Taxonomy and evolutionary relationships within species of section Rimosae (Inocybe) based on ITS, LSU and mtSSU sequence data.

The present study aimed at elucidating the structure of Inocybe subg. Inosperma sect. Rimosae but included also representatives from subg. Mallocybe and the genus Auritella. Phylogenetic relationships were inferred using ITS, LSU and mtSSU sequence data. The analyses recovered the ingroup as a monophyletic, strongly supported clade. The results indicate that recognizing Auritella on the genus level renders Inocybe paraphyletic. The species traditionally placed in sect. Rimosae were found to be distributed over two strongly supported clades, Maculata and Rimosae s.s. The Maculata clade clusters with sect. Cervicolores and the two represent subg. Inosperma in a strict sense. Rimosae s.s. emerges as an independent, supported clade well separated from Inosperma s.s. Twenty-one terminal groups were correlated with morphologically distinct species. In addition several taxa on single branches and minor less supported clades were recovered. A key to the identified species of the Maculata and Rimosae s.s. clades which occur in Northwest Europe is provided.

INTRODUCTION

Inocybe is a large genus of agaric fungi with an estimated 500 species world wide (Kirk et al. 2008), a number that is likely to increase considerably when tropical and southern temperate areas are more intensively explored. Intrageneric classifications have been based mainly on spore morphology, the form and distribution of cystidia, and stipe morphology. The spores may be ellipsoid, amygdaliform or nodulose/angular. Many species have incrusted thick-walled pleuro- and cheilocystidia (metuloids). Some large groups completely lack the metuloids but then have numerous thin-walled cheilocystidia. The stipe may be of uniform thickness or have a distinctly bulbous base. A number of classifications combining these and other characters in various ways have been proposed (Heim 1931, Kühner & Romagnesi 1953, Kühner 1980, Kuyper 1986, Singer 1986, Stangl 1989, Kobayashi 2002).

Several phylogenetic analyses of Inocybe using both ribosomal and protein coding genes have been published (Matheny et al. 2002, Matheny 2005, Matheny & Bougher 2006a, Matheny et al. 2009). These studies confirm that Inocybe is monophyletic. In a multi-gene phylogeny of Agaricomycotina Matheny et al. (2006) showed that Inocybe does not belong in Cortinariaceae, where it has traditionally been placed, but has affinities to Crepidotaceae. Matheny (2005) suggested that Inocybe should be recognized at the family level as Inocybeaceae, a family already proposed and described by Jülich (1982).

Matheny (2005) identified five clades within Inocybaceae, which he called Inocybe, Inosperma, Pseudosperma, Mallocybe and Auritella. The Inocybe clade holds the generic type species and includes all species with incrusted cystidia (metuloids) irrespective of spore shape. Mallocybe, recognized as a separate subgenus (Kuyper 1986), includes species with necropigmented basidia and thin-walled cheilocystidia originating from the subhymenium. Auritella was recently separated from Inocybe as an independent genus (Matheny & Bougher 2006a, b) and seems to represent a unique Paleotropical and Southern hemisphere lineage with species from Australia and Africa. The two clades Inosperma and Pseudosperma basically include the species in sections Rimosae and Cervicolores in the classification by Kuyper (1986). Pseudosperma was introduced as a clade name only and not formally assigned any classification status according to ICBN (McNeill et al. 2006).

Section Rimosae is in traditional classifications placed in subg. Inosperma (Kühner 1980, Kuyper 1986, Stangl 1989). The section includes species characterized by radially fibrillose to rimose (squamulose) caps, ellipsoid to phaseoliform spores, and absence of metuloid pleurocystidia but with densely packed, simple, cylindrical, clavate to pyriform hymenial cheilocystidia that make the gill edge look distinctly white in mature specimens. Other characters that may occur are a distinctly bulbous stem base, reddening flesh, yellow to olivaceous tinges on lamellae, and specific odours. Phylogenetic analyses in a recent publication on the biogeography of Inocybaceae (Matheny et al. 2009) included a broader sampling of sect. Rimosae than any previous study, although representatives from Europe were still few. The results indicated that the section is non-monophyletic.

Many species in Rimosae are known to occur on more nutrient rich soils, often on calcareous ground, others prefer more nutrient poor acid soils. Several species are found in disturbed places such as along forest paths and roadsides. They form ectomycorrhizal associations with a broad range of host trees of both gymnosperms and angiosperms (Kuyper 1986, Stangl 1989, Jacobsson 2008). Several species occur in arctic and alpine regions and are then associated with shrubs and herbs such as Salix, Dryas, and Polygonum (Favre 1955, Horak 1987, Kühner 1988, Bon 1997, Ferrari 2006). A few are found on coastal sand dunes associated with Pinus and Salix (Orton 1960, Bon 1984).

The taxonomy in Rimosae has for a long time been confused since many species are described on small differences in macro- and micro-morphology. Authors with a conservative approach recognise 10–20 species (Kuyper 1986, Stangl 1989), others include more than 40 (Bon 1997). In the Nordic countries 8–13 species are usually recognized (Stridvall et al. 1989, Jacobsson 2008). Some European species are likely to show a northern circumpolar distribution (Ryberg et al. 2008). However, since a modern comprehensive treatment of Inocybe in North America is lacking the biogeographic knowledge is incomplete (Kauffman 1924, Stuntz 1947, 1954).

Inocybe rimosa, taken in a wide sense, shows a considerable morphological variation and also a broad ecological range covering all biomes from nemoral deciduous forests to the arctic-alpine zone. Some authors have advocated a narrow species concept and described a number of species and varieties (Heim 1931, Kühner 1988, Bon 1997). Kuyper (1986) choose the opposite strategy and recognized only one species, listing more than 30 species and varieties as synonyms.

The present study had three aims: to examine the phylogenetic structure and position of sect. Rimosae, to identify the number of North European species within Rimosae, and to elucidate the phylogenetic relationship among them.

MATERIALS AND METHODS

Morphological studies

Micro-morphological characters were observed using a Zeiss Axioscope 2, equipped with phase contrast. Spores and cystidia were measured in a 3 % KOH solution at × 400 and 1 000 magnification using microscope photos taken with a Canon G9 digital camera and using software AxioVision (Carl Zeiss AB). Unusually large or small spores were not considered. Collections are deposited in the herbarium of the Department of Plant and Environmental Sciences, University of Gothenburg (GB) if not otherwise indicated. Data on sequenced specimens is provided in Table 1.

Table 1

Data of specimens sequenced in this study.

Species	Original specimen identification	Coll. ID. / Origin	Ecology, substrate	GenBank
				ITS/LSU	mtSSU
Conocybe siliginea		LÖ93-04 / Swe	In a pasture	DQ389731
Crepidotus mollis		EL45-04 / Swe	deciduous wood	AM882996
C. mollis var. calolepis		EL14-08 / Swe	deciduous wood	FJ904178	FJ904242
Inocybe cervicolor		SJ04024 / Swe	Picea forest, calcareous	AM882939,	FJ904185
I. bongardii		EL123-04 / Swe	Quercus, calcareous	AM882941	FJ904186
I. subhirsuta		EL45-05 / Nor	Dryas, Salix, alpine	AM88294	FJ904187
I. cfr calamistrata		KHL13071 / Costa Rica	Quercus	AM882948
I. dulcamara		EL89-06 / Swe	Salix glauca	FJ904122	FJ904181
I. terrigena		EL117-04 / Swe	Picea, calcareous	AM882864	FJ904183
I. fulvipes		EL37-05 / Nor	Dryas, Salix, alpine	AM882858	FJ904184
I. agardhii		EL88-04 / Swe	Salix, calcareous	FJ904123	FJ904182
Naucoria salicis		EL71a-03 / Swe	Alnus, Betula	FJ904180
N. bohemica		EL71b-03 / Swe	Alnus, Betula	FJ904179	FJ904243
N. submelinoides		TAA185174 / Est	Alnus	AM882885
Inocybe adaequata	I. cfr rimosa	PC2008-0014 / GB	Fagus forest	FJ904177	FJ904240
	I. adaequata	MR00022 / Swe	Tilia, Corylus	AM882706	FJ904241
I. arenicola	I. arenicola	RC GB99-014 / Fra	Pinus, Salix, sand dune	FJ904134	FJ904189
	I. arenicola	EL238-06 / Fra	Pinus, sand dune	FJ904133	FJ904188
I. bulbosissima	I. fastigiata var. alpina	EL51-05 / Nor	Dryas, Salix, alpine	AM882764
	I. fastigiata var. alpina	EL66-05 / Nor	Salix reticulata, alpine	AM882765	FJ904224
	I. fastigiata var. alpina	EL37-06 / Swe	Salix polaris, alpine	FJ904161	FJ904223
	I. bulbosissima	EL75-07 / Swe	Salix reticulata, alpine	FJ904160	FJ904222
	I. rimosa	EL88-06 / Swe	Salix lapponum, subalpine	FJ904159	FJ904221
	I. fastigiata var. alpina	EL30-06 / Swe	Salix polaris, alpine	FJ904158	FJ904220
I. cfr cookei	I. cookei	EL104-04 / Swe	Corylus	AM882952
I. cfr flavella	I. rimosa	GK080924 / GB	Quercus, Betula, wet	FJ904129	FJ904196
	I. majalis	PAM05062502 / Fra	Salix, calcareous soil	FJ904128	FJ904195
	I. flavella	EL118-05 / Fin	Salix, Betula, ravine	AM882782
	I. flavella	BJ920829 / Swe	Salix, Betula, hyperit	AM882774
	I. flavella	EL90-04 / Swe	Salix, Betula, calcareous	AM882773
I. cfr rimosa	I. rimosa	EL71-04 / Swe	Fagus, calcareous soil	AM882786	FJ904193
	I. perlata	JD2008-0241 / GB	Fagus, Corylus	FJ904125	FJ904192
	I. cfr rimosa	I116-06 / Australia	deciduous forest	FJ904142
	I. fastigiata	PAM05061101 / Fra	Tilia, calcareous	FJ904155	FJ904216
	I. arenicola	JV26578 / Est	Pinus, calcareous	FJ904154	FJ904215
	I. rimosa	EL127-04 / Swe	Fagus, Quercus, calcareous	AM882768	FJ904219
	I. rimosa	TAA185135 / Est	Pinus, Betula, calcareous	AM882766
	I. rimosa	JV22619 / Est	Quercus, Corylus, calcareous	FJ904157	FJ904218
	I. umbrinella	PC080925 / GB	Pinus, Quercus	FJ904153
	I. rimosa var. umbrinella	JV8125 / Fin	Picea, Tilia, Populus, rich	FJ904152	FJ904214
	I. cft obsoleta	EL81-06 / Swe	Salix glauca, subalpine, wet	FJ904135	FJ904190
I. cfr squamata	I. cft squamata	I93-04 / Australia	deciduous forest	FJ904141
	I. cft squamata	I113-05 / Australia	deciduous forest	FJ904140
	I. squamata	SJ92-010 / Swe	Picea, calcareous	AM882785
	I. squamata	SM92-013 / Swe	Picea, Populus, Betula	AM882783
	I. squamata	SJ92-017 / Swe	Pinus, Populus, park	AM882784
	I. squamata	Stordal18318 / Nor	Picea mixed forest	FJ904139
	I. squamata	JV2609 / Fin	Picea, Populus, Pinus	FJ904138	FJ904203
I. cookei	I. cookei	MR00035 / Swe	Corylus, Quercus	AM882954
	I. cookei	EL191-06 / GB	Corylus, Quercus	FJ904173	FJ904234
	I. cookei	EL70a-03 / Swe	Fagus, Quercus	AM882953
	I. cookei	EL73-05 / Swe	Betula, Quercus	AM882955
	I. cookei	EL109-04 / Swe	Corylus, Quercus	AM882956	FJ904233
I. dulcamaroides	I. dulcamaroides	EL29-08 / USA	Salix reticulata, alpine	FJ904127
	I. dulcamaroides	EL112-06 / Swe	Dryas, alpine	FJ904126	FJ904194
I. erubescens	I. erubescens	TAA185164 / Est	Quercus, Tilia, calcareous	AM882950
	I. erubescens	KGN980714 / Swe	Fagus, Tilia, rich soil	AM882951	FJ904239
	I. erubescens	BH910707 / Swe	Fagus, park	AM882949
I. flavella	I. flavella	EL56-08 / Swe	Corylus, Salix, Alnus, wet	FJ904131	FJ904198
	I. flavella	EL137-05 / Swe	Corylus, Alnus, Quercus wet	AM882776	FJ904199
	I. flavella	LAS89-030 / Swe	Alnus, wet	AM882775
I. xanthocephala	I. xanthocephala	PAM00100606 / Fra	Salix	FJ904130	FJ904197
I. hygrophorus	I. hygrophorus	EL97-06 / Swe	Betula, Salix, subalpine meadow	FJ904137	FJ904202
I. maculata	I. maculata	EL74-05 / Swe	Fagus, Quercus	AM882959
	I. maculata	MR00020 / Swe	Tilia, Corylus, calcareous	AM882958
	I. maculata	EL121-04 / Swe	Fagus, Quercus, calcareous	AM882957	FJ904232
	I. maculata	EL58-03 / Swe	Fagus, rich soil	AM882963
	I. maculata	EL126-04 / Swe	Fagus, Quercus, calcareous	AM882964
	I. maculata	EL182-08 / Slov	Fagus, rich soil	FJ904172
I. maculata forma fulva	I. maculata	EL78-03 / Swe	Mixed trees, pasture	AM882962
	I. maculata	EL166-08 / Swe	Picea, Corylus, calcareous	FJ904171	FJ904231
	I. cfr maculata	EL114-06 / Swe	Dryas, Polygonum, alpine	FJ904170
	I. cfr rimosa	SJ05029 / Swe	Pinus, Alnus	AM882994	FJ904230
	I. maculata forma fulva	EL247-06 / Fra	Pinus, Populus	FJ904169
	I. maculata forma fulva	PAM01100120 / Fra	Betula	FJ904168
	I. maculata	SJ06007 / Swe	Betula	FJ904167
I. melliolens	I. umbrinella	PAM05052303 / Fra	Tilia, calcareous	FJ904148	FJ904211
	I. melliolens	EL224-06 / Fra	Salix, Quercus, wet	FJ904149
I. cfr microfastigiata	I. microfastigiata	EL113-06 / Swe	Dryas, alpine	FJ904156	FJ904217
I. mimica	I. mimica	EBJ961997 / Swe	Pinus, Picea, calcareous	FJ904124	FJ904191
	I. mimica	TK2004-114 / Swe	Pinus, Betula, calcareous	AM882781
I. obsoleta	I. obsoleta	EL17-04 / Swe	Picea, Corylus	AM882769	FJ904204
	I. obsoleta	BJ890915 / Swe	Picea mixed forest	AM882770
I. perlata	I. perlata	BJ940922 / Swe	Fagus, Betula, meadow	AM882772
	I. perlata	EL74-04 / Swe	Corylus, Betula, calcareous	AM882771	FJ904205
I. quietiodor	I. quietiodor	RP980718 / Swe	Fagus, Quercus, park	FJ936169	FJ904238
	I. quietiodor	LAS97-067 / Swe	Fagus, Quercus, calcareous	AM882974
	I. quietiodor	LAS94-023 / Swe	Fagus, Quercus, calcareous	AM882961
	I. quietiodor	PAM01091310 / Fra	Betula, Salix	FJ936168	FJ904237
	I. quietiodor	EL115-04 / Swe	Quercus, Tilia, park	AM882960	FJ904236
	I. quietiodor	JV20202 / Nor	Betula, Alnus, calcareous	FJ904174	FJ904235
I. rhodiola	I. rhodiola	PAM00090117 / Fra	Salix	FJ904176
	I. rhodiola	EL223-06 / Fra	Salix, wet forest	FJ904175
I. rimosa	I. rimosa	AO2008-0250 / GB	Salix	FJ904147	FJ904210
	I. rimosa	EL118-08 / Swe	Picea, Betula, calcareous	FJ904146	FJ904209
	I. rimosa	EL102-04 / Swe	Betula, garden	AM882761
	I. rimosa var. umbrinella	EL211-06 / Fra	Quercus, Carpinus	FJ904145
	I. rimosa	TK97-156 / Swe	Corylus, calcareous	AM882844
	I. fastigiata var. argentata	PAM03110904 / Fra	Quercus	FJ904144	FJ904208
	I. rimosa	EL75-05 / Swe	Fagus, Quercus, park	AM882762	FJ904207
	I. rimosa	SJ04007 / Swe	Tilia	AM882763
	I. fastigiata var. argentata	PAM06112703 / Corsica	Fagus	FJ904143	FJ904206
I. sororia	I. cfr fasigiata	Kuoljok0512 / Swe	Salix, alpine meadow	FJ904150	FJ904212
	I. rimosa coll.	JV15200 / Swe	Salix herbacea, alpine	FJ904151	FJ904213
I. squamata	I. squamata	SJ08003 / Swe	Betula, Pinus	FJ904136	FJ904201
	I. cfr squamata	TK96-109 / Swe	Populus, calcareous	AM882780
	I. cfr squamata	SJ85048 / Nor	Populus, calcareous	AM882778
	I. curreyi	PAM05052301 / Fra	Populus, Picea, park	FJ904132	FJ904200
I. umbrinella	I. rimosa var. brunnea	JV13699 / Fin	Pinus, Populus, Salix	FJ904165	FJ904228
	I. rimosa coll.	JV17954 / Est	Pinus, calcareous	FJ904166	FJ904229
	I. cfr rimosa	PC081010 / GB	Helianthemum, calcareous	FJ904164	FJ904227
	I. cfr rimosa	PC080816 / GB	Fagus, Quercus, calcareous	FJ904163	FJ904226
	I. perlata	PAM01102912 / Fra	Quercus ilex	FJ904162	FJ904225

Taxon sampling

Ninety-nine ingroup specimens were sequenced. They represent the majority of species within section Rimosae that occur in North Europe (Jacobsson 2008). In addition specimens from Estonia, France, Great Britain, Slovakia, USA, and Australia were included. Specimens were selected to represent a broad spectrum of morphological characters and ecology. Eight species of sect. Cervicolores and subg. Mallocybe were also sequenced and included in the analyses. Based on results from earlier molecular phylogenetic studies of Agaricales and Inocybe (Matheny 2005, Matheny & Bougher 2006a, Matheny et al. 2006) species of Conocybe, Crepidotus, Naucoria, Pleuroflammula, and Simocybe, were selected as out-group. ITS and LSU sequences for included species of Auritella, Simocybe, and Pleuroflammula, were taken from GenBank (AY380371, AY380395, AY635766, DQ494696, AY745706, AF205707, DQ494685, AF745706). Two GenBank sequences (DQ917657 ITS, EU600863 LSU) representing specimens identified as Inocybe sororia were included in the separately aligned and analysed dataset for the Rimosae s.s. subclade A (see below).

DNA extraction, PCR and sequencing

Sequences from three regions were generated for the study, the complete ITS region, 1200 base pairs of the 5’ end of the nuclear LSU ribosomal DNA, and the mitochondrial SSU ribosomal DNA. DNA extractions, PCR reactions and sequencing were performed as described in Larsson & Örstadius 2008. Primers used to amplify the complete ITS region and the 5’ end of the LSU region were ITS1F (Gardes & Bruns 1993) and LR21, LR0R and LR7 (Hopple & Vilgalys 1999), the mtSSU MS1 and MS2 (White et al. 1990). Primers used for sequencing were ITS1, ITS3, ITS4, MS1, MS2 (White et al. 1990), Ctb6 (http://plantbio.berkeley.edu/~bruns/), Lr5 and LR3R (Hopple & Vilgalys 1999).

Phylogenetic analyses

Sequences were edited and assembled using Sequencher 3.1 (Gene Codes, Ann Arbor). Sequences were aligned automatically using the software MAFFT (Katoh et al. 2002) and adjusted manually using the data editor in PAUP* (Swofford 2003). Sequences have been deposited in GenBank and accession numbers are listed in Table 1.

Heuristic searches for most parsimonious trees were performed using PAUP*. All transformations were considered unordered and equally weighted. Variable regions with ambiguous alignment were excluded and gaps were treated as missing data. Heuristic searches with 1 000 random-addition sequence replicates and TBR branch swapping were performed. Relative robustness of clades was assessed by the bootstrap method using 1 000 heuristic search replicates with 100 random taxon addition sequence replicate, TBR swapping, saving 100 trees in each replicate.

Bayesian analysis of the datasets was performed using MrBayes 3.2 (Huelsenbeck & Ronquist 2001, Ronquist & Huelsenbeck 2003). MrModelTest 2.2 (Nylander 2004) was used to estimate separate best-fit models of evolution for ITS (1 and 2 combined), 5.8S, LSU, and mtSSU. The Bayesian inference was set up with model parameters estimated separately for each of the four partitions. Four parallel runs using Metropolis-Coupled Markov Chain Monte Carlo (MCMCMC) were implemented instead of the default of two to improve the inference of convergence statistics. To decrease the computational burden two chains, one hot and one cold (temperature difference set to 0.1 to increase the efficiency of metropolis coupling), were used instead of the default of four. Each chain was run for 10 million generations with tree and parameter sampling every 1 000 generations (10 000 trees). Tracer (Drummond & Rambaut 2007) and AWTY (Wilgenbusch et al. 2004) were used to examine when the chains had reached a stationary state and how many generations were appropriate to discard as burn-in. A 50 % majority-rule consensus phylogram was computed from the remaining trees; the proportions of this tree correspond to Bayesian posterior probabilities (BPP). To investigate if there were any conflict between the nuclear and mitochondrial regions, analyses were also made for these partitions separately. It was then checked if there were any conflict between the regions in nodes separating species and with more than 0.7 BPP support (cf. Lutzoni et al. 2004).

To improve the resolution and be able to use the complete ITS region in the phylogenetic analysis realignment of sequence data in the Rimosae s.s. subclade A and Rimosae s.s. subclade D were performed as described above. Heuristic searches for most parsimonious trees and bootstrap analysis were performed as above except that no restriction on saving of trees in the replicates was applied.

RESULTS

For all 99 ingroup specimens the ITS/LSU region was generated. For 54 of these also mtSSU sequences were generated (Table 1). The aligned complete dataset, including sequences downloaded from GenBank, consisted of 119 sequences and 3 461 characters. The majority of the ITS1 and ITS2 regions was found to be too variable to be included in the analyses. After exclusion of ambiguous regions 1 985 characters remained for the analysis. Of these 1 409 were constant, 174 were variable and parsimony uninformative, and 402 were parsimony informative.

The maximum parsimony analysis yielded 63 900 equally most parsimonious trees (length = 1 812, CI = 0.4354, RI = 0.8490). Bootstrap analysis recovered Inocybe s.l. (including Auritella) as monophyletic with 91 % support. It forms together with Crepidotus, Simocybe, and Pleuroflammula a clade with 92 % bootstrap support.

Four major clades within the ingroup received strong support. They are here called Auritella (100 %), Rimosae s.s. (98 %), Mallocybe (100 %), and Inosperma (98 %). The Inosperma clade was further divided in the Cervicolores clade (98 %) and a moderately supported group here called the Maculata clade (72 %).

The Rimosae s.s. clade includes 68 sequences dispersed over 6 strongly supported subclades (Fig. 1A–F) and a number of groups that in most cases seem to correspond to species. Eight of these terminal groups have been identified as Inocybe arenicola, I. mimica, I. dulcamaroides, I. flavella, I. squamata, I. hygrophorus, I. obsoleta, and I. perlata, respectively. One distinct but non-identified clade is reported as Inocybe sp. Specimens identified as I. flavella seem to cover several taxa differing in the shape and size of spores. Small-spored specimens are together lumped as I. cfr flavella but this label seems to cover at least two taxa. Thirty-three sequences cluster to a strongly supported clade that corresponds to I. rimosa s.l. Within such a broadly defined I. rimosa five subclades corresponding to species were recovered. One is the alpine species I. bulbosissima, a second is I. rimosa s.s., a third is I. umbrinella, and the remaining two are tentatively identified as I. cfr sororia and I. melliolens. The specimens originating from deciduous forests in Australia form a strongly supported clade within Rimosae and represent two unidentified species. In addition several taxa on single branches and minor less supported clades were recovered (Fig. 1). These terminals may represent new species, either undescribed or described from other regions but not yet identified.

Fig. 1

Bayesian 50 % majority-rule consensus phylogram. Bayesian posterior probabilities and bootstrap values above 70 % from the maximum parsimony analysis are indicated on branches. Recovered major clades are named and marked with a scale bar and minor supported clades discussed in the text have been numbered A–F. Conocybe siliginea was used to root the tree.

In the Maculata clade six terminal taxa were recovered as strongly supported (Fig. 1). They are identified as Inocybe adaequata, I. rhodiola, I. erubescens, I. quietiodor, I. cookei, and I. maculata, including I. maculata forma fulva.

For the separate regions, MrModelTest suggested GTR+I+G (ITS), K80+I (5.8S), GTR+I+G (LSU), and GTR+I+G (mtSSU) as optimal models; this information was employed in MrBayes. The Tracer and AWTY analysis indicated that 5 million generations would be an appropriate burn in time as a stationary state was reached for all chains well before that. This was also supported by the fact that the standard deviation of split frequencies calculated in MrBayes was below 0.01 well before this point. The last 5 000 trees of each run (20 000 trees in total) were therefore summarized into a 50 % majority-rule consensus phylogram (Fig. 1).

No conflict was found between the nuclear and mitochondrial regions according to the criteria defined in Materials and Methods.

Also in this analysis Inocybe, including Auritella, is recovered as monophyletic with a BPP value of 1.00. The four major clades recovered in the maximum parsimony analysis, are present also in the Bayesian tree, all of them with BPP values 1.00. The Cervicolores (BPP 1.00) and Maculata clades (BPP 1.00) are also strongly supported. All 21 species clades from the maximum parsimony analysis of sect. Rimosae are also supported in the Bayesian analysis (Fig. 1). The Bayesian tree topology is more or less identical to the MP bootstrap tree. However, some additional clades were recovered with strong support, e.g. Crepidotaceae (BPP 0.99).

The realigned dataset of 33 taxa in the Rimosae s.s. subclade A included 2 819 characters. After exclusion of regions with incomplete data, mainly from the mtSSU, 2 362 characters remained for the analysis. Of these 2 126 were constant, 83 were variable and parsimony uninformative, and 153 were parsimony informative. The heuristic searches recovered 4 290 equally most parsimonious trees (length = 340, CI = 0.7853, RI = 0.8933). Fig. 2 illustrates one of these as a mid-point rooted phylogram. The bootstrap analysis recovered the same strongly supported terminal clades as in the complete parsimony analysis but also generated moderate support (78 %) for I. rimosa s.s. Support for the remaining species level clades were: I. melliolens (99 %), I. cfr sororia (98 %), I. bulbosissima (98 %), and I. umbrinella (100 %). The North American sequences representing I. sororia clustered with North European sequences with 100 % support.

Fig. 2

One of the equally most parsimonious trees obtained from the maximum parsimony analysis of the Rimosae s.s. subclade A, presented as a mid-point rooted phylogram. Bootstrap values are indicated on branches.

The realigned dataset of 18 taxa in the Rimosae s.s. subclade D included 2 812 characters. After exclusion of regions with incomplete data, mainly from the mtSSU, 2 421 characters remained for the analysis. Of these 2 266 were constant, 57 were variable and parsimony uninformative, and 98 were parsimony informative. The heuristic search recovered 675 most parsimonious trees (length = 190, CI = 0.8526, RI = 0.9111). Fig. 3 illustrates one of these as a mid-point rooted phylogram. The bootstrap analysis recovered five strongly supported (above 90 %) clades of which four have the same topology as in the complete analyses, viz. I. squamata (100 %), Inocybe sp. (100 %), I. flavella (100 %), and I. cfr flavella A (99 %) while I. cfr flavella B is supported only together with I. hygrophorus (100 %). Inocybe flavella + I. xanthocephala (91 % in the large parsimony analysis) is not supported.

Fig. 3

One of the equally most parsimonious trees obtained from the maximum parsimony analysis of the Rimosae s.s. subclade D, presented as a mid-point rooted phylogram. Bootstrap values are indicated on branches.

DISCUSSION

The present study aimed at elucidating the phylogenetic structure of Inocybe subg. Inosperma sect. Rimosae as defined by Kuyper (1986) but included also representatives from the subg. Mallocybe and the recently erected genus Auritella (Matheny & Bougher 2006a, b). The ingroup was recovered as monophyletic and strongly supported. However, our results indicate that recognizing Auritella on the genus level renders Inocybe paraphyletic. The solution is either to sink Auritella as a subgenus within Inocybe, or to split Inocybe into a number of smaller genera. However, our study was not designed to take a decision on that matter.

We found that the species traditionally placed in sect. Rimosae did not form a monophyletic clade. Instead they are distributed over two strongly supported clades: Maculata and Rimosae s.s. (Fig 1). The Maculata group clusters with sect. Cervicolores and the two combined represent subg. Inosperma in a new stricter sense. A more narrowly defined sect. Rimosae emerges as an independent supported clade well separated from Inosperma s.s.

In the study by Matheny (2006) a division of sect. Rimosae was indicated although only one representative of the Rimosae s.s. clade was included. In a recent biogeographic study of Inocybaceae more representatives of the Rimosae s.s. clade were included and the split topology again supported (Matheny et al. 2009). Matheny uses Pseudosperma as the clade name for what is here called Rimosae s.s.

Through our morphological investigations we were able to correlate 21 terminal groups within clades Maculata and Rimosae s.s. with morphologically distinct species (Fig. 1). The molecular support is based on data from nLSU, mtSSU, 5.8S and a few conservative regions of ITS, leaving out the variable regions of ITS because of aligning problems. The ITS region is the locus that has been most commonly used for species delimitation of fungi (Kõljalg et al. 2005, Kõljalg & Larsson 1998, Larsson & Örstadius 2008, Nilsson et al. 2008). In general Inocybe species show a high sequence divergence in the ITS region. Closely related species often deviate in several substitutions and insertion/deletion events and are therefore easy to identify using simple sequence comparison (Altschul et al. 1997). We also found that the terminal clades we have been able to correlate to morphological species, in general show a low within clade sequence divergence. Notable exceptions to this observation are I. rimosa and related species (Fig. 2 subclade A), and I. flavella and related species (Fig. 3 subclade D).

The Maculata clade is here represented by seven species with thin-walled, often clavate to pyriform cheilocystidia and phaseoliform spores. They usually have specific odours that differ from the spermatic smell typical of most species in the Rimosae s.s. clade. Inocybe maculata smells like raw potatoes or Tuber, I. cookei like honey, I. erubescens like perfumed soap, I adaequata and I. rhodiola like beetroot, and I. quietiodor like Lactarius quietus, that is, both sourish and sweetish. Having specific odours is a character they share with the species in sect. Cervicolores. However, odour is a very difficult and subjective character to use and although a spermatic smell characterizes most species in the Rimosae s.s. clade some species have other characteristics, e.g. I. melliolens Kühner (1988), which smells of honey when drying.

There is a trend in the Maculata clade for the stipe base to be distinctly bulbous. This characteristic is present in I. cookie, I. maculata, I. maculata forma fulva, and I. quietiodor, while in I. adaequata, I. erubescens, and I. rhodiola the bulbous base is not as pronounced. Several species are characterized by a reddening of the flesh (I. adaequata, I. erubescens, I. rhodiola) and this trait occurs also among species in Cervicolores.

Inocybe maculata is known as a variable species with considerable differences in cap colour and presence of velar remnants. These observations correspond to a high divergence among the sequences generated from specimens initially identified as I. maculata in a wide sense. The specimens we have sequenced can be divided in two morphotypes which also seem to correlate with ecological preferences. One type has a chocolate brown cap, often covered with conspicuous, white velar remnants. It grows associated with Carpinus, Corylus, Fagus, Quercus, and Tilia on rich soils, usually on calcareous ground. This type fits with the original description of I. maculata but unfortunately the specimens representing this morphotype do not form a monophyletic clade and the sequence divergence indicates that more than one species are involved (Fig. 1, 4a). The other morphotype is more yellow to reddish brown and has less or no velar remnants on the cap. This type seems to be associated with Betula, Picea, Pinus, Populus, and Salix, but also with Dryas and Polygonum in alpine environments. It fits the concept of I. maculata forma fulva Bon (1991, Fig. 4b) described from a coastal dune area with Populus in northern France. Of the seven specimens sequenced two originate from France, one of them from the same region and ecological setting as the type of forma fulva. Sequence data are uniform for this taxon and clade support is strong (97 %, BPP 1.0).

Fig. 4

a. Inocybe maculata (EL182/08); b. Inocybe maculata forma fulva (EL82/08); c. Inocybe dulcamaroides (EL112/06); d. Inocybe bulbosissima (EL75/07).

The Rimosae s.s. clade includes six strongly supported subclades (Fig. 1A–F) and altogether 15 terminal clades that we could correlate to morphologically distinguishable species. In addition Rimosae s.s. also includes several unidentified minor terminal clades and sequences that occur on single branches (Fig. 1). In general the species in this clade have ellipsoid to indistinctly phaseoliform spores and the cheilocystidia tend to be more cylindrical to clavate than in the Maculata clade. However, the shape of the cheilocystidia is very variable even within the same species. In this clade we also more often find that the apex of the stipe is distinctly white pruinose to flocculose. The spermatic odour and the presence of yellow to olive-yellow pigments in the lamellae are characteristic for many species in Rimosae s.s. The occurrence of oily refracting contents in hyphae is notable and may be connected to the spermatic odour. Kuyper (1986) found a correlation between the intensity of the olive-yellow tone of the lamellae and the strength of the spermatic odour.

Rimosae subclade A. This clade corresponds to I. rimosa in a wide sense and includes 33 specimens (Fig. 2) originating from England, Estonia, France, and Scandinavia. They represent most of the large variation in macro-morphological characters and ecology demonstrated by the many varieties described (Heim 1931, Favre 1955, Kühner 1988, Bon 1997). Five terminal clades with moderately strong to strong support were recovered and are further discussed below. Still other sequences form unsupported groups or occur on single branches (Fig. 1, 2).

Inocybe rimosa s.s. includes nine specimens that show a great variation in cap colours from pale to ochraceous yellow brown to dark brown. In micro-morphology they are all very similar. The specimens originate from different habitats ranging from nemoral deciduous forests to boreal Picea forest. There is a nine base pair insertion in the beginning of the ITS region in five of the specimens but this difference could not be correlated to differences in morphology or ecology.

The specimens named I. melliolens originate from France. In morphology this species looks like a typical I. rimosa s.s. but has a strong smell of honey. We have not seen any specimens with this character from Northern Europe and the species may have a more southern distribution, even if it is described from Dryas vegetation in the French Alps (Bon 1997).

The LSU sequences deposited in GenBank as I. sororia originate from a Northwest American conifer stand (Matheny et al. 2009). They are almost identical to our sequences from two specimens from herb-rich locations and dwarf Salix in the alpine zone of Sweden (Fig. 2). Inocybe sororia is described by Kauffman (1924) from North American frondose forests. Since we have only studied the Swedish material and they deviate from the descriptions given by Kauffman (1924) and Stuntz (1947) we feel uncertain about the identity of our collections.

Inocybe bulbosissima includes specimens associated with dwarf Salix and Dryas in the alpine zone. The species is usually regarded as a variety of I. rimosa and then named I. fastigiata var. alpina. Another alpine species with similar characteristics is I. microfastigiata which is said to differ by smaller spores and a darker brown cap (Kühner 1988). A specimen with such morphology was included in this study and did not cluster with I. bulbosissima (Fig. 2, 4d). More specimens and sequences are required before the circumscription of I. microfastigiata can be clarified.

Inocybe umbrinella includes specimens with warm yellowish to reddish brown caps with a dark centre and contrasting strongly rimose and lighter periphery. The micro-morphology is almost identical to I. rimosa. In Bresadola′s description (Bresadola 1905) it is said to grow in gravely places with Populus nigra. Three of our specimens were collected in rather dry, sandy environments with Helianthemum, Pinus, and Quercus ilex. In our opinion these specimens fit the descriptions by Bresadola (1905) and Enderle & Stangl (1981).

Some other specimens with a dark brown cap clustering with or close to I. rimosa were first determined as I. umbrinella (Table 1). These specimens have fruit-bodies that in general are somewhat smaller than typical I. umbrinella. Inocybe umbrinella may also be confused with I. perlata but I. perlata usually has a dull dark brown cap and a flattened, less pronounced umbo.

. Inocybe obsoleta and I. perlata were recovered as independent species. Both have rather large and robust fruit-bodies and occur in mixed deciduous forests and parks, often on somewhat calcareous and nutrient rich soils. In micro-morphology they are hard to separate from I. rimosa. The best characters for identification are the robust fruit-bodies and the dark greyish brown colour without any yellow flush on the cap in I. perlata and the pale clay yellowish brown cap with distinct, white velum remnants in young specimens of I. obsoleta. Our interpretation of the name I. obsoleta is here based on specimens from North Europe only.

includes three specimens that originate from Australia. We were not able to correlate them to known species and they are provisionally named I. cfr squamata and I. cfr rimosa. They represent two species closely related to I. rimosa but may not occur in North Europe.

includes species that are often encountered in mixed forests with Salix and Populus. We identified I. flavella s.l., I. hygrophorus, I. squamata, and one undescribed species (Inocybe sp.). Inocybe hygrophorus is represented by only one specimen. It was collected in a subalpine meadow in forest with Salix and Betula and fits the original morphological description. Sequence data confirm that it is distinct from I. flavella. It seems to be rare or maybe overlooked as it may be mistaken for I. flavella or I. rimosa.

The two species I. squamata and Inocybe sp. are very similar in macro-morphology, with scattered appressed scales on the cap. The spores are distinctly phaseoliform and narrow in Inocybe sp. while in typical I. squamata they are broadly ellipsoid and only occasionally slightly phaseoliform. We observed that the lamellae were more yellow and the fruit-bodies on average larger in Inocybe sp. This possibly undescribed species also seems to have a boreal distribution judging from the known records from Sweden, Finland, and Norway. Inocybe squamata is, in the Nordic countries, only known from nemoral and hemiboreal regions.

The specimens determined as I. flavella split into three supported subclades. We have included specimens that originate from Sweden, Finland, England, and France. We identified two morpho-types, here named I. flavella and I. cfr flavella. The specimens are all rather similar in macro-morphology, but show a variation in the presence of yellow pigments on lamellae and stipe and in the colour and structure of the cap. No special odour was detected. In micro-morphology a variation in the length and shape of the spores can be observed. In I. flavella s.s. the spores are 10–12 × 5–6 μm and usually not phaseoliform. In the clade named I. cfr flavella the spores are shorter 9–10.5 × 5–6 μm and often more or less phaseoliform. The analysis including the ITS region confirm the high sequence divergence within I. flavella s.l. (Fig. 3). There are several species and forms described as close to I. flavella (Heim 1931, Orton 1960, Kuyper 1986, Kühner 1988, Bon 1997) and additional sequence data is needed to disentangle the entities involved within this clade.

. Inocybe dulcamaroides is an arctic-alpine species associated with Dryas and Salix reticulata. The two sequences representing this species are 100 % identical throughout the ITS region despite originating from Sweden and USA (Montana), respectively. It is reminiscent of I. dulcamara in that it has a short stipe in comparison to its cap diameter. This makes it a morphologically characteristic species but it is little collected and seemingly rare or overlooked (Fig. 4c).

includes the two species, I. arenicola and I. mimica, which are characterized by rather large spores. They cluster together with sequences of three specimens which could not be matched with any species descriptions. The only uniting factor we found in this clade was a preference for calcareous soil conditions. Other morphological and ecological traits show a large variation.

Most of the species discussed in this paper are typified with material that turned out to be less suitable for DNA extraction. Some collections are simply old, others, e.g. those of Marcel Bon, are apparently dried under conditions that did not preserve DNA well. Still others are very scanty, e.g. most of Kühner’s type specimens. This situation is not at all unusual within fungal taxonomy. If we shall be able to take full advantage of the higher precision of species definitions made possible by molecular data, we must make extensive use of the epitypification tool offered by ICBN. Like all typification measures, also the selection of an epitype must be done with utmost care in order to preserve the intentions of the original author. Our preferred method is to first seek a profound understanding of the regional species diversity through intense field work, then match our collections with existing names, compare them to authoritative material, and finally, if necessary, select epitypes from rich, molecularly characterized collections.

This study had a focus on the species that occur in European arctic and alpine environments and in temperate regions of North Europe. Many of the species belonging in sect. Rimosae which are described from North America and from the Mediterranean have yet to be sampled before a more complete understanding of the phylogenetic diversity of the Maculata and Rimosae s.s. clades can be achieved. Only then will it be appropriate to fix names through epitypification.

TAXONOMY

Species of the Maculata clade identified in Northwest Europe

(Britzelm.) Sacc., Syll. Fung. (Abellini) 5: 767. 1887

Specimens examined. Great Britain, England, Bucks, Kings Wood Tylers Green, 19 Aug. 2008, P. Cullington 2008/0014. – Sweden, Bohuslän, Valla, Sundsby, 1 Sept. 1979, SJ79154; Bohuslän, Tanum, Lammö, 29 Sept. 2004, MR00022; Västergötland, Kinnekulle, Medelplana, Råbäcks munkängar, 6 Aug. 1977, SJ77120:

Bres., Fungi Trident. 2, 8–10: 17. 1892

Specimens examined. Great Britain, Scotland, Ledmore oakwood, 14 Sept. 2006, EL191-06. – Norway, Buskerud, Hönefoss, Grunntjern, 28 Aug. 2003, J. Vauras 20202 (TUR-A). – Sweden, Västergötland, Alingsås, Nolhagaparken, 30 Aug. 2003, EL70A-03; Alingsås, Nolhagaparken, 21 Aug. 2005, EL73-05; Alingsås, Nolhagaparken, 2 Sept. 2006, EL150-06; Östad, Österäng, 26 Sept. 2004, EL109-04; Östad, Ekedalen, 26 Sept. 2004, EL104-04; Östad, Östad säteri, Djurgården, 4 Sept. 2008, EL67-08; Göteborg, Botaniska Trädgården, 19 Sept. 1975, SJ386; V. Tunhem, Hunneberg, 4 Oct. 2004, MR00035; Töreboda, Älgarås, Velen, 6 Sept. 2003, EL50-03; Grimmered, Björräsakulle, 17 Aug. 1989, SJ89002.

A. Blytt, Videnskabs-Selskabets Skrifter. I Math.-Naturv. Kl., 6: 54. 1905 (‘1904’)

Specimens examined. Estonia, Saaremaa, Tagamösa, 29 July 2004, TAA185164. – Sweden, Närke, Askersund, Stjärnsund, 14 July 1998, K-G Nilsson; Skåne, Genarp, Häckeberga, 7 July 1991, Bernt Hägg.

Boud., Bull. Soc. Bot. France 32: 283. 1885

Specimens examined. Denmark, Falster, Nykøbing, Fuglsang Storskov, 3 Oct. 2007, EL136-07. – Slovakia, Rimavská Sobota, Drña, 3 Oct. 2008, EL182-08. – Sweden, Bohuslän, Tanum, Lammö, 29 Sept. 2004, MR00020; Halland, Fjärås, Tjolöholm, 20 Sept. 1975, SJ389; Västergötland, Karlsborg, Undenäs, Bölet, 5 Sept. 2003, EL45-03; Västergötland, Berg, Högsböla ängar, Melldalaskogen, 7 Sept. 2003, EL58-03; Västergötland, Medelplana, Råbäcks Munkängar, 5 Sept. 2003, EL41-03; Västergötland, Kinnekulle, Medelplana, Råbäcks Munkängar, 27 Sept. 2004, EL121-04; Västergötland, Kinnekulle, Medelplana, Råbäcks Munkängar, 27 Sept. 2004, EL126-04; Västergötland, Alingsås, Nolhagaparken, 30 Aug. 2003, EL68-03; Västergötland, Alingsås, Nolhagaparken, 21 Aug. 2005, EL74-05.

forma Bon, Doc, Mycol. 21 (no. 81): 47. 1991

Specimens examined. France, Merlimont, Pas de Calais, 3 Nov. 2006, EL247-06; Isére, Chireas, 1 Oct. 2001, PAM01100120 (LIP). – Sweden, Härjedalen, Hamra, Hamrafjället, 18 Aug. 2006, EL114-08; Dalarna, Rättvik, Ö. om Gärdsjöns sydände, 23 Aug. 1982, R. Morander 4320; Dalarna, Rättvik, Rättviksheden, 29 Aug. 2005, SJ05-029; Närke, Lerbäck, Udden, 13 Aug. 1997, K-G Nilsson; Närke, Lekhyttan, Lunnasjön, 13 Sept. 2008, EL166-08; Närke, Lerbäck, Runsala ravin, 11 Sept. 2008, EL134-08; Närke, Kvistbro, Sixtorp, Gammelhyttan, 9 Sept. 2008, EL82-08; Skåne, Hässleholm, Ignaberga, 15 Sept. 2003, EL78-03; Värmland, Övre Ullerud, Torsberget SSO, 17 Aug. 1991, Bo Jansson; Lule Lappmark, Jokkmokk, SSO Messaure, 23 Aug. 2003, S. Kuoljok 0337.

Bon, Doc. Mycol. 6 (no. 24): 46. 1976

Specimens examined. Denmark, Lolland, Flintinge byskov, 4 Oct. 2007, EL142-07. – France, Isére, Claix, Le Penil, 13 Sept. 2001, PAM01091310 (LIP); Namps-au-Val, Fremoutiers, 14 Sept. 1994, RC/F94064 (LIP). – Sweden, Västergötland, Alingsås, 18 July 1998, RP98/048; Västergötland, Götene, Medelplana, 28 Sept. 1997, LAS97/067; Västergötland, Kinnekulle, Medelplana, Råbäck, 10 Sept. 1994, LAS 94/023; Västergötland, Kinnekulle, Österplana, 27 Sept. 2004, EL115-04; Västergötland, Kinnekulle, Medelplana, Råbäcks Munkängar, 8 Sept. 2008, EL73-08.

Bres., Fungi Trident. 1: 80. 1884

Specimens examined. France, Saint-Amand, Dréve des Prés Charniers, 31 Oct. 2006, EL223-06; Isére, Saint Laurent du Pont, 1 Sept. 2000, PAM00090117 (LIP). – Italy, Cuneo, Ceva, Alessio, 12 Sept. 1980, Bon80091207 (LIP).

Species of the Rimosae s.s. clade identified in Northwest Europe

(R. Heim) Bon, Doc. Mycol. 12 (no. 48): 44. 1983 (‘1982’)

Specimens examined. Estonia, Saaremaa: Kaarma, Mändjala, 19 Sept. 2008, J. Vauras 26578 (TUR-A). – France, Merlimont, Pas de Calais, 2 Nov. 2006, EL238-06; Quend-les-Pines, 18 May 1983, Bon83047 (LIP). – Great Britain, Sandscale, Haws-Nears, Barrow on Furness, 31 Aug. 1999, RC/GB99.014.

(Kühner) Bon, Bull. Mycol. Bot. Dauphiné-Savoie 32 (no. 126): 19. 1992

Specimens examined. France, Les Arcs 2000 (73), Lac Marloup, 24 Aug. 2000, Bon (LIP). – Norway, Hordaland, Ulvik, Finse, Sandalsnut, 12 Aug. 2005, EL51-05; Hordaland, Ulvik, Finse, Sandalsnut, 12 Aug. 2005, EL47-05; Hordaland, Ulvik, Finse, Sandalsnut, 14 Aug. 2005, EL66-05. – Sweden, Härjedalen, Tännäs, Hamra, Hamrafjället, 15 Aug. 2006, EL88-06; Torne Lappmark, Jukkasjärvi, Latnjajaure, 3 Aug. 2006, EL30-06; Torne Lappmark, Jukkasjärvi, Latnjajaure, 4 Aug. 2006, EL37-06; Torne Lappmark, Jukkasjärvi, Latnjajaure, 11 Aug. 2007, EL75-07.

Kühner, Doc. Mycol. 19 (no. 74): 18. 1988

Specimens examined. Sweden, Härjedalen, Tännäs, Hamra, Hamrafjället, 18 Aug. 2006, EL112-06. – USA, Montana, Carbon County, Quad Creek, 8 Aug. 2008, EL29-08.

P. Karst., Meddel. Soc. Fauna Fl. Fenn. 16: 100. 1890

Specimens examined. Finland, Etelä-Hämä, Juupajoki, Korkeakoski, 9 Sept. 2005, EL118-05. – France, Abscon, Carriere des Peupliers, 25 Apr. 2005, PAM05042502 (LIP); Isère, Saint Laurent du Pont, 6 Oct. 2000, PAM00100606 (LIP). – Great Britain, England, Haycop, Shropshire, 24 Sept. 2008, G. Kibby. – Sweden, Bohuslän, Torslanda, Sillvik, 3 July 2004, SJ04-005; Bohuslän, Tanum, Kalvö, 16 Sept. 2004, EL90-04; Bohuslän, Munkedal, Foss, 9 Sept. 1989, LAS89/030; Skåne, Fågeltofta, Kronovalls sumpskog, 21 Aug. 2005, EL137-05; Skåne, Fågeltofta, Kulladal, 21 Sept. 2005, EL137-05; Västergötland, Östad, Djurgården, 3 Sept. 2008, EL56-08; Värmland, Övre Ullerud, Torberget, 30 Aug. 1990, B. Jansson; Värmland, Fryksände, Fensbol, 29 Aug. 1992, B. Jansson.

Kühner, Bull. Trimestriel Soc. Mycol. France 71: 169. 1956 (‘1955’)

Specimens examined. France, Beaufort, Savoie, Col des Pris, 28 Aug. 2008, PAM08082801 (LIP).– Sweden, Härjedalen, Tännäs, Sandåsvallen, 16 Aug. 2006, EL97-06.

Massee, Ann. Bot., Lond. 18: 492. 1904

Specimens examined. Sweden, Öland, Penåsa, 8 Sept. 2004, TK2004-114; Gotland, Tjaukle, 7 Oct. 1996, Elsa Bohus-Jensen.

Romagn., Bull. Trimestriel Soc. Mycol. France 74: 145. 1958

Specimens examined. Estonia, Saaremaa, Torgu, Viieristi Nature Reserve, 20 Sept. 2008, J. Vauras 26619 (TUR-A). – Sweden, Bohuslän, Valla, Sundsby, 12 Sept. 1982, SJ82068; Bohuslän, Torslanda, Röds skalgrus-bank, 28 Aug. 2008, SJ08-006; Bohuslän, Resteröd, Ulvesund, 25 July 2004, EL17-04; Bohuslän, Resteröd, Ulvesund, 18 Sept. 2004, EL100-04; Västergötland, Trollhättan, Åkerströms naturreservat, 10 Aug. 1986, LAS86/024; Västergötland, Fors, nära Slumpån, 5 Aug. 1977, Leif Stridvall; Värmland, Karlstad, Trangård, 15 Sept. 1989, Bo Jansson.

(Cooke) Sacc., Syll. Fung. (Abellini) 5: 774. 1887

Specimens examined. Finland, Varsinais-Soumi, Lohja rural commune, Vitkkala, 7 Aug. 1988, J. Vauras 3091. – Sweden, Bohuslän, Uddevalla, Kuröds skalbankar, 15 Sept. 2004, EL74-04; Värmland, Visnum, Värmlands Säby, 22 Sept. 1994, Bo Jansson.

(Bull.: Fr.) P. Kumm., Führer Pilzk. (Zwickau): 78. 1871

Specimens examined. Estonia, Hiiumaa, Käina, Kassari, Sääre, 17 Sept. 2001, J. Vauras 17954 (TUR-A). – Finland, Varsinais-Soumi, Lohja, Virkkala Kyrkstad, 16 July 1998, J. Vauras 13699 (TUR-A); Varsinais-Soumi, Lohja, Virkkala, Pähkinäniemi, 10 Aug. 1993, J. Vauras 8125 (TUR-A). – France, Lille, Faculté du Pharmacie, 23 May 2005, PAM05052303 (LIP); Lille (Nord), 11 June 2005, PAM05061101 (LIP); Pyrénées Atlantiques, Odres, 9 Nov. 2003, PAM03110906 (LIP); Corsica, Bonifacio, La Tonnare, 27 Nov. 2006, PAM06112703 (LIP). – Great Britain, England, Cambs, Fowlmere RSPB reserve, 2008, A. Outen 2008/0250; England, Nesscliffe, Shropshire, 25 Sept. 2008, P. Cullington 08.09.25; England, Essex, Hales Wood, 25 Sept. 2008, J. Darby 2008/0241. – Sweden, Bohuslän, Tanum, Kville Hjärterön, 14 Sept. 2004, EL54-04; Bohuslän, Sotenäs, Hogsäms bokskog, 15 Sept. 2004, EL71-04; Bohuslän, Tanum, Lur, Galtö, 15 Sept. 2004, EL80-04; Torne Lappmark, Jukkasjärvi, Abisko, Björkliden, 17 Aug. 1999, J. Vauras 15200; Västergötland, Göteborg, Sahlgrenska, 13 July 2004, SJ04-007; Västergötland, Alingsås, Kullingsberg, 19 Sept. 2004, EL102-04; Västergötland, Alingsås, Nolhagaparken, 21 Aug. 2005, EL75-05; Västergötland, Meldelplana, Råbäcks Munkängar, 27 Sept. 2004, EL127-04; Öland, Torslunda, S. om Tvetabäcken, 12 Oct. 1997, TK97-156.

J.E. Lange, Dansk Bot. Ark. 2 (no. 7): 39. 1917

Specimens examined. Finland, Varsinais-Soumi, Turku, Ilpoinen, 14 July 1987, J. Vauras 2607 (TUR-A). – France, Monbéqui, Tarn et Garonne, 11 Nov. 2003, PAM03111204 (LIP); Lille, Faculté du Pharmacie, 23 May 2005, PAM05052301 (LIP). – Norway, Oppland, Östra Toten, 5 July 1977, J.Stordal 18318 (O); Oppland, Lunner, 30 July 2004, T.E. Brandrud 102-04 (O); Oslo, Hovedöya, 19 Aug. 1985, SJ85048. – Sweden, Bohuslän, Torslanda, Röds skalgrusbank, 10 Aug. 2008, SJ08-003; Bohuslän, Torslanda, Röds skalgrusbank, 28 Aug. 2008, SJ08-007; Jämtland, Lit, Niklasbodarna, 10 Aug. 1992, SJ 92010; Jämtland, Östersund, Lövbergaparken, 11 Aug. 1992, SJ92-017; Medelpad, Tuna, Uvberget, 19 Aug. 1992, S.Muskos 92-013; Öland, Gräsgård, Löt, SSV Solberga, 27 Aug. 1997, TK96-109.

Bres., Ann. Mycol. 3 (2): 161. 1905

Specimens examined. Estonia, Hiiumaa district, Käina commune, Kassari, 17 Sept. 2001, J. Vauras 17954 (TUR). – Finland, Varsinais-Suomi, Lohja, Virkala, 16 July 1998, J. Vauras 13699 (TUR-A). – France, Ile de Porquerolles, 29 Oct. 2001, PAM01102912 (LIP). – Great Britain, England, Oxon, Watlington Hill, 10 Oct. 2008, P. Cullington 10.10.08; England, Bucks, Kings Wood, 16 Aug. 2008, P. Cullington 16.08.08. – Italy, Alto Adige, Trento, Desert, 3 June 1899, Bresadola (holotype S).

Basidiocarp ± reddening with age or handling……………2

Basidiocarp not reddening……………4

Pileus robust, whitish, slowly turning brick red with age or from damage……………I. erubescens

Frb red brown with a ± vinaceous tinge……………3

Pileus robust, fibrillose, sometimes breaking up into scales, stipe slowly staining somewhat vinaceous. Smell rather strong, reminding of beetroot……………I. adaequata

Pileus more slender, fibrillose–squamose. Stipe becoming distinctly vinaceous red towards the base. Smell weak…………………………I. rhodiola

With evident, distinct smell of various compounds but not spermatic. Spores generally phaseoliform. Cheilocystidia broadly clavate to pyriform……………5

Smell, if present, spermatic. Spores variable, mostly ellipsoid. Cheilocystidia generally slenderly clavate or cylindrical……………8

Pileus predominantly yellow. Stipe with a distinct bulb……………6

Pileus brown or brownish. Stipe equal or subbulbous……………7

Smell of honey. Spores 7–9 × 4–5 μm, distinctly phaseoliform. Cheilocystidia pyriform……………I. cookei

Smell recalling Lactarius quietus, spores 8–11 × 5.5–6.5 μm, less distinctly phaseoliform……………I. quietiodor

Pileus hazel brown to dark brown, often with white, conspicuous velar patches at centre. Growing in nutrient-rich Fagus or Quercus forests……………I. maculata

Pileus brownish, generally with a ± fulvous tinge, white velar patches often less evident. Growing with various broad-leaved or coniferous trees……………I. maculata forma fulva

Pileus with scales (may sometimes disappear)……………9

Pileus fibrillose-rimose, without scales……………11

Spores 12–15 × 6–8 μm, ellipsoid. Pileus > 65 mm, reminding of I. rimosa, gills with a faint olivaceous tinge. Stipe initially whitish, then reddish brown. Under deciduous trees on calcareous soils. Very rare and poorly known…………………………I. mimica

Spores smaller……………10

Spores 8–10 × 5.5–6.5 μm, broadly ellipsoid (Q = 1.4–1.6). Pileus 20–50 mm, yellowish brown; lamellae initially without or with only weak yellow tinge. With deciduous trees on calcareous soils. In temperate or hemiboreal areas…………………………I. squamata

Spores 8.5–11 × 4.5–6 μm, often somewhat phaseoliform (Q = 1.6–2.0). Pileus 30–90 mm, yellowish to reddish brown, outwards more yellow. Lamellae initially pale yellow. With deciduous and coniferous trees, boreal……………Inocybe sp.

Growing with Salix repens or Pinus in dune sand along coasts in western Europe. Pileus 25–70 mm, initially whitish due to thick velipellis, beneath this straw yellow or ochraceous; gills initially white. Stipe solid, often deeply buried in sand. Spores 12–16 × 6–8.5 μm……………I. arenicola

In other habitats. Pileus generally yellow to brown……………12

Pileus flocculose from a thick universal veil, ochraceous brown, 10–25 mm (reminding of I. dulcamara). Cheilocystidia with internal drops of brown pigments, broadly clavate to utriform. Spores 11–14 × 7–8.5 μm. With dwarf Salix in arctic/alpine zones……………I. dulcamaroides

Pileus fibrillose-rimose. Cheilocystidia without drops……………13

Arctic/alpine species with a distinct, white bulb at the base and large spores, 12–15 × 6–8 μm. Pileus 10–35 mm, pale yellow, then ochraceous to reddish brown. Smell spermatic……………I. bulbosissima

Not so……………14

Pileus only finely fibrillose (almost smooth in centre), pale brownish with a yellow tinge, lamellae with a yellowish olive tinge. Stipe almost equal, 50–80 × 8–12 mm. Spores 9–11 × 4.5–5.5 μm, phaseoliform. In mountainous areas, mixed forest with Salix……………I. hygrophorus

Pileus distinctly fibrillose/rimose……………15

Pileus brown without yellow tinges……………16

Pileus not brown, with ± yellow pigments……………17

Large species with acute umbo, reminding of I. rimosa. Pileus 35–100 mm. Stipe 80–120 × 8–13 mm, becoming brownish with age. Spores 10–13 × 6–8 μm. Under deciduous trees in forests and parks……………I. perlata

Smaller species. Pileus 20–45 mm, hazel to cinnamon brown, usually with blunt umbo. Spores 10–13 × 5.5–6.5 μm……………I. umbrinella

Pileus with a distinct white velipellis. Lamellae without olivaceous tinge. Odour absent. Microscopically as I. rimosa. Under deciduous trees on calcareous ground……………I. obsoleta

Pileus without distinct white velipellis……………18

Pileus typically distinctly umbonate and strongly rimose. Lamellae with an olivaceous yellow tinge. Smell spermatic. Spores 9.5–12.5 × 6–7 μm, generally ellipsoid and only exceptionally somewhat phaseoliform……………I. rimosa

Pileus fibrillose-rimulose, not strongly rimose. Lamellae initially clay-coloured lacking or with only a weak yellow tinge. Smell absent. Spores 9–12 × 5–6 μm, often phaseoliform. Cheilocystidia narrow. Pileus more yellow and less rimose……………19

Spores 10–12 × 5–6.5 μm, not phaseoliform. Pileus 20–60 mm, fibrillose, yellowish brown, predominantly yellow towards the margin……………I. flavella

Spores 9–10.5 × 5–6 μm, ± phaseoliform……………I. cfr flavella