703 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 22001133 SOUTSoHuEthAeSaTsEteRrnN NNaAtTurUaRliAstLIST 1V2o(4l.) :1720,3 N–7o2. 24 All Taxa Biodiversity Inventory Survey of Select Soil and Plant Ecological Parameters Associated with Rhododendron Decline in the Great Smoky Mountains and Surrounding Areas Richard Baird1,*, Alicia Wood-Jones1, Jac Varco2, Clarence Watson3, William Starrett1, Glenn Taylor4, and Kristine Johnson4 Abstract - As part of the All Taxa Biodiversity Inventory (ATBI) of Great Smoky Mountains National Park (GRSM), select site parameters associated within upland sites were measured within variable sized pockets of stressed and dead Rhododendron maximum (Great Rhododendron). During the last 20 years, Great Rhododendron, an important shrub in the southern Appalachian Mountains, has been dying in small to larger areas from an unknown cause. With increased visibility of dieback, the study (2006–2009) was conducted at 10 sites in GRSM and 1 in Nantahala National Forest (NNF). Eleven nematode species were identified, with no specific trends across locations or by plot treatments in frequencies of occurrences. Exceptions were Criconemella xenoplax, which was found at all 11 locations and was generally significantly greater in healthy or control than dieback plots, whereas Helicotylenchus sp. was more frequent in dieback plots. Meloidogyne spp., known parasitic nematodes of woody plants and agricultural crops, occurred in over 50% of the locations; Hoploliamus sp. occurred at 40%; and Belomolaimus sp., which are very destructive to root systems, were found at low levels at the NNF site. Also, Heterodera sp. occurred in control and dieback plots in 10 of the locations. Dieback ratings were not significantly correlated to stem diameter; although the finding was not statistically significant, six of seven sites with dieback plots had numerically greater-sized stems than in the controls. No other parameters such as number of clonal units, site aspect, percent slope, or elevation showed any trends at both sites. Nutrient data did not indicate any specific relationships to plot damage or health. This study provides the first comprehensive reporting of nematode species that occur in Great Rhododendron and associated riparian and upland sites. With concerns of global climate impacts, this research provided additional baseline data for the ATBI of GRSM and NNF. Introduction Over the last 15 years, the All Taxa Biodiversity Inventory (ATBI) conducted by Discover Life in America (DLIA) has provided pertinent data to the Great Smoky Mountains National Park (GRSM) personnel for management considerations and monitoring of forest health problems (Baird et al. 2007, 2009). Baseline data of fungal populations have been determined within select forest tree species, such as Tsuga canadensis (L.) Carr. (Eastern Hemlock), that are being impacted by exotic pests. With the uncertainty of a causal agent(s) of an emerging rhododendron 1Department of BCH-EPP, 2Plant and Soil Sciences, Mississippi State University, Mississippi State, MS 39762. 3Arkansas Agricultural Experiment Station, AFLS 212, University of Arkansas, Fayetteville, AR 72701. 4Vegetation Unit, Great Smoky Mountains National Park, 107 Park Headquarters, Gatlinburg, TN 37738. *Corresponding author - Richard Baird, rbaird@plantpath.msstate.edu. R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 704 decline in the GRSM and surrounding national forests, a survey on potential parasitic organisms and site features within the affected upland areas may provide insight as to the cause of the problem. Since no known microbes have been identified as causal agents in preliminary unpublished studies (Kelly Ivor, North Carolina State University, Raleigh, NC, pers. comm.; Steve Oak, USDA Forest Service, Asheville, NC, pers. comm.), parasitic nematode species might be involved, but little to no data exists to confirm this hypothesis. In terms of forest ecology and management, Rhododendron maximum L.(Great Rhododendron) is an important subcanopy shrub to small tree in the GRSM and is considered a keystone species for the Appalachian forests (Nilsen et al. 1999, Yeakley et al. 1994). Great Rhododendron has been shown to play a role in soil horizon and profile formation due to release of degradation products from its slowly decomposing recalcitrant litter (Boettcher and Kalisz 1990). At the turn of the century and early in the history of the GRSM, many forest disturbances occurred, such as loss of the Castanea dentata (Marsh.) Borkh. (American Chestnut), creating successional changes that had an impact on forest conditions (Rivers et al. 1999) and allowing Great Rhododendron to extend its influences well beyond the niche in which it originally occupied. In the late 1980s, areas of dead Great Rhododendron were reported at upland sites in the GRSM, and loss of this species could endanger other organisms (Ownley 1993). Rhododendron spp. habitats are of importance for numerous organisms including a rare orchid, Listera smallii Weigand (Appalachian Twayblade; Radford et al. 1968, Robinette 1974, Vandermast and Van Lear 2002). Research and study about the ecological importance of Great Rhododendron and its possible competitive inhibition effects has been conducted through a number of experiments, further attesting to the major role Great Rhododendron has in the southern Appalachians (Baker and Van Lear 1998, Clinton 2003, Clinton et al. 1994, McGee and Smith 1967, Monk et al. 1985, Nilsen et al. 1999, Smith 1963, Yeakley et al. 1994). Dieback, decline, or loss of this Ericaceae member is of concern. Over the last 25 years, Park Service personnel observed Great Rhododendron plants dying or dead throughout the park, and damaged areas in 14 stands were large enough to elicit concern over the survival of Great Rhododendron within the park (G. Taylor, pers. observ.). Rhododendron dieback was first studied in the early 1990s, with evaluations of aboveground parameters at Cerulean Knob area of GRSM (Ownley 1993, 1994). That study attempted to isolate causal agent(s) from aboveground plant tissues, but the report was negative for any pathogens. The report addressed the need for continued soil and root analysis before any further understanding of the problem could be determined. Following the initial study, Great Rhododendron dieback continued to be a problem throughout GRSM and in other diverse areas such as the Nantahala National Forest (NNF), but its causal factor(s) remained undetermined. The National Park Service (NPS) reported in 1999 and 2000 that the problem possibly was attributed to a fungal agent (e.g., Botryosphaeria sp.) (NPS 2004a, b). Based on his previous study, Ownley (1994) observed that in most cases the entire plant was 705 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 stressed or in decline with no known causes identified. This problem could be better defined as a decline in which a gradual reduction of species health is observed in the aboveground characteristics of the plants (Holliday 2001). Species of rhododendrons are affected with many stressors, and most have only been studied from a horticultural perspective. Abiotic stresses such as mineral deficiencies or excesses, unnatural pH levels, exaggerated environmental conditions, and air pollution can enhance plant stress, especially on marginal forest sites. Similarly, bacteria, viruses, nematodes, algae, and fungi, can be causal pathogenic agents, while others are incidental (Coyier and Roane 1988). These stresses, both abiotic and biotic, interact with the host plant, affecting possible disease resistance. Mineral nutrient excesses and deficiencies can impact survivability of Great Rhododendron (Fink 1999). Stunted growth, chlorosis, discoloration, leaf spots, leaf loss, and dieback can all be observed due to the levels of specific macro- (N, P, K, Ca, Mg) and micro- (Fe, B, Mn, Cu, Zn, Mo, Cl) nutrients. In some cases, secondary effects from biotic factors, such as root diseases and damage to mycorrhizae, reduce the nutrient uptake ability of the host (Fink 1999). Reduced nutrients have also been attributed to the cause of biotic infections by allowing parasites to attack stressed hosts (Agrios 2005). In young rhododendron leaves, a decrease in the facultative concentration of nitrogen may aid in reducing the incidence of Phytophthora dieback caused by an organism belonging to Chromalveolata (= Stramenopila) (Benson and Hoitink 1988). Calcium imbalances affect the meristems and juvenile tissues, which can result in distortions and decay of the shoots, stems, roots, flowers, and fruits (Agrios 2005, Fink 1999). Potassium regulates plant turgor and viability, while magnesium is essential for photosynthesis and, if deficient, may alter the root-to-shoot ratio by a reduction of root mass (Fink 1999). Magnesium deficiency has also been linked with forest decline (Fink 1999). Therefore, it is important to evaluate soil nutrients to ensure a viable diagnosis based on possible abiotic causes. The occurrence of air pollution and its impact on plant physiology and induced disease has been observed for over a century (Agrios 2005). Rain becomes acidified, changing the pH of soils, and certain nutrients can be affected when too much nitrogen is deposited into the soil. Soil pH level is important for rhododendrons, with the best growth occurring in a more acidic soil. An increase in pH towards a more basic soil may allow for the establishment of Phytophthora and the fungus Phymatotrichum spp. (Benson and Hoitink 1988). Although species of Rhododendron are usually very tolerant to pH changes, extremes of these conditions can impinge on their health, causing the plants to become susceptible to other diseases (Agrios 2005, Sinclair et al. 1993). It has been reported that Great Rhododendron cannot assimilate iron in an alkaline soil, which ultimately leads to chlorosis and loss of vital functions; thus, the plants would need an acidic soil with a pH between 4.5 and 5.0 (Bowers 1960). If plenty of organic material is available in the soil, a pH of 5.5 may be suitable. Rhododendron dieback or decline is becoming widespread in the GRSM and surrounding areas. The problem is becoming more evident, with an increasing number of reported sightings. It is unknown to what extent the impact of the loss of rhododendron outside the riparian zones will be since all dieback sites are primarily R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 706 located in upland areas, and no data was available on replacement species. In addition, rhizosphere data on upland sites within GRSM and NNF on associated root organisms were provided as All Taxa Biodiversity Inventory (ATBI) baseline information. Thus, a holistic research approach was conducted to determine if select abiotic and or biotic factors are associated with this problem. The purpose of this study was to evaluate dieback areas to 1) define and further characterize rhododendron dieback symptoms, 2) evaluate site factors including aspect, slope, pH, and soil fertility, 3) quantify nematode diversity/densities for comparisons with disease ratings and to contribute the resulting species lists to the ATBI, and 4) collect rhododendron plant measurements and growth data to determine plant growth variability between dieback and control plots. Materials and Methods Rhododendron decline site evaluations We studied ten areas of rhododendron dieback within GRSM and one in NNF (Fig. 1); all were surveyed between May and August 2006, and we took soil samples Figure 1. Locations of ten Rhododendron maximum (Great Rhododendron) dieback sites in Great Smoky Mountains National Park (1 = Cerulean Knob, 2=Russell Field, 3 = Laurel Falls, 4 = Huskey Gap, 5 = Nolan Divide, 6 = Deep Low Gap, 7 = Brushy Mt., 8 = Greenbrier Pinnacle, 9 = Newton Bald, 10 = Gabe Mt.) and one site in NantahalaNational Forest (11 = Albert Mountain). 707 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 in July and August 2007. Mapped soil series and textures, and dominant forest types are listed for each survey area (Table 1). All soils ranged from shallow to very deep and all were classified as well drained, except Craggey, which was somewhat excessively drained. Textures ranged from loam to sandy loam to sandy clay loam. All soils formed from colluvium, except the Santeetlah and Spivey series, which formed in place from residuum parent material. The forest ecotypes ranged from cove hardwoods at three locations to a Montane cove type. Other sites were mixed or oak forests and occurred in upland sites. Albert Mountain had the most visible rhododendron dieback damage and was located in an upland oak forest type. At each of the 11 rhododendron dieback sites, an attempt was made to establish identical-sized control plots for comparison. Within each site, there were two treatments, dieback and control, with four subplots within each treatment. Control plots were defined as healthy areas that contained no visible dieback. Only seven of the locations contained sufficient areas of healthy plants for control plots to be established near or adjacent to the damaged areas at similar elevations. At each site, we established one dieback and one control plot, each consisting of a 100-m × 100-m area. To collect rhododendron health data, we used a line-transect method to observe and tally the incidence of dieback in each plot per location. From the center of the plot, we established four 57.7-m transects starting 3 m from the midpoint and ending 10 m from each corner (O’Neill et al. 2005). We tallied all plants located Table 1. Soil and forest types of the rhododendron dieback (DB) and control (CDB) study sites used during general survey of site parameters associated with their decline in Great Smoky Mountains National Park and Nantahala National Forest. (See Fig. 1 for map of site locations). Site Soil type Forest type Albert Mt. (DBAM, CDBAM) Burton-Craggey sandy loam Upland oak type (Nantahala N. F.) Gabe Mt. (DBGM, CDBGM) Ditney Unicoli loam, gravelly loam Cove hardwood type Nolan Divide (DBND, CDBND) Oconaluftee-Guyot-Cataloochee Northern hardwood complex; channery loam, loam type Cerulean Knob (DBCK) Ditney Unicoi-loam, gravelly loam Mixed hardwood, acid type Russell Field (DBRF, CDBRF) Breakneck Pullback channery loam, Mesic oak/ sandy loam hardwood type Laurel Falls (DBLF, CDBLF) Snowbird loam-sandy clay loam Montane cove hardwood type Greenbrier Pinnacle (DBGP, Breakneck Pullback channery Cove hardwood type CDBGP) loan, sandy loan Husky Gap (DBHG) Soco-Stecoah channery loam, sandy loam Submesic oak/ hardwood type Newton Bald (DBNB, CDBNB) Soco-Stecoah channery loam, sandy loam Submesic oak/ hardwood type Deep Low Gap (DBDLG) Soco-Stecoah channery loam, sandy loam Mixed hardwood, acid type Brushy Mt. (DBBM) Spivey-Santeetlah loam, boulder sandy Cove hardwood type loam R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 708 within 2 m on either side of the transect. Each Great Rhododendron plant >3 cm diameter at breast height (dbh) at 50 cm above the ground was evaluated as per Ownley (1994). A qualitative ranking for each plant was based on leaf whorls per branch (WH), percentage of leaves with chlorosis (CH), and percentage of dead twigs and branches (TW) (Table 2). We then analyzed these values using the rhododendron dieback rating formula (Ownley 1994): R_Calc = (0.85 / 3)(WH) + (0.85 / 3)(CH) + (0.85 / 3)(TW) Ratings (R_Calc) can range from 1 to 7, with 1 defined as a heal thy plant and 6 or 7 as dead before or after leaf abscission, respectively. We used the R_Calc value to determine the severity of dieback for each of the 11 sites including control plots. We also measured dbh for each plant along each transect. General site evaluations Within each site, there were two treatments, dieback and control, with four subplots within each treatment. In addition to the rating system of the plants, a general site evaluation was conducted from May through August 2007 on each of the 11 locations, including aspect, slope, soil pH, soil nutrients, and nematode levels. We measured and obtained site factors including aspect and slope in each of four 16 × 16 m subplots nested within each 100-m × 100-m plot using sampling methods previously described (O’Neill et al. 2005). At the center of the plot, one subplot was surrounded by three subplots offset from each other and oriented at 120° angles at a distance of 37 m from the centers. Refer to O’Neill et al. (2005) for a plot diagram. Within each large dieback and control subplot, we calculated slope using the percent scale of a Suunto clinometer (Suunto, Carlsbad, CA). Measurements were done 5 m uphill and 5 m downhill from each subplot center. We averaged percent slope measurements to obtain slope in each subplot. We measured litter depth and depth to bedrock within each of the four subplots of all plots established for the study. Inserting a 1.0-m steel rod into the soil to the Table 2. Three criteria used to estimate rhododendron dieback, and the values assigned to different categories. Criteria Value Criteria Value Number of whorls (TH) % leaf chlorosis (CH) ≥4 whorls 1 0.0 % 1 3–4 whorls 3 0.0–10.0% 2 2–3 whorls 4 10.0–25% 3 1–2 whorls 5 >25% 5 Recent deadA 6 DeadB 7 Proportion of dead twigs and branches (TW) <5% 1 5–20.0% 4 >20.0% 5 ARecent dead refers to a dead plant prior to leaf abscission. BDead plants with no leaves attached. 709 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 top of the mineral soil layer (A horizon), we measured litter depths from the length of the steel rod’s penetration. We measured aspect using a Silva Ranger 515. We collected soil samples for soil-testing analysis and nematode surveys in July and August 2007. Prior to soil sampling, we removed forest litter (O horizon) to bare soil (A horizon) from each of the sampling points within each general site factor subplot. We collected a total of eight soil samples from each subplot per location (four each for soil testing analysis and nematode detection). To obtain each sample, we randomly collected nine sub-samples (15 to 20 cm depth starting at O horizon) per subplot, pooled them into a 19-L polypropylene bucket, and thoroughly mixed them to ensure a homogenized sample. A volume of 500 ml was collected from the homogenized samples, placed in a 946 ml Ziploc® plastic bag, and stored until submission to the Extension Soil Testing Laboratory, Mississippi State University (MSU) for soil test analysis (Mississippi State University 2004, 2005). Soils were stored in a cooled ice chest (0 to 10 °C), transferred to refrigeration (4 °C), and prepared for overnight shipping to the respective diagnostic laboratory units at MSU. Note that we disinfected sampling tools and bucket using 0.525% (w/v) aqueous NaOCl between each sampling of the four to eight subplots per location. We placed an additional homogenized sample of 500 ml from each subplot into a nematode soil sample bag obtained from the Disease Diagnostic Laboratory, Entomology and Plant Pathology Department, MSU. Methods for nematode extraction follow those by Baker (1978). In addition, North Carolina-style semi-automatic elutriator and sugar centrifugation was performed to isolate the juvenile, vermiform, and cyst stages of nematodes from the soil (C. Balbalian, Diagnotic Lab., MSU, pers comm.). Number of cysts were determined by placing 240 g of soil onto a Baermann funnel (Baermann funnel method) for five days, and numbers of larvae were counted after hatching (Baker 1978). Routine soil test analysis included pH (1:2 soil:d.i. H2O); soil organic matter; extractable P, Ca, Mg, K, and Na; percent organic material (OM); and effective cation exchange capacity (ECEC; sum of cations extracted) (Cox 2001). Statistical analysis We analyzed data as a series of combined experiments (combined across sites) using the GLM procedure of SAS (SAS Institute, Cary, NC), and separated means using Fisher’s protected least significant difference (LSD). Data were pooled and subjected to stepwise multiple regression analysis to evaluate the effect of the measured variables on dieback rating (R_Calc). We pooled all data and conducted regression analysis to verify and/or modify the existing rhododendron dieback rating formula. Please note that since 7 of the 11 sites had control plots due to availability, some analyses were conducted between plots from just those locations. Results We identified eleven species of nematodes from soil samples collected during the general survey (Table 3). In addition, we recorded cysts of unknown species, but those numbers were low and not common across locations. Of the R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 710 Table 3. Percent frequenciesA (and total numbers) of nematode speciesB present in dieback and control Rhododendron maximum (Great Rhododendron) plots at ten locations during a general survey of site parameters in the southern Appalachian Mountains. M sp. = Meloidogyne sp., P. sp. = Pratylenchus sp., He. sp. = Helicotylenchus sp., P. m. = Paratrichodorus minor, X. a. = Xiphinema americanum, C. x. = Criconemella xenoplax, Ho. sp. = Hoplolaimus sp., Het. sp.= Heterodera sp., T. sp. = Tylenchorhynchus sp., R. r. = Rotylenchulus reniformis, B. sp. = Belonolaimus sp., and Others = unidentified cysts and larvae. Location M. sp. P. sp. He. sp. P. m. X. a. C. x. Ho. sp. Het. sp. T. sp. R. r. B. sp. Others Albert Mt. Dieback 1.2 (79) 3.5 (223) 7.8 (505) <1.0 ( 32) 0.0 (0) 20.8 (1343) 0.0 (0) 0.0 (0) 0.0 (0) <1.0 (40) 0.0 (0) 0.0 (0) Control 0.0 (0) 3.2 (212) 2.5 (166) <1.0 (6) 0.0 (0) 60.2 (3887) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) <1.0 (21) 0.0 (0) Brushy Mt. Dieback 0.0 (0) 0.0 (0) 38.0 (434) 2.1 (24) 0.0 (0) 32.9 (371) 0.0 (0) 25.8 (292) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Cerulean Knob Dieback 0.0 (0) 6.4 (112) 35.4 (615) 0.0 (0) 0.0 (0) 40.4 (702) 5.5 (95) 12.3 (214) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Deep Low Gap Dieback 85.0 (141) 0.0 (0) 12.4 (206) <1.0 (8) <1.0 (20) 60.0 (993) 0.0 (0) 18.1 (299) 0.0 (0) 0.0 (0) 0.0 (0) <1.0 (20) Gabe Mt. Dieback <1.0 (13) 0.0 (0) 18.6 (228) 0.0 (0) 0.0 (0) 12.5 (153) 5.8 (71) 8.6 (105) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Control 0.0 (0) 0.0 (0) 3.9 (48) 0.0 (0) 0.0 (0) 19.9 (245) 0.0 (0) 29.6 (363) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Greenbrier Dieback 0.0 (0) 0.0 (0) 26.4 (907) 0.0 (0) 0.0 (0) 1.9 (410) 0.0 (0) <1.0 (8) <1.0 (8) 0.0 (0) 0.0 (0) 0.0 (0) Control 2.5 (86) 0.0 (0) 31.2 (1071) <1.0 (6) 0.0 (0) 24.3 (835) 0.0 (0) 2.5 (87) <1.0 (16) 0.0 (0) 0.0 (0) 0.0 (0) Husky Gap Dieback 0.0 (0) 0.0 (0) 52.3 (631) 0.0 (0) 0.0 (0) 33.4 (403) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) 13.0 (157) 711 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 Table 3, continued. Location M. sp. P. sp. He. sp. P. m. X. a. C. x. Ho. sp. Het. sp. T. sp. R. r. B. sp. Others Laurel Falls Dieback 0.0 (0) 0.0 (0) 6.8 (237) 0.0 (0) 0.0 (0) 10.1 (412) 5.6 (62) 6.0 (84) 0.0 (0) <1.0 (12) 0.0 (0) 0.0 (0) Control 3.2 (112) 0.0 (0) 48.6 (502) 0.0 (0) 0.0 (0) 15.6 (525) 3.0 (43) <1.0 (29) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Newton Bald Dieback 1.8 (64) 0.0 (0) 39.6 (1426) <1.0 (24) <1.0 (24) 12.9 (464) <1.0 (8) 9.2 (332) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Control 0.0 (0) 0.0 (0) 7.4 (268) 0.0 (0) 0.0 (0) 29.1 (1048) 0.0 (0) <1.0 (24) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Nolan Divide Dieback 0.0 (0) 0.0 (0) 23.6 (892) 0.0 (0) 0.0 (0) 14.2 (539) 0.0 (0) 2.3 (89) 0.0 (0) 0.0 (0) 0.0 (0) 9.8 (370) Control 0.0 (0) 0.0 (0) 10.2 (384) 0.0 (0) <1.0 (8) 33.2 (1253) 0.0 (0) <1.0 (8) 0.0 (0) 0.0 (0) 0.0 (0) 6.2 (236) Russell Field Dieback 0.0 (0) 0.0 (0) 41.3 (898) 0.0 (0) 0.0 (0) 9.8 (214) 2.9 (64) <1.0 (8) <1.0 (8) 0.0 (0) 0.0 (0) 0.0 (0) Control 1.7 (39) 0.0 (0) 22.8 (497) 0.0 (0) 0.0 (0) 9.8 (213) <1.0 (8) 2.5 (87) <1.0 16) 0.0 (0) <1.0 (16) 0.0 (0) A Percent (%) frequencies are based on mean total number of nematodes calculated across dieback and control Great Rhododendron replicate subplot samples for each location. BNine soil core samples were pooled from each replicate of four subplot within each large area plots (100 m × 100 m) per location. R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 712 total species identified, Criconemella xenoplax Raski (ring), Helicotylenchus sp. (spiral), and Heterodera sp. (sheath) were found at almost all locations. In addition, Meloidogyne sp. (root knot) occurred at six of the locations during the survey. For C. xenoplax, population levels were always greater in the control than dieback subplots. Helicotylenchus sp. levels were generally greater in the dieback than control areas, whereas levels varied for Heterodera and Meliodogyne sp. Percent frequencies of Meloidogyne sp. were also greater in dieback subplots (Table 3). We compared mean numbers of nematode species among sites and within dieback and control plots (Table 4). Heterodera sp. had significantly greater mean population levels from soils collected in the dieback subplots and were almost twice the mean totals, whereas C. xenoplax had consistently greater numbers in control plots. For each nematode species, we compared mean totals across sites and within the dieback and control plots (Table 5). Similar results between treatments were noted as observed for data located in Table 4. We rated dieback and control subplots for plant diameter, dieback ratings, and number of clonal units. All seven survey sites containing dieback and control subplots had significantly greater dieback ratings in the dieback compared to control subplots (Table 6). Albert Mountain had dieback ratings of 6.3 in the dieback subplots compared to 2.3 in the adjacent control subplots. Only at the Laurel Fall site were diameter measurements in the dieback subplots significantly greater (7.4) than in control sites (5.2). At Russel Field, mean number of clonal units in the dieback subplots were significantly greater at Russell Field than contro l areas. Table 4. Comparison of mean number for each nematode speciesA (balanced mean data) compared between dieback and control Rhododerndron maximum (Great Rhododendron) subplots at six locations within Great Smoky Mountains National Park and one in Nanatahala National Forest. LSD = Fisher’s protected least significant difference (P < 0.05). Nematode species/type Dieback Control LSD Meloidogyne root knot 2.6 AB 6.5 A 6.4 Pratylenchus lesion 4.0 A 3.9 A 3.9 Helicotylenchus spiral 98.6 A 87.8 A 36.5 Paratrichodorus minor stubby root 1.1 A 0.0.4 A 1.3 Xiphinema americanum dagger 0.1 A 0.0.0 A 0.2 Criconemella xenoplax ring 88.2 B 169.5 A 54.2 Hoplolaimus lance 7.7 A 3.3 A 5.1 Heterodera sheath 21.1 A 10.0.5 B 9.8 Tylenchorhynchus stunt 0.1 A 0.3 A 0.2 Rotylenchulus reniformis reniform 0.0 A 0.0 A 0.0 Belonolaimus sting 0.0.A 0.3 A 0.4 Cysts spp.C (other) 6.2 A 4.3 A 2.1 AAt the seven locations, each treatment (dieback and control) had nine randomly collected soil samples from four subplots located within the 100-m × 100-m plots sampled in 2007. BTreatment means followed by the same letter within nematode species are significantly different (P < 0.05) according to the LSD. CTotal cysts were counted and then placed onto a Baermann funnel (Baermann funnel method) for five days, then the number of hatched larvae were counted, and these larvae believed were believed to be Heterodera spp. 713 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 Table 5. Comparison of mean number nematode species between dieback and healthy Rhodendron maximum sites at 10 locations within Great Smoky Mountains and 1 in Nantahala National Forest. Me. = Meloidogyne root knot, Pr. = Pratylenchus lesion, Hel. = Helicotylenchus spiral, P. m. = Paratrichodorus minor stubby root, X. a. = Xiphinema americanum dagger, C. x. = Criconemella xenoplax ring, Ho. = Hoplolaimus lance, Het. = Heterodera sheath, Ty. = Tylenchorhynch stunt, R. r. = Rotylenchulus reniformis, Be. = Belonolaimu sting, and Others = Cysts spp.C. Nematode species (balanced mean data) LocationsA Me. Pr. Hel. P. m. X. a. C. x. Ho. Het. Ty. R. r. Be. Others Albert Mt. Dieback 4.0 BCB 12.0 B 25.3 D–F 1.6 AB 0.0 B 67.2 CD 0.0 C 0.0 C 0.0 C 0.0 C 0.0 C 0.0 D Control 0.0 C 11.8 B 8.3 E 0.0.4 B 0.0 B 194.4 A–D 0.0 C 0.0 C 0.0 C 0.0 C <1.0 C 0.0 D Brushy Mt. Dieback 0.0 C 0.0 B 10.08.5 C–E 6.0 A 0.0 B 66.8 B–D 0.0 C 73.0 A 0.0 C 0.0 C 0.0 C 2.0 D Caerulean Knob Dieback 0.0 A 28.0 A 153.8 B–D 0.0 B 0.0 B 175.5 A–D 23.8 A 53.5 AB 0.0 C 0.0 C 0.0 C 0.0 D Deep Low Gap Dieback 35.3 A 0.0 B 51.5 D–E 2.0 AB 2.0 A 248.3 A–C 0.0 C 74.8 A 0.0 C 0.0 0.0 B 0.0 D Gabe Mt. Dieback 3.3 BC 0.0 B 57.0 DE 0.0 B 0.0 B 38.3 D 18.3 AB 26.3 BC 0.0 C 0.0 0.0 B 0.0 D Control 0.0 C 0.0 B 12.0 E 0.0 B 0.0 B 61.3 D 0.0 C 90.0.8 A 0.0 C 0.0 0.0 B 0.0 D Greenbrier Dieback 0.0 C 0.0 B 226.8 A–C 2.0 AB 0.0 B 10.02.5 B–D 0.0 C 2.0 C 2.0 B 0.0 0.0 B 0.0 D Control 21.5 AB 0.0 B 267.8 A–C 2.0 AB 0.0 B 20.08.8 AD 0.0 C 21.5 BC 4.0 A 0.0 0.0 B 0.0 D Huskey Gap Dieback 0.0 C 0.0 B 157.8 B–D 4.0 AB 0.0 B 65.3 CD 0.0 C 6.0 C 0.0 C 0.0 0.0 B 39.3 C Laurel Falls Dieback 0.0 C 0.0 B 52.9 DE 0.0 B 0.0 B 10.03.4 B–D 16.3 A–C 24.5 BC 0.0 C 0.0 C 0.0 B 0.0 D Control 13.4 BC 0.0 B 131.7 B–E 0.0 B 0.0 B 134.4 A–D 9.6 A-C 7.5 C 0.0 C 0.0 C 0.0 B 0.0 D R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 714 Table 5, continued. Nematode species (balanced mean data) LocationsA Me. Pr. Hel. P. m. X. a. C. x. Ho. Het. Ty. R. r. Be. Others Newton Bald Dieback 16.0 A–C 0.0 B 356.5 A 6.0 A 2.0 A 116.0 B–D 0.0 C 83.0 A 0.0 C 0.0 C 0.0 B 0.0 D Control 0.0 C 0.0 B 67.0 DE 0.0 B 0.0 B 262.0 AB 0.0 C 6.0 C 0.0 C 0.0 C 0.0 B 0.0 D Nolan Divide Dieback 0.0 C 0.0 B 223.0 A–C 0.0 B 0.0 B 159.8 A–D 0.0 C 21.8 BC 0.0 C 0.0 C 0.0 B 59.0 B Control 0.0 C 0.0 B 146.2 B–E 2.0 AB 0.0 B 313.3 A 0.0 C 2.0 C 0.0 C 0.0 C 0.0 B 92.5 A Russell Field Dieback 9.8 BC 0.0 B 224.5 A–C 0.0 B 0.0 B 53.5 D 16.0 A–C 60.0.6 AB 0.0 C 0.0 C 0.0 B 0.0 D Control 0.0 C 0.0 B 124.3 C–E 0.0 B 0.0 B 53.3 D 2.0 BC 0.0 C 0.0 C 0.0 C 4.0 A 6.0 D LSD 21 12.9 14 4.9 1.1 184.2 17.8 42.7 1.8 0.0 1.6 18.6 A Locations consisted of 100- × 100-m plots subdivided into four subplots where nine core samples were randomly taken per subplot and pooled in 2007. BTreatment means followed by the same letter within nematode spec ies are significantly different (P < 0.5) according to the LSD. CTotal cysts were counted and then placed onto a Baermann funnel (Baermann funnel method) for five days, then counted the number of larvae that hatched. All hatched larve were then counted as Heterodera spp. 715 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 Significant differences in aspect, soil depth, and slope occurred, but no trends were observed relative to dieback and control plots (Table 7). Significant differences occurred within locations and treatments (i.e., dieback and control), but no consistent trends occurred. Aspect was generally uniform among sites except at Albert Mountain, where dieback was greater than at any other locations. At this location, mean slope was minimal at 13.6% and mean soil depth was second greatest at 111.8 cm. Elevation data between dieback and control plots were significantly different across the survey sites (Table 7). Differences were noted between the locations, with Albert Mountain having the highest elevations and dieback levels (data not shown). For the other sites, there was no relationship between dieback levels and elevations. Table 6. Mean numberA comparisons of select Rhododendron maximum (Great Rhododendron) site parameters at seven locations within dieback and control plots. Diameter Dieback R Number of Treatments (cm)B value ratingC clonesD AspectE Depth (cm) Slope Albert Mt. Dieback 8.4 AA 6.3 A 2.6 A 2.0 A 114.7 A 18.4 A Control 7.1 A 2.3 B 3.2 A 1.6 A 18.9 A 8.8 B LSD 2.92 0.72 0.91 0.78 16.12 5.0 Greenbrier Dieback 7.0 A 5.0 A 5.1 A 6.0 A 98.2 A 26.7 A Control 6.1 A 2.2 B 2.5 B 6.1 A 104.1 B 25.5 A LSD 5.0 3.75 0.71 0.00 35.40 2.44 Newton Bald Dieback 8.3 A 6.3 A 2.6 A 4.8 A 95.6 A 31.5 A Control 7.4 A 5.2 B 3.5 A 4.7 A 98.1 B 32.3 A LSD 1.85 0.90 1.40 1.26 4.61 3.71 Russell Field Dieback 5.6 A 4.7 A 4.8 A 5.3 A 95.4 A 16.7 A Control 3.7 A 1.5 B 1.5 B 4.0 B 122.0 A 11.1 B LSD 5.80 2.11 1.93 1.16 27.71 4.83 Gabe Mt. Dieback 8.0 AA 5.3 A 5.0 A 2.7 B 122.0 A 15.0 B Control 10.0.8 A 2.1 B 2.1 A 6.8 A 114.6 B 21.5 A LSD 3.20 1.30 1.40 1.25 17.10 2.64 Laurel Falls Dieback . 7.4 A 5.2 A 3.5 A 5.5 A 97.5 A 27.3 A Control 5.2 A 2.7 B 3.0 A 5.7 A 104.5 A 26.2 A LSD 1.22 0.94 1.35 2.45 15.99 7.47 Nolan Divide Dieback 7.1 A 2.7 A 3.2 A 5.1 A 93.2 A 19.9 A Control 5.7 A 2.3 A 3.0 A 3.5 B 10.06.6 A 10.0.5 A LSD 2.40 0.79 0.95 1.10 36.67 5.00 ATreatment means followed by the same letter within locations are significantly different (P < 0.05) according to the LSD. BDiameter of trees refers to measurements taken as dbh or breast height. CDieback ratings include ratings number of whorls (WH), percent leaf clorosis (CH; which included dead trees), and number of dead twigs on branches (TW). DNumbers of clones refer to mean number that occurred within each subplot. ENorth = 1, northwest = 2, northeast = 3, south = 4, southwest = 5, southeast = 6, east = 7, and west = 8. R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 716 Table 7. Comparison of mean locations parameters between dieback and or control Rhododendron maximum (Great Rhododendron) sites at 10 locations within Great Smoky Mountains National Park and 1 in Nantahala National Forest. Site factorsA Locations Aspect Soil depth (cm) Slope (%) Elevation (meters) Albert Mt. Dieback 2.0 HIB 114.7 AB 18.8 E–G 1541.4 A Control 1.6 I 108.9 A–C 8.8 H 1545.7 A Brushy Mt. Dieback 6.8 A 105.6 A–C 23.9 CD 1033.3 FG Caerulean Knob Dieback 5.4 B–D 103.4 A–C 20.9 DE 1029.6 FG Deep Low Gap Dieback 4.7 ED 122.0 A 11.2 H 1031.4 FG Gabe Mt. Dieback 2.7 GH 122.0 A 15.7 G 977.8 H Control 6.8 A 114.6 AB 21.5 DE 971.7 H Greenbrier Dieback 6.0 AB 98.2 BC 26.7 C 1059.5 F Control 6.0 AB 104.1 A–C 25.5 C 1015.0 G Huskey Gap Dieback 5.0 CD 103.0 A–C 30.3 AB 1191.8 D Laurel Falls Dieback 5.5 B–D 97.5 BC 27.3 BC 952.4 HI Control 5.8 BC 104.5 A–C 26.2 C 940.5 I Newton Bald Dieback 4.8 DE 94.3 C 31.5 A 1141.8 E Control 4.7 DE 98.6 BC 32.3 A 1048.8 FG Nolan Divide Dieback 5.1 CD 93.2 C 19.9 EF 1411.2 B Control 3.5 FG 106.6 A-C 10.5 EF 1421.8 B Russell Field Dieback 5.3 B–D 95.4 BC 16.7 FG 1152.1 E Control 4.0 EF 122.0 A 11.1 H 1315.8 C LSD 0.80 20.74 1.97 3.45 All locations Dieback 4.5 A 104.8 A 22.1 A Control 4.6. A 108.4 A 20.7 B LSD 0.34 6.37 1.35 ANorth = 1, northwest = 2, northeast = 3, south = 4, southwest = 5, southeast = 6, east = 7, and west = 8; depth was determined by replicate readings taken with each subplot per plot; slope is in percent and was determined within each subplot per plot. BTreatment means followed by the same letter within site factors are significantly different (P < 0.05) according to the LSD. 717 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 In an effort to investigate the possibility of soil factors that may be influencing the decline, multiple differences in soil test results were found, but most were influenced by location rather than being classified as a control or dieback site affect (Table 8). For example, soil test results for phosphorous varied from a low of 24 kg/ ha to a high of 92 kg/ha depending on the site. In terms of crop response, however, only three locations—DBGP, CDBGP, and DBRF—would be classified as “low” in available P and potentially responsive to applied P fertilizer. Variability in soil pH from a low of 3.78 to a high of 4.65 could potentially indicate a stress factor, as the lower end of the range is extremely acidic even for highly acid-tolerant species. The four sites with pH values of 4.05 or less were all “dieback” classified, otherwise there did not appear to be a pH trend effect on rhododendron health. Within-site comparisons showed that in four out of seven sites the subplot classified as dieback also had the lower pH for that particular site. Stepwise regression analysis for dependent variable R_calc (disease ratings) as affected by soil nutrient levels and nematode densities showed no clear associations (data not shown). The R2 values ranged from 0.0811 to 0.1252 (P ranged between <0.001 and 0.017) for models involving OM (organic matter), P, Na, and sting and root knot nematodes. Table 8. Selected soil test results within 11 dieback/control sites. P, Ca, Mg, K, and Na all given in kg/ha. ECEC = effective cation exchange capacity (given in centimoles of charge per kg soil). SOM = soil organic matter. Site IDA pHB SOM % P Ca Mg K Na ECEC DBAM 4.40 6.92 92 9 28 143 93 18 CDBAM 4.62 5.92 73 29 22 134 80 15 DBGM 4.05 5.58 40 9 13 121 80 20 CDBGM 4.15 5.24 47 9 14 151 81 19 DBND 4.30 6.31 62 9 30 221 73 18 CDBND 4.20 7.02 90 9 37 190 78 21 DBCK 4.03 5.73 68 37 21 181 95 21 DBRF 4.25 3.85 24 9 14 123 96 14 CDBRF 4.45 5.20 52 9 11 143 72 16 DBGP 3.90 5.24 31 9 18 128 74 18 CDBGP 4.25 3.94 28 9 9 104 94 12 DBNB 4.65 5.09 43 20 39 228 72 12 CDBNB 4.53 4.73 68 27 45 222 69 12 DBDLG 4.53 3.77 32 9 14 127 85 11 DBLF 4.21 6.57 67 20 22 157 65 20 CDBLF 4.10 8.86 69 53 30 160 65 25 DBBM 3.78 8.85 69 9 50 173 69 29 DBHG 4.15 6.17 59 12 29 209 64 19 LSD(0.05) 0.31 2.14 23 NS NS 42 NS 6 ARefer to Table 1 for actual names of each location. B1:2 soil:d.i.H2O. R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 718 Discussion Rhododendron dieback appears to be widespread throughout the GRSM and surrounding NNF in pockets of various sizes but never greater than two ha. Within the riparian zones, Great Rhododendron appears to be expanding, which could partially be due to alleopathic involvement and acidic soils excluding other species (Baker and Van Lear 1998). However, in the uplands areas where dieback is prevalent, Great Rhododendron is being replaced by other woody plant species. A broad diversity of nematode taxa were found on the upland sites; half were saprophytic and other plant pathogens from forest and agricultural ecosystems. Criconemella xeroplax is known to be a parasite of fruit trees such as Prunus and Juglans spp. (Dreistadt et al. 1994, Ruehle 1973). Also, C. xenoplax was reported to damage Ilex crenata Thunk. (Japanese Holly; Aycock et al. 1976), but direct association was never mentioned for rhododendron or any Ericaeae species. Criconemella xenoplax was reported to decrease tree species growth in association with microbes such as fungus or fungus-like taxa including Fusarium solani (Mart.) Sacc. and Pythium irregulare Buis. (Powell et al. 1968). The nematode Helicotylenchus sp. was common among the 11 dieback locations. It was previously reported that increased levels of Helicotylenchus dihystera (Cobb) Sher. and Phytophthora cinnamomi Rands were correlated with increased disease levels in control studies of nursery tree seedlings (Marx and Davey 1969). In a previous study in Europe, Helicotylenchus digonicus Perry, an ectoparasitic species, was commonly associated as a parasite with many hardwoods and a few conifer tree species (Stollarova 1999). In that study, only nematodes diversity and densities were determined and no microbial associates were evaluated. In a concurrent study from two dieback sites, microbes were sequenced and isolated from roots and soil (Baird et al., in press). However, none of the microbes listed above were found using sequence data or culturing, except the root pathogen Ilyonectria radicicola (Gerlach & L. Nilsson) Chaverri & C.G. Salgado, which occurred at Albert Mountain and Laurel Falls Trail sites during the survey (Baird et al., in press). This pathogen and select nematode species such as Xiphinema bakeri Williams are implicated in a disease complex of coniferous tree species (Ruehle 1973). Additional research will be necessary to confirm whether a disease complex between I. radicicola and a nematode species is responsible for rhododendron decline. Meloidogyne sp. occurred in the current study and is known to be parasitic on landscape or woody plants and forest trees (Dreistadt et al. 1994, Ruehle 1973). However, in this study, occurrence of the nematode was scattered and population levels were low and probably below the threshold for visible damage to occur. Even though at six of seven sites stem diameters were numerically greater in dieback plots than controls shown in Table 6, dieback levels were not significantly correlated with stem diameter (data not shown). Contrary to these results, rhododendron plant diameters were significantly greater in dieback plots in a study reported by Ownley (1994). A preliminary study conducted by USDA/Park Service in 1994 compared dbh of 30 rhododendron plants to core sample age. Plant diameter sizes did not show a significant trend to greater age (growth rings) of stems from the 719 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 Cerulean Knob dieback area (G. Taylor, unpubl. data). Based on this small sample size, this preliminary investigation showed a trend that larger diameter plants with greater disease levels might not be age related. Soil pH and other nutrient parameters appear to be factors needing further study. The observation from the analysis suggests that pH alone is not the determining factor, as other sites with more acidic pH did not behave in the same manner. Site characteristics and other soil chemical properties not measured here, such as exchangeable Al levels, may also play a role in dieback progression. Although extractable soil Ca2+ levels did not differ, the quantities are extremely low for plant growth. The impact of low levels of Ca2+ on dieback need further study as well as the possible effects of varying levels of soluble Al and Mn, which were not determined in this study. In conclusion, nematode species diversity and density data obtained from the 11 sites were used to determine their association to rhododendron stands for ATBI database information and to determine if any other site parameters might be involved in dieback. Those data were submitted to the USDA/Department of Interior, ATBI database being collected to catalog all organisms in GRSM. This research is the first comprehensive study of nematodes from various forest types within the park and southern Appalachian Mountains. Since several of the nematode species are known to be parasitic on woody plants, the data has further implications for forest health, especially when considered in conjuction with climate change parameters and nutrient depositions that affect rhizosphere organisms and subsequent tree growth. Concerning dieback, the site factors varied across locations overall, though without any real trends being noted for the 11 sites. Contrary to these results, nematode occurrence does indicate possible association with rhododendron decline, but additional research is needed to confirm these initial findings. Also, it is possible that a nematode-microbe disease complex may cause rhododendron decline within the Great Smoky Mountains and vicinity. Previous literature supports the hypothesis that certain nematode species identified in this study are parasitic on woody plants as discussed in the introduction. Furthermore a disease complex might be involved, especially since many of the areas where decline occurs (e.g., ridge tops) are not within areas of optimal growing condition for rhododendron. Another study that was conducted simultaneously during the current investigation included sampling for microbes (Baird et al., in press). Ilyonectria radicicola, a known pathogen of tree roots, was found at two dieback locations in the southern Appalachian Mountains. Also the low levels of Ca2+ might predispose rhododendron plants to attack by biotic pests. Because these data indicate certain trends, additional soil-root sampling for nematodes, microbial pathogens, and select nutrients is needed to confirm the previous results across many diverse locations and elevations. We recommend that greenhouse studies confirm the initial field observations. Acknowledgments Appreciation is extended to the Great Smoky Mountains Conservation Association for a Carlos C. Campbell Memorial Fellowship Award in 2007. In addition, financial R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 720 support of the project was provided by American Rhododendron Society in 2007. Also, a thank you is extended to Highlands Biological Station for financial support as Grants- In-Aid during 2007 and 2008. Logistical and technical support of the project provided by Department of Interior, Park Service, GRSM, is appreciated. We thank David Pratt, Botany Field Station, University of Tennessee, for providing the laboratory and housing facilities necessary to support the research. Finally, are grateful to Mississippi State University for use of laboratory facilities and supplies not covered by grants. This paper is MAFES publication number 12186. Literature Cited Agrios, G.N. 2005. Plant Pathology. 5th Edition. Elsevier Academic Press, Boston, MA. 922 pp. Aycock, R., K.R. Barker, and D.M. Benson. 1976. Susceptibility of Japanese Holly to Criconemoides xenoplax, Tylenchorhynchus claytoni, and certain other plant-parasitic nematodes. Journal of Nematology 8:26–31. Baird, R.E., C.E. Watson, and S. Woolfolk. 2007. Microfungi from bark of healthy and damaged American Beech, Fraser Fir, and Eastern Hemlock forests during an all taxa biodiversity inventory (ATBI) in Great Smoky Mountains National Park. Southeastern Naturalist 6:67–82. Baird, R.E., S. Woolfolk, and C.E. Watson. 2009. Microfungi of forest litter of healthy American Beech, Fraser Fir, and Eastern Hemlock stands from forests in Great Smoky Mountains National Park. Southeastern Naturalist 8:609–630. Baird, R., A. Wood-Jones, J. Varco, W. Starrett, G. Taylor, and K. Johnson. In press. Rhododendron decline in Great Smoky Mountains and surrounding areas: Intensive site study of biotic and abiotic parameters associated with the decline. Eastern Biologist (In press). Baker, K.R. 1978. Determining nematode population responses to control agents. Pp. 114–127, In E.I. Zehr (Ed.). Methods for Evaluating Plant Fungicides, Nematicides, and Bactericides. American Phytopathological Society, St. Paul, MN. Baker, T.T., and D.H. Van Lear. 1998. Relations between density of rhododendron thickets and diversity of riparian forests. Forest Ecology Management 109:21–32. Benson, D.M., and H.A.J. Hoitink.1988. Phytophthora dieback. Pp. 12–15, In D.L. Coyier and M.K. Roane (Eds.). Compendium of Rhododendron and Azalea Diseases. APS Press, St. Paul, MN. Boettcher, S.E., and P.J. Kalisz. 1990. Single-tree influence on soil properties in the mountains of eastern Kentucky. Ecology 71:1365–1372. Bowers, C.G. 1960. Rhododendrons and Azaleas. 2nd Edition. Macmillian, New York, NY. 525 pp. Clinton, B.D. 2003. Light, temperature, and soil moisture responses to elevation, evergreen understory, and small canopy gaps in the southern Appalachians. Forest Ecology Management 186:243–255. Clinton, B.D., L.R. Boring, and W.T. Swank. 1994. Regeneration patterns in canopy gaps of mixed-oak forests of the southern Appalachians: Influences of topographic position and evergreen understory. American Midland Naturalist 132:308–319. Cox, M.S. 2001. The Lacaster Soil Test Method as an Alternative to Mehlich 3 Soil Test Method. Soil Science 166:484–489. Coyier, D.L., and M.K. Roane (Eds.). 1988. Compendium of Rhododendron and Azalea Diseases. APS Press, St. Paul, MN. 65 pp. 721 R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 Dreistadt, S.H., J.K. Clark, and M.L. Flint. 1994. Pests of landscape trees and shrubs. An Integrated Pest Management Guide. University of California, Division of Natural Resources, Davis, CA. Publication 3359. Fink, S. 1999. Pathological and Regenerative Plant Anatomy. Gebrüder Borntraeger, Berlin, Germany. 1095 pp. Holliday, P. 2001. A Dictionary of Plant Pathology. 2nd Edition. Cambridge University Press, Cambridge, UK. 536 pp. Marx, D.H., and C.B. Davey. 1969. The influence of ectotrophic mycorrhizal fungi on the resistance of pine roots to pathogenic infections. III. Resistance of aseptically formed mycorrhizae to infection by Phytophthora cinnamomi. Phytopathology 59:549–558. McGee, C.E., and R.C. Smith. 1967. Undisturbed rhododendron thickets are not spreading. Journal of Forestry 65:334–336. Mississippi State University. 2004. Soil testing for the farmer. Available online at http:// msucares.com/pubs/infosheets/is0346.htm. Accessed 3 November 2005. Mississippi State University. 2005. Plant disease and nematode diagnostic services. M-1230. Mississippi State Extension Services, Mississippi State, MS. Monk, C.D., D.T. McGinty, and F.P. Day, Jr. 1985. The ecological importance of Kalmia latifolia and Rhododendron maximum in the deciduous forests of the southern Appalachians. Bulletin Torrey Botanical Club 112:187–193. National Park Service, US Department of the Interior (NPS). 2004a. Inventory and prototype monitoring of natural resources in selected national park system units 1998–1999. Available online at http://www.nature.nps.gov/ publications/TechRpt2000-1/glitzi-03. htm#P49_8562. Accessed 3 November 2005. NPS. 2004b. Inventory and Prototype Monitoring of Natural Resources in Selected National Park System Units 1999–2000. Available online at http://www.nature.nps.gov/ publications/TR2001-1/Glitz-04.htm#P262_38085. Accessed 3 November 2005. Nilsen, E.T., J.F. Walker, O.K. Miller, S.W. Semones, T.T. Lei, and B.D. Clinton. 1999. Inhibition of seedling survival under Rhododendron maximum (Ericaceae): Could allelopathy be a cause? American Journal Botany 86:1597–1605. O’Neill, K.P., M.C. Amacher, and C.H. Perry. 2005. Soils as an indicator of forest health: A guide to the collection, analysis, and interpretation of soil indicator data in the forest inventory and analysis program. North Central Research Station, Forest Service, US Department of Agriculture, St. Paul, MN. General Technical Report NC-258. Ownley, B.H. 1993. Investigator’s annual report-Part II. National Park Service, US Department of the Interior, Gatlinburg, TN. GRSM-N-0860083. Ownley, B.H. 1994. Biotic and abiotic factors associated with rhododendron dieback in the Great Smoky Mountains National Park. National Park Services, US Department of the Interior, Gatlinburg, TN. Sub-Agreement to Cooperative Agreement CA-5460-0-9001. Powell, W.M., F.F. Hendrix, Jr., and D.H. Marx. 1968. Chemical control of feeder root necrosis of pecans caused by Pythium species and nematodes. Plant Disease Reporter 52:577–578. Radford, A.E., H.E. Ahles, and C.R. Bell. 1968. Manual of the Vascular Flora of the Carolinas. University of North Carolina Press, Chapel Hill, NC. 1 183 pp. Rivers, C.T., D.H. Van Lear, B.D. Clinton, and T.A. Waldrop. 1999. Community composition in canopy gaps as influenced by presence or absence of Rhododendron maximum. Pp. 57–60, In J.D. Haywood (Ed.). Proceedings of the Tenth Biennial Southern Silvicultural Research Conference; 1999 February 16–18, Shreveport, LA. US Department of Agriculture, Forest Service, Southern Research Station, Asheville, NC. General Technical Report SRS-30. R. Baird, A. Wood-Jones, J. Varco, C. Watson, W. Starrett, G. Taylor, and K. Johnson 2013 Southeastern Naturalist Vol. 12, No. 4 722 Robinette, S.L. 1974. Rhododendron, Rhododendron maximum L. US Department of Agriculture, West Virginia University, Morgantown, WV. Technical Report 9:113–115. Ruehle, J.L. 1973. Nematodes and forest trees: Types of damage to tree roots. Annual Review of Phytopathology 11:99–118. Sinclair, W.A., H.H. Lyon, and W.T. Johnson. 1993. Diseases of Trees and Shrubs. Comstock Publishing Associates, Ithaca, NY. 575 pp. Smith, R. C. 1963. Some aspects of variation in growth and development of Rhododendron maximum L. in western North Carolina. M.Sc. Thesis. North Carolina State University, Raleigh, NC. Stollarova, I. 1999. The occurrence, distribution, and abundance of plant parasitic nematodes in forest and fruit nurseries of Slovakia. Nematologia Mediterranea 27:47–56. Vandermast, D.B., and D.H. Van Lear. 2002. Riparian vegetation in the southern Appalachian mountains following chestnut blight. Forest Ecology Management 155:7–106. Yeakley, I.A., J.L. Meyer, and W.T. Swank. 1994. Hillslope nutrient flux during near-stream vegetation removal, I. A multi-scaled modeling design. Water, Air, and Soil Pollution 77:229–246.