This Article Abstract Full Text (PDF) Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Wilson, A. W. Articles by Desjardin, D. E. Search for Related Content PubMed Articles by Wilson, A. W. Articles by Desjardin, D. E. Agricola Articles by Wilson, A. W. Articles by Desjardin, D. E.

Phylogenetic relationships in the gymnopoid and marasmioid fungi (Basidiomycetes, euagarics clade)

     San Francisco State University, Department of Biology, 1600 Holloway Ave., San Francisco, California 94132

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

 

Three distinct lineages of gymnopoid and marasmioid fungi are recognized in parsimony and Bayesian analyses of nLSU rDNA sequences. One lineage contains the genera Lentinula, Rhodocollybia, Tetrapyrgos, a resurrected and redefined Mycetinis, and two unresolved clades designated /marasmiellus and /gymnopus. /marasmiellus includes the type species of Marasmiellus and is dominated by members of Gymnopus sect. Vestipedes. /gymnopus includes the type species of Gymnopus, Micromphale and Setulipes, and members of Gymnopus sect. Levipedes. A second lineage includes the genera Marasmius s.s. and Crinipellis and represents a redefined /marasmiaceae. A third lineage includes the genera Cylindrobasidium, Flammulina, Gloiocephala, Physalacria, Strobilurus, Xerula and Marasmius sect. Epiphylli and represents /physalacriaceae. One new combination in Rhodocollybia and four new combinations in Mycetinis are proposed. A discussion of the taxonomic implications resulting from the phylogenetic reconstruction is presented.

Key words: fungal systematics, Gymnopus, lentinuloid, Marasmiellus, Marasmius, Mycetinis, nrDNA

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

 
The marasmioid and gymnopoid (= collybioid and lentinuloid) fungi are saprotrophic or rarely plant-pathogenic euagarics (gilled mushrooms) traditionally placed in the Tricholomataceae, with smooth, white, pale yellowish white or pale pinkish white basidiospores. They are a character-rich group with a wide variety of cell types, tissue arrangements and biochemical reactions. Nonetheless generic distinctions among these fungi have been notoriously difficult and controversial. The marasmioid fungi historically were represented by species that formed marcescent (i.e. reviving in situ) basidiomes with rather tough and persistent, convex to conical, striate to sulcate pilei, adnate to adnexed lamellae, and typically tough, filiform stipes. At one time nearly all species with these features were included in the genus Marasmius (hence marasmioid). In comparison the gymnopoid fungi form relatively putrescent basidiomes with less persistent, convex and often non-striate pilei, variably attached lamellae and more robust, nonfiliform stipes. Such species were placed in the genus Collybia (hence they formerly were called the collybioid fungi). Through the years Marasmius sensu lato and Collybia sensu lato have been segregated into numerous genera based on various combinations of morphological characters. Singer (1942, 1948, 1949, 1958, 1960, 1973, 1975a, 1976a, 1986), Corner (1970), Gilliam (1976), Kühner (1980), Horak (1983, 1986), Berthier (1985), Antonín (1987), Antonín and Noordeloos (1993, 1997), Horak and Desjardin (1994), Antonín et al (1997) and Petersen (2000) have been instrumental in redefining Marasmius and the segregate genera Amyloflagellula, Aphyllotus, Campanella, Chaetocalathus, Crinipellis, Gloiocephala, Hymenogloea, Marasmiellus, Micromphale, Rhizomarasmius, Setulipes and Tetrapyrgos, including several reduced genera (Epichnaphus, Flagelloscypha, Manuripia) or clavarioid allies (Hormomitaria, Physalacria, Pseudotyphula). Halling (1983) and Antonín and Noordeloos (1993) pointed out that Collybia has remained poorly defined and contained many species better placed in allied segregate genera (listed below) or in the more distantly related genera Baeospora, Lyophyllum and Tephrocybe. Accordingly, many transfers have been made. The publications of Singer (1975b, 1986), Lennox (1979), Redhead (1980, 1987), Redhead and Ginns (1980), Halling (1983), Redhead et al (1987), Antonín et al (1997), Redhead and Petersen (1999) and Hughes et al (2001) have made major strides in redefining Collybia and the segregate genera Cyptotrama, Dendrocollybia, Flammulina, Gymnopus, Lentinula, Oudemansiella, Rhodocollybia, Strobilurus and Xerula. Not all contemporary agaricologists have followed this segregationist approach to the taxonomy of marasmioid and gymnopoid fungi. Corner (1996) noted no sharp distinctions between Marasmius, Marasmiellus and Collybia s.l., and he accepted all species with fusoid or subacerose basidioles in the genus Marasmius, describing 103 new species from Malesia. We would place Corner’s new Marasmius species in no fewer than five genera. For an overview of the current, generally accepted generic concepts of marasmioid and gymnopoid fungi based on morphological features refer to the publications of Singer (1976, 1986), Antonín and Noordeloos (1993) and Antonín et al (1997).

With the advent of molecular technologies, phylogenetic analyses of marasmioid and gymnopoid fungi based on sequences of nuclear ribosomal DNA are just beginning to help clarify generic and infra-generic circumscriptions. Owings and Desjardin (1997) and Owings (1997 unpubl master’s thesis) were the first to report that Marasmius sensu Singer (1976, 1986) was not monophyletic, based on evidence from internal transcribed spacer regions (ITS), 5.8S and nuclear ribosomal large subunit (nLSU) DNA sequences datasets. Their phylogenetic reconstructions, developed from a limited taxon sampling, reported three well supported lineages within the marasmioid/gymnopoid fungi. One strongly supported lineage (Clade 1, 100% bootstrap support), designated Marasmius sensu stricto, contained Marasmius sects. Marasmius, Sicci and Globulares. A second lineage (Clade 2, 100% BS) contained Marasmius sect. Epiphylli, the genus Strobilurus, a part of the genus Gloiocephala, and the species Marasmius pyrrhocephalus (= Rhizomarasmius pyrrhocephalus (Berk.) R.H. Petersen). A third lineage (Clade 3, 100% BS) contained Marasmius sects. Androsacei (= Setulipes), Rhizomorphigena and Alliacei (syn. sect. Chordales), and members of the genera Marasmiellus, Micromphale and Gymnopus. In an early nLSU rDNA sequences dataset analysis of 154 agaric taxa, Moncalvo et al (2000) resolved these same three clades of marasmioid/gymnopoid fungi (clades C [62% BS], D [98% BS] and A [82% BS] respectively), albeit with a different but correlative subset of taxa. In this analysis, members of the genera Cyptotrama, Flammulina, Rhodotus and Xerula were added to Clade 2 (of Owings and Desjardin 1997) and Lentinula was added to Clade 3. When describing the new genus Rhizomarasmius based on Marasmius pyrrhocephalus Berk., Petersen (2000) referred to Clade 2 as the Xerulaceae Jülich (1981) but noted that Rhodotaceae Kühner (1980) and Physalacriaceae Corner (1970) potentially were competing names for the clade. Moncalvo et al (2002) subsequently presented another nLSU rDNA sequences analysis but from a much larger sampling of euagarics and other homobasidiomycetes (877 taxa). Once again three concordant lineages of marasmioid/gymnopoid fungi were resolved, arising from an unresolved backbone of disparate euagarics. One clade, designated /marasmiaceae (= Clade 1 of Owings and Desjardin [1997]; = Clade C of Moncalvo et al [2000]) with 48% BS contained the genera Campanella, Chaetocalathus, Crinipellis, Tetrapyrgos and Marasmius sects. Marasmius, Sicci and Globulares. Clade /marasmiaceae is not the focus of this paper, although members of the group are included in the analyses and a discussion of the placement of Tetrapyrgos and Campanella are presented below. A second clade in Moncalvo et al (2002), designated /physalacriaceae (= Clade 2 of Owings and Desjardin [1997]; = Clade D of Moncalvo et al [2000]; = Xerulaceae of Petersen [2000]) with 71% BS contained the genera Armillaria, Cyptotrama, Flammulina, Gloiocephala, Oudemansiella, Physalacria, Rhizomarasmius, Rhodotus, Strobilurus and Xerula. Likewise clade /physalacriaceae is not the focus of this paper, although representatives of the clade are included in the analyses. A third clade in Moncalvo et al (2002), designated /omphalotaceae with 87% BS contained two subclades, viz. /omphalotoid (64% BS) and /lentinuloid (88% BS) (= Clade 3 of Owings and Desjardin [1997]; = Clade A of Moncalvo et al [2000]). The clade /lentinuloid is the focus of this paper. This clade represents what we call the gymnopoid fungi (Gymnopus, Lentinula, Rhodocollybia) and part of the marasmioid fungi (Marasmiellus, Micromphale, Setulipes, Marasmius sect. Alliacei). A recent molecular investigation of the /lentinuloid clade by Mata et al (2004) based on ITS (30 species) and nLSU (20 species) DNA sequences datasets indicated that the genera Gymnopus and Marasmiellus were not monophyletic. Based on their limited sampling, they accepted Marasmiellus, Micromphale and Setulipes as synonyms of Gymnopus. Our data presented herein show a somewhat different picture and allow for a different taxonomic approach. To help clarify the generic delimitations within this group, partial nLSU sequences of 69 species, including pertinent type species, were analyzed with equally weighted parsimony and Bayesian methods.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

 
Taxon sampling.— – Eighty-four partial nLSU sequences representing 69 species were used in the analyses, of which 44 were retrieved from GenBank and 40 sequences were generated for this study (TABLE I). Fifty sequences representing Gymnopus, Marasmiellus and Marasmius were used in the analyses, including their type species (G. fusipes, M. juniperinus, M. rotula, respectively) and members representing diverse infrageneric taxa. Nineteen sequences representing the genera Campanella, Lentinula, Micromphale, Rhodocollybia, Setulipes and Tetrapyrgos, of the /lentinuloid clade of Moncalvo et al (2002) also were used. Included were the type species of Lentinula (L. cubensis (Berk. & M.A. Curtis) Earle = L. boryana), Micromphale (M. foetidum), Rhodocollybia (R. maculata) and Setulipes (S. androsaceus). The type species of Campanella (C. buettneri Henn.) and Tetrapyrgos (T. atrocyanea (Métrod) E. Horak) unfortunately are poorly known and no material was available for study or sequencing. Twelve sequences representing the genera Crinipellis, Cylindrobasidium, Flammulina, Gloiocephala, Physalacria, Strobilurus, Trogia and Xerula were added to augment representation of the clades /marasmiaceae and /physalacriaceae of Moncalvo et al (2002). Callistosporium species, clearly outside the marasmioid/gymnopoid lineages (fide Moncalvo et al 2002), were chosen for rooting purposes.

PCR was performed with primers LR0R and LR7 using modified protocols outlined in Vilgalys and Hester (1990). Cycle sequencing reactions were performed with LR0R, LR7, LR3R, LR5 and LR16 primers. PCR and cycle sequencing reactions were performed on ABI 9700 (Applied Bio-systems, California) and MJ Bioworks PTC 200 DNA Engine (Incline Village, Nevada) thermalcyclers. PCR products were purified with GeneClean glassmilk (Q-BIOgene, www.qbiogene.com) or QIAGEN QIAquick PCR Purification Kit (QIAGEN Inc, Valencia, California) and run on an ABI 377 DNA Sequencer Assembly (Applied Biosystems). Sequence editing was performed with Sequencher v 3.1.1 software (GeneCodes Corp., Ann Arbor, Michigan).

Phylogenetic analyses.— – Alignment was performed with the Clustal X 1.81 (Thompson et al 1997) using the default settings, followed by manual alignment with MacClade v 4.03 (Maddison and Maddison 2001). All analyses were performed on a Macintosh G4 (733 MHz).

Equally weighted parsimony analysis was performed with PAUP* v 4.0b (Swofford 2002) with gaps treated as missing data. One hundred heuristic searches were performed with maxtrees set to auto-increase, random taxon addition sequences, with branch swapping set to TBR, and saving 5000 trees 1039 steps in length per replicate. Parsimony bootstrap analysis was performed with 1000 replicates with 10 random taxon sequence additions per replicate, maxtrees set to 100 and branch swapping set to SPR (Mort et al 2000, Salamin et al 2003).

The model of evolution that best reflects our dataset was determined to be general time reversal plus invariant rates and a gamma distribution (GTR + I + ) with MrModeltest v 1.1b (Johan A. Nylander 2002, Uppsala University, Sweden) and Modeltest v 3.06 (Posada and Crandall 1998). MrModeltest creates a NJ tree with the JC69 model to compare likelihood scores under different models of evolution. Bayesian metropolis-coupled Markov-chain Monte Carlo (MCMCMC) analysis was performed with MrBayes v 2.01 (Huelsenbeck and Ronquist 2001) with the GTR + I + model of evolution. The analysis was run with eight chains for 1 000 000 generations, saving every 100th tree. Default temperatures were used for the heated chains. The trees in the bottom 99% of the range of likelihood scores, occurring at the beginning of the Bayesian analysis, were considered part of the burn-in. All trees gathered in the MCMCMC analysis before reaching the top 1% in the range of likelihood scores were excluded. Posterior probability (PP) values were generated in PAUP* based on 50% majority rule trees.

To test the monophyly of Gymnopus, Marasmius and Marasmiellus, a Shimodaira Hasegawa test of constrained topologies was performed (Shimodaira and Hasegawa 1999). Constrained trees were created with MacClade by starting with an unresolved tree then joining all taxa for the genus under consideration into a single branch with the internal topology unresolved. A neighbor joining (NJ) tree with likelihood corrected distances (using GTR + I + model as suggested by MrModeltest) was created in PAUP* for each of the above constraints and one unconstrained tree. A comparison of the likelihood values among the resulting NJ trees was performed with the Shimodaira-Hasegawa test in PAUP* with RELL bootstrapping set to 1000.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

 
Sequence products generated for this study ranged from 877 base pairs (Gymnopus iocephalus) to 1812 bp (Trogia infundibuliformis). This latter sequence has a 413 bp intron that is between 18 and 19 bp upstream of the LR5 primer site. The shortest sequence used in the analyses (viz. 750 bp) was from Flammulina velutipes (AF042641 [GenBank] ). All sequences were aligned and the ends trimmed to create a dataset of 764 bp (TreeBase accession number S1270). The dataset contained 218 parsimony informative characters. The parsimony search found 21 083 equally parsimonious trees on three tree islands, 1036 steps long (CI = 0.361, RI = 0.727; FIG. 1). The Bayesian search converged on a stable average likelihood of –6367.75 near the 10 000th generation. The first 966 trees, trees occurring before a score of –ln L 6511, were eliminated from the analysis as the burn-in. The remaining trees were used to determine Bayesian posterior probability values (FIG. 1).

View larger version (56K): [in this window] [in a new window]   FIG. 1. Bayesian MCMCMC consensus tree using a GTR + I + G model of evolution. Analysis searched for 1 000 000 generations using eight chains. –ln L 6559.94751. Numbers in bold above branches are posterior probability values with a burn-in of 966 trees. Numbers following, in regular type, are parsimony bootstrap percentages generated from 1000 replicates with 10 random addition sequences per replicate using SPR branch swapping. Branches indicated by letters and arrows identify clades discussed in detail in the text. Taxa marked with shapes and numbers are infrageneric sections identified in the key to the left of the tree. Thick lines on branches designate clades that are resolved in the parsimony strict consensus of 21 083 trees.

Two well resolved clades are the sister group of the clade representing letters A–G. Clade H, supported on a branch with 1.0 PP, 69% BS values and resolved in the parsimony strict consensus tree, corresponds to /marasmioid of Moncalvo et al (2002). This clade includes Marasmius sensu stricto (sects. Globulares, Marasmius, Sicci), Crinipellis and the species Marasmiellus palmivorus. The sister group of this clade is Trogia infundibuliformis on a long branch (0.6 PP). The second of the two well resolved groups represents clade /physalacriaceae of Moncalvo et al (2002). It is supported on a branch with 1.0 PP, 99% BS values and is resolved in the parsimony strict consensus tree. The genera Cylindrobasidium, Flammulina, Gloiocephala, Physalacria, Strobilurus and Xerula are represented in this clade.

The Shimodaira-Hasegawa (SH) test was used to compare constrained neighbor joining trees that force monophyly of Gymnopus, Marasmius, and Marasmiellus to an unconstrained NJ tree. All constrained trees were found to be significantly worse than the unconstrained tree (–ln L 7190.96051). The Gymnopus constrained tree is the most similar to the unconstrained tree with a difference in the log likelihood score of 132.248, but it is rejected in the SH-test (P = 0.004). Marasmius is also rejected having the second highest likelihood score with a difference of 157.98052 compared to the unconstrained tree. Marasmiellus has the least significance as a clade with a difference of 307.09981 in the log likelihood score compared to the unconstrained tree. Both Marasmius and Marasmiellus received a P score below 0.001.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

 
The large and morphologically diverse genera Marasmius (with an estimated >1000 species), Gymnopus (estimated >300 spp.) and Marasmiellus (estimated >400 spp.), as currently circumscribed with morphological characters, are not monophyletic in this analysis. Constraining these genera to monophyly produced significantly worse trees. Three major lineages of marasmioid and gymnopoid fungi were resolved with strong statistical support (viz. Clades A, H, I), corresponding roughly to the clades /lentinuloid, /marasmioid and /physalacriaceae respectively of Moncalvo et al (2002). One significant difference was in the placement of clade /tetrapyrgos. In our analyses, /tetrapyrgos formed a discrete clade (1.0PP, 99% BS) that was sister of /lentinuloid (Clade A) with 0.84 PP, whereas in the Moncalvo et al (2002) analysis, /tetrapyrgos (along with the genus Campanella forming clade /tetrapyrgoid) was sister of /marasmioid with relatively low bootstrap support (48%) forming collectively a clade designated /marasmiaceae. Our data suggest that if a family rank were to be accepted for /marasmiaceae, for which the name Marasmiaceae Roze ex Kühner (1980) is available, it would be restricted to the genera Marasmius sensu stricto, Crinipellis and Chaetocalathus, excluding Tetrapyrgos and Campanella. (A discussion of the significant clades in FIG. 1 is provided below).

Clade A.— – This major clade contains the genera Gymnopus, Lentinula, Marasmiellus p.p., Micromphale, Rhodocollybia and Setulipes and corresponds to /lentinuloid of Moncalvo et al (2002) and Mata et al (2004), herein with the addition of Campanella. This clade, combined with Clade G (/tetrapyrgos) and /omphalotoid (of Moncalvo et al 2002) may represent a distinct family of euagarics, wherein the name Omphalotaceae Bresinsky in Kämmerer et al (1985) is available.

Clade B.— – This is dominated by members of Gymnopus sect. Vestipedes and includes Marasmiellus juniperinus, type species of Marasmiellus, and Marasmiellus synodicus (in sect. Dealbati). Because the type of Marasmiellus belongs to this clade, we have designated it /marasmiellus. This clade unfortunately lacks strong support from parsimony strict consensus, posterior probability and bootstrap statistics. Nonetheless morphologically there are some diagnostic features linking representatives of this clade. All members included in this analysis form basidiomes with noninsititious to subinsititious stipe, form unpigmented, acyanophilous, inamyloid and nondextrinoid basidiospores, and form pileipelli composed of a cutis of radially arranged cylindrical hyphae that are nondiverticulate or weakly diverticulate and typically are roughened or covered with annular to zebroid, brownish pigment incrustations (refer to Singer [1973], Desjardin et al [1999], Mata et al [2004] and Wilson et al [2004] for descriptions of species included). The pileipelli are not a well developed Rameales structure of strongly diverticulate hyphae as in Marasmiellus ramealis (sect. Rameales) and many (most?) other Marasmiellus species. Singer (1973, 1975b, 1986) had a rather broad concept of Marasmiellus, one that included pleurotoid and centrally stipitate basidiomes with subinsititious or insititious stipes and pileipelli ranging from nondiverticulate or weakly diverticulate hyphae (sects. Candidi, Dealbati, Tetrachroi) to strongly diverticulate hyphae (= a Rameales-structure; sects. Distantifolia, Rameales, Tricolores). In addition Singer included species with pileo-setae (sect. Stenophylloides) and several heterogeneous sections based on homoplasic characters (e.g. sect. Defibulati lacking clamp connections, sect. Marasmiellus with pleurotoid basidiomes). The primary character distinguishing Marasmiellus from Gymnopus in modern taxonomy is the development of an insititious stipe in the former versus well developed basal mycelium in the latter. Our sequence data suggest that this character has limited phylogenetic significance. The inclusion of many more Marasmiellus species (from sects. Marasmiellus, Dealbati and Tetrachroi sensu Singer [1973]) and Gymnopus species (from sect. Vestipedes and the subsumed sect. Subfumosae sensu Halling [1983]) exhibiting the diagnostic features itemized above may help to resolve this clade and allow for a redefined genus Marasmiellus. Without further analyses of a larger sample of taxa, we do not support acceptance of Marasmiellus (with >400 spp.) as a synonym of Gymnopus (with >300 spp.) as proposed by Mata et al (2004).

Clade C.— – This clade, strongly supported with Bayesian analysis (1.0 PP) and resolved in the parsimony strict consensus tree, represents the genus Rhodocollybia (four species including the type) and Gymnopus spissus. Rhodocollybia is distinguished from all other marasmioid and gymnopoid fungi in forming basidiospores that are cyanophilic, dextrinoid (or a portion of the spores in a deposit are dextrinoid), and are pink to pale pinkish white in deposit. The inclusion of G. spissus on a long branch sister of R. badiialba (0.99 PP) precipitated a re-examination of the isotype specimen of G. spissus (SFSU!). Further analysis indicated that the basidiospores of the latter species are indeed cyanophilic and a few are weakly dextrinoid (after 10 min in Melzer’s reagent), features that were reported incorrectly as acyanophilic and inamyloid in the protologue (Wilson et al 2004). In addition a pileipellis of radially arranged, repent, nondiverticulate, smooth hyphae, in combination with the cyanophilic and dextrinoid basidiospores, are characters diagnostic for Rhodocollybia. G. spissus accordingly is better placed in Rhodocollybia and a formal transfer is proposed below.

Of note, Marasmiellus ramealis plus Campanella eberhardtii formed a clade sister of Rhodocollybia (0.71 PP), although no statistical support joined M. ramealis with C. eberhardtii. Both of the latter species form pileipelli composed of a well developed Rameales structure, distinct from the type of pileipelli formed by members of /marasmiellus. In the analysis of Moncalvo et al (2002), several undetermined Campanella species were sister of Tetrapyrgos in clade /marasmiaceae/tetrapyrgoid. Additional accurately identified species of Campanella clearly must be included in future molecular analyses before its placement in phylogenetic reconstructions can be clarified. Moreover only a single representative of Marasmiellus sect. Rameales was included in this analysis. If additional species with a well developed Rameales structure are added to future analyses, and if they cluster with M. ramealis (discrete from /marasmiellus), the generic name Collybiopsis ( J. Schröt.) Earle (1909:415; type: Agaricus ramealis Bull. : Fr. = Marasmiellus ramealis) may be available for this clade (see Note 1 below).

The placement of a clade containing G. peronatus and G. termiticola (0.97 PP) as part of an unresolved trichotomy along with /marasmiellus and Rhodocollybia + M. ramealis + C. eberhardtii is difficult to explain. Gymnopus peronatus and G. termiticola share few morphological features in common. Gymnopus pernonatus, a north temperate species currently placed in sect. Vestipedes, shares morphology in common with members of /marasmiellus (cf. Antonín and Noordeloos 1997), whereas G. termiticola, known at present only from southeastern Asia, exhibits unusual micromorphology (e.g. large pleurocystidia) and formed a clade distinct from sects. Vestipedes and Levipedes in a recent phylogenetic analysis of Indonesian Gymnopus (Wilson et al 2004). It should be noted that the sequence of G. peronatus in our analysis was from GenBank and we have not examined the voucher specimen or culture from which the sequence was taken so we cannot verify species determination. Further research is warranted to address the apparent incongruence.

Clade D.— – This clade, strongly supported by Bayesian analysis (1.0 PP) and resolved in the parsimony strict consensus tree, contains the type species of Gymnopus, Micromphale and Setulipes. This clade is roughly equivalent to clade /micromphale of Moncalvo et al (2002) and Clade 2 in the nLSU tree of Mata et al (2004, FIG. 1). We have designated this clade /gymnopus because it includes the type species of the oldest available generic name for the group. The clade is dominated by members of Gymnopus sect. Levipedes that share the anatomical feature of pileipelli composed of relatively short, broad, branched, nondiverticulate hyphae forming a Dryophila-structure (cf. Halling 1983, Antonín and Noordeloos 1997). The type species of Gymnopus, G. fusipes, also shares this anatomical feature. Other than developing a rooted stipe and pale pinkish basidiospores, nothing distinguishes G. fusipes (sect. Gymnopus) from members of sect. Levipedes, so their alignment (FIG. 1) is not surprising. The morphology of Micromphale and Setulipes, however, is quite distinct from that of Gymnopus species and does not suggest close relatedness. The former two genera form small, marcescent basidiomes with wiry insititious stipes often accompanied by black wiry rhizomorphs and pileipelli with brown pigment-incrusted hyphae that are nondiverticulate and gelatinized (Micromphale) or diverticulate-knobby and typically nongelatinized (Setulipes). Mata et al (2004, FIG. 1) documented similar results from a smaller sampling of taxa, and based on strong posterior probability (1.0 PP) and bootstrap (70%) values they accepted Micromphale and Setulipes as synonyms of Gymnopus and accordingly transferred their type species into Gymnopus. It should be noted that Setulipes (= Marasmius sect. Androsacei) contains ca. 35 species worldwide and only a single species has been included in any published phylogenetic analyses. How the addition of numerous Setulipes spp. would affect the topology of /gymnopus (or /lentinuloid) is unknown. In our analysis Micromphale perforans was sister of Clade D (0.68 PP) although the branch was not resolved in the parsimony strict consensus tree. We accept the species as belonging to /gymnopus.

Clade E.— – This strongly supported clade represents the genus Lentinula. We have included four species, including the type, and collectively they form a distinct clade sister of /gymnopus (0.76 PP). Our data suggest that this group represents a distinct lineage that we recognize at the genus rank.

Clade F.— – The type species of Marasmius sect. Alliacei, along with three other members of the section and Marasmiellus opacus (currently placed in sect. Rameales) constitute this clade; it is well supported in Bayesian and parsimony analyses. Earle (1909) elevated Schröter’s (1889:558) Marasmius sect. Mycenopsis ("this includes Marasmius Mycena, subsect. Chordales, of the Sylloge" [Earle 1909:414]) to generic rank and called it Mycetinis with M. alliaceus as type species. Donk (1962:193), for reasons similar to those presented for Collybiopsis (see below), had reservations about Earle’s choice of type species. Nonetheless Earle (1909) validly established a new genus based on M. alliaceus, which we consider to correspond to Marasmius sect. Alliacei (sensu Singer 1976, 1986; = sect. Alliateae pro parte Kühner 1933; = sect. Chordales pro parte sensu Gilliam [1975a] and Antonín and Noordeloos [1993]) and to our Clade F. Members of this section share these anatomical features: subinsititious to noninsititious stipes; inamyloid and nondextrinoid, acyanophilic, white or pale cream basidiospores; nondextrinoid tissues; and pileipelli composed of a hymeniform layer of smooth (nondiverticulate), clavate or lobed cells. In addition the majority of species in the section produce a strong garlic-like (alliaceous) or rotten cabbage-like odor and taste, due to the presence of -glutamyl marasmine (Gmelin et al 1976). At first glance, the appearance of Marasmiellus opacus in this clade seems unusual. In a redescription of the species, Desjardin et al (1993) noted that M. opacus forms inamyloid and nondextrinoid, white basidiospores, nondextrinoid tissues, and forms a pileipellis composed of a subhy-meniform layer of clavate to lobed, often thick-walled cells. These features are diagnostic for Marasmius sect. Alliacei. They chose to accept the species in Mar-asmiellus because at maturity the pileus margin shows a Rameales-structure, the stipe base is insititious, and because the species develops abundant rhizomorphs, features more characteristic of Marasmiellus species. The phylogenetic affinity of M. opacus with M. sco-rodonius as inferred from nLSU sequences of Moncalvo et al (2002, the same sequences were used by Mata et al 2004 and this analysis) suggest that M. opacus is better placed in a taxon with members of Marasmius sect. Alliacei. Pileipellis anatomy hints at this placement. Statistical support for this lineage (FIG. 1) is as strong as that for Rhodocollybia, Lentinula, Crinipellis, Gloiocephala, Physalacria and Xerula, genera that currently are accepted as distinct by contemporary agaricologists. Consequently we accept this lineage to represent a distinct genus and we resurrect the name Mycetinis for the group. A generic circumscription and formal transfers that result are presented below. Mata et al (2004) chose a different approach and accepted M. opacus and M. scorodonius as belonging to a more inclusive Gymnopus and made formal transfers accordingly.

Clades D (/gymnopus), E (Lentinula) and F (Mycetinis) form a clade with 0.9 PP. Many members of this clade produce basidiomes with garlic-like, cabbage-like, or otherwise sulfurous odors and flavors. Gmelin et al (1976) reported a new natural dipeptide with a cysteine sulfoxide moiety (-glutamyl marasmine) that was responsible for the garlic odor in M. alliaceus, M. scorodonius and M. prasiosmus (all members of sect. Alliacei). These same odors and flavors are found in Micromphale foetidum, M. perforans, G. iocephalus, G. polyphyllus and several Setulipes spp. (clade /gymnopus), in Marasmius applanatipes, M. copelandii and other allied species (clade Mycetinis), and are present as an aftertaste in the shiitake mushroom (Lentinula edodes). The presence of such substances may represent an apomorphic character for this lineage.

Clade G.— – Three species of Tetrapyrgos plus Marasmiellus candidus constitute this well supported clade that is sister of /lentinuloid (Clade A) with 0.84 PP. The genus Tetrapyrgos is characterized by basidiomes with convex, rugulose-sulcate pilei-colored white or pale grayish, adnexed to adnate lamellae that are often strongly intervenose, and a central to eccentric stipe (sometimes absent) that is subinsititious to insititious, pruinose, with a black base. Distinctive micromorphology includes white, inamyloid and thin-walled basidiospores that are tetrahedral or with a conspicuous lateral bulge, cheilocystidia and pileocystidia that are diverticulate and often with a capitate terminus, conspicuous pleurocystidia (in some species), and a pileipellis of densely diverticulate hyphae. Although once included in Marasmiellus (Singer 1973), the basidiospore and cystidial morphologies of Tetrapyrgos are distinctive and have warranted acceptance as a discrete genus (Horak 1983, 1986). Our data support this contention. The inclusion of M. candidus in the clade is a bit odd, but it should be noted that this species, type of sect. Candidi, and others in the section are morphologically distinct from other Marasmiellus species and share many features with Tetrapyrgos. Section Candidi is characterized by these features: convex, rugulose-sulcate pilei-colored white to pale fuscous; adnexed to adnate lamellae that often are strongly intervenose; a central to eccentric stipe that is subinsititious to insititious, pruinose, usually with a black base; white, inamyloid and thin-walled basidiospores that are fusoid to elongate-ellipsoid; cheilocystidia and pleurocystidia that are elongate-fusoid and nondiverticulate; and a pileipellis of smooth or weakly diverticulate hyphae with elongate-fusoid, nondiverticulate pileocystidia. The elongate-fusoid pleurocystidia and pileocystidia are unique in Marasmiellus. A quick comparison of these features with those of Tetrapyrgos indicates many shared characters and differences primarily in the shape of basidiospores, cystidia and pileipellis hyphae. Of note, herbarium-deposited North American material of M. candidus often have been misdetermined as Tetrapyrgos nigripes (Desjardin pers obs). We reserve judgment on whether Marasmiellus sect. Candidi belongs to a redefined Tetrapyrgos or represents a distinct lineage until further species of the group are added to the analyses.

Clades H and I.— – These two lineages are not the primary focus of this paper so few taxa were included in the analyses. Clade H corresponds to /marasmioid of Moncalvo et al (2002) and includes what we accept as Marasmius sensu stricto (sects. Marasmius, Globulares, Sicci) and the allied genus Crinipellis. This clade is the focus of ongoing research, wherein we also include in this group Marasmius sects. Hygrometrici, Neosessiles and Leveilleani, and the genus Chaetocalathus (Desjardin unpubl). The inclusion of Marasmiellus palmivorus in the clade, a tropical Old World species similar in morphology to members of sect. Candidi, cannot be explained at present. We included one species of Trogia (viz. T. infundibuliformis) in the analyses in an attempt to clarify its affinities with other marasmioid and gymnopoid fungi. Corner (1966, 1991) included a heterogeneous assemblage of species in the genus (cf. comments by Singer 1975b:304 and Redhead 1987), which was circumscribed primarily on the basis of sarcodimitic tissues. The type species of Trogia, T. montagnei Fr. from India, unfortunately remains an understudied entity with an untraceable type specimen. Until the type species is well characterized, the taxonomic boundaries of Trogia will remain unknown. In our analyses T. infundibuliformis, a species phenetically similar to the protologue of T. montagnei, was sister of /marasmioid on a long branch.

Clade I corresponds to /physalacriaceae of Moncalvo et al (2002), although our sampling of taxa are different. We have included the type species of Gloiocephala (G. epiphylla), additional Physalacria spp., and the resupinate Cylindrobasidium laeve (as suggested by Bodensteiner et al 2004). Research by Owings and Desjardin (1997) indicates that Marasmius sect. Epiphylli also belongs to this clade, while research by Moncalvo et al (2000, 2002) suggests the inclusion of the genera Cyptotrama, Oudemansiella, Rhizomarasmius, Rhodotus and possibly Armillaria.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

 
Our data support those of Owings and Desjardin (1997) and Moncalvo et al (2000, 2004) in recognizing three distinct lineages of marasmioid and gymnopoid fungi. One lineage includes what we recognize as the distinct genera Lentinula, Rhodocollybia, Tetrapyrgos, a resurrected Mycetinis and two unresolved clades designated /marasmiellus (including the type species of Marasmiellus) and /gymnopus (including the type species of Gymnopus, Micromphale and Setulipes). Until further species belonging to Gymnopus sects. Vestipedes and Levipedes, Marasmiellus (all sections) and Setulipes are added to the analyses we do not support recognition of a more inclusive Gymnopus (the oldest available epithet for species included in clades A–G). Our data suggest that if one accepts Marasmiellus Micromphale and Setulipes as synonyms of Gymnopus, as was done by Mata et al (2004), then Rhodocollybia, Lentinula and Mycetinis also would have to be accepted as synonyms for a monophyletic Gymnopus. We think this approach is unwarranted at present. We recognize two additional lineages: one corresponding to a restricted /marasmiaceae including the genera Marasmius s.s. and Crinipellis and one corresponding to /physalacriaceae including the genera Cylindrobasidium, Flammulina, Gloiocephala, Physalacria, Strobilurus and Xerula, and with the addition of Marasmius sect. Epiphylli.

Taxonomic considerations:

Note 1. Earle (1909:415) designated Agaricus ramealis as type species for the genus Collybiopsis. Donk (1962) proposed an alternative type for Collybiopsis (viz. Agaricus calopus Pers. : Fr.). Donk (1962:67–69) reported that if A. ramealis and A. calopus were accepted as belonging to the same taxonomic group (i.e. Marasmius sect. Collybia subsect. Calopodes of Fries [1838:379]; = Marasmius subgen. Collybiopsis of Schröter [1889:559] elevated to generic rank by Earle [1909]) then A. calopus was a better selection for type species of the group (because Fries explicitly named his subsect. Calopodes after it). Unfortunately, most contemporary agaricologists accept A. calopus as either a nomen dubium (cf. Antonín and Noordeloos 1993:183) or as representing an odorless form of Marasmius scorodonius. One could argue that Earle’s (1909) choice of A. ramealis for the type of Collybiopsis (a well defined species that was listed by Schröter in his Marasmius subgen. Collybiopsis), being the first species selected as type for the group, is equally valid and a better choice than a nomen dubium. Collybiopsis then would be available for species allied with M. ramealis.

Mycetinis Earle, Bull. New York Bot Gard 5:414. 1909. Type species: Mycetinis alliaceus ( Jacq. : Fr.) Earle (basionym: Agaricus alliaceus Jacq. : Fr., Syst Mycol 1:140. 1821). Basidiomes reviving or not. Pileus convex to campanulate, margin decurved, even to rugulose-striate, glabrous to suede-like. Lamellae free to adnexed or adnate. Stipe central, glabrous to pubescent, insititious to subinsititious or with basal mycelium; rhizomorphs occasionally present. Odor and flavor not distinctive or garlic-like to cabbage-like. Basidiospores smooth, hyaline, inamyloid and nondextrinoid, acyanophilic, thin-walled. Basidioles clavate to subfusoid. Cheilocystidia present, cylindrical to fusoid or clavate, often diverticulate, not subcapitate or capitate. Pleurocystidia absent or rarely present and then similar to cheilocystidia. Tramal tissues non-dextrinoid. Pileipellis subhymeniform to hymeniform (at least when young), composed of sphaeropedunculate, clavate, subcylindrical or irregular cells, sometimes lobed or with broad knobs, thin-walled to thick-walled, non-dextrinoid; pileocystidia absent. Clamp connections present.

Mycetinis applanatipes (Desjardin) A.W. Wilson and Desjardin, comb. nov. (basionym: Marasmius applanatipes Desjardin. Mycologia 77:899. 1985).

Mycetinis copelandii (Peck) A.W. Wilson and Desjardin, comb. nov. (basionym: Marasmius copelandii Peck. Bull. Torrey Bot Club 31:182. 1904).

Mycetinis opacus (Berk. & M.A. Curtis) A.W. Wilson and Desjardin, comb. nov. (basionym: Marasmius opacus Berk. & M.A. Curtis. Hooker J Bot 1:99. 1849).

Mycetinis scorodonius (Fr. : Fr.) A.W. Wilson and Desjardin, comb. nov. (basionym: Agaricus scorodonius Fr. : Fr. Syst Mycol 1:130. 1821).

Rhodocollybia spissa (A.W. Wilson, Desjardin & E. Horak) A.W. Wilson & Desjardin, comb. nov. (basionym: Gymnopus spissus A.W. Wilson, Desjardin & E. Horak. Sydowia 56:204. 2004).

	ACKNOWLEDGMENTS

 
We thank Drs David Hibbett and Manfred Binder for the use of unpublished nLSU-rDNA sequences, Drs Ron Petersen and Karen Hughes for early use of their now published sequences, and Drs Roy Halling and Vladimir Antonín for contributing herbarium material. A special thanks to Dr David Hibbett for allowing the use of his laboratory facilities in the completion of this study. We are grateful to Drs David Hibbett, Brandon Matheny and Manfred Binder for their critical review of the manuscript before submission. We thank the Mycological Society of San Francisco for the Harry D. Thiers Scholarship that helped finance part of this study. This research was financed in part by National Science Foundation grants DEB-9705083 and DEB-0118776 to Desjardin.

	FOOTNOTES

 
Accepted for publication January 24, 2005.

¹ Current address: Clark University Biology, 950 Main St., Worcester, MA 01610. E-mail: anwilson{at}clarku.edu

² Corresponding author. E-mail: ded{at}sfsu.edu

	LITERATURE CITED

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS LITERATURE CITED

 
Antonín V. 1987. Setulipes, a new genus of marasmioid fungi (Tricholomatales). Ceska Mycol 41:85–87.

———, Halling RE, Noordeloos ME. 1997. Generic concepts within the groups of Marasmius and Collybia sensu lato. Mycotaxon 63:359–368.

———, Noordeloos ME. 1993. A monograph of Marasmius, Collybia and related genera in Europe. Part 1: Marasmius, Setulipes, and Marasmiellus. Lib Bot 8:1–229.

———, ———. 1997. A monograph of Marasmius, Collybia and related genera in Europe. Part 2: Collybia, Gymnopus, Rhodocollybia, Crinipellis, Chaetocalathus, and additions to Marasmiellus. Lib Bot 17:1–256.

Berthier J. 1985. Les Physalacriaceae du globe. Biblio Mycol 98:1–128.

Bodensteiner P, Binder M, Moncalvo J-M, Agerer R, Hibbett DS. 2004. Phylogenetic relationships of cyphelloid homobasidiomycetes. Mol Phylogen Evol 33:501–515.[CrossRef][Medline]

Corner EJH. 1966. A monograph of cantharelloid fungi. England: Oxford University Press. 255 p.

———. 1970. Supplement to "A monograph of Clavaria and allied genera." Beih Nova Hedwigia 33:1–299.

———. 1991. Trogia (Basidiomycetes). Singapore Botanic Gardens, Singapore. 100 p.

———. 1996. The agaric genera Marasmius, Chaetocalathus, Crinipellis, Heimiomyces, Resupinatus, Xerula, and Xerulina in Malesia. Beih Nova Hedwigia 111:1–175.

Desjardin DE, Gordon SA, Petersen RH. 1993. Observations on two rhizomorph-forming species of Marasmiellus. Mycol Res 97:111–122.

———, Halling RE, Hemmes DE. 1999. Agaricales of the Hawaiian Islands. 5: the genera Rhodocollybia and Gymnopus. Mycologia 91:166–177.[CrossRef]

Donk MA. 1962. The generic names proposed for Agaricaceae. Beih Nova Hedwigia 5:1–317.

Earle FS. 1909. The genera of North American gill fungi. Bull New York Bot Gard 5:373–451.

Fries EM. 1838. Epicrisis systematis mycologici. Upsala, Sweden. 610 p.

Gilliam MS. 1975. Marasmius section Chordales in the northeastern United States and adjacent Canada. Contr Univ Michigan Herb 11:25–40.

———. 1976. The genus Marasmius in the northeastern United States and adjacent Canada. Mycotaxon 4:1–144.

Gmelin R, Luxa H-H, Roth K, Höfle G. 1976. Dipeptide precursor of garlic odour in Marasmius species. Phytochemistry 15:1717–1721.[CrossRef]

Halling RE. 1983. The genus Collybia (Agaricales) in the northeastern United States and adjacent Canada. Mycol Mem 8:1–148.

Horak E. 1983. Neufunde und Bemerkungen zu einem emendierten Gattungskonzept von Pterospora Métrod (Agaricales). Sydowia 36:125–138.

———. 1986. Tetrapyrgos Horak (nom. et gen. nov.) replacing Pterospora Métrod (1949; nom. preocc.). Sydowia 39:101–103.

———, Desjardin DE. 1994. Reduced marasmioid and mycenoid agarics from Australasia. Australian J Syst Bot 7:153–170.[CrossRef]

Huelsenbeck JP, Ronquist F. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17:754–755.[Abstract/Free Full Text]

Hughes KW, Petersen RH, Johnson JE, Moncalvo J-M, Vilgalys R, Redhead SA, Thomas T, McGhee LL. 2001. Infragenic phylogeny of Collybia s. str., based on sequences of ribosomal ITS and LSU regions. Mycol Res 105:164–172.[CrossRef]

Jülich W. 1981. Higher taxa of Basidiomycetes. Biblio Mycol 85:1–485.

Kämmerer A, Besl H, Bresinsky A. 1985. Omphalotaceae fam. nov. und Paxillaceae, ein chemotaxonomischer Vergleich zwier Pilzfamilien der Boletales. Plant Syst Evol 150(1–2):101–117.[CrossRef]

Kühner R. 1933. Etudes sur le genre Marasmius. Botaniste 25:57–116.

———. 1980. Les hyménomycetes agaricoides (Agaricales, Trichomatales, Pluteales, Russulales). Bull Soc Linn Lyon 49 (num spec):1–1027.

Lennox JW. 1979. Collybioid genera in the Pacific Northwest. Mycotaxon 9:117–231.

Maddison DR, Maddison WP. 2001. MacClade 4, version 4.03. Sunderland, Massachusetts, USA: Sinaur.

Mata JL, Hughes KW, Petersen RH. 2004. Phylogenetic placement of Marasmiellus juniperinus. Mycoscience 45:214–221.

Moncalvo J-M, Lutzoni FM, Rehner SA, Johnson J, Vilgalys R. 2000. Phylogenetic relationships of agaric fungi based on nuclear large subunit ribosomal DNA sequences. Syst Biol 49:278–305.[CrossRef][Medline]

———, Vilgalys R, Redhead SA, Johnson JE, James TY, Aime MC, Hofstetter V, Verduin SJW, Larsson E, Baroni TJ, Thorn RG, Jacobsson S, Clémençon H, Miller Jr OK. 2002. One hundred and seventeen clades of euagarics. Mol Phylogen Evol 23:357–400.[CrossRef][Medline]

Mort ME, Soltis PS, Soltis DE, Mabry ML. 2000. Comparison of three methods for estimating internal support on phylogenetic trees. Syst Biol 49:160–171.[CrossRef][Medline]

Owings P. 1997. Evolutionary relationships within the genus Marasmius inferred by morphological and nrDNA sequence comparisons [Master’s thesis]. San Francisco, California: San Francisco State University. 92 p.

———, Desjardin DE. 1997. A molecular phylogeny of Marasmius and selected segregate genera. Inoculum 48(3): 29.

Petersen RH. 2000. Rhizomarasmius, gen. nov. (Xerulaceae, Agaricales). Mycotaxon 75:333–342.

Posada D, Crandall KA. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14(9):817–818.[Abstract/Free Full Text]

Redhead SA. 1980. The genus Strobilurus (Agaricales) in Canada with notes on extralimital species. Can J Bot 58:68–83.

———. 1987. The Xerulaceae (Basidiomycetes), a family with sarcodimitic tissues. Can J Bot 65:1551–1562.

———, Ginns J. 1980. Cyptotrama asprata (Agaricales) from North America and notes on the five other species of Cyptotrama sect. Xerulina. Can J Bot 58:731–740.

———, ———, Shoemaker RA. 1987. The Xerula (Collybia, Oudemansiella) radicata complex in Canada. Mycotaxon 30:357–405.

———, Petersen RH. 1999. New species, varieties and combinations in the genus Flammulina. Mycotaxon 71:285–294.

Rogers SO, Bendich AJ. 1994. Extraction of total cellular DNA from plants, algae and fungi. D1:1–8. In: Gelvin SB, Schilperoort RA, eds. Plant Molecular Biology Manual. 3rd ed. Dordrcht, The Netherlands: Kluwer Academic Press.

Salamin N, Mark CW, Hodkinson TR, Savolainen V. 2003. Assessing internal support with large phylogenetic DNA matrices. Mol Phylogen Evol 27:528–539.[CrossRef][Medline]

Schröter J. 1889. Pilze: Agaricacei. In Cohn: Kryptogamenflora von Schlesien 3:513–684.

Shimodaira H, Hasegawa M. 1999. Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Mol Biol Evol 16(8):1114–1116.

Singer R. 1942. A monographic study of the genera Crinipellis and Chaetocalathus. Lilloa 8:441–534.

———. 1948. Diagnoses fungorum novorum Agaricalium. Sydowia 2:26–42.

———. 1949 (1951). The Agaricales (mushrooms) in modern taxonomy. Lilloa 22:1–832.

———. 1958. New genera of fungi. VIII. Notes concerning the sections of the genus Marasmius. Mycologia 50:103–110.[CrossRef]

———. 1960. Monographs of South American Basidiomycetes, especially those of the east slope of the Andes and Brazil. 3. Reduced marasmioid genera in South America. Sydowia 14:258–280.

———. 1973. The genera Marasmiellus, Crepidotus and Simocybe in the Neotropics. Beih Nova Hedwigia 44:1–339.

———. 1975a. The neotropical species of Campanella and Aphyllotus. Nova Hedwigia 26:847–895.

———. 1975b. The Agaricales in modern taxonomy. 3rd ed. J. Cramer, Vaduz. 912 p.

———. 1976. Marasmieae (Basidiomycetes—Tricholomataceae). Fl Neotropica 17:1–348.

———. 1986. The Agaricales in modern taxonomy. 4th ed. Koeltz Scientific Books, Germany. 981 p.

Swofford DL. 2002. PAUP*. Phylogenetic Analysis Using Parisimony (*and other methods), version 4. Sunderland, Massachusetts: Sinauer.

Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. 1997. The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl Acid Res 24:4876–4882.

Vilgalys R, Hester M. 1990. Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. J Bact 172(8):4238–4246.[Abstract/Free Full Text]

Wilson AW, Desjardin DE, Horak E. 2004. Agaricales of Indonesia. 5. The genus Gymnopus from Java and Bali. Sydowia 56:137–210.

This Article Abstract Full Text (PDF) Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Wilson, A. W. Articles by Desjardin, D. E. Search for Related Content PubMed Articles by Wilson, A. W. Articles by Desjardin, D. E. Agricola Articles by Wilson, A. W. Articles by Desjardin, D. E.