1Recent research has demonstrated important linkages between above- and below-ground components of terrestrial ecosystems, but the relationships between aerial parasitic plants, such as dwarf mistletoes, and below-ground organisms, such as mycorrhizal fungi, have not been examined in detail.
2We examined the relationship between dwarf mistletoe infection, host vigour and the ectomycorrhizal colonization and fungal community composition of pinyon pine (Pinus edulis) using a combination of field observations and glasshouse studies.
3High levels of dwarf mistletoe infection were not associated with increased mortality or needle loss of infected pinyons, but infected trees had lower shoot growth.
4Ectomycorrhizal colonization was positively associated with dwarf mistletoe infection severity at two sites in two years. In addition, the ectomycorrhizal fungal community structure of trees with low, intermediate and high levels of dwarf mistletoe were significantly different, primarily due to a shift in the dominance of ascomycete fungi.
5These higher levels of ectomycorrhizal colonization were associated with increased fungal inoculum under the crowns of pinyons heavily infected with dwarf mistletoe. In addition, 33% more pinyon seedlings were found in the understories of mistletoe-infected trees than uninfected trees.
6These findings point to complex multi-trophic interactions between the above-ground and below-ground communities of pinyon pine, and suggest that a detailed understanding of host-parasite relationships may require study of other symbionts associated with the host.
The above- and below-ground components of terrestrial ecosystems have been shown to interact across a broad range of ecosystems (Wardle et al. 2004). While the majority of studies examining these interactions have focused on herbivores, below-ground organisms are likely also to impact or be impacted by aerial parasitic plants, such as mistletoes. Like herbivores, parasitic plants alter the nutrient balance of their host plants (Pennings & Callaway 2002), which in turn may influence below-ground organisms, particularly those intimately associated with the roots of plants, such as mycorrhizal fungi. Although mistletoes have significant impacts on their host plants in numerous ecosystems (Watson 2001), few studies have examined interactions between mistletoes and mycorrhizal fungi (but see Gehring & Whitham 1992), and no studies have explored the relationship between mistletoe parasitism and below-ground fungal community composition.
Dwarf mistletoes (Viscaceae: Arceuthobium) are aerial haustoral hemiparasites found on members of the Pinaceae and Cupressaceae that are dependent on the host plant for water and mineral nutrients and a portion of their carbon requirements (Hawksworth & Weins 1996). High levels of infection have been shown to result in reduced height and diameter growth (Mathiasen et al. 1990), suppressed reproductive output and germination success (Sproule 1996), altered water and nutrient relations (Lamont 1983; Sala et al. 2001) and increased mortality (Mathiasen et al. 1990) of the host plant. As a result of these negative impacts, trees supporting high levels of dwarf mistletoe are often more susceptible to secondary agents, such as fire, insect herbivores and disease (Geils & Hawksworth 2002; Negron & Wilson 2003). Dwarf mistletoes also affect forest canopy structure by altering biomass allocation in their host plants (Mathiasen 1996), and appear to have the capacity to alter the community structure and composition of forest ecosystems (Godfree et al. 2002). Dwarf mistletoes are also considered to be a keystone resource in forest ecosystems (Watson 2001; Press & Phoenix 2005).
Both dwarf mistletoes and EM fungi have been shown to have large impacts on their host plants, but the tripartite host-parasite-EM fungus relationship has not been examined. However, because EM fungi benefit the host plant by increasing the uptake of water and mineral nutrients removed by dwarf mistletoe, and also require significant amounts of carbon (Smith & Read 1997), interactions are likely. We examined the relationship between dwarf mistletoe (Arceuthobium divaricatum Engelm.) and pinyon pine (Pinus edulis Engelm.) and its dependent EM fungi. We addressed two questions. First, is infection by A. divaricatum associated with decreased host vigour, as measured by mortality, needle retention and shoot growth? Secondly, what is the relationship between A. divaricatum and EM fungi? Specifically, are high levels of dwarf mistletoe infection associated with alterations in colon-ization levels and in the fungal communities of mature trees, and does this result in differences in EM fungal inoculum in the understorey of infected trees that may act as nurses for establishing seedlings? Because dwarf mistletoes and EM fungi are found together on numerous dominant plant species in a broad range of ecosystems (Hawksworth & Weins 1996; Smith & Read 1997), these findings have ramifications for numerous forest systems.
This research was conducted across 85 km of the pinyon-juniper woodland north of the San Francisco Peaks near Flagstaff, Arizona. Dominant overstorey plants included pinyon pine (Pinus edulis Engelm.) and one-seed juniper (Juniperus monosperma Engelm.). Common shrub species included rabbitbrush (Chrysothamnus nauseosus Pall. Britt) and Apache plume (Fallugia paradoxa D. Don ex Torr.). Soils within study sites included Typic Ustorthents, Typic Haplustalfs, Vitrandic Ustochrepts, Typic Haploborolls and Lithic Ustochrepts (USDA) and elevation ranged from 1975 to 2247 m a.s.l. Precipitation for northern Arizona (AZ Division 2) from September to August was 39.8, 16.6 and 35.9 cm for 2001, 2002 and 2003, respectively (http://www.noaa.gov).
dwarf mistletoe and host plant vigor
To determine if parasitism by dwarf mistletoe was associated with increased pinyon mortality, we quantified tree mortality that resulted from a severe drought in Arizona in 2002 (Mueller et al. 2005). In August to September of 2002, three 50 × 10 m plots were placed randomly in eight sites with known incidence of mistletoe infection. Within each plot all pinyons were examined and classified as living or dead, and dwarf mistletoe noted as present or absent as determined by the presence of aerial mistletoe shoots and/or swellings of pinyon branches. We did not attempt to quantify mistletoe abundance at these sites because tree mortality resulted in mistletoe mortality and abscission of some aerial shoots. The average number of pinyons per site within all three transects was 69.8 ± 8.4. The percentage mortality of trees with and without dwarf mistletoe was calculated and compared across all sites using a paired t-test. To compare pinyon mortality and the incidence of dwarf mistletoe at the stand level, a regression analysis was performed using SigmaPlot v. 9.0 (SigmaPlot 2004).
Annual shoot growth was also measured on living trees at two of the above eight sites where both infected and uninfected trees were present. Dwarf mistletoe infection severity was measured using the six-class system (Hawksworth 1977). This system divides trees into thirds and classifies each section according to the percentage of branches infected. A dwarf mistletoe rating (DMR) of 0 describes a tree with no visible dwarf mistletoe shoots or branch swelling, while a rating of 6 describes a tree with more than half of the branches infected in each third of the tree. We selected 10 trees classified as high mistletoe (DMR of 5–6) and 10 classified as low mistletoe (DMR of 0–1). Six branches per tree were selected at random, and shoot growth from 1995 to 2002 was estimated using bud scars to define years. Shoot growth was averaged across all six branches, and growth for high and low mistletoe pinyons was compared using a repeated measures analysis of variance in SPSS vs. 12.0. The number of needle cohorts remaining on the branches was also counted as an additional measure of tree vigour and compared using a t-test in SPSS vs. 12.0.
dwarf mistletoe–em relationship
We randomly selected three of the eight survey sites to examine the interaction between dwarf mistletoe infection and EM colonization. Due to severe levels of drought-induced pinyon mortality in both 2002 and 2003, measurements could only be taken at each site during a single year in the study period (2001–03). At two sites, a quantitative measurement of mistletoe infection severity was made by calculating the mistletoe biomass to tree volume ratio for each of 10 or 11 trees. The external shoots of all dwarf mistletoes were removed manually, and shoots were oven dried at 70 °C for 48 hours and weighed to the nearest hundredth of a gram. The width and height of each tree was measured to the nearest cm, and tree volume was approximated as the volume of a cone using the formula πr2 h/3 where r = width of the crown and h = tree height. Mistletoe infection had no obvious impact on tree architecture so that the use of a cone was appropriate for all trees regardless of level of mistletoe parasitism. At the third site, a gradient of dwarf mistletoe infection was examined using high (DMR 5 or 6, n = 18), intermediate (DMR = 3, n = 8) and low (DMR = 0 or 1, n = 19) levels of dwarf mistletoe infection. Due to the low occurrence of uninfected trees at this site we were unable to analyse these as a separate group, and trees with DMR of 0 were included in the low infection group. Although this method is not likely to be as accurate in describing the relationship between EM and mistletoes as the quantitative measure described above, it allowed us to determine whether patterns observed using the easily measured six-class rating system were similar to those obtained using destructive sampling. This method also allowed us to compare EM community composition quantitatively among the high, intermediate and low mistletoe groups.
Ectomycorrhizal colonization measurements were made in August of 2001–03, following the onset of monsoonal rains, as pinyon EM fungi have been shown to respond positively to increased precipitation (Swaty et al. 1998). Approximately 200 cm of pinyon root was collected on the east side of all study trees and scored for EM colonization according to the methods of Gehring & Whitham (1991). All short roots were identified and classified as living or dead and mycorrhizal or non-mycorrhizal. Percentage colonization was calculated as the number of living EM root tips divided by the total number of short roots available for colonization. Percentage EM colonization was compared using a one-way anova for the gradient of infection (high vs. intermediate vs. low) in SPSS vs. 12.0. A regression analysis was used to analyse the relationship between mistletoe biomass and EM colonization in SigmaPlot v. 9.0.
Ectomycorrhizal fungal community composition was analysed along the qualitative infection gradient only. Ectomycorrhizal root tips were categorized into morphologic groups (morphotypes) as described by Horton & Bruns (1998) based on branching pattern, colour and texture, together with the presence, colour and texture of emanating hyphae and rhizomorphs. Three to five tips of each morphotype per tree were placed in microcentrifuge tubes and stored at −20 °C until DNA was extracted for molecular analyses using the techniques described by Gardes & Bruns (1993) and modified by Gehring et al. (1998) for pinyons. The internal transcribed spacer (ITS) region of the fungal genome, located between the 18S and 28S rRNA genes, was amplified using PCR with the ITS1F and ITS4 primer pair (Gardes & Bruns 1993). The amplified ITS region was characterized using restriction enzyme digestion with HinfI and MboI (Invitrogen) followed by visualization using agarose gel electrophoresis. Digestion with two enzymes frequently allows distinction between EM fungal species (Gardes & Bruns 1996; Dahlberg et al. 1997) and these two enzymes have been shown to discriminate among fungal species in Pinus edulis (Gehring et al. 1998). Digital images of gels were recorded and analysed using a Kodak EDAS 290 gel documentation system and accompanying software. Root tip restriction fragment length polymorphism (RFLP) patterns were compared with existing EM root tip and sporocarp data bases for pinyons to identify fungi (Gehring et al. 1998; Haskins & Gehring 2004; Swaty et al. 2004). Unknown RFLP patterns were sequenced at the DNA Sequencing Facility at the University of Arizona on an ABI 3730xl Genetic Analyser and analysed using SeqMan 5.05 software (1998–2002, DNASTAR Inc.). To compare fungal community composition among the three groups, we used non-metric multidimensional scaling (NMDS), an ordination technique, to analyse the RFLP community from each tree using the program DECODA (Fensham et al. 2000). We used analysis of similarity (ANOSIM; Clarke 1993; Dungey et al. 2000) to test for significant differences in EM fungal community composition. Pair-wise comparisons were used to test for differences among the three groups. Diversity was calculated using the Shannon-Wiener index, and species richness and diversity were compared among the three groups using an anova. In addition, because variation in the prevalence of ascomycete vs. basidiomycete fungi has been observed in association with interspecific competition (Haskins & Gehring 2004) and herbivory (Brown et al. 2001), we also compared the percentage of EM fungi in the subdivision Ascomycotina among groups using an anova in SPSS vs. 12.0.
dwarf mistletoe, em fungal inoculum and seedling recruitment
Because previous studies on root parasites have shown that parasitism impacted mycorrhizal colonization (Davies & Graves 1998; Gworgwor & Weber 2003), which is important in the establishment of pinyons (Gehring & Whitham 1994a), we quantified the fungal inoculum under the crowns of 10 pinyons with high vs. low levels of dwarf mistletoe infection at sites one and two using a bioassay. Two soil cores (7.5 cm width by 15 cm depth) were taken midway between the dripline and trunk of pinyons and transferred intact into pots in the field. Surface-sterilized pinyon seeds of mass 0.24–0.29 g were placed in the soil cores in growth chambers on 12:12 hour light:dark photoperiods. Seeds were watered every other day until germination, and twice per week subsequently. Six months after germination, seedlings were harvested, and EM colonization, root length, root to shoot ratios and total dry biomass were measured. The relatively short duration of this bioassay was selected to minimize changes in fungal activity due to the presence of host plant roots or altered soil condition (Brundrett et al. 1996) and thus to evaluate more accurately the amount of fungal propagules present in the soil (Brundrett & Abbot 1995). Measurements of seedlings from soil taken from the understorey of the same tree were averaged. Root length was measured using WinRhizo version 2003a. Seedlings were separated at the root crown, and root and shoot portions were oven dried at 70 °C for 48 hours and weighed to the nearest hundredth of a gram. Ectomycorrhizal colonization was analysed following the methods described above. All other seedling parameters were compared using a manova followed by contrasts in SPSS vs. 12.0.
In addition, because variation in levels of EM inoculum associated with mistletoe infection would be predicted to result in altered pinyon seedling recruitment (Lovelock & Miller 2002), we counted the number of pinyon seedlings located in the understorey of the mistletoe-infected and uninfected trees in all eight survey sites. The average number of seedlings associated with infected and uninfected trees at all sites was compared using a paired t-test.
host plant mortality and performance
Infection by A. divaricatum was not associated with increased mortality of pinyon pine during the severe drought of 2002. We found no difference in percentage mortality between infected and uninfected pinyons across the eight survey sites (mean values 51.3 ± 7.50% and 48.7 ± 7.39%, respectively, t = 0.37, P = 0.36). In addition, the incidence of dwarf mistletoe infection was not significantly correlated with stand-level pinyon mortality (r2 = 0.09, F = 0.01, P = 0.894).
Similarly, we found no differences between high and low mistletoe trees in the number of needle cohorts retained (6.83 ± 0.647 vs. 7.41 ± 0.447, t = 0.73, P = 0.238). However, pinyons with high levels of dwarf mistletoe infection had significantly lower shoot growth than trees with low levels of infection (time, F = 20.591, P < 0.001; treatment, F = 4.845, P = 0.04; time × treatment, F = 8.365, P < 0.001) (Fig. 1).
dwarf mistletoe–em relationship
Ectomycorrhizal colonization was positively correlated with dwarf mistletoe infection severity at two sites during two years. We found large differences in mean EM colonization between the two sites (25.4 ± 2.17% vs. 67.7 ± 4.84%), which were likely to be the result of differences in soil type (Gehring & Whitham 1994a), but when mistletoe biomass was compared with EM colonization, a positive linear relationship was found at both sites (r2 = 0.513, F1,9 = 9.49, P = 0.013 and r2 = 0.575, F1,8 = 10.81, P = 0.011, respectively, Fig. 2). In agreement with these findings, EM colonization increased along the qualitative gradient of dwarf mistletoe infection (F2,42 = 3.92, P = 0.04), such that trees with high levels of dwarf mistletoe supported 26% higher EM colonization than trees with low levels Fig. 3). Ectomycorrhizal colonization of pinyons with intermediate levels of mistletoe infection was not significantly different from pinyons with high or low levels of mistletoe.
Fungal communities based on RFLP data suggested that the species composition of EM fungi differed among trees with high, intermediate and low levels of dwarf mistletoe infection (R = 0.526, P < 0.001) (Fig. 4). Pair-wise comparisons showed that the community compositions of all three groups were distinct. Twelve unique RFLP patterns were observed, and trees with low, intermediate and high levels of dwarf mistletoe had eight, five and nine RFLP types, respectively. Four RFLP types were restricted to low mistletoe trees, one to intermediate trees, and four to high mistletoe trees. Both richness (F2,42 = 7.8, P = 0.001) and diversity (F2,42 = 4.87, P = 0.013) differed with infection level (Table 1).
Table 1. Pinyons with high, intermediate and low levels of dwarf mistletoe differed in EM fungal richness and diversity, and the percentage of the EM community composed of ascomycete fungi. Data presented are means ± 1 SE. Different letters indicate significant differences (Tukey) at α≤ 0.05
Based on comparisons with sporocarp RFLPs, comparison with root tip RFLPs from previous studies and DNA sequencing, four RFLP patterns could be identified to species (Cenococcum geophilum, Geopora cooperi, Rhizopogon rubescens and Tomentella pilosa), and constituted 29.8% of the total EM tips. One RFLP type was identified to family (Thelephoraceae, Blast ID AF377065.1) and one was classified as a member of the Basidiomycotina (Gehring et al. 1998; Haskins & Gehring 2004; Swaty et al. 2004). Four RFLP types were identified to order (Pezizales) and matched the same GenBank sequence, but with varying affinity (Blast ID AF266709 with 92, 94, 97 and 99% sequence similarity), and one RFLP pattern was identified as an unknown member of the Basidiomycotina. Due to poor sequence quality, two RFLP types could not be identified by sequencing, but these unknowns composed only 2.6% of all tips examined on the three groups of trees combined.
The observed community changes can be partially attributed to differences among groups in the dominance of EM fungal species within the subdivisions Basidiomycotina and Ascomycotina. Pinyons with high and intermediate levels of dwarf mistletoe infection supported significantly greater amounts of ascomycete fungi than pinyons with low levels of infection (F2,42 = 16.7, P < 0.001) (Table 1).
dwarf mistletoe, ectomycorrhizal inoculum and seedling associations
Bioassay measurements from the growth chamber experiment showed that EM fungal inoculum was 41% greater under high dwarf mistletoe pinyons than low dwarf mistletoe pinyons (t = 2.17, P = 0.019). Colonization of seedlings grown in soil from the understorey of infected and uninfected pinyons was 65.3 ± 4.37% and 46.4 ± 6.74%, respectively. No differences in root length, root to shoot ratios or total dry biomass were found for bioassay seedlings (Wilks’ lambda = 0.924, F = 0.655, P = 0.588).
In agreement with these findings, the number of seedlings associated with dwarf mistletoe infected trees in the field tended to be higher than the number of seedlings associated with uninfected trees (t = 1.54, P = 0.08). Across all transects, the number of seedlings under an infected tree was 0.40 ± 0.07, and the number of seedlings per uninfected tree was 0.30 ± 0.06.
dwarf mistletoe and tree vigour
Although dwarf mistletoe infection is generally associated with increased host mortality (Hawksworth & Weins 1996), dwarf mistletoe infection of pinyon was not associated with increased pinyon mortality during the extreme drought in 2002. In contrast, Negron & Wilson (2003) found that dwarf mistletoe infection was a significant predictor of stand-level pinyon mortality following the severe drought of 1996, but this discrepancy could be due to higher mortality within stands with the most severe levels of dwarf mistletoe infection, as opposed to just the presence of dwarf mistletoe. However, at the stand level, the incidence of dwarf mistletoe infection accurately predicts the severity (DMR) within that stand (r2 = 0.993, n = 14 sites) (Mueller 2004). Within the eight stands surveyed, pinyon mortality was not significantly correlated with the incidence of dwarf mistletoe, suggesting that even stands with the highest levels of dwarf mistletoe did not experience higher levels of mortality.
In contrast to tree mortality, we found that high mistletoe trees had significantly lower shoot growth than low mistletoe trees. However, differences in growth varied across years; during periods of severe drought (i.e. 1996 and 2002), differences were absent (Fig. 1), which argues that the impacts of dwarf mistletoe vary in response to environmental conditions. Kirkpatric (1989) found that differences in transpiration between dwarf mistletoes and their host plants increased during drought. Although these findings contrast with our growth data, they suggest that the relationship between dwarf mistletoes and their hosts changes in response to climate.
dwarf mistletoe and em colonization
We consistently found a positive correlation between dwarf mistletoe infection severity and EM colonization, but the mechanism driving this relationship is unknown. Infection by below-ground root parasites has been shown to result in decreased arbuscular mycorrhizal colonization (Davies & Graves 1998), and EM and arbuscular mycorrhizal fungi have been shown to impact parasitic plants both positively (Sanders et al. 1993; Salonen et al. 2000, 2001) and negatively (Lendzemo & Kuyper 2001; Gworgwor & Weber 2003). Because previous studies indicate that mycorrhizal fungi can impact plant parasites, and vice versa, we suggest three possible hypotheses to explain the observed relationship. First, establishment and reproduction of dwarf mistletoe could be higher on pinyons that support high levels of EM. Salonen et al. (2000) found that reproduction of the root parasite Melampyrum pratense was highest on Pinus sylvestris hosts with high levels of EM colonization, suggesting that parasitic plants can benefit from increased nutrient acquisition by EM. As a result, the positive relationship observed between dwarf mistletoe and EM may be the result of the beneficial effects of EM on dwarf mistletoe growth and reproduction.
Alternatively, the increased nutrient demands of host plants that result from mistletoe parasitism may lead to increased EM colonization. Because EM fungi provide the same nutrients (water, minerals) removed by dwarf mistletoe, increased allocation below ground to these fungi could result from the mistletoe nutrient sink. Pennings & Callaway (2002) proposed that the antagonistic vs. beneficial impacts of mycorrhizal fungi on parasitic plants depend largely on whether mycorrhizal fungi increase host plant defences or host biomass and nutrient quality. We propose the additional possibility that, instead of altering plant resistance to dwarf mistletoe, high levels of EM colonization of highly infected pinyons may result in increased host tolerance to parasitism by dwarf mistletoe. For example, infection by arbuscular mycorrhizal fungi counteracted the negative effects of the parasite Striga hermonthica in one Sorgum bicolor cultivar (Lendzemo & Kuyper 2001). Increased levels of mycorrhizal colonization may also contribute to the lower occurrence of adverse impacts of dwarf mistletoe on pinyon than on other hosts (Hawksworth & Weins 1996).
dwarf mistletoe em fungal community
In addition to quantitative differences in EM fungi associated with dwarf mistletoe infection, we found differences in fungal community composition. Approximately 5000 species of fungi form EM associations (Molina et al. 1992), and a wide range of functional roles of these fungi has been demonstrated within and across habitats (Allen et al. 1995). Mycorrhizal fungal diversity can promote increased host nutrient uptake (Baxter & Dighton 2001) and increased plant productivity (Jonsson et al. 2001). As a result, shifts in EM diversity and community composition could have important consequences for the host plant.
We observed changes in both diversity and community composition associated with mistletoe infection, with the latter due largely to greater ascomycete abundance on trees with medium or high mistletoe infection than on trees with little or no mistletoe. Previous studies on pinyons have also shown shifts in the dominance of ascomycete fungi in response to insect herbivory (Brown et al. 2001), site stress (Gehring et al. 1998) and below-ground competition (Haskins & Gehring 2004). The higher proportions of ascomycete fungi associated with pinyons supporting intermediate and high levels of dwarf mistletoe (Table 1) could similarly be due to increased host stress induced by mistletoe parasitism and this may have contributed to the observed differences in EM fungal communities (Fig. 4). In addition, similarities in the dominance of ascomycete fungi on pinyons infected with dwarf mistletoe and pinyons susceptible to insect attack (Brown et al. 2001) suggest that the impacts of a plant parasite and insect herbivore on the EM community composition of pinyon could be similar (e.g. Pennings & Callaway 2002). However, while high levels of mistletoe parasitism were associated with increased EM fungal species richness and diversity, similar patterns were not observed in response to insect herbivory (Gehring & Whitham 2002). This discrepancy suggests that the relationships between herbivores or parasites and EM fungi are complex and understanding them will require further study in the field and in the laboratory, where parasites and EM fungi can be more readily manipulated.
extended effects of dwarf mistletoe
We found that mature pinyons with high levels of dwarf mistletoe supported higher levels of EM, and also had enhanced levels of EM fungal inoculum in their crown understoreys. Although we found no differences in growth between seedlings grown in high vs. low dwarf mistletoe soil, Gehring & Whitham (1994b) found that EM colonization was positively linearly associated with increased pinyon seedling growth over a longer time period. In addition, bioassays are more useful for estimating fungal inoculum potential than they are for determining growth responses to EM colonization due to their short durations and pot sizes that may reduce the benefits seedlings receive from EM colonization (Brundrett et al. 1996). Mycorrhizal fungi may also provide benefits other than nutrient uptake, such as protection from pathogens, which can be more important than growth benefits in field settings (e.g. Newsham et al. 1995).
Proximity to established EM vegetation often results in increased EM colonization of associated seedlings (Simard et al. 1997; Hagerman et al. 2001; Horton et al. 1999; Dickie et al. 2002; Dickie et al. 2004; Nara & Hogetsu 2004). Although we could not establish causation in this study, pinyons with high levels of mistletoe also had high levels of EM fungal inoculum associated with them. As a result, seedling recruitment may be higher in association with dwarf mistletoe infected pinyons. Because seedlings commonly associate with conspecific nurses in areas of high pinyon mortality (Mueller et al. 2005), high mistletoe trees could serve as islands of inoculum in these sites. This hypothesis is supported by data indicating that trees infected with dwarf mistletoe supported 33% more conspecific seedlings than uninfected trees.
In conclusion, the relationships observed between dwarf mistletoe and EM fungi suggest that the apparently simple host-parasite relationship between mistletoes and pinyons is much more complex when mycorrhizal fungi are considered. Future research should examine the potential for indirect mutualisms between parasitic plants and their hosts mediated by mycorrhizal fungi.
We thank Tom Whitham, Tom Kolb, Galena Nelson, Melissa Dunseath, Sarah Husband, Scott Hayden, Theresa McHugh, Crescent Scudder, Adrian Stone, Karla Kennedy, Beth Franklin, Sierra Delance, Cory Helton and Kobe Moby for field, laboratory and miscellaneous assistance and two anonymous referees for helpful comments on this manuscript. This research was funded by NSF grants DEB-0087017 and DEB-0415563 and a Merriam Powell Center for Environmental Research Fellowship.