Cortinarius subgenus Leprocybe in Europe: expanded Sanger and Next Generation Sequencing unveil unexpected diversity in the Mediterranean.

Molecular phylogenies in the past decade have demonstrated that the described diversity of Cortinarius is still underestimated, especially outside continental and boreal ecoregions where the genus has been historically investigated. We tackled this issue by revisiting the so far unresolved subgenus Leprocybe, and focused on the largely unexplored Mediterranean hotspot of biodiversity. The sequencing and phylogenetic analysis of 161 vouchered collections from Austria, Cyprus, France, Germany, Italy and Spain, including 16 types, allowed for the delineation of 11 species in this lineage, three of them recognised as new to science and formally introduced as C. jimenezianus, C. selinolens and C. viridans spp. nov., respectively. Interestingly, the newly described species exhibit a strict Mediterranean distribution, and one of them is putatively endemic to the island of Cyprus, highlighting the remarkable potential of this neglected ecoregion to uncover further undescribed diversity of Cortinarius in the future. The present work also unveils 23 synonymies in this subgenus, as well as previously undetected crypticism within C. venetus. Next Generation Sequencing carried out on three old and contaminated holotypes, successfully decrypts their phylogenetic identity, including that of C. leproleptopus, finally settling the long-standing controversy over the taxonomic status of this species. A brief overview of each species in the subgenus is lastly provided and a key is proposed to facilitate the identification of presently known European taxa of Leprocybe in the field. Citation: Bidaud A, Loizides M, Armada F, et al. 2021. Cortinarius subgenus Leprocybe in Europe: expanded Sanger and Next Generation Sequencing unveil unexpected diversity in the Mediterranean. Persoonia 46: 188-215. https://doi.org/10.3767/persoonia.2021.46.07.

INTRODUCTION

Cortinarius is the largest genus of macrofungi in the order Agaricales, widely distributed in the Holarctic and Austral realms (Tedersoo & Nara 2010, Garnica et al. 2016, Soop et al. 2019a). Over 4 500 species names of Cortinarius have been published to date, mostly from Europe and North America (http://www.catalogueoflife.org, http://www.mycobank.org, accessed 31 Oct. 2019). However, many of these appear to be synonyms, while the phylogenetic and taxonomic relationships within many groups are still poorly understood, and the number of taxa in the genus is expected to significantly increase in the coming years (Garnica et al. 2016, Soop et al. 2019a). As ectomycorrhizal (EcM) symbionts, Cortinarius species are integral components of woodland ecosystems, enhancing plant productivity and facilitating the cycling of vital nutrients such as nitrates and phosphates (Smith & Read 2010, Bödeker et al. 2014, Martin et al. 2016, Pellitier & Zak 2017). Most frequently reported tree-hosts are members of the Fagaceae, Pinaceae and Salicaceae, but several Cortinarius taxa are also associated with sclerophyllous shrubs of the Cistaceae family, most notably Cistus (Henry & Contu 1986, Comandini et al. 2006, Vila & Llimona 2006, Loizides 2016) and Helianthemum (Cheype 2014, Liimatainen & Ainsworth 2018).

In spite of some noteworthy systematic advances in recent years, our knowledge of species diversity in Cortinarius remains incomplete and many sections that have been historically defined based on morphological or biochemical taxonomic criteria, have yet to be phylogenetically revised (Niskanen et al. 2012, Soop et al. 2019a). European monographic treatments of the genus have traditionally focussed on continental and northern European ecoregions, and species from southern European/Mediterranean regions have been for the most part poorly represented. Nevertheless, a number of studies and some notable pioneers from France, Italy and Spain (e.g., Chevassut & Henry 1975, Chevassut & Trescol 1986, Ortega & Mahiques 1995, Consiglio & Moënne-Loccoz 2004), have significantly contributed to documenting the diversity of Mediterranean cortinarii in regional bulletins. Among several novelties introduced from the Mediterranean, many were later included in two major monographs, the Atlas des Cortinaires (ADC), published as 24 ‘Pars’ between 1990 and 2017 (e.g., Moënne-Loccoz & Reumaux 1990, Bidaud et al. 2017) and Il Genere Cortinarius in Italia published as 6 ‘parte’ between 2003 and 2012 (e.g., Consiglio et al. 2003, Consiglio 2012). Still, perhaps with the exception of some impressive fleshy species belonging to subg. Phlegmacium, the Mediterranean basin has not been as thoroughly investigated as other parts of the continent, despite the fact that this region is a well-known hotspot of biodiversity (Meìdail & Queìzel 1997, Myers et al. 2000, Meìdail & Myers 2004).

More recently, a number of studies employed molecular tools to unveil overlooked diversity in Mediterranean cortinarii and test the phylogenetic autonomy of previously published species described on the basis of morphology (Ortega et al. 2008, Suárez-Santiago et al. 2009, Clericuzio et al. 2017, Bellanger et al. 2018, Mahiques et al. 2018, Soop et al. 2019b). In the present paper, we investigate subg. Leprocybe (= sect. Veneti), which includes species akin to C. cotoneus and C. venetus. Basidiomata in this group display dry and tomentose pilei cuticles with green-olivaceous to yellow-rusty tinges, subglobose to broadly ellipsoid warty spores, and unique UV-fluorescent pigments known as ‘leprocybins’ (Moser 1969, Brandrud et al. 1990, 1992, 1994, 1998, 2014, Consiglio et al. 2003, 2004, Bidaud et al. 2005, Niskanen & Kytövuori 2012). Currently, approx. 30 published names from Europe, North America and Costa Rica fit this circumscription and are thus included in this subgenus (Table 1, 2). Because some defining features are shared by C. callisteus, C. limonius, C. orellanus and other related taxa, a subg. Leprocybe in a broad sense was initially proposed by Moser (1969). However, phylogenetic studies have since indicated that such ‘leprocyboid’ fungi do not constitute a monophyletic arrangement and, taking into consideration the novelties from the southern hemisphere, this group is in need of an in-depth taxonomic revision (Peintner et al. 2004, Ammirati et al. 2012, Stensrud et al. 2014, Niskanen et al. 2016, Soop et al. 2019a).

Table 1

Cortinarius species classified in subg. Leprocybe (= sect. Veneti) by the main European authors.

This study	Niskanen & Kytövuori (2012) *	Bidaud et al. (2005)	Consiglio et al. (2003, 2004, 2006)	Moser (1969)
Cortinarius cotoneus	C. cotoneus	C. conopus, C. cotoneus, C. cotoneus var. mellinoides, C. perannulatus, C. psittacinus p.p., C. subnotatus	C. cotoneus p.p. ?	C. cotoneus, C. subannulatus
C. jimenezianus sp. nov.	absent	absent	absent	absent
C. leproleptopus	absent	C. leproleptopus, C. leptopus, C. melleifolius var. basiluteus p.p., C. psittacinus p.p.	C. leproleptopus	C. mellinus p.p., C. psittacinus p.p. ?
C. melanotus	C. melanotus	C. colymbadinus, C. melanotus, C. submelanotus, C. zonatus	C. melanotus	C. melanotus
C. pescolanensis	absent	C. argyrocognitus, C. junquilleovestitus, C. mellinus, C. psittacinus p.p.	absent	C. mellinus p.p.
C. phrygianus	C. phrygianus	C. phrygianus (sect. Pholidei)	absent	C. phrygianus
C. selinolens sp. nov.	absent	absent	absent	absent
C. squamivenetus	absent	absent	absent	absent
C. subcotoneus	absent	C. borealis, C. melleifolius var. basiluteus p.p., C. subcotoneus, C. xantholamellatus	C. cotoneus p.p.?	C. sublanatus ss Fries ?
C. venetus	C. venetus	C. cf. nothoraphanoides, C. pseudovenetus, C. venetus, C. venetus var. viridis	C. venetus	C. nothoraphanoides, C. venetus
C. viridans sp. nov.	absent	absent	absent	absent

* includes Brandrud et al. 1990, 1991, 1994 and 1998. Bold names indicate sequenced species, p.p. = pro parte.

Table 2

Material included in the phylogenetic analysis.

GenBankSpecies	Voucher Id	Leg	Collection date	Putative host tree	Country	Locality	GenBank ITS*
Cortinarius atkinseae	TENN071879 (holotype)	R.M. Atkins & P.B. Matheny	18/09/2017	Fagus, Quercus, Carya	USA	Tennessee	MG663233
Cortinarius aureopigmentatus	JFA 11940	R. Halling & J.F. Ammirati	07/06/1996	Quercus	Costa Rica	San José	EF420140
Cortinarius clandestinus	JMB165	J.M. Birkebak	01/05/2004	Pseudotsuga, Tsuga	USA	Washington	FJ717551
Cortinarius cotoneus	PML 5260 (holotype of C. cotoneus var. mellinoides)	G. Redeuilh	07/10/1994	Broadleaved trees	France	Yvelines	MW010117
	PML 5429 (holotype of C. perannulatus)	J. Cavet	14/08/1994	Fagus, Abies	France	Isère	MW010116
	HMAS260331	NA	NA	NA	China	NA	KX513578
	HMAS254210	NA	NA	NA	China	NA	KX513580
	QL060 (ecto.)	NA	NA	Quercus	China	NA	HM105543
	ZWL560 (ecto.)	NA	NA	Quercus	China	Shaanxi	KX444284
	TU105155	A. Kollom	24/08/2011	Quercus	Estonia	Harju	UDB015877
	AB 04-10-338 (as C. subnotatus)	G. Raffini	05/10/2004	Fagus	France	Ain	MW010155
	AB 06-09-61	A. Bidaud	01/09/2006	Fagus, Carpinus	France	Ain	MW010160
	AB 06-09-69	A. Bidaud	03/09/2006	Picea	France	Ain	MW010161
	AB 09-09-277	A. Bidaud & R. Fillion	29/09/2009	Fagus, Quercus	France	Ain	MW010173
	AB 11-08-49	A. Bidaud	07/08/2011	Carpinus, Quercus	France	Ain	MW010175
	AB 11-09-83	A. Bidaud	21/09/2011	Carpinus, Quercus	France	Ain	MW010176
	AB 11-10-206	A. Bidaud & R. Fillion	17/10/2011	Fagus, Quercus	France	Ain	MW010177
	AB 12-10-96	A. Bidaud	02/10/2012	Carpinus	France	Ain	MW010179
	AB 09-10-336	A. Bidaud	25/10/2009	Carpinus, Quercus	France	Côte-d’or	MW010174
	PML 5427 (as C. conopus)	A. Faurite-Gendron	06/10/1994	Fagus, Abies	France	Drôme	MW010136
	AB 06-10-223	A. Bidaud	10/10/2006	Fagus	France	Drôme	MW010162
	PML 5430 (as C. subnotatus)	A. Bidaud	21/10/1998	Fagus, Pinus	France	Hautes-Alpes	MW010138
	AB 04-09-81 (as C. psittacinus)	A. Bidaud & R. Fillion	05/09/2004	Broadleaved trees	France	Haute-Savoie	MW010153
	PML 5417 (as C. conopus)	A. Bidaud	13/10/1999	Fagus, Abies	France	Isère	MW010132
	AB 05-09-66	A. Bidaud	02/09/2005	Fagus, Abies	France	Isère	MW010157
	AB 05-10-241	A. Bidaud	04/10/2005	Fagus, Abies	France	Isère	MW010159
	JMB2013092505	J.-M. Bellanger	25/09/2013	Abies, Fagus	France	Isère	MW010108
	GS2	G. Saar	30/08/2004	Picea, Abies, Pinus	Germany	Baden-Würtemberg	MW010093
	GS3	G. Saar	17/09/2008	Picea, Abies, Pinus, Fagus	Germany	Baden-Würtemberg	MW010094
	GH20131018a	G. Hensel	18/10/2013	Carpinus	Germany	Saxony-Anhalt	MW010083
	GH20110930	G. Hensel	30/09/2011	Corylus	Germany	Saxony-Anhalt	MW010086
	GH20050820	G. Hensel	20/08/2005	Quercus, Carpinus	Germany	Saxony-Anhalt	MW010087
	MH4836-2	M. Huth	29/10/2009	Quercus, Carpinus	Germany	Saxony-Anhalt	MW010091
	GH20140811	G. Hensel	11/08/2014	Carpinus	Germany	Saxony-Anhalt	MW010113
	GH20140914	G. Hensel	14/09/2014	Corylus	Germany	Saxony-Anhalt	MW010114
	SSt01-100	G. Schmidt-Stohn	09/10/2001	Fagus	Germany	Schleswig-Holstein	MW010080
	Ir22 (ecto.)	NA	NA	Fagales	Iran	NA	FR852015
	ecto.	NA	NA	Fagaceae	Iran	Nowshahr	UDB005580
	IB19860257	M. Moser	20/09/1986	Tilia	Italy	Trento	UDB001029
	OS579	NA	NA	NA	Norway	Oslo	KC842423
Cortinarius flavifolius	TENN071880	A.M. Overholt	11/09/2017	Pinus, Fagus, Quercus, Carya	USA	Tennessee	MG663246
Cortinarius fuscoflavidus	ecto.	NA	NA	NA	Canada	British Columbia	UDB032444
	OSC 80195	NA	NA	NA	USA	Oregon	EU669233
	115-1 (ecto.)	NA	NA	Pseudotsuga	USA	Oregon	DQ365640
Cortinarius hugheseae	TENN:TLB11 (as C. cotoneus)	T.L. Bailey & P.B. Matheny	12/10/2015	Pinus, Fagus	USA	Tennessee	MF686506
Cortinarius jimenezianus	JA-CUSSTA 9402 (holotype)	F. Armada & J.D. Reyes	17/12/2018	Cistus ladanifer, Q. ilex	Spain	Jaén	MW010204
	JDRG26120401	J.D. Reyes	26/12/2004	Cistus ladanifer, Pinus	Spain	Jaén	MW010205
	AMI3777	A. Mateos Izquierdo	09/12/2014	Cistus ladanifer	Spain	Jaén	MW010050
	JDRG17121801/FA4792	F. Armada & J.D. Reyes	17/12/2018	Cistus ladanifer, Q. ilex	Spain	Jaén	MW010195
Cortinarius leproleptopus	R. Henry 84.109 (holotype)	probably G. Chevassut	probably 1984	probably Quercus	France	probably Hérault	MW010056
	MCVE763/E.C. 94190 (epitype)	G. Consiglio & B. Bigazzi	09/10/1994	Castanea, Q. pubescens, Fagus, Pinus	Italy	Reggio Emilia	MW010049
	G00126720/AF-99-10-17CC (holotype of C. cepistipes)	A. Favre	17/10/1999	Q. robur	France	Haute-Savoie	MW010197
	AB 99-10-257 (paratype of C. cepistipes)	A. Bidaud	16/10/1999	Quercus	France	Ain	MW010147
	PAM00103106 (paratype of C. cepistipes)	J.-C. Hermite	31/10/2000	Quercus	France	Bouches-du-Rhône	MW010052
	PAM99101703 (paratype of C. cepistipes)	J.-C. Deïana, R. Fillion, P.-A. Moreau	17/10/1999	Q. robur, Carpinus	France	Haute-Savoie	MW010187
	PAM99111001 (paratype of C. cepistipes)	P.-A. Moreau	10/11/1999	Q. ilex, Q. pubescens	France	Var	MW010188
	ML21211CL	M. Loizides	01/12/2012	Q. alnifolia, Pinus, Arbutus	Cyprus	Mavro Dasos	MW010208
	AB 08-09-199	A. Bidaud	20/09/2008	Quercus, Carpinus	France	Ain	MW010164
	AB 03-11-81 (as C. melleifolius var. basiluteus)	A. Bidaud	05/11/2003	Q. ilex	France	Ardèche	MW010151
	AB 18-11-159	A. Bidaud	15/11/2018	Q. ilex	France	Ardèche	MW010194
	FR2312003	P.-A. Moreau	NA	NA	France	Bouches-du-Rhône	MW010065
	AB 08-09-215	A. Bidaud & F. Armada	22/09/2008	Fagus	France	Drôme	MW010165
	AB 08-10-320	A. Bidaud	17/10/2008	Q. ilex	France	Drôme	MW010166
	AB 08-10-395	A. Bidaud	31/10/2008	Q. ilex	France	Gard	MW010171
	PML 2949 (as C. psittacinus)	A. Bidaud & R. Fillion	30/09/1992	Broadleaved trees	France	Haute-Savoie	MW010128
	PML 5420	P. Bertéa	31/10/1999	Quercus	France	Hérault	MW010133
	JMB2010110201	J.-M. Bellanger	02/11/2010	Q. ilex	France	Hérault	MW010207
	OTU27 (ecto.)	E. Taschen & F. Richard	02/2011	Q. ilex	France	Pyrénées-Orientales	KM247711
	GE01-049	NA	13/10/2001	NA	France	Val-de-Marne	MW010150
	PAM99111002	P.-A. Moreau	10/11/1999	Q. ilex, Q. pubescens	France	Var	MW010196
	PML 5421	A. Bidaud & M. Heullant	02/11/1993	Quercus	France	Vaucluse	MW010134
	JMB2011112628	J.-M. Bellanger	26/11/2011	Q. pubescens, Q. ilex, Pinus	France	Vaucluse	MW010068
	JMB2011112629	J.-M. Bellanger	26/11/2011	Q. pubescens, Q. ilex, Pinus	France	Vaucluse	MW010067
	GS9	G. Saar	01/10/2012	Fagus	Germany	Baden-Würtemberg	MW010100
	SSt12-039	G. Saar	27/09/2012	Fagus	Germany	Bavaria	MW010077
	GH20140907	G. Hensel	07/09/2014	Corylus	Germany	Bavaria	MW010112
	GS1	G. Saar	27/09/2012	Fagus	Germany	Bavaria	MW010092
	GS4	G. Saar	26/09/2012	Fagus, Carpinus, Quercus	Germany	Bavaria	MW010095
	GS8	G. Saar	08/10/1998	Fagus	Germany	Bavaria	MW010099
	FR2014330	M. Huth	29/10/2009	Quercus, Carpinus	Germany	Saxony-Anhalt	MW010053
	FR2014332	M. Huth	14/10/2009	Quercus	Germany	Saxony-Anhalt	MW010074
	MH4835-4	M. Huth	12/10/2013	Quercus, Carpinus	Germany	Saxony-Anhalt	MW010076
	SSt13-173	G. Schmidt-Stohn	11/10/2013	Carpinus	Germany	Saxony-Anhalt	MW010079
	GH20091024	G. Hensel	24/10/2009	Betula, Carpinus	Germany	Saxony-Anhalt	MW010085
	GH20131018b	G. Hensel	18/10/2013	Quercus	Germany	Saxony-Anhalt	MW010089
	GH20140911b	G. Hensel	11/09/2014	Carpinus, Quercus	Germany	Saxony-Anhalt	MW010115
	ST40	S. Toninelli	11/10/2014	Q. cerris	Italy	Grosseto	MW010181
	IB19960212 (as C. mellinus)	M. Moser	06/10/1986	Q. cerris	Italy	Parma	UDB001105
	MES4373	R. Mahiques Santandreu	02/11/2007	Q. ilex	Spain	Alicante	MW010048
Cortinarius melanotus	PML 5454 (holotype of C. submelanotus)	B. Rainon	11/11/2000	Broadleaved trees	France	Rhône	MW010120
	XC2005-20 (holotype of C. zonatus)	P. Reumaux	24/10/2004	Quercus	France	Ardennes	MW010121
	PML 5461	A. Bidaud	27/10/1990	Fagus, Carpinus, Castanea, Abies	France	Ain	MW010142
	PML 5455 (as C. perannulatus)	A. Bidaud	01/11/1997	Abies, Castanea	France	Ain	MW010141
	AB 04-09-70	A. Faurite-Gendron	02/09/2004	Abies	France	Ain	MW010152
	AB 05-09-83	T. Trémont	04/09/2005	Abies	France	Ain	MW010158
	GC08101103	G. Corriol	11/10/2008	Abies	France	Aude	MW010190
	FR2019579	J.-M. Bellanger	10/11/2019	Fagus, Quercus, Castanea	France	Aveyron	MW010200
	FR2012264	NA	02/11/2012	NA	France	Gard	MW010070
	PAM19110411	N. Suberbielle & P.-A. Moreau	06/11/2019	Q. suber, Pinus, Arbutus	France	Haute-Corse	MW010201
	PML 5462 (as C. submelanotus)	P. Chapon	16/10/1997	Abies	France	Haute-Loire	MW010143
	GC13092202	G. Corriol	22/09/2013	Fagus, Abies	France	Hautes-Pyrénées	MW010191
	GC18102206	G. Corriol	22/10/2018	Abies	France	Hautes-Pyrénées	MW010193
	AB 06-10-254	A. Bidaud & A. Faurite-Gendron	13/10/2006	Fagus, Picea, Pinus	France	Isère	MW010163
	PML 5428 (as C. colymbadinus)	G. Delavenat	24/09/1999	Abies	France	Jura	MW010137
	PML 2344	P. Moënne-Loccoz	10/10/1991	Fagus, Abies	France	Savoie	MW010125
	GS10	G. Saar	29/09/2012	Abies	Germany	Baden-Würtemberg	MW010101
	GS12	G. Saar	01/11/2012	Fagus, Quercus	Germany	Baden-Würtemberg	MW010103
	GS14	G. Saar	08/10/2004	Abies, Picea, Pinus	Germany	Baden-Würtemberg	MW010105
	GS11	G. Saar	13/10/2013	Fagus, Quercus	Germany	Bavaria	MW010102
	GS5	G. Saar	07/10/1998	Carpinus, Quercus	Germany	North Rhine-Westphalia	MW010096
	MH4735-4	M. Huth	08/09/2006	Quercus	Germany	Saxony-Anhalt	MW010073
	FR2014334	M. Huth	29/10/2009	Fagus	Germany	Saxony-Anhalt	MW010075
	SSt10-253	G. Schmidt-Stohn	04/10/2010	Quercus, Carpinus	Germany	Saxony-Anhalt	MW010078
	JM281014	J. Mora Gómez	28/10/2014	Pinus, Cistus	Spain	Guadalajara	MW010063
	JA-CUSSTA 2774	L. Romero de la Osa	30/11/2002	Q. suber, Castanea, Cistus	Spain	Huelva	MW010058
Cortinarius parkeri	WTU:ADP000502-1 (holotype)	D. Parker	22/04/2013	Conifers	USA	Washington	JN976987
Cortinarius pescolanensis	MCVE29054 (holotype)	B. Picillo & M. Marchionni	14/10/2012	Abies, Fagus	Italy	Molise	NR_153070
	BP13/291 (isotype)	B. Picillo & M. Marchionni	14/10/2012	Abies, Fagus	Italy	Molise	KX010945
	PML 5448 (as C. argyrocognitus ad int)	A. Faurite-Gendron	06/10/1994	Abies, Fagus	France	Drôme	MW010139
	PML 5426 (as C. junquilleovestitus ad int)	T. Tremont	05/09/1999	Abies, Picea	France	Jura	MW010135
	PML 3262 (as C. psittacinus)	J. Garin	09/10/1992	Abies, Fagus, Pinus	France	Drôme	MW010129
	PML 5449 (as C. mellinus)	A. Bidaud	01/10/1995	Abies, Fagus	France	Isère	MW010140
	JMB2013092404	J.-M. Bellanger	24/09/2013	Abies, Fagus	France	Isère	MW010107
	JMB2013092503	J.-M. Bellanger	25/09/2013	Abies, Fagus	France	Isère	MW010109
	PAM02082506	P.-A. Moreau	25/08/2002	Abies, Fagus	France	Savoie	MW010186
	JB-8114/13	J. Ballarà	21/09/2013	Abies, Fagus, Pinus	Spain	Lérida	KY657256
Cortinarius phrygianus	TU105353	T. Ploompuu	27/09/2012	Picea, Pinus	Estonia	Lääne	UDB018645
	PML 5451/AB 97-10-332	G. Chamonaz	16/10/1997	Picea, Pinus	France	Ain	MW010146
	KS-CO584	K. Soop	10/09/1992	Pinus	Sweden	Dalarna	KJ421031
	AT2002139	A.F.S. Taylor	17/08/2002	Picea	Sweden	Västerbotten	UDB000688
Cortinarius selinolens	MPU1116858 (holotype)	A. Bidaud	31/10/2008	Q. ilex	France	Gard	MW010172
	AB 11-11-377/JMB2011112503	A. Bidaud & J.-M. Bellanger	25/11/2011	Q. ilex	France	Hérault	MW010071
	JDRG02121005	J.D. Reyes	02/12/2010	Q. ilex	Spain	Jaén	MW010059
	FR2013185	F. Richard	01/11/2010	Arbutus	Tunisia	Jendouba	MW010072
Cortinarius squamivenetus	H:TN 07-119 (holotype)	T. Niskanen, I. Kytövuori & K. Liimatainen	01/09/2007	Picea, Pinus, Populus	Finland	Koillismaa	KP114458
Cortinarius subcotoneus	PML 2143 (holotype)	A. Bidaud	09/11/1990	Q. pubescens	France	Drôme	MW010122
	PML 5412 (holotype of C. borealis)	C. Guyot	23/09/1993	Q. pubescens, Carpinus	France	Isère	MW010119
	PML 5410 (holotype of C. xantholamellatus)	C. Guyot	03/11/1990	Q. pubescens	France	Drôme	MW010054
	JLC001029-8 (as C. melleifolius var. basiluteus)	J.-L. Cheype	29/10/2000	Q. ilex, Q. pubescens	France	Alpes de Haute-Provence	MW010149
	AB 08-10-331	A. Bidaud	20/10/2018	Q. ilex	France	Ardèche	MW010167
	AB 08-10-387	A. Bidaud	27/10/2018	Q. ilex	France	Ardèche	MW010168
	AB 08-10-388	A. Bidaud	27/10/2018	Q. ilex	France	Ardèche	MW010169
	AB 08-10-394	E. Bidaud & G. Raffini	31/10/2018	Q. ilex	France	Gard	MW010170
	AB 99-11-324	P. Pellicier	01/11/1999	Q. ilex	France	Hérault	MW010148
	XC2013-109	T.E. Brandrud	22/10/2013	Q. ilex	France	Hérault	MW010180
	FR2012097	F. Richard	25/11/2011	Q. ilex, Q. pubescens	France	Hérault	MW010051
	SSt13-211	G. Schmidt-Stohn	23/10/2013	Q. ilex	France	Hérault	MW010081
	GH20131023	G. Hensel	23/10/2013	Q. ilex	France	Hérault	MW010082
	PML 5411	A. Bidaud	29/10/1992	Q. ilex	France	Vaucluse	MW010131
	JMB2011112632	J.-M. Bellanger	26/11/2011	Q. pubescens, Q. ilex, Pinus	France	Vaucluse	MW010069
	JMB2011112636	J.-M. Bellanger	26/11/2011	Q. pubescens, Q. ilex, Pinus	France	Vaucluse	MW010066
	GS7	G. Saar	16/10/2012	Fagus	Germany	Baden-Würtemberg	MW010098
	GS15	G. Saar	21/10/2012	Fagus, Abies	Germany	Baden-Würtemberg	MW010106
	TUB011826 (as C. cotoneus)	NA	NA	NA	Germany	NA	AY669597
	GH20100827	G. Hensel	27/08/2010	Carpinus	Germany	Saxony-Anhalt	MW010088
	GH20140911a	G. Hensel	11/09/2014	Quercus	Germany	Saxony-Anhalt	MW010111
	GS6	G. Saar	25/10/2012	Fagus, Quercus	Germany	Schleswig-Holstein	MW010097
	ST42	S. Toninelli	19/10/2014	Q. ilex	Italy	Grosseto	MW010182
	JDRG08120801	J.D. Reyes	08/12/2008	Q. suber, Cistus ladanifer	Spain	Jaén	MW010060
	JDRG17110810	J.D. Reyes	17/11/2008	Q. ilex	Spain	Jaén	MW010061
Cortinarius venetus	PML 245 (holotype of C. venetus var. viridis)	P. Moënne-Loccoz	18/09/1985	Picea	France	Haute-Savoie	MW010118
	R. Henry 3412 (holotype of C. pseudovenetus)	R. Henry	NA	NA	France	NA	MW010055
	IB19650108 (as C. nothoraphanoides)	M. Moser	15/09/1965	Pinus, Picea	Austria	Tyrol	MW010183
	HMAS268596	NA	NA	NA	China	NA	KX513586
	HMAS274352	NA	NA	NA	China	NA	KX513585
	HMAS274611	NA	NA	NA	China	NA	KX513584
	TAAM128287	A. Kollom	14/09/2001	Pinus	Estonia	Hiiu	UDB019877
	TAAM128694	A. Kollom	16/10/2007	Picea	Estonia	Järva	UDB015960
	TAAM128704	A. Kollom	16/10/2007	Picea	Estonia	Järva	UDB019876
	TU106430	V. Liiv	10/10/2008	Mixed forest	Estonia	Saare	UDB011758
	TU106896	V. Liiv	28/09/2010	Mixed forest	Estonia	Saare	UDB011272
	TU120024	V. Liiv	01/09/2014	Picea	Estonia	Tartu	UDB023575
	AB 12-09-62	A. Bidaud	22/09/2012	Picea	France	Ain	MW010178
	PML 5464/AB 00-08-13	T. Trémont	08/08/2000	Picea	France	Ain	MW010144
	AB 04-09-251 (as C. cf. nothoraphanoides)	A. Faurite-Gendron	25/09/2004	Picea, Fagus	France	Doubs	MW010154
	PML 3265 (as C. cf. nothoraphanoides)	J. Garin	04/10/1992	Fagus, Abies	France	Drôme	MW010130
	PML 5466 (as C. cf. nothoraphanoides)	A. Bidaud	05/10/1994	Fagus, Pinus	France	Hautes-Alpes	MW010145
	PML 1665	P. Moënne-Loccoz	12/07/1990	Picea	France	Haute-Savoie	MW010123
	PML 1781 (as C. pseudovenetus)	P. Moënne-Loccoz	01/11/1990	Picea	France	Haute-Savoie	MW010124
	PML 2393 (as C. pseudovenetus)	P. Moënne-Loccoz	10/11/1991	Picea	France	Haute-Savoie	MW010126
	GC18090805	G. Corriol	08/09/2018	Fagus, Abies	France	Hautes-Pyrénées	MW010192
	GC18091404	G. Corriol	14/09/2018	Abies	France	Hautes-Pyrénées	MW010189
	AB 05-09-65	A. Bidaud	02/09/2005	Abies, Fagus	France	Isère	MW010156
	JMB2013092405	J.-M. Bellanger	24/09/2013	Fagus, Abies	France	Isère	MW010110
	GH20100927	G. Hensel	27/09/2010	Picea	Germany	Baden-Würtemberg	MW010090
	GS13	G. Saar	11/11/1999	Abies, Picea, Pinus	Germany	Baden-Würtemberg	MW010104
	GH20110814	G. Hensel	14/08/2011	Picea	Germany	Bavaria	MW010084
	TUB 011645 (as C. venetus var. montanus)	NA	NA	NA	Germany	NA	EF600897
	IB19970371	M. Moser	23/09/1997	Fagus	Italy	Abruzzo	UDB001076
	IB19650159 (as C. nothoraphanoides)	I. Gruber & M. Moser	19/09/1965	Conifers	Italy	Bolzano	MW010184
	MCVE1691	E. Campo	30/09/1993	Fagus	Italy	Pordenone	JF907870
	IB19680089 (as C. nothoraphanoides)	D. Lamoure & M. Moser	10/09/1968	Conifers	Italy	Trentino	MW010185
	2M06	N. Matsushita	04/10/2016	NA	Japan	Iwate	LC373240
	YM2442 (ecto.)	NA	09/2011	Tsuga	Japan	Yamanashi	AB848452
	OS580	NA	NA	NA	Norway	Oslo	KC842422
	T42	K. Høiland	05/09/1994	Mixed forest	Norway	Telemark	UDB000134
	MES4179	J. Ballarà & R. Mahiques	30/09/2005	Pinus	Spain	Barcelona	MW010064
	JDRG20120201	J.D. Reyes	20/12/2002	Cistus ladanifer, Pinus	Spain	Jaén	MW010062
	JDRG28101801	J.D. Reyes	28/10/2018	Pinus, Quercus	Spain	Jaén	MW010206
	TU105219	A. Kollom	26/09/2011	Pinus	Sweden	Fleringe	UDB018218
	AT2004115	A.F.S. Taylor	18/07/2004	Mixed forest	Sweden	Uppsala	UDB000717
Cortinarius viridans	MPU1116859 (holotype)	J.-M. Bellanger & M. Loizides	29/11/2019	Pinus, Cedrus, Q. alnifolia, Arbutus	Cyprus	Trooditissa	MW010202
	ML81119CV	M. Loizides	09/11/2018	Pinus, Cedrus, Q. alnifolia, Arbutus	Cyprus	Trooditissa	MW010199
	ML911192CV2	P.-A. Moreau	29/11/2019	Pinus, Cedrus, Q. alnifolia, Arbutus	Cyprus	Trooditissa	MW010203
	ML911192CV3	P.-A. Moreau	29/11/2019	Pinus, Cedrus, Q. alnifolia, Arbutus	Cyprus	Trooditissa	MW010057
	ML810192CV	M. Loizides	29/10/2018	Cedrus, Pinus	Cyprus	Troodos	MW010198
Cortinarius betuletorum	TU105268	A. Kollom	05/09/2012	Picea, Pinus, Betula, Corylus	Estonia	Võru	UDB018300
Cortinarius colymbadinus	CFP1130 (neotype)	T.E. Brandrud, H. Lindstrom, H. Marklund & S. Muskos	16/08/1992	Picea	Sweden	Jämtland	JX127302
Cortinarius olivaceofuscus	TAAM128766	A. Kollom	24/09/2008	Tilia	Estonia	Hiiu	UDB015963
Cortinarius raphanoides	PML 2507	A. Bidaud	20/09/1992	Betula	France	Isère	MW010127
Cortinarius rubicundulus	O-F-301662	G. Bollingmo	31/08/2012	NA	Norway	Ringerike	UDB036906
Cortinarius rubicundulus aff.	F15898 (as C. rubicundulus)	NA	NA	NA	Canada	British Columbia	FJ157080
	TO4 (as C. sp.)	NA	NA	NA	Canada	British Columbia	FJ717594
	IB20010177-A5-Sab (as C. psittacinus)	U. Peintner & M. Moser	22/09/2001	Fagus	Italy	Abruzzo	UDB001106
Cortinarius valgus	LM761 (ecto.)	NA	NA	Q. robur	Germany	NA	KM576359
Cortinarius veronicae	PDD68468 (holotype)	K. Soop	21/04/1997	Nothofagus	New-Zealand	Canterbury	KC017355
Cortinarius veronicoides	MEL2120747	R.H. Jones	05/06/1999	NA	Australia	Victoria	GQ890324

* In bold: sequences newly generated for the present work. NA: not available. The dashed line delimits outgroup sequences.

Several Leprocybe have been reported from Mediterranean ecosystems, either as range-restricted specialists (e.g., C. leproleptopus, C. subcotoneus, C. xantholamellatus), or as biogeographically widespread species also present in other ecoregions (e.g., C. cepistipes, C. cotoneus, C. melanotus) (Favre & Moreau 2001, Bidaud et al. 2005). Extensive taxon sampling and phylogenetic assessment of concerned collections are nonetheless lacking, therefore the full range of intraspecific ecological plasticity, as well as diagnostic criteria between taxa, remain ambiguous. Being the first Leprocybe to be described from thermophilous Mediterranean Quercus woodlands, Cortinarius leproleptopus is one such case. This species has long been known as ‘the Mediterranean Leprocybe’ and was believed to be confined to Mediterranean and sub-Mediterranean ecoregions, occurring in exclusive association with evergreen oaks (Chevassut & Henry 1988, Consiglio et al. 2004, Bidaud et al. 2005). As a result, when slightly deviant collections outside the Mediterranean region were found, the new taxon C. cepistipes was controversially introduced to accommodate them (Favre & Moreau 2001), sparking a heated debate (Bidaud et al. 2005, Favre & Moreau 2006).

Resolving this taxonomic controversy through an in-depth phylogenetic analysis of a wide sampling of all concerned species, has been the starting point of the present study. Beyond this primary objective, we further aimed at phylogenetically revising the group by

obtaining informative sequences from as many European holotypes assigned to subg. Leprocybe as possible;

defining evolutionary limits and phylogenetic relationships among identified lineages;

clarifying the status of a number of Mediterranean collections not comfortably fitting the concepts of currently described taxa in the subgenus;

identifying reliable diagnostic traits to enable the recognition of phylogenetically confirmed species by morphological criteria; and

updating the ecology and biogeography for all concerned taxa.

Because type collections for a number of critical taxa in the subgenus are few decades old and not well preserved for DNA analyses, with some of them contaminated by parasitic moulds, we further sought to evaluate whether Next Generation Sequencing (NGS), including metabarcoding/Illumina methodology, could provide an alternative to conventional Sanger sequencing in resolving old holotypes. This promising approach has been successfully applied to yield DNA barcodes from century-old type specimens of insects (Prosser et al. 2015), and its application was more recently validated in Ascomycota (Forin et al. 2018). However, the latter study did not include type material so, prior to the present work, no study has been published using NGS to barcode nomenclaturally-linked material in higher fungi.

MATERIALS AND METHODS

Taxon sampling

For the purpose of this study, we assembled and sequenced a dataset of 161 vouchered specimens from Austria, Cyprus, France, Germany, Italy, Spain and Tunisia, including 16 Typus samples and 37 collections previously published in the Atlas des Cortinaires (Bidaud et al. 2005). Together with 56 sequences selected from GenBank and UNITE, our dataset encompasses 217 ITS sequences overall, including 11 sequences selected as outgroup. Information on these collections and the GenBank accessions of their sequences are provided in Table 2. Herbaria abbreviations follow Index Herbariorum [constantly updated] (http://sweetgum.nybg.org/science/ih/).

Morphological studies

Freshly collected basidiomata were photographed in situ and the habitat, altitude, soil characteristics and nearby trees and shrubs were documented. Detailed observations on the size, colour and odour were made on fresh material and at least one basidioma from each collection was longitudinally sectioned to observe context colour. Basidioma slenderness was assessed, using the ratio S = , adapted from Kuyper (1986). When S < 4, we deduced that the habitus of the mushroom was often qualified as stout. Additional macromorphological characters were observed on both fresh and dehydrated material. Potassium hydroxide (KOH), silver nitrate (AgNO3), gaïacol (Gaiac), phenolaniline, Tl4 and metol, were applied to test possible colour changes on the pileus, context and stipe. Colours are coded according to Séguy (1936, ‘Ség.’) and Cailleux (1981, ‘Caill.’). Micro-anatomical features were studied in water (fresh material) or in KOH 5 % (exsiccatae), at ×400 and ×1000 magnifications, under a trinocular Leitz Ortholux II, an Infinity plan achromatic trinocular, and an Olympus CH-2 research, microscopes. Spore measurements were obtained from naturally discharged spores on a glass slide, or spores deposited on the cortina or the apex of the stipe (fresh material). When fresh material was not available, spores were studied and measured from dried lamellae (exsiccata), always mounted in water. A minimum of 30 mature spores were measured from each basidioma and the median spore quotient (Qm) was calculated. For each new species, a sporogram was produced, following methods and art style applied in the Atlas des Cortinaires (Bidaud et al. 2005). To better observe and image detailed epispore ornamentation, spores from the novel taxa proposed in this work were coated with gold in a sputter coater (Quorum Q150R S Plus) and subsequently examined under a Hitachi S-3000-N scanning electron microscope (SEM), with photographs taken at 15 kV (Y. Ruiz, Real Jardin Botánico, Madrid, Spain).

DNA extraction, amplification and Sanger sequencing

DNA extraction and PCR amplification were conducted with the REDExtract-N-Amptm Plant PCR Kit (Sigma-Aldrich, St. Louis, MO, USA), following the manufacturer’s instructions. The internal transcribed spacers and 5.8S rDNA (ITS) was amplified from each collection, with primers ITS-1F, ITS-4b, ITS-4, ITS-2 and ITS-3 (White et al. 1990, Gardes & Bruns 1993), as described in Liimatainen et al. (2017). Amplicons were purified and sequenced by Eurofins Genomics, Ebersberg, Germany. Raw sequence data were edited and assembled with Codon Code Aligner v. 4.1.1 (CodonCode Corp., Centerville, MA, USA) and deposited in GenBank under the accession numbers indicated in Table 2.

Phylogenetic analyses

Phylogenetic analyses were all performed online at www.phylogeny.fr (Dereeper et al. 2008) and on the CIPRES Science Gateway (www.phylo.org/index.php/). Multiple sequence alignment was carried out with MUSCLE v. 3.7 (Edgar 2004). Maximum likelihood (ML) phylogenetic analyses were performed with PhyML v. 3.0 (Guindon et al. 2010), using the GTR + I + Γ model of evolution. Branch support was assessed using the non-parametric, Shimodaira-Hasegawa, version of the approximate likelihood-ratio test (SH-aLRT), implemented in the latest release of PhyML and which ensures high accuracy when SH-aLRT > 0.8 (Anisimova et al. 2011, Bellanger et al. 2015). Bayesian inference (BI) of phylogeny was performed using MrBayes v. 3.1.2 (Ronquist & Huelsenbeck 2003). Two runs of four Monte Carlo Markov Chains each were performed for 1 000 000 generations, with stationarity convergence estimated by the Potential Scale Reduction Factor = 1 (Gelman & Rubin 1992). Tree and parameters were sampled every 1 000 generations (1 000 trees). The initial burn-in was set to 25 % (250 trees). Branch support was assessed by Bayesian posterior probabilities (BPP). Trees were built using FigTree v. 1.4.2 (http://tree.bio.ed.ac.uk/software/figtree/) and edited with Inkscape v. 0.91 (https://inkscape.org/fr/).

Illumina sequencing and dataset analysis

Amplicon libraries from R. Henry 3412 (holotype of Cortinarius pseudovenetus), R. Henry 84.109 (holotype of Cortinarius leproleptopus) and PML 5410 (holotype of Cortinarius xantholamellatus) were prepared in two steps. A first PCR reaction was conducted as described in Liimatainen et al. (2017) but with the Phusiontm polymerase, on one microliter of extract prepared with solutions E and D of the REDExtract-N-Amptm Plant PCR Kit (Sigma-Aldrich, St. Louis, MO, USA), with primers ITS3 and ITS4 (White et al. 1990) modified by the addition in 5’ of the following linkers: TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG and GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG, respectively. A second PCR reaction (10 cycles), priming on these linkers, introduced Illumina adapters and sample-specific short indices. Libraries were then sequenced on a MiSeq analyser, at the GenSeq platform of the Labex CeMEB (Montpellier). Because the expected size of the target amplicon is ~400 nucleotides, a 2 × 250 nt paired-end strategy has been adopted to ensure a sufficient overlap (~100 nt) of most forward and reverse reads. More than 12 000 reads were generated for each sample, from both amplicon ends. After an automated quality-filtering step performed by the Illumina software, forward and reverse reads were imported in CodonCode Aligner v. 4.1.1 for further processing. Firstly, reads shorter than 150 nt were discarded. Secondly, full-length sequences were assembled by merging forward and reverse reads, using the following settings: end-to-end alignment, minimal length of overlap = 10 nt, percentage of identity > 70 %, maximum unaligned sequence ends = 100 %. Full-length sequences were then clustered as molecular Operational Taxonomic Units (mOTUs), using a 98 % identity threshold (Garnica et al. 2016). The Cortinarius fraction and taxonomic identity of the most represented species in each dataset were then estimated by Blast analysis of mOTUs. Position and frequency of single nucleotide polymorphisms (SNPs) were recorded for each sample (Fig. 1). The threshold used for mapping the most frequent SNPs was set to 0.5 % within each mOTU, corresponding to 13 occurrences in PML 5410 and 17 occurrences in R. Henry 3412 and R. Henry 84.109. This cut-off is arbitrary, but conservative enough to ensure that mapped SNPs do not result from PCR mutations or sequencing errors. According to reported fidelity of the Phusiontm polymerase (~5 × 10-7 error/bp/duplication, McInerney et al. 2014) and the mean error rate of the Illumina Miseq sequencing after quality filtering (0.24 %, Pfeiffer et al. 2018), the number of fake polymorphisms should not exceed one per read. SNPs represented only once in each dataset (singletons) were not considered in statistics about ITS rDNA polymorphism in the three genomes.

Fig. 1

Mapping of the most frequent intragenomic SNPs along the ITS barcode of three Cortinarius species. Consensus sequence of the 5.8S-ITS2 rDNA fragment of the holotypes of: a. C. pseudovenetus; b. C. leproleptopus; c. C. xantholamellatus. Generated by Illumina Miseq sequencing, with SNPs found in at least 0.5 % of the total number of reads within each mOTU indicated as lower case letters above the sequence. Grey and red boxes mark, respectively, the 5.8S rDNA and positions distinguishing each species from their closest neighbour. The number of polymorphic sites is in: a. 48; b. 56; c. 50. Note that the 5.8S rDNA is almost as polymorphic as the ITS2 rDNA and that species diagnostic positions are not (with one exception) touched by these frequent SNPs.

RESULTS

Phylogenetic boundaries and species diversity in subg. Leprocybe

At the European scale, the phylogenetic analysis of our ITS dataset resolves 11 supported species in subg. Leprocybe, each represented by 1 to 40 sequences (Fig. 2). The minimal phylogenetic distance between them ranges from 5 (C. cotoneus/C. subcotoneus pair) to 23 (C. venetus/C. cotoneus pair) differences, representing c. 0.6 % to 3.5 % of sequence divergence, respectively.

Fig. 2

ITS phylogeny of Cortinarius subg. Leprocybe. Combined Bayesian and Maximum likelihood analyses of 217 ITS sequences falling in subg. Leprocybe, including 11 sequences belonging to sect. Veronicae, sect. Rubicunduli and other distantly related lineages in the genus. Branches with strong statistical support (SH-aLRT > 0.8 and BPP ≥ 95%) are highlighted as thick lines, others display support values as SH-aLRT/% BPP. Typus collections are highlighted in bold, those originating from the Mediterranean area (as defined in e.g., Quézel & Médail 2003) are written in blue, those taxonomically described in Bidaud et al. (2005) are marked by ‘ADC’ and those illustrated in Fig. 3, 4, 5, 6 and 7 of this study are marked by a camera symbol. Species limits, retained name and major tree hosts for the 11 European species known to date are indicated on the right of each relevant clade, with grey highlight for the three new species here introduced (cf. Taxonomy). Indicated sections are from the companion paper Ammirati et al. 2021.

Three species, all from the Mediterranean area, are here identified as new to science and described as C. jimenezianus, C. selinolens and C. viridans spp. nov. (Taxonomy). Our sampling efforts yield the first molecular signature, as barcoding sequence, for the historical and well-known taxon C. melanotus, so far oddly absent from public ITS sequence databases, and identifies the first confirmed collection of the rare C. phrygianus outside northern Europe (see also Ammirati et al. 2021). Our analysis also detects crypticism within C. venetus, with two phylogenetically close but distinct European lineages (Fig. 2). Both lineages appear widespread (though only one of those is represented in Spain), but a thorough morphological analysis of collections nesting in these two subclades has failed to identify any macro-, micro-anatomical, ecological or further biogeographical differences between them. Because the two lineages appear to be fully cryptic based on present data (Fig. 3), we therefore cautiously refrain from proposing a new taxon and tentatively include the two clades within the phylogenetic boundaries of C. venetus, at least until some morphological or ecological differences between them can be established (Fig. 2). Eight identified clades overlap with at least one published name, or with a shared and not critical interpretation of a historical binomial (see notes under C. leproleptopus and C. phrygianus, though), and can thus be readily treated as ‘stabilized’ species. Overall, the 11 species delineated here allow naming at least eight UNITE species hypotheses (Nilsson et al. 2018; Taxonomy).

Fig. 3

Cortinarius venetus, a genetically polymorphic pan-European greenish species. a–d. In situ photography of fresh basidiomata; e. transmitted light microscopy imaging; f. scanning electron microscopy imaging of spores of collections genetically belonging in the smallest subclade depicted in Fig. 2 (from: a. AB 05-09-65; b. JMB2013092405; c, e. GC18091404; d, f. JDRG20120201). — Scale bars: a–d = 5 cm, e = 50 μm, f = 5 μm. — Photos by: a. A. Bidaud; b. J.-M. Bellanger; c, e. G. Corriol; d. J.D. Reyes.

Phylogenetic identity of species resolved by Next Generation Sequencing

Our Sanger sequencing attempts failed to produce any sequences for the holotypes of C. pseudovenetus (R. Henry 3412), C. leproleptopus (R. Henry 84.109) and C. xantholamellatus (PML 5410)*, therefore we submitted these three collections to next generation sequencing of the ITS2 domain (Illumina method, cf. Materials and Methods). This powerful method generated more than 12 000 DNA reads out of each sample, of which 68.1 to 81.3 % were full-length fungal sequences of good quality (Table 3). Clustering the latter as molecular Operational Taxonomic Units (mOTUs), revealed that traces of Cortinarius DNA have been preserved in these three collections but, consistent with our Sanger sequencing attempts, they are heavily contaminated by Ascomycota species. Specifically, 3521, 3688 and 2496 full-length and good-quality sequences of Cortinarius have been produced out of R. Henry 3412, R. Henry 84.109 and PML 5410, respectively, representing 16.1 % to 29 % of total reads (Table 3). As expected from the taxonomic history of C. pseudovenetus and the phylogenetic position of collections identified as C. leproleptopus and C. xantholamellatus in our dataset, these three binomials fall in the clades of C. venetus, C. leproleptopus and C. subcotoneus, respectively (see Fig. 2 and Discussion).

Table 3

Illumina sequencing data.

Voucher	R. Henry 3412	R. Henry 84.109	PML 5410
Voucher Id	Holotype of C. pseudovenetus	Holotype of C. leproleptopus	Holotype of C. xantholamellatus
Total nb of reads N0	12189	12709	15542
Nb of paired reads, size- and quality-filtered N1	9908 (81.3 % of N0)	10117 (79.6 % of N0)	10583 (68.1 % of N0)
Nb of Cortinarius paired reads, size- and quality-filtered	3521 (28.9 % of N0, 35.5 % of N1)	3688 (29 % of N0, 36.5 % of N1)	2496 (16.1 % of N0, 23.6 % of N1)
Nb and size of mOTUs 98 %	27, 2-4674 sequences	28, 2-3406 sequences	9, 2-4677 sequences
Size of Cortinarius mOTU 98 % (frequency rank)	3319 sequences (rank #2, 33.5 % of N1)	3345 sequences (rank #2, 33.1 % of N1)	2405 sequences (rank #3, 22.7 % of N1)
% of Cortinarius polymorphic sites, excluding singletons	64.9 % (261/402 sites)	62.6 % (249/398 sites)	60.9 % (245/402 sites)
Taxonomy of the biggest (rank #1) mOTU 98 %	Aspergillus sp.	Aspergillus sp.	Cladosporium sp.
Best Blast hit of the Cortinarius mOTU 98 %	EF600897 (Cortinarius ‘venetus var. montanus’)	KM247711 (Cortinarius ‘sp.’)	AY669597 (Cortinarius ‘cotoneus’)
Revised taxonomy	C. venetus	C. leproleptopus	C. subcotoneus

Beyond these taxonomic findings, deep sequencing of this material allowed positioning single nucleotide polymorphisms (SNPs) in each individual collection, yielding an unprecedented molecular landscape of the ITS rDNA intragenomic diversity within uncultivated higher Fungi. Excluding singletons from the analysis, which may represent PCR or sequencing artefacts (Materials and Methods), 245 to 261 polymorphic sites out of the ~400 nucleotides of the amplified barcode sequence can be identified in each sample, representing 61 % to 65 % of total amplicon length (Table 3). Mapping the most frequent SNPs within each mOTU identifies 48 to 56 highly polymorphic sites, mostly found in the variable ITS2 domain and its junction to the 5.8S rDNA, but also within the 5.8S rDNA, supposedly much less variable due to structural constraints of the ribosome (Fig. 1). Of note, with one exception (a T in R. Henry 3412), none of the positions distinguishing the three species from their sister species (highlighted in red in Fig. 1) is highly polymorphic, indicating that most diagnostic SNPs in the ITS2 domain have been fixed in the three pairs of sister species, further supporting their reproductive isolation and evolutionary autonomy.

Uncovered synonymies

In an ongoing effort to assign a molecular signature to each of the numerous taxa published in the ADC, 37 collections treated in the French monograph are here shown to fall within subg. Leprocybe, representing 21 of the 38 taxa included in subsect. Leprocybe by Bidaud et al. (2005) (Table 1, Fig. 2). Nineteen of these names unambiguously (but redundantly) fall within the clades of C. cotoneus, C. leproleptopus, C. melanotus, C. pescolanensis, C. subcotoneus and C. venetus, establishing a number of synonyms for these six confirmed species and clarifying long-standing controversies (see Taxonomy). On the other hand, collections of C. melleifolius var. basiluteus and C. psittacinus are represented in two and three clades, respectively, indicating that previous interpretations of these two taxa are ambiguous and in need of revision.

Biogeographical and ecological patterns

Most of the species delineated in our phylogeny appear to be segregated according to putative host plants, biogeography and in some cases substrate specificities, but a couple of taxa revealed to have a broader ecology and distribution than previously thought (Fig. 2). Cortinarius cotoneus, C. subcotoneus, C. leproleptopus and C. selinolens are all firmly associated to broadleaved trees (with the interaction of C. cotoneus with Fagaceae further confirmed by EcM sequences), while C. pescolanensis, C. phrygianus, C. squamivenetus and C. viridans seem to be linked to conifers. The latter is notable for its so far exclusive association with the endemic cedar of Cyprus (Cedrus brevifolia), occurring on igneous and serpentinized substrates of the Troodos ophiolite. Similarly, the also newly described C. jimenezianus from Andalusia (Spain), is unique in this subgenus by its apparently strict association with Cistaceae shrubs, as it has so far been exclusively collected under gum rockroses (Cistus ladanifer). Based on our expanded molecular sampling, C. cotoneus, C. phrygianus and C. squamivenetus are absent from the Mediterranean region, while C. jimenezianus, C. viridans and meridional collections of C. pescolanensis and C. venetus seem to be restricted to the meso-, supra- or oro-Mediterranean zones, where cooler temperatures and more humid conditions exist. Cortinarius selinolens, on the other hand, displays strict tropism and a marked preference for xerophilous woodlands of this ecoregion. In contrast to the above, C. leproleptopus, C. melanotus and C. subcotoneus revealed to be widely distributed in warm and dry areas of the Mediterranean but are also found in Germany, associated with a broad range of mesophilous deciduous tree-hosts.

Overall, this study represents an extensive revision of subg. Leprocybe of Cortinarius in Europe, including the phylogenetic placement of 16 previously unsequenced holotypes, epitypes and paratypes, the molecular confirmation of 23 synonymies, and the description of 3 species new to science.

TAXONOMY

M.M. Moser, Z. Pilzk. 35 (3+4): 223. 1969

= Cortinarius sect. Veneti Konrad & Maubl., Icones Selectae Fungorum VI: 170. 1937, [invalid].

= Cortinarius sect. Leprocybe M.M. Moser, Z. Pilzk. 35 (3+4): 223. 1969.

= Cortinarius subsect. Leprocybe Bidaud, Moënne-Locc. & Reumaux, in Bidaud et al., Documents Mycologiques 24 (95): 40. 1994.

Type. Cortinarius cotoneus Fr., Epicrisis Systematis Mycologici: 289. 1838 [MB 228460].

Basidiomata with yellowish, greenish to olivaceous colours; pileus (20–)40–100 mm, tomentose to fibrillose or squamulose, not or only slightly hygrophanous; stipe cylindrical or clavate, sometimes bulbous; universal veil yellow, olivaceous yellow, olivaceous green or blackish brown; odour usually of radish or parsley; KOH-reaction in most species red to red brown on the flesh; spores subglobose; fluorescent yellow. According to recent phylogenetic studies at the global scale, the subgenus is represented in coniferous and broadleaved forests of the Northern Hemisphere and also includes sect. Veronicae and sect. Persplendidi, so far represented in Australia and New Zealand (Soop et al. 2019a). In Europe, 11 species can currently be delineated, occurring from southern Spain to Cyprus in the Mediterranean, to Fennoscandia in northern Europe.

Armada & J.D. Reyes, sp. nov. — MycoBank MB 837444; Fig. 4

Fig. 4

Cortinarius jimenezianus, a Cistus-associated Leprocybe from southern Spain. a–b. In situ photography of fresh basidiomata; c. transmitted light microscopy imaging of the suprapellis; d. UV-light fluorescent imaging of a dried specimen; e. scanning electron microscopy imaging of spores; f. sporogram (from: a, e–f. FA4796 (isotype); b. JDRG26120401; d. JDRG17121801/FA4792). — Scale bars: a–b = 5 cm, c = 50 μm, d = 1 cm; e = 5 μm; f = 10 μm. — Photos by: a, c. F. Armada; b. J.D. Reyes.

Etymology. Dedicated to Felipe Jiménez, honorary President of the Mycological Society ‘Lactarius’ of Jaén.

Diagnosis — Pileus 13–70 mm latus, conico campanulatus, dein applanato convexus, interdum in umbone subdepressus, valde mammosus, carnosus; margine inflexa dein recta. Indumentum siccum, primum leviter lubricosum, dein tomentosum, levis et fulgidum in vetustis basidiomatis; color variabilis, viridis, luteo-olivaceus, brunneo-olivaceum, brunneo-rufus, in margine viridis. Velum universale luteum. Lamellae 3–8 mm latae, adnatae sinuatae, luteae, luteo-olivaceae vel brunneo-olivascentes, dein brunneo-rufae. Stipes 15–90 × 4–8 mm, firmus, plenus et carnosus, ad basim attenuatus, luteo-viridis, ad apicem lutescens, deorsum cremeo-rufus velo luteo-citrino, subannulato praeditus. Cortina tenuis. Mycelium sulphurinum. Caro in pileo brunneo-olivacea, in cortice stipitis luteo-rufescens, in basi brunnea. Odor primum leviter pomaceus, dein raphanicus; sporae ovoideo-subglobosae, vel ellipsoideae, leviter ellipsoidae, laeviter verrucosae vel cristatae (6.8–)7–7.5(–8) × (5.3–)5.5–6.5(–7) μm.

Typus. SPAIN, Jaén, Parque Natural de Despeñaperros, Santa Elena, Raso de la Venta, 895 m asl, under Cistus ladanifer and Quercus ilex on acidic soil, 17 Dec. 2018, F. Armada & J.D. Reyes (holotype, JA-CUSSTA 9402), isotype FA4796 in F. Armada and J.D. Reyes personal herbaria. GenBank ITS MW010204.

Pileus 13–70 mm diam, broadly conical to campanulate at first, gradually expanding to convex or applanate and weakly to moderately umbonate, slightly viscid and finely tomentose when young, but progressively dry and silky or glabrous when ageing, bright olive-green (Ség. 296–301) or greenish yellow to olive-brown (Ség. 315–427), rapidly discolouring reddish brown to dark brown from the centre outwards (Ség. 232). Margin incurved, corrugated when ageing, with persisting greenish hues. Lamellae 8 mm wide, bright olive-yellow (Ség. 261–263), then olive-brown (Ség. 216), rather crowded, adnate to emarginate, nearly decurrent, moderately thick. Stipe 15–90 × 4–8 mm, rather slender, equal or slightly bulbous and attenuated at the base, entirely hollow when ageing, covered by a cortinoid veil later fleeting to leave a ± well-defined ring-like zone at the apex, greenish yellow or lemon yellow (Ség. 270–320), yellowish in the upper part (Ség. 264), yellowish cream at base (Ség. 185). Cortina concolorous, mycelium sulfur yellow. Context olivaceous brown in the pileus (Ség. 435), but olive-cream in the stipe, sometimes with reddish hues at the base (Ség. 296). Fluorescence in UV (360 nm). Smell pleasant when cut, fruity. Taste bitter.

Spores subglobose to ovoid or broadly ellipsoid, dextrinoid, measuring (6.8–)7–7.5(–8) × (5.3–)5.5–6.5(–7) μm (Q = 1.1–1.4; Qm = 1.2), ornamented by coarse, cristate or often truncated warts frequently coalescing to form short discontinuous ridges. Basidia cylindrical, tetrasporic, 30–38 × 7–9.5 μm, with 3 μm long sterigmata. Marginal cells 5–14 μm wide, cylindric-clavate to sublageniform, thin walled, septate, on average 30 × 8.5 μm. Suprapellis a trichodermium, made up of 3–5 μm wide, parallel hyphae. Subpellis composed of allantoid and distinctly inflated hyphae 5–18 (up to 30) μm wide. Pigment both vacuolar and parietal, generally smooth but encrusted in places, dark yellow in the suprapellis, light yellow in the subpellis. Clamp connections present.

Chemical reactions — KOH blood-red on the pileus cuticle, olive (+++) in the pileus and at base of stipe; AgNO3 (+++) greenish yellow in the pileus; Gaiac null in the pileus but ++ in the stipe; Phenolaniline null.

Ecology & Distribution — Fruiting in densely gregarious or fasciculate groups late in the season (December). So far restricted to a few localities of the Parco Natural de Despeñaperros in southern Spain, between 770 m and 915 m asl, under Cistus ladanifer on acidic soil.

Other collections studied. SPAIN, Jaén, Parque Natural de Despeñaperros, Santa Elena, Llano de las Américas, 770 m asl, under Cistus ladanifer and Quercus ilex on acidic soil, 26 Dec. 2004, J.D. Reyes, JDRG26120401; ibid., under Cistus ladanifer, 9 Dec. 2014, A. Mateos Izquierdo, AMI3777; ibid., Umbria de Magaña, 916 m asl, under Cistus ladanifer and Quercus ilex on acidic soil, 17 Dec. 2018, F. Armada & J.D. Reyes, JDRG17121801/FA4792.

Notes — In the field, this species immediately evokes C. venetus, especially its variety ‘viridis’ (Bidaud et al. 2005), because of the olivaceous-yellow veil. The two species display similarly sized and shaped spores, as well as overall anatomy. However, the flesh of C. venetus is not reported as bitter and the latter species fruits under conifers or Fagus in Europe, with no confirmed collection in the Mediterranean area prior to this work. Cortinarius olivaceofuscus, currently treated in subg. Dermocybe, may also superficially resemble C. jimenezianus, but the spores of this species are more elongated, measuring 6.5–8 × 4–5 μm (Q = 1.6–1.7) and the ecology is different (associated with Carpinus, Abies or Picea in continental mesophilous forests). Phylogenetically, C. jimenezianus is closest to C. melanotus and the newly described Cyprian species C. viridans, from which it differs by 21 substitutions and indels.

Bidaud & Bellanger, sp. nov. — Fig. 5; MycoBank MB 837445

Fig. 5

Cortinarius selinolens, a quercophilous Mediterranean C. melanotus lookalike. a–b. In situ photography of fresh basidiomata; c. transmitted light microscopy imaging of the suprapellis; d. UV-light fluorescent imaging of a dried specimen; e. scanning electron microscopy imaging of spores; f. sporogram (from: a, c, f. AB 08-10-396 (isotype); b, e. JDRG02121005; d. AB 11-11-377/JMB2011112503). — Scale bars: a–b = 5 cm, c = 30 μm, d = 1 cm; e = 5 μm; f = 10 μm. — Photos by: a, c. A. Bidaud; b. J.D. Reyes.

Etymology. From the Latin names Petroselinum (parsley) and olens (smelling), referring to the typical odour of the basidiomata of this species.

Diagnosis — Pileus 40–50 mm latus, in centro valde carnosus, convexus dein applanatus, late umbonatus. Margo inflexa vel infracta. Indumentum siccum, non hygrophanum, tomentosum vel squamulosum – squamae in umbone plus minusve hirtae – fusco-brunneum vel umbrino brunneum (Caill. S67–S75) marginem versus olivaceo-brunneum (Caill. P77 usque N79). Lamellae 4–9 mm latae, satis strictae dein distantes, sinuatae-emarginatae, luteo-olivaceae (Caill. N85) dein luteo-olivaceae. Acies integra vel crenulata. Stipes 40–70 × 7–13(–19) μm, plus minusve bulbosus vel subfusiformis, ad apicem albus, luteolus, velo citrino usque ad cortinam brunneo-olivadeam vestitus, ad basim aetate rufescens. Mycelium cremeo-album. Caro in pileo stipitis corticeque alba, ad apicem luteo-olivaceo marmorata, in medulla virido-lutea, ad basim rufescens. Odor petroselini. Sapor dulcis. Sporae (7–)7.5–8.5(–9) × (6–)6.5–7(–7.5) μm ovoideo-globosae, verrucis obtusis praeditae.

Typus. FRANCE, Gard, Rousson, Serre des Pastres, 250 m asl, under Quercus ilex on calcareous soil, 31 Oct. 2008, A. Bidaud & J. Chabrol (holotype, MPU1116858), isotype AB 08-10-396 in A. Bidaud personal herbarium. GenBank ITS MW010172.

Pileus 40–50 mm diam, at first hemispherical to convex, gradually expanding to plano-convex with a broad, pronounced and somewhat flattened at the top umbo, very thick-fleshed towards the middle; margin roughly incurved. Cuticle dry, non-hygrophanous, finely tomentose to squamulose, more densely squamulose and dark grey-brown or umber-brown in the centre (Caill. S67–S75), paler olive-brown to brown towards the margin (Caill. P77 to N79). Lamellae 4–9 mm wide, sinuate-emarginate, moderately crowded and straw-yellow to olive-yellow (Caill. N85) when young, becoming more distant and olivaceous brown at maturity, with smooth or crenate edges. Stipe 40–70 × 7–13(–19) mm, at first squat and strongly clavate, soon becoming more elongated and cylindrical, dirty white and with rusty-brown cortina remnants in the upper part, straw-coloured to yellowish and often with brown or ochraceous stains below; base remaining bulbous (onion-shaped) to subfusiform and extensively brown- to reddish stained with age. Veil fibrillose, lemon-yellow, soon fleeting, leaving streaks of olive-brown or rusty-brown cortinoid remnants on the stipe. Basal mycelium whitish cream. Context whitish in pileus and cortex of the stipe, yellowish green in centre, and reddish at the base. Fluorescence in UV (360 nm). Odour of parsley.

Spores ovoid to subglobose, dextrinoid, measuring (7–)7.5–8.5(–9) μm × (6–)6.5–7(–7.5) μm (Q = 1.1–1.3; Qm = 1.2), with fine obtuse warts, some coalescing to form short isolated ridges; hilar appendage 1–1.5 μm long. Basidia tetrasporic, clavate, 29–36 × 8–9 μm; sterigmata 4–5 μm long. Marginal cells clavate, 5–7 μm wide, intermixed with tufts of 8–11 μm wide clavate or subcapitate hairs. Suprapellis rather thin, made up of erected bundles of 10–20 μm wide, septate hyphae, terminating in 28–60 μm long acute elements. Subpellis poorly developed, made up of short septate cells, 10–15 μm long, radially arranged. Vacuolar pigment olive-yellow and parietal pigment brown-yellow, generally smooth but encrusted by places.

Chemical reactions — KOH olive-yellow in the pileus, vinaceous-pink at base, on pileipellis venetian-red or blackish in the centre, mahogany-brown on margin; AgNO3 sulfur-yellow in the context; Gaiac, Phenolaniline, Tl4, Metol: null.

Ecology & Distribution — Fruiting solitary or gregariously during autumn and winter; so far restricted to thermophilous Quercus ilex or Arbutus unedo woodlands of the western Mediterranean region.

Other collections studied. FRANCE, Hérault, Notre-Dame-de-Londres, 250 m asl, under Quercus ilex on calcareous soil, 25 Nov. 2011, J.-M. Bellanger & A. Bidaud, AB 11-11-377/JMB2011112503. – SPAIN, Jaén, Parque Natural de Sierra Mágina, Cambil, 1150 m asl, under Quercus ilex on calcareous soil, 2 Dec. 2010, J.D. Reyes, JDRG02121005. – TUNISIA, Jendouba, Souk el Jema, under Arbutus unedo, 1 Nov. 2010, F. Richard, FR2013185.

Notes — The relatively stout habitus and markedly bulbous stipe base of this species may evoke C. cotoneus in the field, but the latter has larger spores and, as shown in the present work, seems to be absent from the Mediterranean. Because of its similar colours, C. phrygianus can also be compared with C. selinolens, but the flesh of this rare species is odourless, its spores are considerably smaller (5.5–6.5 × 4.5–5 μm) and the ecology is different, typically found in association with Pinus on calcareous soil (Niskanen & Kytövuori 2012). Conversely, C. subcotoneus and C. leproleptopus occur in thermophilous Mediterranean woodlands like C. selinolens, but they are different morphologically and can hardly be confused with this species. Due to the parsley scent and scaly pileus, C. selinolens is morphologically closer to C. melanotus, but the stipe of the latter species is usually less bulbous and covered by a brown veil, the pileus displays more obvious greenish hues, and spores are smaller, measuring 6.5–7.5(–8) × 5–5.5(–6) μm (Niskanen & Kytövuori 2012). Phylogenetically, C. selinolens is sister to C. subcotoneus, from which it differs by 5 substitutions and 2 indels.

Bellanger & Loizides, sp. nov. — Fig. 6; MycoBank MB 837446

Fig. 6

Cortinarius viridans, a Cyprian endemic Leprocybe associated with Cedrus brevifolia. a–b. In situ photography of fresh basidiomata; c. transmitted light microscopy imaging of the suprapellis; d. UV-light fluorescent imaging of a dried specimen; e. scanning electron microscopy imaging of spores; f. sporogram (from: a, c. ML911192CV1 (isotype); d–f. ML810192CV). — Scale bars: a–b = 5 cm, c = 50 μm, d = 1 cm; e = 5 μm; f = 10 μm.— Photos by: a. P.-A. Moreau; b–c. M. Loizides.

Etymology. Viridans [Adj.] = green, referring to the olivaceous hues and greenish tinges of basidiomata.

Diagnosis — Pileus 20–60(–70) mm latus parvus vel medius, tomentoso-fibrillosus vel squamulosus, brunneo-olivaceus. Lamellae ochraeco-luteae. Stipes ad basim acuminatus, apicem versus cortina aurantio-brunnea egregie subannulatus. Sporae ovoideae vel subglobosae, verrucis truncatis, saepe curte cretatis, 6.5–7.5(–8) × 5.5–6.5(–7) μm. Ex insula Cypro sub Cedrus brevifolia, in solo vulcanico.

Typus. CYPRUS, Trooditissa, 1400 m asl, under Cedrus brevifolia on igneous, acidic soil, 29 Nov. 2019, J.-M. Bellanger & M. Loizides (holotype, MPU1116859), isotype ML911192CV1 in M. Loizides personal herbarium. GenBank ITS MW010202.

Pileus 20–60(–70) mm diam, hemispherical to campanulate at first, gradually expanding to convex but usually maintaining a broad, flattened umbo, dry, coarsely and densely tomentose to fibrillose when young, becoming more distinctly fibrillose-squamulose with age, occasionally silky and adpressly squamulose, umber-brown, greenish brown or olive-brown (Ség. 277–278 to Ség. 283–295); fibrills dark brown to olive-grey, often black towards the centre; margin persistently incurved, not significantly expanding or undulating, not hygrophanous, frequently with remnants of cortina attached. Lamellae 2–4 mm wide, olive-yellow, gradually becoming ochre-yellow and finally tobacco-brown, adnate to adnexed or somewhat emarginate at maturity, moderately thick; edges smooth, concolorous or slightly paler. Stipe 40–70 × 10–30 mm, clavate to ventricose and typically deeply rooting into the substrate, slightly bulbous or attenuated towards the base, covered in a cortinoid veil later fleeting to leave a thick annular zone at the apex; pure white above the annular zone, dirty white or straw-coloured below and usually covered in conspicuous, brownish, fibrillose or squamulose bands; cortina thick, abundant, brown to orange in colour; basal mycelium white. Context straw-coloured or concolorous with the stipe. Fluorescence in UV (360 nm). Smell at first weakly raphanoid, but then fruity when cut, somewhat reminiscent of apricots. Taste mild, fungoid.

Spores subglobose to ovoid, sometimes also broadly ellipsoid, measuring 6.5–7.5(–8) × 5.5–6.5(–7) μm (Qm = 1.2), without a germ pore, ornamented by coarse truncated warts frequently coalescing to form short, discontinuous ridges; hilar appendage poorly developed, < 1 μm long, often inconspicuous. Basidia tetrasporic, measuring 28–42 × 6–8 μm, irregularly cylindrical to clavate, thick-walled, with granular or vacuolar content and a basal clamp; sterigmata acute, 3–5 μm long. Marginal cells abundant, similar in size and shape to the basidia, with a basal clamp. Hymenophoral trama composed of non-dextrinoid, ± parallel, hyaline and thick-walled hyphae 4–18 μm wide, often somewhat constricted at the septa; septa frequently clamped. Suprapellis composed of periclinal, subhyaline to brown (KOH) cylindrical or allantoid (inflated) hyphae 6–17 μm wide, often terminating in clavate or somewhat attenuated and slightly ascending hyphal ends, sometimes coarsely encrusted; most septae with clamps. Subpellis extensively gelatinized in parts, composed of tightly packed, poorly differentiated hyphae. Clamp connections frequent throughout, but sometimes inconspicuous.

Chemical reactions — KOH brown or vinaceous-brown on the pileus and stipe; Gaiac instantly reddish purple (+++) on the pileus; Phenolaniline null.

Ecology & Distribution — Fruiting gregariously or in fasciculate clusters between late October and December, on serpentine and igneous substrates under Cedrus brevifolia. Rare, known only from a couple of localities on the island of Cyprus, at high elevations ranging from 1400 m to 1600 m asl.

Other collections studied. CYPRUS, Troodos, 1600 m asl, under C. brevifolia on serpentine soil, 29 Oct. 2018, M. Loizides, ML810192CV; Trooditissa, 1400 m asl, under C. brevifolia on igneous, acidic soil, 9 Nov. 2018, M. Loizides, ML81119CV; ibid., 29 Nov. 2019, P.-A. Moreau, ML911192CV2 and ML911192CV3.

Notes — Considering its unique ecological niche and biogeographical distribution, this rare species can hardly be confused with other species in the field. Provided the ectomycorrhizal association with Cedrus is confirmed by sequencing of the root tips in the future, C. viridans would represent the first report of an endemic Basidiomycota species associated with an endemic tree-host in the Mediterranean. From a morphological perspective, C. viridans closely resembles C. melanotus, both macromorphologically and microanatomically. However, the latter species usually displays a pileus which is more tomentose than fibrillose, with more prominent dark scales towards the center, and has an odour of parsley. Cortinarius selinolens is also similar, but produces stouter basidiomata, also with a parsley odour, and has larger spores measuring (7–)7.5–8.5(–9) × (6–)6.5–7(–7.5) μm. This species also has a very different ecology and is so far known only from thermo-Mediterranean Quercus ilex and Arbutus unedo matorral in the western Mediterranean. Cortinarius phrygianus can also superficially resemble C. viridans, but has much smaller spores measuring 5.5–6.5 × 4.5–5 μm, and a completely different ecology (see below). So far, none of the above species have been documented in Cyprus, with C. leproleptopus being the only other member of subg. Leprocybe currently represented on the island. Phylogenetically, C. viridans is not polymorphic at the ITS locus and is sister to C. melanotus, with 4 substitutions and 2 indels further distinguishing the two species.

Fr., Epicrisis Systematis Mycologici: 289. 1838 — MycoBank MB 228460. UNITE SH1503798.08FU (1.5 %)

≡ Dermocybe cotonea (Fr.) Ricken, Die Blätterpilze 1: 160. 1915.

= Cortinarius perannulatus Bidaud, in Bidaud et al., Atlas des Cortinaires 15: 1031. 2005.

= Cortinarius cotoneus var. mellinoides Bidaud, Moënne-Locc. & Reumaux, in Bidaud et al., Atlas des Cortinaires 15: 1031. 2005.

= Cortinarius subannulatus Jul. Schäff. & M.M. Moser, Beih. Sydowia 1: 226. 1957, p.p.

= Cortinarius subnotatus Fr., Epicrisis Systematis Mycologici: 290. 1838, sensu Bidaud et al. 2005, p.p.

= Cortinarius psittacinus M.M. Moser, Z. Pilzk. 35: 245. 1969, sensu Bidaud et al. 2005, p.p.

= Cortinarius conopus (Pers.) Hlaváček, Mykol. Sborn. 64 (5): 247. 1987, sensu Bidaud et al. 2005, p.p.

Typus. SWEDEN, Gotland, Vallstena sn, Alvena lindängar, broad-leaved deciduous forest, calcareous soil (Corylus, Quercus, Tilia), 1 Oct 1990, coll. Brandrud et al., CFP1032 (neotype, F44846 in S, designated in Ammirati et al. 2021).

Selected descriptions — Niskanen & Kytövuori 2012: 765; Bidaud et al. 2005: f. 763 & 763bis (as C. cotoneus var. mellinoides); Brandrud et al. 1992: B01.

Selected illustrations — Bidaud et al. 2005: Pl. 561 (PML 5429, as C. perannulatus), Pl. 570 (PML 5260, as C. cotoneus var. mellinoides), Pl. 572 (PML 5417, PML 5427, as C. conopus), Pl. 573 (PML 5430, AB 04-10-338, as C. subnotatus), Pl. 574 (AB 04-09-81, as C. psittacinus).

Notes — As here phylogenetically circumscribed, C. cotoneus is a broadly distributed species likely associated to Fagaceae trees, present in continental and northern Europe. The taxonomic study of North American Leprocybe performed in the companion paper reveals that the species is present in North America. However, the unique sequence so far published from the New World and annotated as C. cotoneus, MF686506, actually belongs to a distinct species (C. hugheseae, cf. Ammirati et al. 2021). European sequences are remarkably conserved at the ITS locus (not considering UDB001029, that displays signs of poor quality, data not shown and Fig. 2), but the clade also includes slightly divergent sequences from northern Iran and China. A deeper taxon sampling in central and eastern Asia will be necessary to clarify the extent of gene flow within this lineage, and thus identify species boundaries across the Eurasiatic transect.

An unexpected outcome of our analyses is the apparent absence of this species in the Mediterranean area, as all collections identified as C. cotoneus in Spain, Italy or southern France turned out to belong to C. subcotoneus, or to the newly described C. selinolens. Considering this updated biogeography, the usually robust habitus of the species and its large spores, C. cotoneus can hardly be misidentified by field taxonomists. The present study shows that C. perannulatus and C. cotoneus var. mellinoides, both published in the ADC in 2005, are conspecific with C. cotoneus. In the same monograph, several collections described as distinct species, namely C. conopus, C. psittacinus and C. subnotatus, are also here shown to be later synonyms of this species. An old collection identified as C. subannulatus by Moser (IB19990237) also falls in this clade (U. Peintner & R. Kuhnert, pers. comm.), indicating that the latter binomial may also represent a late synonym of C. cotoneus. In its revised boundaries, this species is closest to C. selinolens and C. subcotoneus, and differs from each one of them by 5 substitutions.

Chevassut & Rob. Henry, Doc. Mycol. 19 (73): 47. 1988 — Fig. 7; MycoBank MB 134786. UNITE SH1662309.08FU (1 %)

Fig. 7

Cortinarius leproleptopus, a pan-European yellowish Leprocybe. In situ photography of fresh basidiomata from: a–d. Germany; e. Cyprus. Note the green-olivaceous tinges on some German collections, absent from the original diagnosis describing Mediterranean collections. Reaction to KOH on the flesh and pileus is shown in c and d (from: a. GH20140907; b. FR2014332; c. GH20140911b; d. SSt12-039; e. ML21211CL). — Scale bars = 5 cm.— Photos by: a. G. Hensel; b. M. Huth; c. M. Huth & G. Hensel; d. G. Saar; e. M. Loizides.

≡ Cortinarius leptopus Reumaux, in Bidaud et al., Atlas des Cortinaires 15: 1031. 2005, [illeg.].

? = Cortinarius leproleptopus var. basiluteus Chevassut & Rob. Henry, Doc. Mycol. 24 (94): 57. 1994.

= Cortinarius melleifolius var. basiluteus (Chevassut & Rob. Henry) Bidaud & Reumaux, in Bidaud et al., Atlas des Cortinaires 15: 1031. 2005, p.p.

= Cortinarius cepistipes A. Favre & P.-A. Moreau, Rivista di Micologia 1: 17. 2001.

= Cortinarius mellinus Britzelm., Bot. Centralbl. 62 (10): 308. 1895, sensu Moser 1969, p.p.

= Cortinarius psittacinus M.M. Moser, Z. Pilzk. 35: 245. 1969, sensu Bidaud et al. 2005, p.p.

Typus. FRANCE, Hérault, Quissac, under Quercus ilex, 19 Nov. 1984, G. Chevassut (holotype, R. Henry 84.109 in PC), isotype R. Henry 84.110 in PC, paratypes Chev. 3341, 3342 in PC. GenBank ITS MW010056. – ITALY, Reggio Emilia, Castelnuevo ne’ Monti, Borra del Prato, under Castanea sativa, Quercus pubescens, Fagus sylvatica and Pinus sylvestris, 9 Oct. 1994, G. Consiglio & B. Bigazzi, E.C. 94190 (epitype, MCVE763). GenBank ITS MW010049.

Selected illustrations — This study: Fig. 7; Bidaud et al. 2005: Pl. 574 (PML 5420, 5421), Pl. 575 (PML 2949, as C. psittacinus); Consiglio et al. 2004: B8; Favre & Moreau 2001: 19, 21 (as C. cepistipes) and 22.

Emended description — Pileus 20–60(–80) mm diam, convex-campanulate with an incurved margin when young, entirely covered with squamous fibrils, golden yellow to greenish yellow on fresh basidiomata, becoming honey-rusty with age, red to reddish brown with KOH. Lamellae adnate or emarginated, rather spaced, concolorous. Stipe 50–80 × 5–12 mm, cylindrical, clavate or clearly bulbous (up to 30 mm diam), fibrillous, yellowish, naked or covered variable amounts of a yellow partial veil. Context yellowish, odour of raw potato, of radish or dusty-earthy, with weak or null reaction to KOH.

Spores ovoid to subglobose 6–8.5(–9.5) × 5.5–7 μm (Q = 1.1–1.4), ornamented by almost isolated warts. Suprapellis hymenodermic or trichodermic, made up of parallel bundles of hyphae 3–12 mm wide, more or less erected. Subpellis usually well developed, pseudoparenchymatous, made up of cylindrical elements up to 100 × 20 μm.

Ecology & Distribution — Solitary or gregarious, fruiting rather late in the autumn under broadleaved trees of the Fagaceae family, occurring mostly, but not exclusively, on calcareous soils, from Mediterranean Quercus ilex, Q. pubescens, Q. alnifolia or Castanea sativa woodlands to ± thermophilous Q. robur and Fagus sylvatica forests in continental regions. So far confirmed in Cyprus, France, Germany, Italy and Spain.

Notes (see also ‘Discussion’ for further comments) — This species has been originally described as a yellow, strictly meridional Leprocybe without greenish tones, but with a cylindrical stipe and an ecology restricted to thermophilous and calcareous Quercus woodlands of the Mediterranean area (Chevassut & Henry 1988). As revised here, C. leproleptopus is more polymorphic and displays a much broader distributional and ecological range. In particular, the stipe can be cylindrical to neatly bulbous and may be glabrous or covered by abundant yellow veil remnants. The dominant colour of basidiomata is yellow, but greenish hues may be present on the pileus and lamellae of some fresh collections (Fig. 7). Microscopically, the hyphae of the hypodermis are usually well differentiated and spores rarely exceed 8.5 × 7 μm. This species is indeed present in the western Mediterranean region (Spain, France, Italy), but also occurs considerably further east, in Cyprus and north, in the French Alps as well as Germany. The conspecificity of C. leproleptopus var. basiluteus with the present species is likely, but cannot be ascertained because no original material of this taxon could be traced and sequenced. From a phylogenetic point of view, C. leproleptopus is remarkably stable at the ITS locus (0 differences between the 40 available sequences) and differs from its sister species C. pescolanensis by 6 substitutions.

Kalchbr., in Fries, Hymenomycetes Europaei (Upsaliae): 365. 1874 — MycoBank MB 219634

≡ Gomphos melanotus (Kalchbr.) Kuntze, Revis. Gen. Pl. (Leipzig) 2: 854. 1891.

= Cortinarius submelanotus Bidaud, in Bidaud et al., Atlas des Cortinaires 15: 1032. 2005.

= Cortinarius zonatus Reumaux, in Bidaud et al., Atlas des Cortinaires 15: 1033. 2005.

= Cortinarius colymbadinus Fr., Epicrisis Systematis Mycologici: 289. 1838, sensu Bidaud et al. 2005.

Typus. Kalchbrenner, Icones selectae Hymenomycetum Hungariae 3, 1875, t. 27, f. 2 (holotype, iconotype). See also Ammirati et al. (2021).

Selected descriptions — Niskanen & Kytövuori 2012: 766; Consiglio et al. 2006: D128; Bidaud et al. 2005: f. 756, f. 756bis (as C. submelanotus), f. 759 (as C. zonatus); Brandrud et al. 1998: D56.

Selected illustrations — Bidaud et al. 2005: Pl. 558 (PML 5428, as C. colymbadinus), Pl. 559 (PML 5461), Pl. 560 (PML 5454, as C. submelanotus, PML 2344), Pl. 563 (PML 5455, as C. perannulatus, XC 2005-20, as C. zonatus).

Notes — Prior to the present work, no ITS sequence of this historical species was publicly available. Our dataset shows that, as currently interpreted by taxonomists, C. melanotus is confirmed in Spain and is not rare in France and Germany. Since this species was described from Central Europe and is reported in Fennoscandia (Fries 1874, Brandrud et al. 1998, Niskanen & Kytövuori 2012), it is probably present throughout most of the continent (Table 2, Fig. 2). Cortinarius melanotus is often claimed to be associated with Abies or Pinus spp. in calcareous soils (Brandrud et al. 1998, Niskanen & Kytövuori 2012, Eyssartier & Roux 2017). However, the clade includes several collections from broadleaved woodlands and at least some fruiting on acidic soils (Bidaud et al. 2005; Table 2). The unveiled conspecificity of C. submelanotus with C. melanotus in the present work, was already suspected by Bidaud et al. (2005) who considered their new species as a putative variant of C. melanotus. The colour of veil remnants on the stipe (brown vs yellow), should thus not be overemphasised in taxonomic terms, since collections with both velar colours nest in the same clade. However, the placement of C. zonatus in the latter was less expected and indicates that the species can occasionally produce odourless basidiomata with crown-like colours on the pileus. The interpretation of C. colymbadinus by the French authors of the ADC also falls in this clade, but this binomial has recently been neotypified and is now assigned to a distinct lineage of subg. Telamonia (Dima et al. 2014; Fig. 2). Phylogenetically, C. melanotus displays limited polymorphism at the ITS locus, with 4 variable sites, and is separated from its closest species C. viridans by 4 substitutions and 2 indels.

Picillo & Marchionni, Micologia e Vegetazione Mediterranea 31 (1): 63. 2016 — MycoBank MB 816349. UNITE SH1662079.08FU (1 %)

= Cortinarius mellinus Britzelm., Bot. Centralbl. 62 (10): 308. 1895, sensu Bidaud et al. (2005).

= Cortinarius psittacinus M.M. Moser, Z. Pilzk. 35: 245. 1969, sensu Bidaud et al. (2005), p.p.

= Cortinarius junquilleovestitus Bidaud, in Bidaud et al., Atlas des Cortinaires 15: f. 765. 2005, [invalid].

= Cortinarius argyrocognitus Bidaud, in Bidaud et al., Atlas des Cortinaires 15: f. 767. 2005, [invalid].

Typus. ITALY, Molise, Pescopennataro, Fonte della Gallina, under Abies alba with some Fagus sylvatica trees on basic soil, 14 Oct. 2012, B. Picillo & M. Marchionni (holotype, MCVE29054), isotype BP13/291 in B. Picillo personal herbarium. GenBank ITS KX010944/NR_153070 and KX010945.

Selected descriptions — Ballarà et al. 2017: 39; Picillo & Marchionni 2016: 63; Bidaud et al. 2005: f. 765 (as C. junquilleovestitus), f. 767 (as C. argyrocognitus), f. 768 (as C. mellinus).

Selected illustrations — Ballarà et al. 2017: 40; Picillo & Marchionni 2016: 64; Bidaud et al. 2005: Pl. 572 (as C. junquilleovestitus), Pl. 575 (PML 3262, as C. psittacinus), Pl. 576 (as C. argyrocognitus), Pl. 577 (PML 5449, as C. mellinus).

Notes — This taxon was recently published from Italy but had been repeatedly described under different names in the ADC, namely C. mellinus, C. junquilleovestitus ad int and C. argyrocognitus ad int. All phylogenetically confirmed collections of C. pescolanensis so far originate from mountainous forests of France, Italy and Spain, dominated by Abies spp., more or less mixed with scattered Fagus or Pinus trees, and occurring exclusively on calcareous bedrock (Table 2 and data not shown). Basidiomata are intermediate in size between C. venetus and C. cotoneus, while the pileus and the veil are predominantly ochraceous-orange in colour, lacking green hues, and the pileus center is distinctly darker by persistent dark scales. Phylogenetically, the species is remarkably stable at the ITS locus (1 polymorphic position) and differs from its closest neighbour C. leproleptopus by 6 substitutions.

(Fr.) Trog, Verzeichnis schweizerischer Schwämme (1844) — MycoBank MB 152271. UNITE SH1662078.08FU (1 %)

Cortinarius sublanatus var. phrygianus Fr., Epicrisis Systematis Mycologici: 283. 1838.

≡ Cortinarius phrygianus Fr., Hymenomycetes Europaei (Upsaliae): 365. 1874.

≡ Gomphos phrygianus (Fr.) Kuntze, Revis. Gen. Pl. (Leipzig) 2: 854. 1891.

= Cortinarius arenatus (Pers.) Fr., Epicrisis Systematis Mycologici: 283. 1838, sensu Melot in Brandrud et al. (2014).

Typus. Sweden. Darlana, Rättvik sn, Enån, coniferous forest, Pinus, 14 Sept. 1988, coll. Brandrud et al., CFP774 (neotype F44888 in S, designated in Ammirati et al. 2021).

Selected descriptions — Niskanen & Kytövuori 2012: 766; Bidaud et al. 2005: f. 738; Brandrud et al. 1990: A30.

Selected illustrations — Soop et al. 2019a: f. 3d; Bidaud et al. 2005: Pl. 541; Brandrud et al. 1990: A30.

Notes — This iconic but rare Friesian species has been reported by Nordic authors to be largely confined to Fennoscandia (Brandrud et al. 1990). Consistently, it has so far been confirmed only from dry Pinus or Picea woodlands of Sweden and Estonia. A few older reports from Austria, Germany and Switzerland do exist, but in the absence of molecular analysis, they may refer to different species, especially Cortinarius melanotus (Brandrud & Bendiksen 2014). In France, C. phrygianus has been first interpreted by Henry (1977) as a rare species fruiting in broadleaved forests of the eastern part of the country. The authors of the ADC named this latter taxon C. hirtipes and instead applied C. phrygianus to a single collection from the Ain department (Bidaud et al. 2005), here found to be conspecific with the Swedish and Estonian sequenced material (Table 2, Fig. 2). Such finding represents the first confirmation of C. phrygianus outside Nordic European countries, and incidentely yields some credit to previous, non-sequenced continental reports. The collection of this species described in 1990 in the CFP has been lately renamed as C. arenatus (Brandrud et al. 2014: plate A30), a binomial variously interpreted by previous authors (Bidaud et al. 2005). We tentatively ascribe the better-known name C. phrygianus to the present taxon, until the synonymy of the two names is fully established. In the field, the species can only be confused with C. melanotus, which shares similar macro-morphology, although the latter usually displays more obvious greenish hues, typically smells of parsley and has larger spores measuring 6.5–7.5(–8) × 5–5.5(–6) μm vs 5.5–6.5 × 4.5–5 μm (Niskanen & Kytövuori 2012). Phylogenetically, C. phrygianus is remarkably stable at the ITS locus and well separated from all other known species in the subgenus (e.g., 19 substitutions and 9 indels relative to C. leproleptopus). More collections of this species are, however, needed to better understand its natural variability, at molecular, morpho-anatomical as well as ecological levels.

Kytöv., Liimat. & Niskanen, Index Fungorum 196 (1): 2. 2014 — MycoBank MB 550843. UNITE SH2122528.08FU (< 0.5 %).

Typus. FINLAND, Koillismaa, Kuusamo, Oulanka National Park, Camping pale, Puukkorinne rich mixed forest (Picea, Pinus, Populus), 1 Sept. 2007, T. Niskanen, I. Kytövuori & K. Liimatainen, TN 07-119 (holotype in H), isotype in NY. GenBank ITS KP114458.

Selected description — Liimatainen 2014: 2.

Selected illustration — Cf. Ammirati et al. (2021).

Notes — This species was so far known only from the northern boreal zone of Finland, where it fruits early in the autumn in coniferous forests on calcareous ground. Ammirati et al. (2021) describe two additional collections from Quebec (Canada). No photographic representation of C. squamivenetus has been published to date, but judging from descriptions, this taxon would be mostly similar to C. venetus, from which it would differ by black appressed fibrillose scales and slightly smaller spores (6.4 × 5.5 μm vs 7 × 5.8 μm, on average). A broader taxon sampling of this species is obviously necessary, to better understand the full range of its morpho-anatomical variability and precise its ecological/biogeographical range. From an evolutionary perspective, it is difficult to position C. squamivenetus relative to other species solely by ITS phylogeny, because in the phylogram depicted in Fig. 2, it lies in an unsupported basal position, with more than 20 differences with other clades (e.g., 15 substitutions and 7 indels relative to C. venetus).

Bidaud, Atlas des Cortinaires 15: 1032. 2005 — MycoBank MB 357354. UNITE SH1856985.08FU (0.5 %)

= Cortinarius borealis Bidaud, in Bidaud et al., Atlas des Cortinaires 15: 1031. 2005.

= Cortinarius xantholamellatus Bidaud, in Bidaud et al., Atlas des Cortinaires 15: 1033. 2005.

? = Cortinarius sublanatus var. xanthophyllus Maire, Fungi Catalaunici: Contributions à l’étude de la Flore Mycologique de la Catalogne: 104. 1933.

= Cortinarius melleifolius var. basiluteus (Chevassut & Rob. Henry) Bidaud & Reumaux, in Bidaud et al., Atlas des Cortinaires 15: 1031. 2005, p.p.

Typus. FRANCE, Drôme, La Roche-Saint-Secret, 350 m asl, under Quercus pubescens in calcareous soil, 9 Nov. 1990, A. Bidaud, PML 2143 (holotype in PC). GenBank ITS MW010122.

Selected descriptions — Bidaud et al. 2005: f. 762 (as C. borealis), f. 764 (including C. xantholamellatus).

Selected illustrations — Bidaud et al. 2005: Pl. 567 (as C. borealis), Pl. 570, Pl. 571 (as C. xantholamellatus), Pl. 578 (JLC001029-8, as C. melleifolius var. basiluteus).

Notes — This taxon has been initially defined as a C. cotoneus-like species with different veil remnants on the stipe and slightly smaller spores. Our analyses provide an additional important difference between the two species, biogeographical in nature: while C. cotoneus seems to be absent from the Mediterranean area, C. subcotoneus is almost exclusively fruiting in this ecoregion, with the notable exception of a few German localities. The species is closely associated with Quercus spp. in meridional localities and with Fagaceae in Germany. The synonymy of C. subcotoneus with C. xantholamellatus, C. borealis and one published collection of C. melleifolius var. basiluteus, all unravelled here, indicates that yellow pigments can occasionally be present in basidiomata, at least veil and lamellae, and that, in rare cases, some blue hues can be detected at the top of the stipe. Bidaud et al. (2005) proposed that C. xantholamellatus may be conspecific with C. sublanatus var. xanthophyllus, a variant described by R. Maire from Spain under Quercus suber and Castanea sativa, on acidic soil. However, no collection of C. subcotoneus has so far been confirmed outside calciphilous Quercus woodlands in the Mediterranean area, casting some doubt on such a synonymy. Phylogenetically, the species is moderately polymorphic, with 3 identified substitutions and 2 indel positions, and can be distinguished from its sister species C. selinolens by 5 substitutions and 2 indels.

(Fr.) Fr., Epicrisis Systematis Mycologici: 291. 1838 — Fig. 3; MycoBank MB 143469. UNITE SH1855806.08FU (0.5 %) + SH1856058.08FU (0.5 %)

≡ Agaricus raphanoides (var.) b venetus Fr., Observ. Mycol. 2: 76. 1818.

≡ Dermocybe veneta (Fr.) Ricken, Die Blätterpilze 1: 162. 1915.

= Cortinarius pseudovenetus Rob. Henry, Bull. Soc. Mycol. France 77 (2): 116. 1961.

= Cortinarius venetus var. montanus M.M. Moser, Z. Pilzk.: 25. 1970.

= Cortinarius venetus var. viridis Moënne-Locc., in Bidaud et al., Atlas des Cortinaires 15: 1033. 2005.

= Cortinarius nothoraphanoides (ined.) M.M. Moser, Guida alla determinazione dei funghi (Polyporales, Boletales, Agaricales, Russulales): 374. 1986, [invalid].

Typus. SWEDEN, Uppland, Uppsala-näs sn, Dalkarlskärret, coniferous forest (Pinus, Picea), 18 Aug. 1986, Lindström et al., CFP112 (neotype, F44498 in S, designated in Ammirati et al. 2021).

Selected descriptions — Niskanen & Kytövuori 2012: 766; Bidaud et al. 2005: f. 760 (as C. pseudovenetus), f. 760bis (as C. cf. nothoraphanoides), f. 761 (as C. venetus and C. venetus var. viridis); Consiglio et al. 2004: B193; Brandrud et al. 1990: A15.

Selected illustrations — This study: Fig. 3; Bidaud et al. 2005: Pl. 564 (PML 2393, as C. pseudovenetus), Pl. 565 (PML 1781, as C. pseudovenetus, PML 1665), Pl. 566 (PML 245, as C. venetus var. viridis), Pl. 567 (PML 3265, PML 5466, AB 04-09-251, as C. cf. nothoraphanoides).

Notes — As revised here, C. venetus is a European species occurring as two haplotypes currently lacking biogeographical, ecological or morpho-anatomical distinguishing features (Fig. 2, 3). The species fruits under coniferous trees or Fagus and produces relatively small basidiomata that usually display obvious green hues on the pileus (from green yellowish to olive-brown), with a hygrophanous tendency unusual in the subgenus and frequent ochraceous oxidation when ageing. This species delivers spores of 7 × 5.8 μm on average but one collection studied by us yielded smaller spores of 5.4–6.5 × 4.7–5.2 μm, in addition to displaying atypical red staining of the stipe base (GC18091404, Fig. 3c, e). Risk of confusion exists with the newly described C. jimenezianus in southern Spain, where the two species co-occur. The late fruiting of the latter, in addition to its bitter taste and association with rockroses should provide sufficient clues in distinguishing the two lookalikes. Our work further establishes the synonymy of C. pseudovenetus with C. venetus and supports the varietal rank of C. venetus var. viridis, indicating that the extent of green hues on young basidiomata and that of oxidation in older ones, should not be granted too much taxonomic value. Consistent with the reported synonymy of C. pseudovenetus with C. venetus var. montanus (Bidaud et al. 2005), a short sequence of the holotype of this variety could be obtained and falls within the C. venetus clade as well (U. Peintner & R. Kuhnert, pers. comm.). Lastly, authentic material of C. nothoraphanoides (ined.) and 3 collections published in the ADC as C. cf. nothoraphanoides, belong to the same clade, strongly suggesting that this invalidly published binomial actually represents a synonym of C. venetus. From an evolutionary point of view, C. venetus is closest to several East Asiatic sequences and to the western North American C. fucoflavidus (cf. Ammirati et al. 2021).

DISCUSSION

Revisiting species boundaries in subgenus Leprocybe

Until the simultaneous publication of Cortinarius Flora Photographica CFP (Brandrud et al. 1990) and the Atlas des Cortinaires ADC in the late 20th century (Moënne-Loccoz & Reumaux 1990), taxonomic boundaries within subg. Leprocybe (= sect. Veneti) were for the most part consensual. These two major contemporary monographs were built upon rather opposing views of species concepts, and authors adopted contrasting taxonomic choices to delimit species within this lineage. As a result, only four species were accepted in the CFP (Brandrud et al. 1990, 1992, 1998), as opposed to 38 taxa in the ADC (Bidaud et al. 2005). The four names included in the CFP, C. cotoneus, C. melanotus, C. phrygianus and C. venetus, are typified in the companion paper of this study (Ammirati et al. 2021), with sequenced collections from the CFP monograph corresponding with the four clades as delineated here. In marked contrast, the phylogenetic identity of the ADC taxa remained entirely unresolved prior to the present work, with the exception of C. raphanodiabolicus sensu ADC, which has been recently synonymised with C. sciophylloides in subg. Telamonia (Liimatainen et al. 2017).

Our sequencing efforts on this taxonomically valuable material, allowed us to phylogenetically position 21 of these 38 taxa within subg. Leprocybe (Table 1, Fig. 2). In addition, we could demonstrate that C. raphanoides sensu ADC, as well as C. squamulosus sensu ADC and C. ignipes sensu ADC, belong, respectively, to sect. Brunneotincti of subg. Telamonia (Fig. 2), and dermocyboid lineages unrelated to subg. Leprocybe (data not shown). Conversely, it is here confirmed that C. phrygianus, treated in sect. Pholidei in the ADC but in sect. Veneti in Funga Nordica, actually belongs to subg. Leprocybe (Fig. 2). The remaining 13 taxa described by the French authors in the lineage and still lacking a molecular signature are C. alutaceoolivascens, C. cotoneus, C. depexus, C. depexus var. luminosus, C. raphanoides var. carnecyaneus, C. fulvoraphanoides, C. leptopus, C. melleifollius, C. notatus, C. pseudodecoloratus, C. subannulatus, C. sublanatoides and C. valgus (Bidaud et al. 2005). Considering nonetheless the phylogenetic positioning of successfully sequenced collections, and especially those of C. raphanoides and C. raphanodiabolicus in subg. Telamonia, it seems unlikely that any of these phylogenetically unresolved taxa represent additional biological species in subg. Leprocybe.

Controversy over Cortinarius leproleptopus finally settled

One of the main questions we sought to answer in the present study, concerns the phylogenetic identity of C. leproleptopus and C. cepistipes. The former, described by Chevassut & Henry in 1988, is a striking yellow species fruiting under evergreen oaks in thermophilous woodlands of the Mediterranean, resembling a slender C. cotoneus without olivaceous tinges and with smaller spores. A few years later, the same authors introduced the variety basiluteus, occurring in the same habitat, but displaying a more bulbous stipe covered at its base with conspicuous remnants of a yellow cortina (Chevassut & Henry 1994). However, following additional harvests of both taxa in their original area of distribution and further collections from the French Alps, Favre & Moreau (2001) concluded that the taxon is chimeric. After revising the type material of both C. leproleptopus and of var. basiluteus, these authors argued that the diagnosis and the conserved material actually correspond to two distinct species, which can be mostly separated by the extent of development of the hypodermis. For the one represented by the exsiccata (exhibiting a banal filamentous hypodermis), they published a revised description out of a collection from the French Provence (PAM99111002), and epitypified the name with an Italian collection by G. Consiglio (MCVE763/E.C. 94190), while for the one described in the protologue (with a more differentiated hypodermis) and most new harvests of theirs, they introduced the new name C. cepistipes.

This move prompted strong criticism from the authors of the ADC (Bidaud et al. 2005), who rejected the analysis of Favre & Moreau and controversially introduced a new (but invalid) binomial, C. leptopus, in an attempt to reinstate the original species-concept described by Chevassut & Henry under a different name. The ADC authors also considered C. cepistipes to be a later synonym of C. psittacinus, a species that features a subcellular hypodermis as well. Favre & Moreau (2006) subsequently pointed out that C. leptopus is indeed invalid, because it was based on the same type as C. leproleptopus, and also contested the synonymy of C. cepistipes with C. psittacinus. Nonetheless, the controversy over the autonomy of C. cepistipes and the status of var. basiluteus, persisted to this day, as original material from both taxa remained unsequenced. The conspecificity demonstrated here of (i) the holotype and epitype of C. leproleptopus, (ii) the holotype and four paratypes of C. cepistipes, (iii) PAM99111002, and (iv) the two published collections of C. leproleptopus sensu ADC, establishes that C. leproleptopus and C. cepistipes are conspecific, with nomenclatural priority laying with the former binomial. Additional collections from diverse habitats and geographical regions sequenced for the present work, also illustrate considerable ecological plasticity for this species, significantly extending its distribution range to the eastern limits of the Mediterranean basin (Cyprus) and continental Europe (Germany).

In contrast, the taxonomic status of C. leproleptopus var. basiluteus remains uncertain, since both collections cited in the original diagnosis by Chevassut & Henry (1994), the holotype n° 3594 and the paratype n° 91.83, could unfortunately not be located. A third authentic collection annotated ‘C. leproleptopus var. basiluteus’ exists in the two French mycologists’ private herbaria, but its voucher id ‘Holot. Hry 84109’ is cited in the protologue of the type and is thus nomenclaturally linked to var. leproleptopus, not var. basiluteus. Favre & Moreau (2001) could not explain this oddity but cite C. leproleptopus var. basiluteus as a synonym of C. leproleptopus. Four years later, considering the colour of the pileus and the shape of the stipe, the authors of the ADC recombined the taxon as C. melleifolius var. basiluteus, and added two collections to support their interpretation: AB 03-11-81 and JLC001029-8 (Bidaud et al. 2005). Unfortunately, this material belongs to two distinct clades, C. leproleptopus and C. subcotoneus, respectively (Table 2, Fig. 2), making this taxon chimeric and too ambiguous to be reliably interpreted based on current data. However, given the remarkable morphological plasticity unveiled here in C. leproleptopus, the possibility that this variant represents yet another phenotype of the former, seems highly probable.

A question indirectly raised by this long-standing debate finally needs to be addressed and finds here a partial answer: what is Cortinarius psittacinus? The three collections cited in the ADC under this name fall in three distinct clades (C. cotoneus, C. leproleptopus and C. pescolanensis; Table 1, Fig. 2), indicating that, as pointed out by Favre & Moreau (2006), the authors of the ADC had a mixed interpretation of Moser’s species. But so did Moser himself, it seems. Indeed, although the holotype of C. psittacinus could unfortunately not be sequenced (R. Kuhnert & U. Peintner, pers. comm.), the plate illustrating the species in the Farbatlas (n° 110, Moser & Jülich 2000) undoubtedly represents a member of subg. Leprocybe, akin to C. cotoneus and C. leproleptopus. However, the only available sequence of this species, obtained from an Italian harvest from Moser himself (UDB001106), falls in sect. Rubicunduli, in sister clade relationship to C. rubicundulus (Fig. 2). At first glance, this finding comes as a surprise, because this species is phylogenetically unrelated to subg. Leprocybe (Garnica et al. 2005, Stensrud et al. 2014, Soop et al. 2019a) and, from a morphoanatomical point of view, displays only superficial resemblance to members of this lineage. However, although included in subg. Phlegmacium by some contemporary authors (Bidaud et al. 1999), C. rubicundulus is treated in subg. Leprocybe by Moser (1978), or in a wide concept of subg. Cortinarius by Nordic mycologists (Brandrud et al. 1990). In their typical forms, C. rubicundulus and C. leproleptopus are easily distinguished from one-another, at least by the strong and rapid yellowing of basidiomata upon bruising and obvious bundles of cystidia on gill edges for the former. The two species may be confused in the field, however, since they can both be found in the same Fagaceae woodlands, share the same stout habit and squamulose pilei and, when ageing, they may both develop rusty-brown colours. It is also worth noting that the Italian sequenced collection does not actually belong to C. rubicundulus, but to a closely related clade (referred to as C. aff. rubicundulus in Fig. 2), of which no morphological description or representation are available. Thus, we can only speculate as to whether Moser actually misidentified a slightly deviating collection of C. rubicundulus, or collected a cryptic, sister species of the latter possibly lacking some of the obvious diagnostic differences with C. leproleptopus. In the current state of knowledge though, C. psittacinus does not qualify as an earlier name for C. leproleptopus, nor do older binomials such as C. conopus, C. depexus, C. notatus or C. subnotatus, all matching some features of C. leproleptopus, but also those of C. cotoneus and other confirmed species in the subgenus (Favre & Moreau 2001, Bidaud et al. 2005).

Mediterranean Leprocybe: ecological patterns and specificities

A striking pattern emerging from the results of the present study is the strict Mediterranean distribution of three previously undescribed species, each of them defined by highly specialised ecological preferences (habitat, specificity). Cortinarius jimenezianus is so far known only from matorral on siliceous soils in southern Spain, where it appears to occur in exclusive association with Cistus ladanifer. Cistaceae shrubs are prominent components of the Mediterranean biome, often hosting very specific ectomycorrhizal fungi, including several species of Cortinarius (Henry & Contu 1986, Vila & Llimona 2006, Ortega et al. 2007, Loizides & Kyriakou 2011, Loizides 2016). This highly specialised ecological niche, formed in heavily eroded and degraded Mediterranean landscapes, is replaced by Helianthemum heathlands in northern latitudes, where a different group of Cortinarius species thrive in treeless communities in association with perennial plants of genus Helianthemum (Barden 2007, Henrici 2010, 2011, Kibby 2015, Dima et al. 2016). Although a limited number of species appear to overlap between these two ecosystems (e.g., C. cisticola, C. decipiens), the majority of Cortinarius diversity in Mediterranean Cistus matorral is distinctly different from that of Helianthemum heathlands (Liimatainen & Ainsworth 2018). Considering the morphological similarity between C. jimenezianus and C. venetus, Mediterranean collections assigned to the latter species fruiting near rockroses (e.g., those reported in Bidaud et al. 2007), should be carefully re-examined and sequenced, in order to confirm their phylogenetic identity.

Cortinarius selinolens is in all likelihood associated to Quercus ilex, even though a single collection of this species originates from a pure Arbutus unedo woodland in Tunisia. Trees of the Ericaceae family are known to form arbutoid mycorrhizae (AM) with a broad range of EcM fungi, often acting as ‘substitute’ hosts for pine- or oak-associated fungi (Molina & Trappe 1982, Kennedy et al. 2012). Consistent with this pattern, three additional Cortinarius species (C. luteocingulatus, C. natalis and an unknown species akin to C. luteoperonatus), as well as several Basidiomycota typically linked to thermophilous Quercus forests, were also collected in the same Tunisian forest together with C. selinolens (F. Richard, pers. comm.). Like some Mediterranean collections of C. venetus may in fact reveal to belong to C. jimenezianus, reports of C. melanotus from sclerophyllous woodlands (e.g., Ortega et al. 2007), may actually refer to C. selinolens instead, due to some morphological similarities and, more importantly, a reported odour of parsley.

Cortinarius viridans has so far been exclusively collected under Cedrus brevifolia, the endemic cedar of Cyprus, on igneous and serpentine substrates of the Troodos ophiolite. The unique geographical position, altitudinal range and insular character of Cyprus, have contributed in shaping a diverse range of ecosystems and associated fungal diversity, reflected in high numbers of putatively endemic or range-restricted species (Crous et al. 2017, 2018, Loizides et al. 2016, 2019, 2020, Vidal et al. 2019). Ultramafic bedrocks, from which serpentines are derived, often support unique and highly specific plant communities (Whittaker 1954, Brooks 1987), but it is less clear whether these iron-rich, nutrient-poor substrates can act as a strong abiotic filter to EcM and AM fungi (Branco 2010, Southworth et al. 2014, Kohout et al. 2015). Interestingly, Cortinarius species have been shown to dominate EcM lineages found in ultramafic soils of New Caledonia (Carriconde et al. 2019), and it is thus possible that the highly specialised ecological and pedoclimatic preferences of C. viridans, are indicative of a narrowly endemic distribution of this species confined within the Troodos ophiolitic massif. The taxonomic autonomy of Cedrus brevifolia relative to C. lebani has been controversial but it is postulated that the former represents recently diverging populations of the latter, supporting an ongoing speciation process in Cyprus (Bou Dagher-Kharrat et al. 2006, Eliades et al. 2011, Jasinska et al. 2012). Further sampling for possible Leprocybe occurrences under C. lebani in nearby Lebanon as well as in igneous and serpentine-associated ecosystems at a continental scale, will be necessary to confirm this hypothesis in the future.

The neglected Mediterranean hotspot – an emerging area of interest

Besides these novelties, our findings suggest that the Mediterranean functions as a major ecological filter in this lineage, with most species being either absent from this ecoregion, or tightly linked to it. This was somewhat unexpected because many meridional species in the genus occur both in Mediterranean Quercus ilex forests and in relatively thermophilous deciduous forests of much northern continental Europe (Brandrud 2000), pretty much like C. leproleptopus, C. melanotus and C. subcotoneus. Such ecological plasticity has been the root of much taxonomic confusion in the past, including the controversial interpretations of extra-Mediterranean collections of C. leproleptopus, because Mediterranean species were a priori assumed to be distinct from their extra-Mediterranean counterparts, and phenotypic polymorphism was often interpreted as diagnostic features. In Cortinarius, such bias is not rare and is responsible, for instance, for the failure to recognize C. haasii (extra-Mediterranean) in C. aurantiorufus (Mediterranean), when the latter species was first introduced (Bidaud et al. 2001, Frøslev et al. 2007, Bellanger 2015).

It becomes apparent from the results of the present study, that in the absence of a wide phylogenetic taxon sampling from both Mediterranean and continental ecoregions, it is virtually impossible to draw reliable taxonomic conclusions on the ecological, morphological and distributional boundaries of such versatile polymorphic species. Moreover, the evolutionary history of subg. Leprocybe as unfolding here, suggests that the Mediterranean has been, and probably still is, a key area of species radiation within this lineage (Fig. 8), confirming that this often-neglected ecoregion is indeed a remarkable hotspot of diversity for higher Fungi. Such findings are of paramount significance in the broader context of climate change, as Mediterranean ecosystems are forecasted to undergo dramatic changes due to accelerated climate warming, including mass aridification, desertification, extensive loss of habitat, shifts in biodiversity composition and perhaps even mass species extinctions (Giorgi 2006, IPCC 2007, Giorgi & Lionello 2008, Richard et al. 2011, Ágreda et al. 2015, Büntgen et al. 2015, Loizides et al. 2019).

Fig. 8

Biogeographical distribution of Leprocybe species in the Mediterranean. Mapping of sequenced collections of the eight confirmed species in subg. Leprocybe around the Mediterranean basin. Colours delimit vegetation types in the Mediterranean, as defined in Quézel & Médail (2003). Note that C. pescolanensis is present in a single locality within the area, at high elevations, and that C. cotoneus, C. phrygianus and C. squamivenetus are so far not represented in the Mediterranean.

Next generation sequencing of old type material – what future in fungal taxonomy?

Nomenclatural instability has hampered fungal taxonomy for decades. The advent of molecular phylogenetic tools at the turn of the millennium, has revolutionised our ability to resolve cryptic or poorly understood species, but still failed to stabilise taxonomy, prompting some scientists to controversially advocate abandoning fungal nomenclature altogether (Money 2013). The inability to reliably assign a single Linnaean binomial to molecularly identified biological species, can be mostly attributed to two fundamental problems:

the insufficient representation of type collections in public sequence databases, to which names are legitimately attached; and

the inherent difficulties to identify species limits by phylogenetic methods, especially when using the cheap and straightforward ITS rDNA barcoding method (Kiss 2012).

By its ability to sort out original DNA from contamination DNA, next generation sequencing addresses these two shortcomings and is thus expected to soon integrate into the fungal taxonomist’s toolbox. First, as demonstrated here and in another recent work (Forin et al. 2018), old and heavily contaminated herbarium material can be ‘salvaged’ by deep sequencing, due to the potent sensitivity of the method, which is several magnitudes higher than that of Sanger sequencing. Generating and publishing ITS barcodes of ancient holotypes conserved in European and North American fungaria, will undoubtedly mark a milestone in stabilizing fungal taxonomy, as old and previously ambiguous names will be unequivocally deciphered and made available for currently recognised clades. Second, by documenting the intragenomic diversity of the ITS barcode for analysed species, NGS allows selecting SNPs relevant for defining species limits. Indeed, the limited number of nucleotide differences between two closely related clades is often difficult to translate into an informative taxonomy, such as for instance trying to interpret between two recently derived species and a single polymorphic one, the two often representing competing options in the absence of any other criterion. Because polymorphisms tend to fix between two populations upon their reproductive isolation, documenting genetic flow at polymorphic sites is crucial in the decision-making process. In Sanger sequencing of diploid organisms, heterozygocities represent hallmarks of genetic flow between two haplotypes and, as such, they disqualify concerned sites for species diagnosis. However, insufficient sampling, the iterative and stochastic nature of the PCR that tend to asymmetrically amplify alleles over cycles, and excessive data processing of public sequences (Selosse et al. 2016), all converge to mask heterozygocities. The ITS polymorphisms unveiled by NGS at the individual genome scale, alleviate this issue and provide an efficient alternative in disqualifying variable sites for species limit delineation.

Beyond these obvious benefits for fungal taxonomy, the molecular landscape of the ITS allelic diversity within a single fungal genome yields data out of reach of classical Sanger sequencing, of interest to better understand the complex mechanisms of concerted evolution (Eickbush & Eickbush 2007). This process is responsible for sequence homogenization of the multiple alleles of DNA arrays such as the rDNA cistron within individual genomes, limiting sequence variation at the intraspecific scale. However, just a handful of studies have addressed this evolutionary mechanism in fungi and only a very recent one reports on higher Fungi (Tremble et al. 2020). Deciphering how concerted evolution controls the allelic diversity of multiple copy genes in these organisms, and how the process is modulated (or not) by environmental cues, certainly represents a promising research avenue. Although such research lays beyond the scope of the present work, the high intragenomic ITS polymorphism unveiled here in three species of Cortinarius was unexpected and contrasts with observations in yeast strains maintained in the lab (Ganley & Kobayashi 2011). However, it is consistent with what has been reported in plant pathogenic fungi and, more recently in Boletus edulis (Simon & Weiß 2008, Tremble et al. 2020), suggesting that concerted evolution may be relaxed in some or all lineages of uncultured higher fungi. More work will shed light on the regulatory pathways controlling this universal evolutionary process in Basidiomycota, in relation to relevant ecological constraints.

1. In boreal coniferous forests . . . . . . . . . . . . . . . . . . . . . . 2

1. In continental or mountainous mesophilous forests, absent from the Mediterranean . . . . . . . . . . . . . . . . . . . . . . . . . 3

1. In Mediterranean ecosystems . . . . . . . . . . . . . . . . . . . 9

2. Average spore size < 6 × 5 μm, conspicuous brown veil bands on stipe . . . . . . . . . . . . . . . . . . . . C. phrygianus

2. Average spore size > 6 × 5 μm, conspicuous brown scales on pileus . . . . . . . . . . . . . . . . . . . . . . C. squamivenetus

3. Slender species (S > 4) . . . . . . . . . . . . . . . . . . . . . . . . . 4

3. Stout species (S < 4)* . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

4. Brown scales on pileus, never hygrophanous, olivaceous-brown dominating hues, parsley odour typical or at least frequent . . . . . . . . . . . . . . . . . . . . . . . . . . . C. melanotus

4. Smooth and often hygrophanous pileus, green dominating hues, odour of parsley absent . . . . . . . . . . . . C. venetus

5. In deciduous forests . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

5. In coniferous forests . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

6. Average spore size > 8.5 × 7 μm . . . . . . . . . C. cotoneus

6. Average spore size < 8.5 × 7 μm . . . . . . . . . . . . . . . . . . 7

7. Average spore size = 8 × 6.5 μm, olivaceous-brown dominating hues . . . . . . . . . . . . . . . . . . . . . . . C. subcotoneus

7. Average spore size = 7.5 × 6 μm, yellow dominating hues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. leproleptopus

8. Average spore size < 6 × 5 μm, conspicuous brown veil bands on stipe . . . . . . . . . . . . . . . . . . . . . . C. phrygianus

8. Average spore size > 6 × 5 μm, meridional . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. pescolanensis

9. Known from Cyprus, under Cedrus brevifolia . . C. viridans

9. Known from Spain, under Cistus ladanifer . . . . . . . . . . 10

9. Ecology different . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

10. Flesh bitter, cespitose or densely gregarious . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. jimenezianus

10. Flesh mild, solitary to gregarious, rarely cespitose . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. venetus

11. Abies-dominated, supra-mediterranean ecosystems . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. pescolanensis

11. Ecology different . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

12. Stout species (S < 4) . . . . . . . . . . . . . . . . . . . . . . . . . . 13

12. Slender species (S > 4) . . . . . . . . . . . . . . . . . . . . . . . . 15

13. Average spore size < 8 × 6 μm, yellow dominating hues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. leproleptopus

13. Average spore size > 8 × 6 μm, olivaceous-brown dominating hues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

14. Scaly pileus and parsley odour, evoking C. melanotus . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. selinolens

14. Tomentose pileus and rapahanoid odour, evoking C. cotoneus . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. subcotoneus

15. Brown scales on pileus, never hygrophanous, olivaceous-brown dominating hues, parsley odour frequent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. melanotus

15. Smooth and often hygrophanous pileus, green dominating hues, never with an odour of parsley . . . . . . . C. venetus