Ectomycorrhizal community structure in a xeric Quercus woodland based on rDNA sequence analysis of sporocarps and pooled roots


Author for correspondence: Matthew E. Smith Tel: +1 530 7520862 Fax: +1 530 7525809 Email:


  • • Quercus woodlands are key components of California's wild landscapes, yet little is known about ectomycorrhizal (EM) fungi in these ecosystems.
  • • We examined the EM community associated with Quercus douglasii using sporocarp surveys and by pooling EM roots and subjecting them to DNA extraction, polymerase chain reaction (PCR), cloning, restriction fragment length polymorphism (RFLP) screening and DNA sequencing.
  • • Ectomycorrhizal root symbionts were sampled four times in 2003–04. During this time, the below-ground community structure was relatively stable; we found no evidence of taxa adapted to winter or spring conditions and only one species varied widely in occurrence between years.
  • • The EM community from sporocarps and roots was diverse (161 species), rich in Ascomycota (46 species), and dominated by fungi with cryptic sporocarps. This included a large number of resupinate and hypogeous taxa, many of which were detected both above- and below-ground.


The last decade has seen explosive growth in the number of ectomycorrhizal (EM) fungal community ecology studies (Horton & Bruns, 2001). The heightened interest is partly due to the decreased cost and increased availability of powerful molecular tools first employed by Gardes & Bruns (1996) to elucidate below-ground EM dynamics. Since then studies have used similar techniques to clarify various aspects of EM ecology on roots, including effects of disturbance (Taylor & Bruns, 1999), nitrogen deposition (Avis et al., 2003), soil type (Gehring et al., 1998), coarse woody debris (Tedersoo et al., 2003), host plant (Kennedy et al., 2003) and succession (Nara et al., 2003).

Despite this increase in research, several gaps or limitations remain in our understanding of ectomycorrhizal community ecology. For example, many EM fungi on roots remain taxonomically unknown when sporocarps cannot be matched to roots using restriction fragment length polymorphism (RFLP) of the internal transcribed spacer (ITS) region (Horton & Bruns, 2001). Although ITS sequencing is now commonly used for identification, this region provides poor resolution for fungal groups underrepresented in public databases (e.g. hypogeous fungi) (Izzo et al., 2005b). Two recent EM community studies used rDNA sequences from both ITS and the large subunit (28s) to clarify the phylogenetic position of fungi not clearly resolved by ITS alone (Tedersoo et al., 2003; Cline et al., 2005). Because of the conserved nature of the 28s rDNA, this approach provides higher-level taxonomic information for unknowns while retaining the enhanced resolution provided by ITS.

Ectomycorrhizal studies also suffer from the common practice of assessing the diversity of sporocarp-producing EM fungi based primarily on macroscopic, epigeous fruiting bodies. Whether sporocarps are used to generate ITS profiles for comparison to roots (Yamada & Katsuya, 2001) or to detect ecological patterns (Richard et al., 2004), reported species assemblages may provide a skewed view of EM communities because they disregard important hypogeous and resupinate taxa as well as most Ascomycota (Köljalg et al., 2000; Weiss et al., 2004; Izzo et al., 2005b; Tedersoo et al., 2006). Comparisons between EM roots and primarily epigeous sporocarps may contribute to the widely held view that there is a disconnect between taxa on roots and those present as reproductive structures (Izzo et al., 2005b).

Another common limitation of EM studies has been the use of single sampling dates to assess the diversity and dynamics of below-ground fungal communities; such studies can only provide ‘snapshots’ of communities and cannot fully account for temporal dynamics (Buée et al., 2005). Bruns (1995) hypothesized that EM communities may respond rapidly to seasonal fluctuations in temperature, moisture, and light. While we know that there can be significant species turnover at an annual scale in some ecosystems (Izzo et al., 2005a) and that EM communities change radically over longer time scales (Taylor & Bruns, 1999), few studies have tracked EM fungi over shorter times.

Much of our knowledge about fungal communities on EM roots is inferred from studies of conifer-dominated ecosystems (Horton & Bruns, 2001 and references therein) but three recent studies have used molecular methods to examine EM fungal associates of the geographically widespread and ecologically diverse genus Quercus (Avis et al., 2003; Richard et al., 2005; Walker et al., 2005). In California, Quercus-dominated woodlands and savannas are key ecosystems, covering approximately one-third of the land area and hosting the greatest species diversity of any biome in the state (Standiford, 2002). However, few of California's Quercus-dominated habitats are protected by inclusion in natural reserves, and Quercus spp. are under increasing pressure from human activities such as fire suppression, urban development and introduced pathogens (Pavlik et al., 1991; Standiford, 2002). Although Quercus are obligate ectomycorrhizal hosts (Smith & Read, 1997), little is known about their native EM mycota in California. No previous studies have examined EM communities on Quercus roots in California but data from taxonomic studies of sporocarps suggest that EM fungi associated with Quercus are diverse, with a high number of putative endemics (Parks, 1921; Nishida, 1989; Smith et al., 2006a; Smith et al., 2006b).

In this study we characterized the EM fungal community associated with a drought-tolerant, deciduous, California endemic oak, Quercus douglasii, based on rDNA sequencing from sporocarps and pooled EM roots. To document the temporal dynamics of the EM fungal community, we sampled EM roots on four dates over 2 yr and collected sporocarps over 4 yr. We attempted to accurately depict the EM fungal sporocarp community by sampling equitably from different morphological types including both hypogeous and resupinate taxa.

Materials and Methods

Study site

All research was conducted in the Koch Natural Area at the University of California Sierra Foothill Research and Extension Center (UCSFREC) in the Sierra Nevada foothills of Yuba County, CA, USA (39°17′ N, 121°17′ W). The site has a Mediterranean climate characterized by cool, wet winters and hot, dry summers with terrain that consists of hills 400–600 m above sea level. Precipitation generally occurs between October and May (annual mean 71 cm, range 23–132 cm). Temperature varies greatly between seasons (annual mean 17.8°C, range 10–43°C). Soils are derived from basic metavolcanic rocks and are classified as fine, mixed, active thermic Mollic Haploxeralfs (Argonaut series) from the Alfisol order. Soils are low in carbonates and pH ranges between 5.7 and 6.2.

The site is dominated by Quercus douglasii Hook & Arn. (blue oak) but two other EM species, Quercus wislizeni A. DC. (interior live oak) and Pinus sabiniana Douglas (foothill pine), are locally abundant. Little disturbance has occurred since 1960. Before that time the area was subjected to small-scale logging, thinning, cattle grazing and frequent fires (McClaran, 1988, UCSFREC:

EM root sampling

A 32 × 32 m plot was established in a blue oak woodland to investigate EM community structure over time at a small spatial scale (Fig. 1). Forty-one Q. douglasii were the only trees and known EM hosts on the plot, although Q. wislizeni and P. sabiniana were present in the surrounding woodland. The nearest alternative EM host was a mature Q. wislizeni c. 15 m from the plot edge. Based on fire history and tree sizes, we estimated the stand of Q. douglasii to be ≥ 50 yr old. However, because diameter at breast height and age are weakly correlated in Q. douglasii, the exact age of the stand is not known (Standiford, 2002). Understory vegetation consisted of assorted nonectomycorrhizal herbaceous plants and scattered Toxicodendron diversilobum. A thin litter layer (mean depth 0.75 cm, range 0–2 cm) was irregularly distributed below the trees, but otherwise no soil stratification was noted in the upper 40 cm of soil.

Figure 1.

Diagram of sampling design for ectomycorrhizal (EM) roots of Quercus douglasii with 32 m2 plot subdivided into 8 × 8 m squares (1–16). Light grey area within square 7 was intensively sampled. Open circles represent the approximate location of Q. douglasii trees. Dark grey 25-cm radius circle depicts where EM cores were extracted at random locations within each sampling square at four times (winter and spring; 2003 and 2004).

We sampled soil cores at two different times (winter and spring) in two different years (2003 and 2004) for a total of four sampling dates. The 32 m2 plot was subdivided into 16 8 × 8 m squares (Fig. 1) and on 5 March 2003 we randomly sampled one soil core from within each of the 16 sampling squares. We returned on three subsequent dates (21 May 2003, 29 February 2004, 3 May 2004) and randomly sampled a core from each 8 × 8 m square within a 25 cm radius from the first core (Fig. 1). This yielded 64 soil cores, of which 59 were successfully polymerase chain reaction (PCR)-amplified and used for both EM community and spatiotemporal analysis. On each of the four dates we also sampled 8–10 soil cores from a 2 × 2 m area within square 7 (Fig. 1), yielding a total of 35 soil cores. Data from these 35 soil cores were used to characterize the EM community on roots of Q. douglasii but were not used for spatiotemporal analysis.

Since weather and soil conditions varied between years, sampling dates were selected based on empirical observations of soil moisture, root activity, and above ground phenology of Q. douglasii. Winter samples were collected after significant rainfall occurred, but before leaf production by Q. douglasii. This period was selected because exploratory sampling showed that few EM roots were active in early winter (M. Smith, pers. obs.). Sampling during the summer drought was impractical because of hardening of the upper soil layers. Spring samples were collected after leaf out and flowering of Q. douglasii, when the soil was rapidly drying but not yet impenetrable (see the Supplementary Material, Fig. S1).

Litter was removed and soil cores of c. 900 cm3 (8.5 cm deep × 11.5 cm diameter) were collected, stored at 4°C and processed within 15 d. Shallow soil layers were examined because fine roots of Q. douglasii are most abundant in the top 20 cm (Milikin & Bledsoe, 1999). During soil collection, care was taken to minimize soil disturbance and replace leaf litter since any disruption could affect the remaining EM community. Soil samples were gently wet-sieved over a 0.12 mm sieve with deionized water and all detectable fine roots were removed with forceps under a dissecting microscope (×10–×20 magnification). Roots were trimmed into 6-cm sections and gently rinsed. Almost all live root tips were ectomycorrhizal; roots with signs of decay were discarded. The first 400–500 healthy EM tips encountered under the microscope were severed from fine roots, cleaned, and placed in a Petri dish with water. Roots were considered healthy and colonized by EM fungi when they lacked root hairs and displayed signs of a fungal mantle (color change or emanating hyphae). To achieve randomized sampling, the Petri dish was viewed under the dissecting microscope and swirled in a circular motion to redistribute the EM tips. The first three healthy EM tips encountered were placed in a 1.5 ml microcentrifuge tube. The process was repeated until c. 100 root tips were obtained. EM tips were rinsed again with clean water then 100 root tips were lyophilized together and frozen at −80°C until DNA extraction.

Sporocarp sampling

Sporocarps of EM fungi were collected within a c. 200 m radius of our EM sampling plot on c. 50 expeditions between December 2000 and April 2005. Sporocarps were categorized into three morphological types: epigeous, hypogeous, and resupinate. Taxa whose sporocarps were found at the soil surface, but below the litter, were considered hypogeous regardless of morphology. To collect cryptic sporocarps, a garden cultivator was used to remove litter and soil beneath randomly selected, mature Quercus spp. Cryptic sporocarps were not collected inside the root plot because we did not want to disturb EM fungi on roots. For molecular analysis, a small piece of fresh tissue was removed from each sporocarp and stored in cetyltrimethylammonium bromide (CTAB) buffer at −20°C until DNA extraction. Voucher specimens of sporocarps have been deposited at the herbaria of UC Berkeley (UC) and Oregon State University (OSC) in the USA.

Molecular protocols

Sporocarp samples were ground with a micropestle and DNA was extracted with a modified cetyltrimethylammonium bromide (CTAB) method (Gardes & Bruns, 1993) or Qiagen Stool Kit (Qiagen Inc., Valencia, CA, USA). PCR was performed with forward primers ITS1F and LROR in various combinations with reverse primers ITS4, ITS4B and LR3 (White et al., 1990; Gardes & Bruns, 1993; Hopple & Vilgalys, 1994). The PCR conditions were optimized for each primer set, but the general protocol was: 94°C for 5 min followed by 25 cycles of 1 min at 94°C, 1 min at 55°C and 2 min at 72°C, followed by 7 min at 72°C.

The PCR products were visualized on 1.5% agarose gels with SYBR Green I (Molecular Probes, Eugene, OR, USA). When necessary, PCR products were gel purified with Bio-Rad Freeze ‘n’ Squeeze Tubes (Bio-Rad, Hercules, CA, USA) and reamplified with appropriate primers. The PCR products were cleaned with Microcon-PCR Spin Tubes or Montage PCR96 Filter Plates (Millipore, Billerica, MA, USA). Sequencing was performed as needed with the same primers as above with the ABI Big Dye Terminator Sequencing Kit (v3.1). Sequences were read with an ABI13730xl capillary sequencer (Applied Biosystems, Foster City, CA, USA) at the CAES Genomics Facility, UC Davis. Sequences were edited with sequencher v.4.1 (Gene Codes Inc., Ann Arbor, MI, USA).

Pooled roots were treated differently from sporocarps. Before analysing roots, we tested three primer combinations – (i) ITS1F and LR3, (ii) ITS1F and LR5, and (iii) ITS1F and nLSU1221r (Hopple & Vilgalys, 1994; Schadt et al., 2003) – to determine whether different primers detected the same EM fungi. Although all primer sets recovered most of the same EM fungi in preliminary tests, we used the ITS1F/LR3 primer set because it detected several rare species not found by the others (M. Smith, unpublished). Pooled lyophilized roots were ground with a micropestle and DNA extracted using a modified CTAB method (Gardes & Bruns, 1993) and then purified with a MO-BIO Soil DNA kit (MO-BIO Laboratories, Solana Beach, CA, USA). We used 1–4 µl of DNA template and high resolution Taq polymerase (Invitrogen, Carlsbad, CA, USA) for PCR from roots. The PCR protocol was similar to that used for sporocarps except it had an initial denaturation step of 10 min, included 20 instead of 25 cycles and had 4-min extension steps instead of 2-min extension steps. Amplicons were visualized as above. Four successful replicate reactions per sample were mixed together then cloned with a TOPO-TA kit (Invitrogen). At least 48 successful clones per reaction were grown overnight in Luria–Bertani media amended with 100 µg ml−1 of ampicillin. We selected 48 clones because previous work by Landeweert et al. (2003) and Anderson et al. (2003) found that analysis of 30–50 clones was sufficient to detect most common species of fungi from diverse DNA mixtures. Cloned fragments were reamplified in PCR reactions with c. 0.5 µl of the bacterial suspension as template with the same protocol as for sporocarps. Amplicons were digested separately with the restriction enzymes AluI and HinfI and run simultaneously on a 1.5% agarose gel. Paired digests of each clone were run in adjacent lanes and visualized together. Gel photographs were manually scored to detect all unique RFLP patterns within a core; patterns were not compared across cores. One to four representative clones of each RFLP type in each core were sequenced with ITS1F following the protocol used for sporocarps. Further sequencing was performed as needed with ITS4, LROR, and LR3.

DNA sequence analysis and taxonomic determination of EM fungi

The ITS sequences were examined by blast searches against GenBank and an in-house database using bioedit (Altschul et al., 1997; Hall, 1999). Since almost half of the species detected on roots were also present as sporocarps, we used a traditional species concept to designate taxon names. The EM root sequences matching sporocarps were designated either by a genus and species epithet or, in cases where species identity was unclear, by a genus name followed by an ‘SRC’ (Sierra Research Center) or ‘SF’ (Sierra Field) voucher number. Sequences found on roots but not as sporocarps were named based on data from blast searches, sequence alignments and advice from taxonomic experts. We used a conservative naming approach whereby a sequence with affinities to two genera within a family was designated by the family name, followed by a number. For example, a sequence similar to Peziza and Hydnotryopsis, but not matching a sporocarp, was designated as Pezizaceae followed by a number. To confirm identifications, unknown sequences were aligned with known sequences in clustalx (Chenna et al., 2003) and macclade (Swofford, 2001). When ITS sequences did not provide sufficient information, we sequenced representative clones with LR3 and/or LROR and then subjected them to similar analyses.

For most taxonomic groups, ITS sequences were considered representative of the same species if they differed by < 3% across the length of ITS1, 5.8 s, and ITS2. This criterion assumes that a 0.2–1.2% error rate may be generated by PCR, cloning, and unidirectional sequencing, and that ITS heterogeneity of c. 1.5% exists at small spatial scales for some species (Horton, 2002; Izzo et al., 2005a). This criterion is also similar to that of previous studies (Izzo et al., 2005a; O’Brien et al., 2005). A few phylogenetic groups had either a large number of ITS types (e.g. Thelephoraceae) or appeared to represent species complexes (e.g. G. harknessii). For these groups we designated names based on previously identified lineages (Douhan & Rizzo, 2005; Smith et al., 2006a), well supported groups, as identified by neighbor-joining analyses in paup* 4.0b10 (Swofford, 2001), and consultation with taxonomic experts.

To detect PCR artifacts and sequences of nonEM fungi, we used a combination of sequence alignments and blast searches. The ITS sequences found in multiple cores or matching sporocarps were considered authentic sequences but root sequences detected only once were considered possible chimeras. These sequences were subjected to blast followed by alignment with taxonomically similar clones and a second alignment with all sequences from the same core. It is difficult to detect chimeras comprised of DNA from closely related taxa (Qiu et al., 2001), so it is possible that the methods described did not detect all chimeras in two speciose groups, the family Thelephoraceae and the genus Inocybe. For these groups, we assembled alignments with all sequences and searched for chimeras using the maxchi and chimaera programs of rdp (recombination detection program) (Martin & Rybicki, 2000). Sequences were designated nonEM taxa if they: (i) showed low blast affinity to known EM fungi; (ii) showed high blast affinity to saprobic or parasitic fungi or nonfungal sequences; and (iii) were not identified as chimeras.

Diversity and cluster analyses

We used four diversity measures to characterize the community of EM fungi on roots. Richness (S) is the total number of species. Simpson's Diversity Index (1–D) is a measure of the ‘likelihood that two randomly chosen individuals will be of different species’; values of 1–D range between 0 (no diversity) and 1 (infinite diversity). Pielou's J is a measure of evenness, if all species occur at the same proportion then J = 1. The Shannon–Wiener Index (H) is a measure that accounts for evenness and richness. We did not consider EM root tips from a core as independent units and thus did not consider clones from a core as independent units (see Cline et al., 2005). Accordingly, we calculated diversity indices based on frequency rather than abundance. Frequency is defined as the number of soil cores in which a species or group was detected and relative frequency is the total number of occurrences of a species or group divided by the total number of occurrences of all taxa. Abundance data was only used as a secondary measure for calculating the importance of different EM groups. Abundance is defined as the total number of clones of a species or group and relative abundance is defined as the total number of clones of a species or group divided by the total number of clones of all taxa.

Hierarchical cluster analysis is an agglomerative method that creates groups from multivariate data using a dissimilarity matrix. We used cluster analysis to categorize soil cores based on species composition of EM fungi. We limited cluster analysis to cores sampled at the largest spatial scale (n = 59, Fig. 1); intensively sampled cores from square 7 were not included. We tested whether species composition was influenced more by date or geographic position. Because cluster analyses are influenced by the selected linkage method and distance measure, we performed cluster analysis with two distance measures (Sørrenson and Euclidean distance) and four linkage methods (flexible Beta, centroid, group average, and Ward's method). The Sørrenson distance is influenced by rare species while Euclidean distance downweights rare species (McCune & Mefford, 1999). We selected four linkage methods that compared groups based on different criteria and produced dendrograms with varying levels of chaining. Because Ward's method and the Sørrenson distance are incompatible, Ward's method was only used with Euclidean distance (McCune & Mefford, 1999).

Statistical analysis of clustering was performed by MRPP (Multi-Response Permutation Procedure). This is a nonparametric statistic that uses a Monte Carlo method to examine whether user defined groups are statistically different from random association. Since there are no distributional assumptions, MRPP is appropriate for community analysis (McCune & Mefford, 1999). We tested whether data grouped by each of four categories (core position, year, season, or sampling time (year × season)) were statistically different from random associations. Group weighting was calculated using the default method of pc-ord with 1000 permutations. All diversity indices, cluster analyses, and statistical tests were performed with pc-ord version 4.20 (McCune & Mefford, 1999).


Fungal community on EM roots

During winter and spring of two consecutive years, 9400 EM root tips of Q. douglasii were sampled from 94 soil cores in a 32 m2 plot. Restriction fragment length polymorphisms were used to screen 4532 clones; 4473 of which represented EM taxa. We recovered a total of 92 EM taxa on roots (1–14 spp. per core, mean 5.56 spp. per core) (Table 1). From each of the four sampling times we analysed 20–26 soil cores and recovered 42–59 EM taxa. No arbuscular mycorrhizal (AM) fungi were detected. The EM community was dominated by several frequently occurring species with a large number of rare taxa (Table 1) and the species accumulation curve showed a characteristic exponential rate of increase similar to other EM ecosystems (data not shown). Commonly used diversity indices (Shannon–Wiener Diversity Index (H) = 1.077, Evenness (J) = 0.63, and Simpson's Diversity Index (1–D) = 0.525) suggest that overall diversity was high. The ITS sequences from 42 of 92 taxa (45.7%) were matched to locally collected sporocarps (sclerotia for Cenococcum), and the identities of most other taxa on roots were well-resolved to the genus or species level ( Table 1). However, several taxa were only resolved at the Family or Order level. Of the 92 taxa, 36 were Ascomycota (39.1%) and 56 were Basidiomycota (60.9%). Species in the families Thelephoraceae, Sebacinales, Tuberaceae, and Cortinariaceae were the most frequent and abundant. Thelephoraceae, Pyronemataceae, and Cortinariaceae were particularly speciose with 17, 17, and 16 species, respectively (Table 2). Taxonomic groups tended to be ranked similarly whether by relative frequency or relative abundance, although this pattern was not true for every group (Table 2). Table 1 shows the frequency (number of cores) of each taxon. Reporting of abundance for individual taxa was problematic because RFLP did not resolve all species within the Thelephoraceae, the Cortinarius flexipes group or between Sebacinales2 and Sebacinales4 (one core only). Thus, 666 clones were not resolved at the species level, 605 of which were Thelephoraceae. When grouped by sporocarp type, 31 taxa had epigeous or likely epigeous sporocarps (33.7%), 24 had hypogeous or likely hypogeous sporocarps (26.1%), 23 had resupinate or likely resupinate sporocarps (25.0%) and 14 had an unknown fruiting form or reproduce asexually (15.2%) (Table 2). A large number of Ascomycota and hypogeous taxa were detected relative to other studies of EM fungi associated with hosts in the Fagaceae (see the Supplementary Material, Table S1). In addition to EM fungi, we encountered 51 clones from 20 cores that were either chimeras or nonEM taxa. Chimeras were detected in nine cores and 18 nonEM taxa were identified from 13 cores (Table S2).

Table 1.  Ectomycorrhizal (EM) fungi detected as sporocarps or on the roots of Quercus douglasii
Taxon NameVoucher1AccessionEM cores2Phylum3Sporocarp type4rDNAClosest blast match5% Similarity
  • 1

    Taxa collected as sporocarps have voucher numbers: SRC (Sierra Research Center), SF (Sierra Field), and R- (Russula). –, Taxa where no sporocarp was found.

  • 2

    The number of soil cores where each taxon was detected (out of a total of 94 cores). Taxa on roots and as sporocarps have both a voucher and EM core number.

  • 3

    Asco, Ascomycota; Basidio, Basidiomycota.

  • 4

    All taxa have been assigned to one of four sporocarp categories: epigeous (E), hypogeous (H), resupinate (R) or unknown/asexual (U/A).

  • 5

    blast results are not shown when sporocarps were identified to species or genus by taxonomic experts or when blast results were uninformative. In cases where the best blast matches were unidentified environmental samples, an identified sequence with a slightly lower blast score is shown for clarity.

  • 6

    Putatively matched to EM roots: sporocarps of H. lacunosa were found but no ITS sequence could be obtained.

  • 7

    Lineages as identified in Douhan & Rizzo (2005).

  • 8

    Lineages as identified in Smith et al. (2006a).

  • 9

    Some root associated Ascomycota were previously considered ericoid mycorrhizas or root endophytes but have recently been shown to form ectomycorrhizas (Vrålstad, 2004 and references therein), we have included several here as putative Quercus ectomycorrhizal symbionts.

Amanita sp. (715)src715DQ9746890BasidioEITS
Amanita laneisrc437DQ9746930BasidioEITS
Amanita novinuptasrc669DQ9746900BasidioEITS
Amanita vaginatasrc385DQ9746910BasidioEITS
Amanita velosasrc431DQ9746920BasidioEITS
Balsamia sp.src395DQ9747300AscoHITS
Boletus amygdalinussrc491DQ9747050BasidioEITS
Boletus flaviporussrc443DQ9747060BasidioEITS
Boletus zelleriisrc444DQ9747040BasidioEITS
Cantharellus californicus nom. prov.src46DQ9746880BasidioEITS
Cenoccocum geophilum (Lineage 1)7AY818603, AY8185621AscoU/AITS, SSU intron
Cenoccocum geophilum (Lineage 2)7AY818592, AY81856030AscoU/AITS, SSU intron
Cenoccocum geophilum (Lineage 3)7AY818587, AY81856412AscoU/AITS, SSU intron
Clavariadelphus sp. (src121)src121DQ9747090BasidioEITSClavariadelphus truncatus (DQ097871)430/450 (95%)
Clavulina cf. cristatasrc75DQ9747100BasidioEITSClavulina sp. (DQ202266)578/604 (95%)
Clavulina cf. rugosasrc661DQ9747120BasidioEITSClavulina sp. (AY918958)485/529 (91%)
Clavulina sp. (SF-11)SF-11DQ9747114BasidioEITSClavulina cristata (AJ889929)538/587 (91%)
Clavulina sp. 1EF0251121BasidioEITSClavulina sp. (AY918958)565/566 (99%)
Cortinarius cf. glaucopusDQ9747231BasidioEITSCortinarius glaucopus (AY174785)586/604 (97%)
Cortinarius cf. flexipes (src48)src48DQ9747140BasidioEITSCortinarius cedriolens (AY083179)367/379 (96%)
Cortinarius cf. flexipes (src6)src6DQ9747150BasidioEITSCortinarius flexipes (AJ889971)476/498 (95%)
Cortinarius cf. flexipes (src94)src94DQ9747160BasidioEITSCortinarius hinnuleus (AY083184)502/525 (95%)
Cortinarius cf. flexipes ITS type 1DQ9747246BasidioEITSCortinarius paleaceus (AJ889974)440/475 (92%)
Cortinarius cf. flexipes ITS type 2DQ9747254BasidioEITSCortinarius flexipes (AJ889971)630/654 (96%)
Cortinarius cf. flexipes ITS type 3DQ9747221BasidioEITSCortinarius cf. decipiens (AJ889947)349/359 (97%)
Cortinarius cf. flexipes ITS type 4DQ9747261BasidioEITSCortinarius flexipes (AJ889971)573/593 (96%)
Cortinarius cf. glaucopussrc262DQ9747170BasidioEITSCortinarius glaucopus (AY174785)586/604 (97%)
Cortinarius sp. (src174)src174DQ9747190BasidioEITSCortinarius caninus (U56024.1)586/612 (95%)
Cortinarius sp. (src28) (subgen Phlegmacium)src28DQ9747180BasidioEITSCortinarius citrinus (AY174825)629/650 (96%)
Cortinarius sp.1 (subgen. Phlegmacium)DQ9747211BasidioEITSCortinarius citrinus (AY174825)629/650 (96%)
Cortinarius cf. trivialissrc611DQ9747200BasidioEITSCortinarius favrei (AF325575)396/401 (98%)
Elaphomyces muricatussrc641DQ9747400AscoHITS
Entoloma sp. (src741)src741DQ9746940BasidioEITSEntoloma abortivum (AF357019)556/583 (95%)
Entoloma sp. (src742)src742DQ9746952BasidioEITSEntoloma prunuloides (DQ457633)429/462 (92%)
Ericoid ascomycete19DQ9748271AscoU/AITS, 28sCapronia pilosella (AF050255)552/599 (92%)
Gautieria cf. crispasrc627DQ9747323BasidioHITS
Genabea cerebriformissrc637DQ2068644AscoHITS, 28s
Genea arenariasrc398DQ2068451AscoHITS, 28s
Genea gardnerisrc831DQ206850, DQ2069654AscoHITS, 28s
Genea cf. harknessii (lineage1)8DQ2183015AscoHITS, 28s
Genea cf. harknessii (lineage2)8DQ2183007AscoHITS, 28s
Genea cf. harknessii (src616)8src616DQ2068577AscoHITS, 28s
Genea cf. harknessii (src665)8src665DQ2068592AscoHITS, 28s
Genea cf. harknessii (src830)8src830DQ2068613AscoHITS, 28s
Genea sp. (src680)8src680DQ206849, DQ2069670AscoHITS, 28s
Genea sp. (src680-like)8DQ206855, DQ2069703AscoHITS, 28s
Geopora cooperii var. gilkeyisrc515DQ9747310AscoHITS
Gilkeya compacta8src718DQ2068620AscoHITS, 28s
Gymnomyces xerophilussrc433AY6031020BasidioHITS
Hebeloma cf. oculatumsrc875DQ9746961BasidioEITSHebeloma oculatum (AY311525)613/630 (97%)
Hebeloma cf. velutipessrc303DQ9746972BasidioEITSHebeloma velutipes (AY818351)593/620 (95%)
Helotiales19DQ9748261AscoU/AITS, 28sMeliniomyces bicolor (AY394885)587/614 (95%)
Helotiales29DQ9748242AscoU/AITSHydrocina chaetocladia (AY789413)310/331 (93%)
Helvella cf. crispasrc1940AscoE
Helvella cf. lacunosa6DQ9748312AscoEITS, 28sHelvella lacunosa367/375 (97%)
DQ974834    (U42681) 
Hydnobolites californicussrc667DQ9747331AscoHITS
Hydnobolites sp. 1DQ9748182AscoHITS
Hydnoplicata purpurea nom. prov.src701DQ9747340AscoHITS
Hygrophorus cf. gliocyclussrc744DQ9747280BasidioEITS
Hygrophorus cf. roseiburneussrc29DQ9747270BasidioEITS
Hygrophorus chrysodonsrc782DQ9748091BasidioEITS, 28s
Hygrophorus eburneussrc13DQ9747291BasidioEITS
Hynotryopsis setchelliisrc870DQ9747350AscoHITS
Hysterangium cf. separabilesrc649DQ9748104BasidioHITS
Hysterangium sp. (src642)src642DQ9747360BasidioHITS
Inocybe cf. armeniacasrc225DQ9748030BasidioEITS, 28s
Inocybe cf. fraudanssrc156DQ9748045BasidioEITS, 28s
Inocybe cf. godeyi (src252)src252DQ9748053BasidioEITS, 28sInocybe godeyi (AY038316)556/572 (97%)
Inocybe cf. maculatasrc516DQ9748061BasidioEITS, 28s
Inocybe cf. pudicasrc554DQ9747410BasidioEITS, 28sInocybe pudica (AY228341)594/612 (97%)
Inocybe cf. pusio (src527)src527DQ9747420BasidioEITS, 28s
Inocybe cf. rimosa (src514)src514DQ9748010BasidioEITS, 28s
Inocybe cf. sororiasrc60DQ9748029BasidioEITS, 28sInocybe sp. PBM 2206 (AY732213)603/623 (96%)
Inocybe sp. 1DQ97481311BasidioEITS, 28sInocybe godeyi (AY038316)616/640 (96%)
Inocybe sp. 2 (cf. sindonia)DQ97481221BasidioEITS, 28sInocybe sindonia (AY380393)649/684 (94%)
Inocybe sp. 3 (cf. lanatodisca)DQ97481118BasidioEITS, 28sInocybe lanatodisca (AY380382)579/617 (93%)
Inocybe sp. 4 (cf. flocculosa)DQ9748152BasidioEITS, 28sInocybe cf. flocculosa (AF335452)1209/1235 (97%)
Inocybe sp. 5DQ9748161BasidioEITS, 28sInocybe godeyi (AY038316)595/615 (96%)
Inocybe sp. 6DQ9747431BasidioEITSInocybe relicina (AF325664)201/206 (97%)
Laccaria cf. bicolorsrc65 646/646 100%DQ9746982BasidioEITSLaccaria cf. bicolor (AY228356) 
Laccaria cf. laccatasrc671DQ9746990BasidioEITSLaccaria laccata (AF204814)592/615 (96%)
Lactarius alnicolasrc439DQ9747441BasidioEITS
Lactarius pallescens var. pallescenssrc743DQ9747470BasidioEITS
Lactarius substriatussrc438DQ9747461BasidioEITS
Lactarius xanthogalactussrc299DQ9747450BasidioEITS
Marcelleina sp. (src731)src731DQ9748258AscoHITS, 28s
Melanogaster cf. tuberiformissrc419DQ9747370BasidioHITS
Octavianina sp. 1DQ9747482BasidioHITSOctavianina sp. (AY918953)535/550 (97%)
Otidea umbrinasrc520DQ9747381AscoEITSOtidea umbrina (AF086586)515/515 100%
Pachyphloeus sp. 1DQ9748301AscoHITSPachyphloeus sp.426/453 (94%)
 DQ974833    (AY920528) 
Pachyphloeus sp. 2DQ9748191AscoHITS, 28sPachyphloeus sp. (AY920528)580/622 (93%)
Peziza‘hypogeous’ sp.src696DQ9747392AscoHITS, 28s
Peziza infossasrc683DQ97481712AscoHITS, 28s
Pezizaceae1 (cf. Scabropezia)DQ9747493AscoU/AITS, 28sScabropezia flavovirens (AY500556)534/575 (92%)
Pezizaceae2DQ9746871AscoUAITSPezizales EM (AJ893245)593/602 (98%)
Pezizaceae3DQ9747501AscoUAITS, 28sMarcelleina pseudoanthracina (AY500538)426/464 (91%)
Phialophora sp. 19DQ9748221AscoU/AITS, 28sPhialophora verrucosa (AF050283)1050/1144 (91%)
Pyronemataceae1DQ9747514AscoU/AITS, 28sOtidea umbrina (AY500540)374/405 (92%)
Pyronemataceae2 (cf. Genea)DQ9748321AscoU/AITS, 28sHumaria hemsiphaerica (AY789389)438/481 (91%)
Pyronemataceae3 (cf. Genea)DQ9747521AscoU/AITS, 28sHumaria hemsiphaerica (AY789389)515/558 (92%)
Pyronemataceae4 (src840)src840DQ9748239AscoHITS, 28sAleuria aurantia (AY544654)183/184 (99%)
Ramaria sp. (src827)src827DQ9747130BasidioEITSRamaria flavobrunnescens (AY102864)293/327 (89%)
Russula albidulasrc2DQ97476011BasidioEITS
Russula basifurcatasrc553DQ9748290BasidioEITS
Russula brevipes var. acriorsrc442DQ9747531BasidioEITS
Russula cf. amoenolensDQ9747631BasidioEITSRussula pectinatoides (AY061732)638/658 (96%)
Russula cf. crassotunicatasrc23DQ9747540BasidioEITSRussula emetica (AY061673)623/644 (96%)
Russula cremoricolorR-3006DQ9747550BasidioEITS
Russula cyanoxanthaR-3010DQ9747580BasidioEITS
Russula exalbicansR-4006DQ9747590BasidioEITS
Russula rhodopodaR-4001DQ9747570BasidioEITS
Russula sp.1 (grisea group)R-3029DQ9747620BasidioEITSRussula parazurea (AY061704)601/635 (94%)
Russula sp.2 (grisea group)src204DQ9747610BasidioEITSRussula aeruginea (AF418612)549/580 (94%)
Russula tenuicepsR-4005DQ9747560BasidioEITS
Sebacina cf. epigaeasrc723DQ97476712BasidioRITS, 28sSebacina epigaea (AF490397)537/585 (91%)
Sebacina sp. (src657)src657DQ9747683BasidioRITS, 28sTremellodendron pallidum (AF384862)1111/1171 (94%)
Sebacina sp. (src725)src725DQ9747690BasidioRITSSebacina calcea (AJ427408)231/244 (94%)
Sebacina sp. (src835)src835DQ9747700BasidioRITS, 28sSebacina epigaea (AJ966754)500/522 (95%)
Sebacinales1DQ97476424BasidioRITS, 28sSebacina endomycorrhiza (AF440648)1109/1157 (95%)
Sebacinales2DQ97476619BasidioRITS, 28sSebacina endomycorrhiza (AF440651)1133/1179 (96%)
Sebacinales3DQ9747654BasidioRITS, 28sSebacina incrustans (MW 573)1134/1168 (97%)
Sebacinales4EF0273861BasidioRITSSebacinales EM (AY656955)528/541 (97%)
Sistotrema sp. 1DQ9748211BasidioU/AITS, 28sSistotrema coronilla (AF506475)441/461 (95%)
Tarzetta sp. (src844)src844DQ9748202AscoHITS, 28s
Thelephora cf. anthocephalasrc614DQ9747715BasidioRITS
Thelephoraceae1DQ9747874BasidioRITSThelephora caryophyllea (AJ889980)490/523 (93%)
Thelephoraceae10DQ9747971BasidioRITSTomentella sp. (DQ068972)469/508 (92%)
Thelephoraceae11DQ9747961BasidioRITSTomentella ferruginea (AF272909)545/555 (98%)
Thelephoraceae12 (fuscocinerea group)DQ9747951BasidioRITSTomentella stuposa (AY010277)471/511 (92%)
Thelephoraceae13DQ9748281BasidioRITSTomentella lapidum (AF272941)387/407 (95%)
Thelephoraceae2DQ9747942BasidioRITSThelephoraceae EM (AJ893307)620/635 (97%)
Thelephoraceae3DQ9747932BasidioRITSTomentella sp. N44 (AJ534916)636/690 (92%)
Thelephoraceae4DQ9747924BasidioRITSTomentella cinerascens (AF272915)528/534 (98%)
Thelephoraceae5DQ9747913BasidioRITSTomentella galzinii (AJ421255)433/465 (93%)
Thelephoraceae6DQ97479010BasidioRITSTomentella atramentaria (AF272904)450/470 (95%)
Thelephoraceae7DQ9747893BasidioRITSTomentella badia (AF272937)441/465 (94%)
Thelephoraceae9DQ97478820BasidioRITSThelephora caryophyllea (AJ889980)584/624 (93%)
Tomentella cf. atramentariasrc753DQ9747720BasidioRITS
Tomentella cf. coerulea1 (src839)src839DQ9747730BasidioRITS
Tomentella cf. coerulea2 (src630)src630DQ9747740BasidioRITS
Tomentella cf. ellisiisrc846DQ9747750BasidioRITS
Tomentella cf. fuscocinereasrc813DQ9747764BasidioRITS
Tomentella cf. lateritiasrc833DQ9747770BasidioRITS
Tomentella cf. nitellinasrc675DQ97477812BasidioRITS
Tomentella sp. (src821)src821DQ9747790BasidioRITSTomentella galzinii (AJ421255)521/538 (96%)
Tomentella sp. (src822)src822DQ9747800BasidioRITSThelephoraceae EM (AJ893317)641/662 (96%)
Tomentella sp. (src823)src823DQ9747810BasidioRITSThelephora americana (U83487)598/648 (92%)
Tomentella sp. (src824)src824DQ97478232BasidioRITSTomentella fusco-cinerea (AF272942)390/418 (93%)
Tomentella sp. (src834)src834DQ9747830BasidioRITSTomentella atramentaria (AF272904)444/470 (94%)
Tomentella sp. (src857)src857DQ97478433BasidioRITSTomentella lapidum (AF272941)515/546 (94%)
Tomentella sp. (src859)src859DQ9747850BasidioRITSTomentella botryoides (AF272912)392/410 (95%)
Tomentella sp. (src862)src862DQ9747860BasidioRITSTomentella subclavigera (AY010275)539/585 (92%)
Tricholoma cf. flavovirensSF-4DQ9747030BasidioEITS
Tricholoma portentosumsrc522DQ9747000BasidioEITS
Tricholoma saponeaceumsrc582DQ9747010BasidioEITS
Tricholoma ustalesrc352DQ9747020BasidioEITS
Tuber californicumsrc880DQ9747990AscoHITS
Tuber cf. candidumsrc625DQ97480743AscoHITS, 28s
Tuber cf. whetstonenseDQ97480011AscoHITSTuber whetstonense (AY830855)499/511 (97%)
Tuber sp. (src664)src664DQ9748087AscoHITS, 28sTuber oligospermum (AY515306)374/388 (96%)
Tuber sp. (src709)src709DQ9747980AscoHITSTuber quercicola (AY918957)230/246 (93%)
Table 2.  Relative importance of different ectomycorrhizal fungal groups on the roots of Quercus douglasii as indicated by the number of species, the relative frequency, and the relative abundance
Fungal groupNumber of species (n = 92)Relative frequency (%)Relative abundance (%)
  • Relative frequency is defined as the total number of occurrences of the particular taxonomic or morphological group divided by the total number of occurrences of all taxa. Relative abundance is defined as the total number of clones of the particular group divided by the total number of clones of all taxa.

  • 1

    Taxa were placed in family level groups based on a combination of data from blast searches, sequence alignments and the advice of taxonomic experts.

  • 2

    Sebacinales is listed at the Order level because of taxonomic uncertainties within this group (Weiss et al., 2004).

  • 3

    Cenococcum is listed at the genus level because it is an asexual ascomycete of uncertain taxonomic placement (Douhan & Rizzo, 2005).

  • 4

    Other Basidiomycota includes species from several disparate genera: Clavulina, Entoloma, Gautieria, Hygrophorus, Hysterangium, Octavianina and Sistotrema.

  • 5

    Other Ascomycota includes species within the genera Helvella and Phialophora as well as three poorly resolved taxa (Helotiales1, Helotiales2 and Ericoid ascomycete1).

Taxonomic group1
 Sebacinales2 612.017.1
 Tuberaceae 311.616.1
 Pyronemataceae1710.9 7.0
 Cenococcum3 3 8.2 2.5
 Pezizaceae 8 5.5 3.0
 Other Basidiomycota410 4.6 3.6
 Russulaceae 5 2.9 1.5
 Other Ascomycota5 5 1.7 0.4
Sporocarp type
 Unknown/asexual1411.5 4.3

Fungal community of EM sporocarps

Sporocarps of 108 EM species were collected over a 4-yr period; 83 were Basidiomycota and 25 were Ascomycota (Table 1). Many taxa were rare and encountered only one or a few times. When grouped by sporocarp type, 60 species produced epigeous sporocarps (55.6%), 28 produced hypogeous sporocarps (26.4%) and 20 produced resupinate sporocarps (18.5%). Most Ascomyota formed hypogeous sporocarps, and most hypogeous sporocarps were Ascomycota; only five Basidiomycota formed hypogeous sporocarps. By contrast, most epigeous species were Basidiomycota; only three Ascomycota were epigeous. All resupinate sporocarps were Basidiomycota; we detected 15 Tomentella spp. and four Sebacinales spp. The most speciose genera based on sporocarps were Tomentella (15 spp.), Russula (11 spp.), and Inocybe (8 spp.).

Spatiotemporal patterns in the EM community

Approximately 92% of the common EM fungal species (those found in four or more soil cores) were detected on EM roots in both seasons and in both years (Fig. 2). One exceptional species, Inocybe sororia, was not detected in 2003 but was present in nine cores during 2004. The opposite pattern was never observed; none of the common species from 2003 were absent during 2004. Among infrequently occurring species (taxa found in 2–4 cores) much wider variation was observed between seasons and years. Thirteen species were found only in 2003 or 2004 and six species were detected only in winter or spring (Fig. 2). There were no obvious groupings based on taxonomy or sporocarp type.

Figure 2.

Frequency of most commonly detected ectomycorrhizal (EM) fungi on roots by (a) year (2003 or 2004) and (b) season (winter or spring). Data from 94 soil cores.

The stability of species composition over time suggests persistence of EM taxa in small habitat patches, a pattern that was notable for several infrequently occurring species. Table 3 depicts nine species found on multiple sampling dates, but only in one or a few limited areas of the plot. Six of these species persisted in the same location(s) during both years, while the other three were found during two consecutive seasons of the same year.

Table 3.  Evidence of persistence in space and time of nine species of ectomycorrhizal (EM) fungi on Quercus douglasii roots
Taxon nameTotal number of occurrencesNumber of sampling grids2003120041Persistence between years2
Winter (n = 24)Spring (n = 20)Winter (n = 26)Spring (n = 24)
  • These taxa were found three to seven times over a 2-yr period but were restricted to small, 8 × 8 m plots.

  • 1

    Numbers indicate 8 × 8 m sampling squares where the species were detected; see Fig. 1.

  • 2

    +, Species that persisted between years; –, species found only in 1 yr.

  • 3

    These taxa also found in one or two other cores on a single sampling date (these data have been omitted for clarity).

Thelephoraceae5311 11+
Pezizaceae 1 (Scabropezia clade)3112 1212+
Helvella cf. lacunosa21  33
Entoloma sp. (src742)21  44
Laccaria cf. bicolor2177  
Pyronemataceae14213 13, 1010+
Genea gardneri342 101010+
Tuber sp. (src664)3631033, 103+
Genea cf. harknessii (src616)374 6, 116, 1111+

Hierarchical cluster analysis

Regardless of whether rare species were considered equally or not, cluster analyses produced dendrograms with similar patterns. Only the dendrogram created with the Sørrenson distance measure and flexible Beta linkage method is shown (Fig. 3). Soil cores from within the same 25-cm radius tended to have similar species composition, regardless of sampling date. Consistent with this pattern, MRPP analysis indicated that species composition was correlated with core position (Euclidean P = 0.001, Sørrenson P = 0.001). By contrast, MRPP constrained by season (all samples from a season are pooled, regardless of year) or sampling date (each year × season combination treated separately) were not significantly different from random association (season: Euclidean P = 0.707, Sørrenson P = 0.515; sampling date: Euclidean P = 0.464, Sørrenson P = 0.475). However, when MRPP was constrained by year (all samples pooled by year regardless of season) the two distance measures produced conflicting results (Euclidean P = 0.048, Sørrenson P = 0.150), suggesting a weak effect due to year.

Figure 3.

Dendrogram depicting hierarchical cluster analysis of soil cores grouped by similarity of species composition of ectomycorrhizal fungi, generated using the Sørrenson distance measure and flexible Beta linkage method. Data are from 59 cores sampled at the largest spatial scale (8 × 8 m plots). For each number in the dendrogram, the first number designates the location within the sampling grid and the number preceded by a ‘T’ designates the sampling time (1–4). Black rectangles indicate where cores from within the same 25-cm radius, but at sampled at different times, had similar or identical species composition as indicated by adjacent positions in the dendrogram. Grey rectangles indicate cores that were close in terms of species composition, but were not directly adjacent in the dendrogram.


The EM community associated with Quercus douglasii

Although the diversity of EM fungi in this system was comparable to that of more mesic habitats (Avis et al., 2003; Walker et al., 2005), the Mediterranean climate of California's Sierra Nevada foothills strongly influenced the phylogenetic affinities and sporocarp forms of the EM fungi associated with Q. douglasii. Two of the most striking features of this community were that many of the EM species on roots (c. 40%) were Ascomycota and that many (c. 26%) produced hypogeous sporocarps. We found species of Pezizales in > 80% of soil cores and encountered several pezizalean genera that have rarely or never been detected on roots (e.g. Genabea, Hydnobolites, Hydnotryopsis, Pachyphloeus and Tarzetta) (Tedersoo et al., 2006). Despite high ascomycete diversity, the EM community on roots was dominated overall by basidiomycetes, particularly species with inconspicuous sporocarps (e.g. Inocybe, Thelephoraceae). Except for Inocybe, agaricoid taxa were abundant and diverse as sporocarps but either infrequent or altogether absent on roots (e.g. Amanita, Boletus and Hygrophorus). Even Russula and Lactarius, two dominant genera on EM roots worldwide (Horton & Bruns, 2001), were poorly represented on Q. douglasii. Although the reasons for this phenomenon are unclear, preliminary data suggest that agaricoid taxa are poorly adapted for the dry microclimate beneath deciduous Q. douglasii and instead associate preferentially with two co-occurring evergreen hosts that have thicker litter (Q. wislizeni and P. sabiniana) (M. Smith and M. Morris, unpublished).

Previous studies have reported a pervasive disconnect between EM fungi detected on roots and those detected as sporocarps (Gardes & Bruns, 1996; van der Heijden et al., 1999; Horton & Bruns, 2001; Yamada & Katsuya, 2001; Richard et al., 2005). Few studies, however, have intensively sampled both EM roots and sporocarps simultaneously; in particular, most studies have overlooked inconspicuous sporocarps (Izzo et al., 2005b). We understand that some taxa produce abundant sporocarps and others produce them sparingly (Luoma et al., 1991; Gardes & Bruns, 1996) but the reported disconnect between EM roots and sporocarps has often been interpreted to mean that species with ‘missing sporocarps’ are present only on EM roots or in soil (van der Heijden et al., 1999; Yamada & Katsuya, 2001). There are certainly cases where true discrepancies exist between ‘above-ground’ and ‘below-ground’ biomass; we detected at least one species that produced extremely abundant sporocarps, but was rare on EM roots (Hygrophorus eburneus; M. Smith, pers. obs.). However, in their seminal work Gardes & Bruns (1996) noted that many taxa found on roots might produce inconspicuous sporocarps below-ground or within leaf litter. In this study, we matched EM roots to sporocarps for c. 45% of the EM root community, despite the cryptic and ephemeral nature of most fungi in this arid woodland. Had we surveyed only epigeous sporocarps, the level of correspondence between fruiting bodies and roots would have been quite low (c. 20%). It is possible that unmatched taxa on roots did not produce sporocarps or produced them rarely (Gardes & Bruns, 1996; Straatsma et al., 2001), but an alternative explanation is that their sporocarps were present but difficult to detect and collect. Moreover, many ‘unmatched’ taxa were from hyperdiverse taxonomic groups (e.g. Thelephoraceae, Inocybe) or cryptic species complexes (e.g. Genea harknesii, Cortinarius flexipes group) where species are difficult or impossible to discriminate morphologically. Given these results we suggest that the purported disconnect between EM roots and sporocarps may be heavily influenced by sampling difficulties and the inherent visual bias toward epigeous species. It is possible that the inclusion of hypogeous and resupinate taxa in future sporocarp surveys will decrease this perceived disconnect.

Although the Mediterranean ecosystem and unique mycorrhizal host certainly influenced the composition of EM fungi detected in this study, it is also possible that the pooled root technique affected our perception of the community. For example, members of the C. geophilum species complex were uncommon relative to many other EM studies (Avis et al., 2003); a finding that may result from the small size of C. geophilum EM tips, which may yield less rDNA template, less PCR product and fewer clones than species with larger tips. Despite its potential pitfalls, we used a pooled root technique to avoid the low PCR success rate previously reported from individual EM roots of Fagaceae (e.g. 49–70%)(Avis et al., 2003; Walker et al., 2005). We reasoned that our pooling technique should detect the most common EM taxa but eliminate the need for morphotyping and the problem of poor DNA sequencing due to contamination or dual EM colonization.

Spatiotemporal patterns of EM fungi on Quercus douglasii roots

Because of the strong seasonality of the Mediterranean climate in the Sierra Nevada foothills, we expected significant seasonal variation within the EM community on roots and greater species-turnover between years. Although winter and spring sampling times were separated by only 3 months, these seasons represented biological extremes. When winter soil cores were collected, temperatures were cool, soils were saturated, and shoots and leaves were dormant. When spring cores were collected, temperatures were hot, soils were much drier, and shoots and leaves were growing (see the Supplementary Material, Fig. S1).

Despite these considerable seasonal differences, several lines of evidence suggest that the EM fungal community structure on roots of Q. douglasii remained relatively stable over a 2-yr period. First, the same taxa were dominant on all four sampling dates and only one common species, I. sororia, varied substantially between years (Fig. 2). Second, hierarchical cluster analysis indicated that cores collected within a 25-cm radius contained similar species regardless of the sampling date, suggesting that most taxa either persisted in or recolonized small areas (Fig. 3). Third, persistence was further highlighted by several rare taxa that were detected only on roots within the same 25 cm radius but on successive sampling dates (Table 3).

The hypothesis that the EM community would be temporally dynamic was influenced by results from two recent studies. Izzo et al. (2005a) examined the EM root community in a montane, Abies-dominated California forest and found that although species composition remained relatively stable at large spatial scales over 3 yr, common species went locally extinct within many plots. In a second study, Buée et al. (2005) intensively sampled morphotypes of Fagus in France over 1 yr and found seasonal shifts in the EM community, particularly for two dominant species. In contrast to the results of Izzo et al. (2005a), none of the common species in our plot were ephemeral and we never detected ‘local extinction’. Similarly, we did not detect the wide seasonal variation seen by Buée et al. (2005).

Although sampling design probably accounts for some of the differences between our study and those of Izzo et al. (2005a) and Buée et al. (2005), differences in other factors, such as EM host species, may have affected the dynamics of the EM root communities. Two recent studies by Querejeta et al. (2003, 2006) highlight how deep-rooted, drought-adapted hosts, such as Q. douglasii, might enhance the survival and persistence of EM fungi during prolonged drought. Querejeta et al. (2003, 2007) showed that Quercus pull water from deep soil layers by hydraulic lift and translocate the water directly to EM hyphae in the rhizosphere. Consequently, despite xeric conditions above-ground, EM fungi connected to Q. douglasii roots may be buffered from the effects of drought. Sustaining EM fungi in this way presumably benefits Quercus hosts because viable symbionts can remain active longer into the summer and grow quickly at the onset of winter rains.

While we did not find evidence for frequent species turnover or strong seasonal changes in the EM root community, our cluster analysis indicated that species composition was more similar within years than between years. This suggests that the EM community overall changed via ‘slow drift’ rather than ‘rapid turnover’, at least at the site and spatiotemporal scale we examined. However, despite the large number of soil cores and rDNA sequences we examined, it seems likely that sampling with increased frequency or over more years would have enhanced our view of EM temporal dynamics. When our results are viewed with the studies mentioned above, they highlight how environmental conditions, host species, experimental design, or other unknown factors might affect our understanding of temporal dynamics within EM communities. More research from diverse EM systems is needed before we can make general and definitive statements about temporal dynamics within EM fungal below-ground communities.


This work was made possible by valuable contributions of many taxonomic experts including J. Trappe, K. Hansen, B. Perry, J. Ammirati, B. Matheny, U. Köljalg, S. Dunham, M. Weiss, J. Frank, S. Miller, and R. M. Davis. We thank M. Morris, K. Huryn, R. Albright, E. Smith and F. Smith for laboratory and field assistance. UCSFREC staff provided logistical support for fieldwork, A. Fremier supplied invaluable assistance during data analysis and R. Vilgalys suggested candidate 28s primers. Helpful comments on previous drafts were provided by A. Izzo, C. Bledsoe, T. Gordon and several anonymous reviewers. Research was supported by grants to D. Rizzo by the National Science Foundation (DEB-99-81711) and to M. Smith by the Mycological Society of America and the North American Truffling Society.