however, wolf spiders changed trophic position to a secondary predator, feeding on large harvestmen and ants. Indeed, these other generalist predators collectively constituted more than half the diet of wolf spiders in GM plots. The results of both linear and concentration-dependent mixing models confirmed the relative trophic positioning of these 3 predators in relationship to the generalist consumers aphids, small harvestmen, and cabbage loopers in GM 2012 P.A.P. deHart and S.E. Strand 585 plots. While ants and large harvestmen fall well within the range of potential prey sources, wolf spiders are trophically enriched relative to the other generalist predators, underscoring their consumption of higher-trophic-level organisms. While the isotopic shift in nutrient acquisition between UI and GM regions could be due to a change in prey type, the prey organisms observed solely in GM areas (aphids and cabbage loopers) were also observed in varying quantities in the diet of predators collected in UI plots. This pattern is likely due to predators captured in UI plots consuming aphids and cabbage loopers in GM plots, then moving to UI plots, where they were then captured in our study. In contrast, organisms found only in UI plots (springtails) represented the smallest potential input to predator diet. This pattern indicates a potential behavioral change in response to environmental factors that can have a serious impact on organism survival (Bennett and Hobson 2009, deHart 2006, Kaufman et al. 2010, McNabb et al. 2001, Wise et al. 2006). Additionally, these dietary trends may continue to change over time. Factors such as life-stage modifications, organism age, and seasonal changes may also cause differences over time in the nutrient flow (Bennett and Hobson 2009, McNabb et al. 2001, Oelbermann and Scheu 2002), stimulating shifts in the trophic position of organisms throughout an ecosystem. As has been suggested in prior studies, the mechanism for the variation in both δD and δ18O may be more dependent on organismal water source than dietary inputs (Bowen et al. 2005; DeNiro and Epstein 1981a, b; Estep and Dabrowski 1980), and so warrants further investigation of fine-scale water inputs to this ecosystem. Wolf spiders are again the most enriched of the organisms in both δ18O and δD. If the mechanism of enrichment for δ18O and δD is similar to enrichment of δ15N and δ13C, then that pattern would further confirm the trophic position of wolf spiders in the forest understory system. The relationship between the other organisms is more complex, however, because enrichment values for δ18O and δD for ants and large harvestmen differ more between UI and GM plots than was observed with δ15N and δ13C. This finding may be due to the differences in soil moisture between the 2 plot types. Complicating this interpretation is that δ18O and δD are taken up by the organisms at different rates. On average, organisms from GM plots are more enriched in δ18O, but significantly depleted in δD. A portion of this discrepancy could be due to microclimate changes in the soil utilized by some arthropods as a nutrient source. These changes could yield an increased abundance of isotopically depleted soil, thus depleting the signature of those arthropods dependent on that pool of nutrients. Widely varying ratios for these isotopes could be due to variable lipid composition in typical prey items, not necessarily differential prey selection (DeNiro and Epstein 1981b). Future studies examining a wider array of primary consumers, including a more significant representation from the small harvestmen, may untangle the specific mechanisms driving these patterns in δ18O and δD. In conclusion, this research provides evidence that Garlic Mustard significantly alters the diet, distribution, and behavior of arthropods, and that invasive species can have cascading effects up trophic levels. In the presence of this invasive species, generalist predators experience a measurable shift in their diet 586 Southeastern Naturalist Vol. 11, No. 4 from that of a true generalist predator to the more energetically expensive role of higher-order predator, reliant on intraguild predation and cannibalism. These changes are likely due less to organisms utilizing Garlic Mustard as the primary food source, but more to an overall decrease in abundance of preferred prey, causing arthropods to spend more time and distance searching for food. While the degree to which arthropod physiology is affected by an invasive plant species is still yet to be determined, these behavioral shifts are clear, and are beginning to change the flow of energy throughout the ecosystem. This study provides essential baseline isotopic values for forest arthropods and also demonstrates the utility of SIA and multi-isotopic methodologies to address complex questions in both arthropod biology and the impacts of invasive species. Acknowledgments Research funding to P.A.P. deHart was provided by a Grants-in-Aid award from the Research Committee and the Department of Biology at the Virginia Military Institute. Support and stipend funds to S.E. Strand were provided by the Summer Undergraduate Research Institute and the Swope Summer Research Program. Equipment and sample processing funds to S.E. Strand were additionally provided by the Virginia Military Institute Research Labs through the Wetmore Research Fund for undergraduate research and the Department of Biology at the Virginia Military Institute. Anne Alerding (VMI Biology) was essential to the conception and progress of this research. Richard Rowe (VMI Biology) provided review and critical comments. Field assistance was provided by Matt Elliott, Roy Hunter, and Matthew Waalkes. Literature Cited Alerding, A.B., R.M. Hunter, M.R. Waalkes, and S.E. Strand. 2011. Garlic Mustard (Alliaria petiolata, Brassicaceae) juveniles as pH modulators of a detritivore food web. Unpublished manuscript to the Virginia Military Institute Biology Department, Lexington, VA. 18 pp. Bennett, P.M., and K.A. Hobson. 2009. Trophic structure of a boreal forest arthropod community revealed by stable isotope (δ13C, δ15N) analyses. Entomological Science 12:17–24. Bowen, G.J., L.I. Wassenaar, and K.A. Hobson. 2005. Global application of stable hydrogen and oxygen isotopes to wildlife forensics. Oecologia 143:337–348. Bultman, T.L., and D.J. DeWitt. 2008. Effect of an invasive ground-cover plant on the abundance and diversity of a forest-floor spider assemblage. Bio Invasions 10:749–756. deHart, P.A.P. 2006. A multiple stable-isotope study of Steller Sea Lions and Bowhead Whales: Signals of a changing northern environment. Ph.D. Dissertation. University of Alaska Fairbanks, Fairbanks, AK. 146 pp. DeNiro, M.J., and S. Epstein. 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochimica et Cosmochimica Acta 42:495–506. DeNiro, M.J., and S. Epstein. 1981a. Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta 45:341–351. DeNiro, M.J., and S. Epstein. 1981b. Hydrogen isotope ratios of mouse tissues are influenced by a variety of factors other than diet. Science 214:1374–1375. Estep, M.F., and H. Dabrowski. 1980. Tracing food webs with stable hydrogen isotopes. Science 209(4464):1537–8. 2012 P.A.P. deHart and S.E. Strand 587 Gardner, W.H. 1986. Methods of soil analysis, Part 1. Physical and mineralogical methods - agronomy monograph. American Society of Agronomy and Soil Science Society of America. 9(2):493–544. Grey, J., and R.I. Jones. 2001. Seasonal changes in the importance of the source of organic matter to the diet of zooplankton in Loch Ness, as indicated by stable isotope analysis. American Society of Limnology and Oceanography 46(3)505–531. Hobson, K.A. 1999. Tracing origins and migration of wildlife using stable isotopes: A review. Oecologia 120: 314–326. Kaufman, M.G., K.S. Pelz-Stelinski, D.A. Yee, S.A. Juliano, P.H. Ostrom, and E.D. Walker. 2010. Stable-isotope analysis reveals detrital resource base sources of the Tree Hole Mosquito, Aedes triseriatus. Ecological Entomology 35:586–593. Keeler, M.S., and F.S. Chew. 2008. Escaping an evolutionary trap: Preference and performance of a native insect on an exotic invasive host. Oecologia 156:559–568. Martinez, N.D., B.A. Hawkins, H.A. Dawah, and B.P. Feifarek. 1999. Effects of sampling effort on the characterization of food web structure. Ecology 80:1044–1055. McLean, E.O., A.L. Page, R.H. Miler, and D.R. Keeny. 1982. Soil pH and lime requirement. Methods of soil analysis, Part 2. Chemical and microbiological properties - agronomy monograph. America Society of Agronomy and Soil Science Society of America. 9(2):199–225. McNabb, D.M., J. Halaj, and D.H. Wise. 2001. Inferring trophic positions of generalists predators and their linkage to the detrital food web in agroecosystems: A stableisotope analysis. Pedobiologia 45:289–297. Myers, C.V., and R.C. Anderson. 2002. Patterns of seed-mass variation and their effects on seedling traits in Alliaria petiolata. American Midland Naturalist 150(2) 231–245. Oelbermann, K., and S. Scheu. 2002. Stable isotope enrichment (δ15N and δ13C) in a generalist predator (Paradosa lugubris, Aureae: Lycosidae): Effects of prey quality. Oecologia 130:337–344. Ostrom, P.H., M. Colunga-Garcia, and S.H. Gage. 1997. Establishing pathways of energy flow for insect predators using stable-isotope ratios: Field and laboratory evidence. Oecologia 109:108–113. Peterson, B.J., and B. Fry. 1987. Stable isotopes in ecosystem studies. Annual Review of Ecology and Systematics 18:293–320. Phillips, D.L., and J.W. Gregg. 2003. Source partitioning using stable isotopes: Coping with too many sources. Oecologia 136(2):261–269. Phillips, D.L., and P.L. Koch. 2002. Incorporating concentration dependence in stableisotope mixing models. Oecologia 130:114–125. Post, D.M. 2002. Using stable isotopes to estimate trophic positon: Models, methods, and assumptions. Ecology 83(3):703–718. Roberts, K.J., and R.C. Anderson. 2001. Effect of Garlic Mustard (Alliaria petiolata (Beib. Cavara & Grande)) extracts on plants and arbuscular mycorrhizal (AM) fungi. American Midland Naturalist 146(1):146–152. Rodgers, V.L., B.E. Wolfe, L.K. Werden, and A.C. Finzi. 2007. The invasive species Alliaria petiolata (Garlic Mustard) increases soil nutrient availability in northern hardwoodconifer forests. Oecologia 157(3):459–471. Rudgers, J.A., and K. Clay. 2008. An invasive plant-fungal mutualism reduces arthropod diversity. Ecology Letters 11:831–840 Rypstra, A.L., and F. Samu. 2005. Size-dependent intraguild predation and cannibalism in coexisting wolf spiders (Araneae, Lycosidae). The Journal of Arachnology 33:390–397. 588 Southeastern Naturalist Vol. 11, No. 4 Stinson, K.A., S.A. Campbell, J.R. Powell, B.E. Wolfe, R.M. Callaway, G.C. Thelen, S.G. Hallett, D. Prati, and J.N. Klironomos. 2006. Invasive plant suppresses the growth of native trees seedlings by disrupting belowground mutualisms. PLoS Biol 4(5):e140. Stobbs, L.W., and J.G. Van Schagen. 1987. Occurrence and characterization of a turnip mosaic virus isolate infecting Alliaria petiolata in Ontario, Canada. Plant Disease 71:965–968. Strand, S.E. 2010. Garlic Mustard and the observed effect on arthropod diversity using stable-isotope analysis. Research paper for the Summer Undergraduate Research Institute, Virginia Military Institute, Lexington, VA. 16 pp. Tieszen, L.L. 1978. Carbon-isotope fractionation in biological material. Nature 276:97–98. Tieszen, L.L., T.W. Boutton, K.G. Tesdahl, and N.A. Slade. 1983.Fractionation and turnover of stable carbon isotopes in animal tissues: Implications for δ13C analysis of diet. Oecologia 57:32–37. Wikum, D.A. 1978. Application of the Braun-Blanquet cover-abundance scale for vegetation analysis in land development studies. Environmental Management 2(4):323–325. Wise, D.H., D.M. Moldenhauer, and J. Halaj. 2006. Using stable isotopes to reveal shifts in prey consumption by generalist predators. Ecological Applications 16(3):865–876. Wolfe, B.E., V.L. Rodgers, K.A. Stinson, and A. Pringle. 2008. The invasive plant Alliaria petiolata (Garlic Mustard) inhibits ectomycorrhizal fungi in its introduced range. Journal of Ecology 96:777–783. Yates, C.N., and S.D. Murphy. 2008. Observations of herbivore attack on Garlic Mustard (Alliaria petiolata) in Southwestern Ontario, Canada. Biological Invasions 10:757–760.