Page  51 ï~~2003 THE MICHIGAN BOTANIST 51 ALLELOPATHY IN RHAMNUS CATHARTICA, EUROPEAN BUCKTHORN Scott Seltzner and Thomas L. Eddy 376 Palmer Avenue 426 Walker Avenue Green Lake, Wisconsin 54941 Green Lake, Wisconsin 54941 ABSTRACT Rhamnus cathartica L., common or European buckthorn, is a thoroughly naturalized introduction in much of the United States and southern Canada. Since the mid-19th century, when R. cathartica was introduced into Wisconsin as an ornamental hedge, it has spread extensively, colonizing disturbed forest and savanna habitats. In this study, allelopathy was examined as one contributing factor to the invasive character exhibited by R. cathartica by measuring the effects of plant exudates obtained from fruit, leaves, bark, and roots on seed germination in alfalfa (Medicago sativa L.). The data show that the exudate from the drupes exerts the greatest inhibitory effect on seed germination, with less effect by leaf exudate. Bark and root exudates show no significant effects on seed germination. INTRODUCTION The genus Rhamnus includes about 100 species (Gleason & Cronquist 1991); four species occur in Wisconsin: Rhamnus alnifolia L'H6r., R. cathartica L., R. frangula L., and R. lanceolata Pursh var. glabrata Gleason (WIS 2002). Rhamnus cathartica and R. frangula are non-native, originating from Eurasia, but widely naturalized. Rhamnus cathartica, common or European buckthorn, the subject of this study, is adapted to mainly well-drained upland soils, while R. frangula establishes more frequently in wetlands. The floristic rating for R. cathartica as a "wetlands indicator" species is low, being designated FACU (facultative upland), with an estimated probability occurrence in non-wetlands of 67%-99% (WIS 2002). R. frangula, which is rated FAC+ (facultative), is equally likely to occur in wetlands or non-wetlands with an estimated probability 34%-66%. Both species are aggressive pioneer competitors that can form crowded thickets that exclude native plants, thereby constituting a threat to Wisconsin's native flora (Kline 1999; Taft & Solecki 1990; WDNR 2002). Rhamnus cathartica is distributed mainly in the southern half of Wisconsin, readily colonizing disturbed woodland understories, notably open oak forests and savanna habitats (Fig. 1). Kline (1999) refers to R. cathartica as "an obnoxious weed-the shrub equivalent of purple loosestrife." The species regularly naturalizes in thickets, hedgerows, old fields, pastures, and roadsides. In old fields of central New York State, however, Gill and Marks (1991) report that post-dispersal predation and frost heaving make R. cathartica establishment a "very low probability event." Nonetheless, our observations of R. cathartica in Green Lake County (east central Wisconsin) show that the plant is quite com

Page  52 ï~~52 THE MICHIGAN BOTANIST Vol. 42 52 THE MICHIGAN BOTANIST Vol. 42 FIGURE 1. Invasive understory of Rhamnus cathartica in oak woods at Highknocker Park, Green Lake, WI (5 January 2003). mon in various disturbed habitats, including an abandoned apple orchard (Fig. 2). From the time of its introduction to North America as an ornamental in the mid-19th century, R. cathartica has spread extensively, ranging west from Nova Scotia to Saskatchewan, south to Missouri and east to Virginia (Fig. 3). As early as 1849 both R. cathartica and R. frangula were planted in Wisconsin as hedgerow cultivars (WDNR 2002). Just as with R. frangula, R. cathartica is an alternate host to crown rust, a fungal parasite of oats, as well as the soybean aphid, which overwinters in the buds of R. carthartica (Ginns 1986; Hartzler & Pope 2002) Gleason & Cronquist (1991) describe R. cathartica as a shrub or small tree growing in height to six meters. Short thorns terminate some of the branches; the leaves are elliptical and mostly opposite, 3-6 centimeters in length, with finely serrated margins. Three to four pairs of raised secondary veins converge near the blade apex. The unisexual flowers are 4-merous and open with the leaves, May through June. The fruit is a dark purple or black-colored fleshy drupe, 5-6 mm in diameter, typically containing four stones (Figures 4 and 5). Unlike native woody plants, the green foliage of R. cathartica persists well after the first frosts, thus making it highly visible and easy to identify. Several factors contribute to the success of R. cathartica outside its native range: lack of natural predators; wide habitat tolerance; rapid growth rate and vigorous vegetative regeneration; prolific fruit/seed production and potential for long-distance seed dispersal; and phenotypic plasticity that enables R. cathartica

Page  53 ï~~2003 THE MICHIGAN BOTANIST 53 2003 THE MICHIGAN BOTANIST 53 FIGURE 2. Colonies of Rhamnus cathartica in an abandoned apple orchard (5 January 2003). to exploit varying environmental conditions, notably in its response to light (Gale 2000; WDNR 2002). Since the leaves of R. cathartica emerge earlier and senesce later, they are present an average 58 days longer compared to that of two native species, Cornus racemosa Lam. and Prunus serotina Ehrh. (Harrington et al. 1989). Thus, besides maximizing light for photosynthesis by an extended growing season, R. cathartica can eventually shade out native groundlayer species that develop after buckthorn leafs out (Dugal 1992; Gourley & Howell 1984). Furthermore, high densities of seeds beneath parent trees can result in 162 to 215 buckthorn seedlings/m2, or more than 2 million seedlings/hectare (Moriarty 1998). The impact on native plant communities by dense groundcovers of R. cathartica is potentially devastating. In addition to competing for space, nutrients, and light, decomposition of R. cathartica leaf litter may further contribute to its ability to dominate disturbed habitats. Heneghan et al. (2002) report that rapid decomposition of buckthorn leaf litter, which is high in nitrogen, can modify soil fertility to the degree that increased soil nitrogen may favor buckthorn growth. Reproduction by R. cathartica is primarily by seed with great potential for long-distance dispersal by frugivorus songbirds, the main dispersal agents. The drupes are ingested when alternative food sources become scarce (Gourley & Howell 1984). The presence of anthraquinone in the fruit functions as a laxative, permitting swift passage of the seed, presumably before seed viability diminishes (Archibold et al. 1997).

Page  54 ï~~54 THE MICHIGAN BOTANIST Vol. 42 54 THE MICHIGAN BOTANIST Vol. 42 FIGURE 3. Distribution of Rhamnus cathartica in North America. The single-hatched areas represent regions where R. cathartica is cited in floras and other literature sources. The small circles and solid dark areas represent recent and historic collections, as well as sight and literature records (IPCAN 2002). Allelopathy Allelopathy, the chemical inhibition of one species by another, has also been suggested as a factor contributing to the invasive character exhibited by R. cathartica (Heidorn 1991). At Pipestone National Monument, Minnesota, Boudreau and Wilson (1992) report that allelopathic compounds within the fruit and leaves might function to inhibit seed germination and growth by other plants. Similar conclusions by Krebach and Wilson (1996) suggest that the fruit of R. cathartica contains allelochemicals that may retard growth by competing plants. Krebach noted that fruit exudates obtained in the summer and fall inhibited germination of ryegrass (Lolium sp.) seeds, but after frost the fruits no longer had any noticeable effects (Dr. C. Wilson, personal communication, 17 June 2002). According to Wilson (personal communication), young buckthom leaves do not exert as strong an allelopathic effect as was observed through the summer and fall. Krock and Williams (2002) examined the effects of leaf and root tissue exudates from R. frangula (glossy buckthorn), another invasive exotic. They measured the effects of exudates on seed germination and growth in lettuce (Lactuca sativa L.) and radish (Raphanus sativus L.), but did not observe any evidence of allelopathy. It should be noted, however, that exudates from the fruit of R. frangula were not applied in their seed trials.

Page  55 ï~~2003 THE MICHIGAN BOTANIST 55 2003 THE MICHIGAN BOTANIST 55 FIGURE 4. Winter drupe of Rhamnus cathartica (O10x). In a study of R. cathartica near Saskatoon, Saskatchewan, Archibold et al. (1997) report that 90% of the buckthorn fruit eventually falls beneath the parent trees. Moriarty (1998), a wildlife biologist with Hennepin Parks, Minneapolis/St. Paul, reports that there can be about 807 seeds/m2 beneath mature buckthorn. The seeds remain viable in the soil one to five years (USGS 2002). On the assumption that each drupe contains four stones, this amounts to 202 drupes/m2, a potentially vast reservoir of allelochemicals. If the drupes do contain allelochemicals, a combination of aggressive colonization and allelopathy via fallen fruits and leaf litter beneath parent trees may be an adaptive strategy for excluding competing species. There are suggestions of allelopathy in R. cathartica by various workers (Heidorn 1991; Boudreau & Wilson 1992; Krebach & Wilson 1996; Larson 2002; Lerman n.d; Coder 1998; B.N.I. 2002), but published studies that measure the allelopathic effects by R. cathartica under controlled laboratory conditions are wanting. We hypothesized that fallen fruits and leaves of R. cathartica contain secondary metabolites that intensify interspecific competition by interfering with germination and growth by other species. Based on the literature (Boudreau & Wilson 1992; Krebach & Wilson 1996), we hypothesized that exudates prepared from the drupes and leaves might produce the greatest interference with seed germination by another species.

Page  56 ï~~56 THE MICHIGAN BOTANIST Vol. 42 56 THE MICHIGAN BOTANIST Vol. 42 FIGURE 5. The four one-seeded stones of a typical drupe in Rhamnus cathartica (10x). METHODS In this investigation we 1) studied what parts of R. cathartica produce the greatest (and least) allelopathic effects on seed germination by another species, and 2) examined the effects on germination by seeds treated with varying exudate concentrations from different parts of R. cathartica. Fresh plant material from R. cathartica was collected from three different locations in Green Lake County, Wisconsin. Exudate sources included drupes, leaves, bark, and roots. Drupes, leaves and bark were collected in the fall and separately bagged, while roots were collected in the spring. Samples of plant parts from the three different locations were combined, weighed, and then fully dried to determine the original moisture content. The percentage water content for each part is: drupes, 51.2%; leaves, 69.5%; bark, 37.8%; and roots, 17.1%. Full-strength (100% concentrate) exudates were prepared by processing the plant material in a food blender, then eluting the mixture with a minimum volume of distilled water. The weight ratio of plant material to distilled water was: drupes, 2:1; leaves, 1:2; bark, 1:2; and roots, 1:1. The percentage of distilled water added to each part included: drupes, 26%; leaves, 66%; bark, 69%; and roots, 52%. Solutions from the blended mixtures were obtained by squeezing the macerated pulp to collect the exudate, which was then suction-filtered with a vacuum pump. For comparison of exudates obtained at different temperatures, filtrates were prepared at room temperature (18-200C) and in a warm waterbath (49-51 Â~C). Exudates were refrigerated until used. The average pH of exudates from each part was: drupes, pH 5.2; leaves, pH 6.5; bark, pH 3.4; and roots, pH 5.0. Alfalfa (Medicago sativa) seed was used to test the effects of R. cathartica allelopathy on germination because of its availability, uniformity, and fast germination time. We agree with Inderjit

Page  57 ï~~2003 THE MICHIGAN BOTANIST 57 (1996) that the "allelopathic bioassay must be ecologically realistic..." and that "experiments must be designed with conditions resembling those found in natural systems." However, for the scope of this investigation we chose M. sativa, a non-native species for the reasons cited above. Our objective was to determine if exudates from R. cathartica exhibit an allelopathic effect on any seed germination; measuring effects of exudates on native plants remains an objective for the future. Seeds were placed on cut paper toweling in Petri dishes (100 seeds/dish/10 trials) and treated with 50 drops (3.4mL) of varying exudate concentrations (25%, 50%, 75%, 100%) from each plant part. Seeds moistened with distilled water served as a control. All germination dishes were placed in a lighted Climatarium at a temperature between 24-300C for five days, and then sprouts counted. Seed germination data from exudates obtained from room temperature and warm waterbath preparations were combined when it was determined that minor statistical differences (in the leaves) to no statistical differences in other plant parts could be detected. There were 2000 seeds tested for germination per control and for each exudate source per concentration. RESULTS AND DISCUSSION A summary of the data shows that exudates from the drupes of R. cathartica exhibit the greatest inhibitory effect on alfalfa seed germination (Table 1; Fig. 6). Moreover, a marked decrease in seed germination corresponds with an increase in exudate concentration, e.g. at 100% concentration, 1 seed per 2000 germinated. Exudates from the leaves of R. cathartica also demonstrate mild allelopathic effects on seed germination, while the effects of exudates from bark and roots were slight to none. Analyses (InStat 1998) of germination by seeds treated with varying exudate concentrations from each source (drupes, leaves, bark, roots) were conducted using nonparametric repeated measures ANOVA (Friedman test). Increased concentrations of exudates from drupes and leaves, which produced notable inhibition of seed germination, measure P values far below the threshold (0.05). Thus, variation is significantly greater than expected by chance, or to put it another way, there is a high probability that increased exudate concentrations account for decreased seed germination. Exudates obtained from bark and roots, however, measure above the P value threshold (0.05). Increasing bark and root exudate concentrations do not appear to decrease seed germination. Dunn's multiple comparisons test (post-test) on seed germination variance among the different exudate concentrations closely agrees with the Friedman test results. The P values generated by the Friedman test for each exudate source are shown in Table 2. We also investigated seed germination data variance among drupes, leaves, TABLE 1. Summary of alfalfa seed germination trials Total Seeds Germinated/Exudate Concentration Exudate Source Control 25% 50% 75% 100% Drupes 1846 1708 256 45 1 Leaves 1864 1853 1718 1432 1167 Bark 1877 1860 1842 1823 1802 Roots 1851 1854 1859 1861 1837 1Based on 2000 seeds per trial

Page  58 ï~~58 THE MICHIGAN BOTANIST Vol. 42 58 THE MICHIGAN BOTANIST Vol. 42 u ate Concentrations FIGURE 6. Effect of various exudates and concentrations on germination of alfalfa seeds. bark, and roots. Table 3 summarizes statistics (e.g. mean, SD, SEM, median) for the average total seeds germinated per exudate source. Based on nonparametric repeated measures ANOVA (Friedman test) of the seed germination data variance among drupes, leaves, barks and roots, the P value equals 0.0016, which is considered very significant. In other words, the null hypothesis is rejected and variations among the germination data are significantly greater than expected by chance. Again, a post-test (Dunn's) corresponds closely with the Friedman test, i.e., variance is significantly greater than by chance alone. In addition to testing variance produced by different plant part exudates, the effect on germination by exudates prepared from "cold" and "warm" baths was examined. Paired t tests that compare results among seeds treated with exudates from cold and warm water baths show no significant differences among drupes, bark, and roots. In leaves, however, seed germination from cold- and warm-extracted exudates reveals a modest statistical difference in germination results (Table 4; Fig. 7). Note that the P value of "Leaves" is 0.0494, slightly below the threshold value of 0.05. TABLE 2. Friedman test results comparing exudate concentrations on alfalfa seed germination. Drupes P value < 0.0001, considered extremely significant. Variation is significantly greater than expected by chance. Leaves P value < 0.0001, considered extremely significant. Variation is significantly greater than expected by chance. Bark P value = 0.1185, considered not significant. Variation is not significantly greater than expected by chance. Roots P value + 0.4821, considered not significant. Variation is not significantly greater than expected by chance.

Page  59 ï~~2003 THE MICHIGAN BOTANIST 59 TABLE 3. Statistical summary for the average total seeds germinated per exudate source. Seed Germination Summary Drupes Leaves Bark Roots Mean 770.9 1606.8 1840.8 1852.4 Standard Deviation 924.77 301.29 29.592 9.476 Sample Size* 5 5 5 5 Standard Error of Mean 413.57 134.74 13.234 4.238 Lower 95% confidence limit -377.17 1232.8 1804.1 1840.6 Upper 95% confidence limit 1919.0 1980.8 1877.5 1864.2 Minimum 0.5000 1167.0 1802.0 1837,0 Neduab (50th percentile) 255.50 1718.0 1842.0 1854.0 Maximum 1846.0 1864.0 1877.0 1861.0 Normality test KS 0.3113 0.2440 0.1418 0.2413 Normality text P value >0.10 >0.10 >0.10 >0.10 Passed normality test? Yes Yes Yes Yes *Sample size is based on five trials (Control, 25%, 50%, 75%, 100%) with each trial consisting of 2000 seeds tested, i.e. total 10,000 seeds per exudate source. TABLE 4. Paired t test results comparing effects of cold-water-extracted versus warm-water- extracted exudates on germination. Exudate Source T value Two-tailed P value Drupes 0.3638 with 4 degrees of freedom 0.7344 Leaves 2.787 wih 4 degrees of freedom 0.0494 Bark 2.389 with 4 degrees of freedom 0.0752 Roots 0.8480 with 4 degrees of freedom 0.4442 FIGURE 7. Effects of cold-extracted versus warm-extracted exudates.

Page  60 ï~~60 THE MICHIGAN BOTANIST Vol. 42 CONCLUSIONS AND FUTURE STUDIES In this study, R. cathartica exhibits allelopathy toward seed germination in alfalfa, Medicago sativa. The data show that the exudate from the drupes exerts the greatest inhibitory effect on seed germination, with less effect by leaf exudate. Bark and root exudates show no significant effects on seed germination. We recommend that future studies of allelopathy in R. cathartica incorporate experimental designs that demonstrate allelopathy as an essential ecological mechanism of plant interference by the following: 1) isolate and identify allelopathic compounds, 2) account for the significance of exudate mixtures, because according to Einhellig (1995) the chances of only one compound causing allelopathic effects are remote, 3) collect data on the fate and persistence of allelochemicals in soil and their interactions with abiotic and biotic factors of the natural environment, and 4) obtain and test exudates from plant parts throughout the growing season to improve understanding of the temporal aspect of allelopathy. For comparison, we recommend that future investigations include similar bioassay studies on allelopathy in R. frangula, which like R. cathartica exhibits weedy tendencies. Laboratory bioassays (Krock & Williams 2002) indicate that exudates from leaf and root tissues of R. frangula produced stimulatory, not inhibitory effects on seed germination in Lactuca sativa and Raphanus sativus. However, the allelopathic effects by drupe exudates from R. frangula were not investigated. ACKNOWLEDGMENTS This study was undertaken by Scott Seltzner as partial fulfillment of a senior biology course requirement at Green Lake School District, Green Lake, Wisconsin. Thanks go to Sara Wichlacz, Ripon College, for proofreading the manuscript. The photography is by co-author, T. L. Eddy. LITERATURE CITED Archibold, O. W., D. Brooks, & L. Delanoy. 1997. An investigation of the invasive shrub European buckthorn, Rhamnus cathartica L., near Saskatoon, Saskatchewan. Canadian Field-Naturalist 111(4): 617-621. B.N.I. (Biocontrol News and Information). 2002. Buckthorn biocontrol for North America [available online] June 2002, Volume 23 No. 2. Retrieved 3 January 2003 from the WWW: http://pest. Boudreau, D. and G. Wilson. 1992. Buckthorn research and control at Pipestone Monument. Restoration & Management Notes 10: 94-95. Coder, K. D. 1998. Exotic trees in the United States: naturalized or escaped from cultivation [available online]. Retrieved 3 January 2003 from the WWW: service/library/for98-027/ Dugal, A. W. 1992. Leitrim Albion Road wetlands part 2. Trail and Landscape 26: 64-94. Einhellig, F. A. 1989. Interactive effects of allelochemicals and environmental stress. In Phytochemical Ecology: Allelochemicals, Mycotoxins, and Insect Pheromones and Allomones (Chou, C. H., Waller, G. R., eds.), 101-116. Institute of Botany, Academia Sinica Monograph Series No. 9, Taipei, ROC. Gale, S. W. 2000. Control of the invasive exotic Rhamnus cathartica in North American Forests [available online]. Restoration and Reclamation Review, Volume 6-Fall 2000. Retrieved 15 November 2002 from the WWW:

Page  61 ï~~2003 THE MICHIGAN BOTANIST 61 Gill, D. & P. L. Marks. 1991. Tree and shrub seedling colonization of old fields in central New York. Ecological Monographs 6:183-205. Gleason, H A. & A. Cronquist. 1991. Manual of Vascular Plants of Northeastern United States and Adjacent Canada (2nd edition). The New York Botanical Garden. Gourley, L. C. & E. Howell. 1984. Factors in buckthorn invasion documented; control measures checked. Restoration & Management Notes 2: 87. InStat (GraphPad Software). 1998. InStat guide to choosing and interpreting statistical tests, GraphPad Software, Inc., San Diego CA ( Harrington, R. A., B. J. Brown & P. B. Reich. 1989. Ecophysiology of exotic and native shrubs in southern Wisconsin. Oecologia 80: 356-367. Hartzler, B. & R. Pope. 2002. Buckthorn control and soybean aphids. Iowa State University, Weed Science [available online]. Retrieved 3 January 2003from the WWW: http://www.weeds. Heidorn, R. 1991. Vegetation management guideline: exotic buckthorns common buckthorn (Rhamnus cathartica L.), glossy buckthorn (Rhamnus frangula L.), Dahurian buckthorn (Rhamnus davurica Pall.). Natural Areas Journal 11: 216-217. Heneghan, L., C. Clay, & C. Brundage. 2002. Rapid decomposition of buckthorn litter may change soil nutrient levels. Ecological Restoration, 20(2):108-111. Inderjit. 1996. Plant phenolics in allelopathy. Botanical Review 62: 186-202. Inderjit & K. M. M. Dakshini 1995. On laboratory bioassays in allelopathy. Botanical Review 61: 28-44. IPCAN (Invasive Plants of Canada). 2002. Invasive exotic plants of Canada - common buckthorn (Fact Sheet No. 7) [available online]. Retrieved 6 January 2003 from the WWW: Kline, J. 1999. Wisconsin plant of the week - Rhamnus cathartica [available online]. Retrieved 21 December 2002 from the WWW: plantpages_71.html Krock, S. L. & C. E. Williams. August 2002. Allelopathic potential of the alien shrub glossy buckthorn, Rhamnusfrangula: a laboratory bioassay. Journal of the Pennsylvania Academy of Science. 76(1): 17-21 Larson, J. R. 2002. Buckthorn has become a pernicious invader. Minnesota Plant Press 21(2) [available online]. Retrieved 21 December 2002 from the WWW: mnps/papers/Larson2002212.html Lerman, M. M. (n.d.). Buckthorn taking over. Retrieved 3 January 2003 from the WWW: Moriarty, J. J. 1998. The Trouble with backyard buckthorn [available online]. Minnesota Conservation Volunteer (July-August 1998). Retrieved 5 January 2003 from the WWW: Taft. J. B. & M. K. Solecki. 1993. Vascular flora of the wetland and prairie communities of Gavin Bog and Prairie Natural Preserve, Lake County, Illinois. Rhodora 92: 142-165. USGS (United States Geological Survey). 2002. Southwest exotic plant information clearinghouse [available online]. Retrieved 29 December 2002 from the WWW: WIS (University of Wisconsin-Madison Herbarium). 2002. Wisconsin Vascular Plant Checklist [available online]. Retrieved 30 December 2002 from the WWW: http:/ wisflora Wisconsin Department of Natural Resources. 2002. Invasive species-non-native plants, common buckthorn [available online]. Retrieved 30 December 2002 from the WWW: