MULLIGAN, M. R., AND D. L. GORCHOV. (Department of Botany, Miami University, Oxford, OH 45056). Population loss of goldenseal, Hydrastis canadensis L. (Ranunculaceae), in Ohio. J. Torrey Bot. Soc. 131: 305-310. 2004.-Goldenseal, Hydrastis canadensis L., is harvested from forests in the eastern U.S. for its rhizome, which is considered to have medicinal properties. While listed as rare or threatened in many states, its status in Ohio has not been assessed. To establish the status of historic goldenseal populations, we assessed 71 sites where voucher specimens had been collected from 1845 to 1998. Of these sites, 13% were deforested and no longer supported populations. Goldenseal was found on 65% of the remaining forested sites. Nearly half of documented goldenseal populations have become extinct, suggesting an overall decline of goldenseal in Ohio. The major cause of extinction appears to differ among Ecoregions, with deforestation important in the Eastern Corn Belt Plains, herbivory by white-tailed deer in Erie/Ontario Drift and Lake Plain, and overcollection in the Western Allegheny Plateau.
Key words: goldenseal, Hydrastis canadensis, wild harvested plants, herbal plants, herbarium vouchers, local extinction, rarity, white-tailed deer.
Goldenseal, Hydrastis canadensis L. (Ranunculaceae), is a slow-growing perennial herb that is harvested for the medicinal properties of its rhizome (Foster 1991). Its historical range extends from Ontario south to Alabama and west to Kansas (Davis 1999). It still occurs in patches of moist soils in deciduous forests throughout its historical range, but the core range now appears to consist of Ohio, Indiana, Kentucky and West Virginia (Sinclair and Catling 2000a).
Populations of goldenseal have been dramatically reduced as a result of collection for medicinal use and deforestation during and since the mid 1800s (Lloyd and Lloyd 1884-5; Gagnon 1999). It is listed as rare in 7 states, threatened in 3, and "of concern" in 2. Furthermore, goldenseal is ranked "vulnerable" in 5 states, "imperiled" in 8 states and "critically imperiled" in 5 states by Natural Heritage Programs. However, even in areas where goldenseal is more prevalent, serious reductions of populations have been reported and attributed to overharvesting and deforestation (Davis 1999). These two impacts, as well as agricultural expansion, road intrusion, urbanization and recreational use, were listed as reasons why goldenseal is increasingly difficult to find in forests where plants were formerly abundant (Liebmann et al. 1998). Sinclair and Catling (2000a) proposed that the current distribution in Canada, small isolated patches, might be a result of loss of disturbance (such as flooding and fire) that benefits goldenseal, and extinction or extirpation of seed dispersers. Although many populations are small, inbreeding is probably not a cause of decline, as goldenseal reproduces vegetatively as well as sexually, and is self-compatible (Sinclair et al. 2000, Sanders and McGraw 2003, Christensen and Gorchov unpubl.).
Goldenseal was listed in 1997 on Appendix II of the Convention for International Trade on Endangered Species (CITES). The CITES program requires exporters of goldenseal rhizomes (but not finished products) to obtain permits, and encourages dealers and diggers to monitor populations and harvest sustainably (Robbins 2000). This listing recognizes that international trade may have a significant impact on wild goldenseal if the plant is not adequately managed (liebmann et al. 1998).
Much of the information and government decisions concerning goldenseal are based on anecdotal rather than quantitative information on population status (Robbins 2000). Without organized or reported monitoring of historic wild populations of goldenseal in North America, it is difficult to quantify the extent of potential extirpation (Bannerman 1997). A recent survey of goldenseal in the central Appalachian region (consisting of populations mostly from West Virginia, with a few from Kentucky, Maryland, Ohio and Pennsylvania) encountered goldenseal infrequently, and found no particular elevation, aspect, land use or vegetation parameters that significantly affected presence or density of goldenseal (McGraw et al. 2003).
Although Ohio comprises a substantial part of the core of goldenseal's range, no comprehensive assessment has been made of the status of goldenseal populations in the state, where the Natural Heritage Program ranks it only as "reported." Only one site in Ohio, Wayne National Forest (WNF), was included in McGraw et al.'s (2003) survey. Harvesting occurs on private land and in WNF, where a US$10 permit allows collection of 5 lb dry weight (Erin Larson, Forest Botanist, WNF, pers.comm. 5/11/03), which is equivalent to an estimated 5000 rhizomes (Christensen and Gorchov unpublished). Although collecting is illegal on all other public lands, some land managers report cases of poaching, a practice which is difficult to eliminate (Mulligan 2003). However, even though populations are not eliminated in a single harvest (Van der Voort et al. 2003), populations have been projected to go extinct if only 10% of the plants are harvested annually (Christensen and Gorchov 2002).
The purpose of this study was to establish the degree of population loss of goldenseal in Ohio and to examine some of the possible reasons for its decline.
Materials and Methods. SITE SELECTION AND LOCATION. The status of goldenseal in Ohio was assessed by determining what proportion of sites that contained goldenseal populations historically still support populations today. Loans of goldenseal collections were requested from all open herbaria in Ohio and received from BGSU, BHO, CINC, CLM, JHWU, KE, MU and OS. A total of 269 voucher specimens were received and examined for locality information. Records that gave specific locations (
These sites were located using topographic maps and plat books.Landowners, land managers, or other knowledgeable individuals were then questioned by phone regarding site history and whether the site was still forested. If a site was determined to have been cleared or developed after the date of the voucher specimen, that population was scored extinct due to deforestation. For stands that were forested but currently lacked goldenseal, we evaluated whether the stand had been cleared subsequent to the year goldenseal was recorded by interviewing land managers, and where necessary, examining aerial photographs from each 10-year interval subsequent to the collection date.
To examine trends by region, sites were stratified by Ecoregion, using U.S. Environmental Protection Agency's Level III Ecoregions (Woods et al. 1998). A map of sites by Ecoregion was generated using ArcMap. Sites were also classified by ownership as private and public sectors. Privately owned sites were classified as corporate or private landowner. Public sites were classified as national, state, state-university, county, and municipality.
SAMPLING. Forested sites were surveyed for goldenseal between June 3rd and July 24th, 2002. Each site was searched by two people, allowing one hour per hectare up to a maximum of 4 hours per site. We traversed the area in parallel lines approximately 2-5 m apart. If goldenseal was found within 0.5 km of the original site description, the population was scored as extant. The number of goldenseal ramets with 1 leaf (non-reproductive), and 2-3 leaves (reproductive) were counted, following Gagnon's (1999) protocol for monitoring goldenseal populations. Counts were terminated at 200, in which case the total number of plants in the population was estimated. In sites that contained obviously over 200 plants, counts were not initiated, but the total was estimated. Observations were made on the structure of each stand, and in most stands the dominant species of herb, shrub, sub-canopy and canopy layers were recorded.
Herb-layer dominants were used to assess which forested sites had experienced excessive herbivory by white-tailed deer, enabling us to make inferences about the role of deer on goldenseal extinction. The abundance of deer in Ohio has increased nearly exponentially, from near zero in the 1940s to over 500,000 in 1996 (Iverson and Iverson 1999), and goldenseal is one of the most heavily browsed species of forest herbs at high deer densities (Frankland 2000; Frankland and Nelson 2001). We compared the dominant herbaceous species to records of percent grazing for 55 herb species in an Illinois forest (Frankland 2000). We considered the 10 species with the highest percent grazing in Frankland's (2000) study to be "preferred" by deer, and the 24 species with 0% grazing to be "avoided" by deer.
