The genus Melanconis (Diaporthales)

Abstract The genus Melanconis (Melanconidaceae, Diaporthales) in the strict sense is here re-evaluated regarding phylogenetic structure, taxonomy, distribution and ecology. Using a matrix of sequences from ITS, LSU, ms204, rpb2, tef1 and tub2, eight species are recognised and their phylogenetic positions are determined. Based on phylogenetic, morphological and geographical differentiation, Melanconis marginalis is subdivided into four subspecies. Melanconis italica is reduced to a subspecies of Melanconis marginalis. The two species Melanconis larissae from Betula sp. and M. pacifica from Alnus rubra are described as new. Melanconis alni and M. stilbostoma are lectotypified and M. alni, M. marginalis and M. stilbostoma are epitypified. All GenBank sequences deposited as Melanconis alni are shown to actually represent M. marginalis and those as M. marginalis belong to the newly described M. pacifica. Currently, Alnus and Betula are the sole host genera of Melanconis. All species and subspecies are (re-)described and illustrated. In addition, the neotypification of Melanconium pterocaryae is here validated.


Introduction
Melanconis, the type genus of the family Melanconidaceae (Diaporthales), was originally described by Tulasne (1856) with M. stilbostoma as its generic type, but without a generic diagnosis. His inclusion of species like M. spodiaea made the genus heterogeneous from the beginning. Since then, many species names have been erected in the genus. In his generic revision, Wehmeyer (1941) treated the genus in a very wide sense, organising the species in subgenera and sections, which themselves were heterogeneous, containing species of genera like Chapeckia, Coryneum (Pseudovalsa), Macrodiaporthe, Massariovalsa, Melanconiella or Pseudovalsella. Barr (1978) accepted Melanconis roughly in the sense of Wehmeyer´s subgenus Eumelanconis, which included Melanconiella. In this sense, the genus Melanconis was one of many genera of the large family Melanconidaceae and was defined by a distinct ectostromatic disc, a more or less well-developed entostroma, twocelled hyaline or brown ascospores with or without appendages, in combination with melanconium-or discosporium-like asexual morphs (Barr 1978). The first phylogenetic analyses of the Diaporthales (Castlebury et al. 2002; see also Jaklitsch et al. 2016, Senanayake et al. 2018, however, suggested that Melanconidaceae should be confined to its type genus Melanconis with a restricted number of species. This phylogenetic generic concept corresponds, apart from a few exceptions, with Wehmeyer's (1941) section Stilbostomae of his subgenus Eumelanconis. Subsequently, many names have been combined in other genera in various families following morphological and/or phylogenetic analyses (Barr 1978;Jaklitsch and Voglmayr 2004;Voglmayr and Jaklitsch 2008;De Silva et al. 2009). Melanconiella was extensively studied by Voglmayr et al. (2012), who determined that species of Melanconis cause more conspicuous bumps in the host bark than those of Melanconiella and form light-coloured, white or yellowish ectostromatic discs. Wehmeyer (1941) had used this trait to distinguish his section Stilbostomae from his Chrysostromae, which are characterised by dark coloured discs. Although light coloured discs are not uncommon in Melanconiella, Wehmeyer's (1941) section Chrysostromae of his subgenus Eumelanconis basically matches the phylogenetically conceived genus Melanconiella, except for a few species, which belong elsewhere. For some of these species, the new genus Juglanconis was established in the new family Juglanconidaceae . Two other species were segregated from Melanconis to Alnecium and Phaeodiaporthe by Voglmayr and Jaklitsch (2014). Voglmayr et al. (2012) found an unexpectedly high species diversity in Melanconiella, particularly on Carpinus spp. and showed that its species either have a melanconium-or a discosporina-like asexual morph, but never both morph types. They gave also information of taxonomic placement of other Melanconis spp. Here we treat the residual species of Melanconis in the strict sense.

Sample sources
All isolates included in this study originated from ascospores or conidia of freshly collected specimens derived from recently dead branches or twigs. Details of the strains including NCBI GenBank accession numbers of gene sequences used to compute the phylogenetic trees are listed in Table 1. Strain acronyms, other than those of official culture collections, are used here primarily as strain identifiers throughout the work. Representative isolates have been deposited at the Westerdijk Fungal Biodiversity Centre (CBS-KNAW), Utrecht, The Netherlands. Details of the specimens, used for morphological investigations, are listed in the Taxonomy section under the respective descriptions. Herbarium acronyms are according to Thiers (2019). Freshly collected specimens have been deposited in the Fungarium of the Department of Botany and Biodiversity Research, University of Vienna (WU) and in the Fungarium of the Natural History Museum of Vienna (W).

Morphology
Microscopic observations were made in tap water, except where noted. Morphological analyses of microscopic characters were carried out as described by Jaklitsch (2009). Methods of microscopy included stereomicroscopy using a Nikon SMZ 1500 and Nomarski differential interference contrast (DIC), using the compound microscopes Nikon Eclipse E600 or Zeiss Axio Imager.A1 equipped with a Zeiss Axiocam 506 colour digital camera. Images and data were gathered using a Nikon Coolpix 4500 or a Nikon DS-U2 digital camera and measured by using the NIS-Elements D v. 3.0 or 3.22.15 or Zeiss ZEN Blue Edition software packages. For certain images of ascomata, the stacking software Zerene Stacker v. 1.04 (Zerene Systems LLC, Richland, WA, USA) was used. Measurements are reported as maxima and minima in parentheses and the range representing the mean plus and minus the standard deviation of the number of measurements given in parentheses.

Phylogenetic analyses
The newly generated sequences were aligned with the Melanconis sequences of Fan et al. (2016Fan et al. ( , 2018) and a few additional GenBank sequences. Species of Juglanconis were selected as outgroup ; the GenBank accession numbers of the sequences, used in the phylogenetic analyses, are given in Table 1. All alignments were produced with the server version of MAFFT (www.ebi.ac.uk/Tools/mafft), checked and refined using BioEdit v. 7.2.6 (Hall 1999). For phylogenetic analyses, all sequence alignments (ITS, LSU, ms204, rpb2, tef1 and tub2) were combined.
Maximum Likelihood (ML) analyses were performed with RAxML (Stamatakis 2006) as implemented in raxmlGUI 1.3 (Silvestro and Michalak 2012), using the ML + rapid bootstrap setting and the GTRGAMMA substitution model with 1000 bootstrap replicates. The matrix was partitioned for the different gene regions and substitution model parameters were calculated separately for them.
Maximum Parsimony (MP) analyses were performed with PAUP v. 4.0a166 (Swofford 2002). All molecular characters were unordered and given equal weight; analyses were performed with gaps treated as missing data; the COLLAPSE command was set to MINBRLEN. MP analysis of the combined multilocus matrix was done, using a parsimony ratchet approach. For this, a nexus file was prepared using PRAP v. 2.0b3 (Müller 2004), implementing 10000 ratchet replicates with 25% of randomly chosen positions upweighted to 2, which were then run with PAUP. MP bootstrap analyses were performed with 1000 replicates, using 5 rounds of random sequence addition and subsequent TBR branch swapping (MULTREES option in effect, steepest descent option not in effect) during each bootstrap replicate, with each replicate limited to 100000 rearrangements.
In the Results and Discussion sections, bootstrap values (BS) below 70% are considered low, between 70-90% medium and above 90% high.

Revision of Melanconis sequences deposited in GenBank
Comparison of our sequences with GenBank sequences revealed that all accessions of Melanconis alni and M. marginalis, deposited in GenBank, were misidentified. All GenBank accessions of M. alni were shown to actually represent M. marginalis, while the single isolate of M. marginalis turned out to be a new species, described as M. pacifica below. These misidentifications were also confirmed by morphological re-investigation of specimens from which these sequences were generated.

Phylogenetic analyses
Of the 6052 characters included in the combined multilocus analyses, 925 were parsimony informative (133 from ITS-LSU, 142 from ms204, 214 from rpb2, 245 from tef1 and 191 from tub2). The best ML tree (lnL = −18240.558) revealed by RAxML is shown as Fig. 1. The MP analysis revealed 3394 MP trees 1647 steps long, which were identical except for some differences within species and a polytomy at the M. groenlandica-M. larissae-M.stilbostoma node (not shown). Tree topology of the MP strict consensus tree was compatible with the ML tree, except for a sister group relationship of M. marginalis subsp. europaea and M. marginalis subsp. marginalis and some minor topological differences within species and subspecies (not shown).
All Within Melanconis marginalis, two main subclades were evident with ML and MP BS above 85%, one containing accessions from eastern Canada, Alaska, Japan and the Russian Far East and another with accessions from Central Europe; in addition to these two main subclades, the Melanconis marginalis clade contained two deviating lineages, an Italian collection from ?Alnus cordata described as M. italica by Senanayake et al. (2017) and two accessions from eastern Tyrol from Alnus alnobetula. In light of this geographical differentiation, a substantial genetic variability within these clades (Fig. 1) and minor morphological differences, these four lineages are formally recognised on the subspecies level.

Culture characteristics
Culture images of seven studied Melanconis species, grown on MEA and CMD, are illustrated in Figure 2. Culture descriptions are given under the respective species.  Figure 1. Phylogram of the best ML tree (lnL = −18240.558) revealed by RAxML from an analysis of the ITS-LSU-ms204-rpb2-tef1-tub2 matrix of Melanconis, with 5 species of Juglanconis (Juglanconidaceae) selected as outgroup. ML and MP bootstrap support above 50% are given at the first and second position, respectively, above or below the branches. Strain numbers are given following the taxon names; strains formatted in bold were sequenced in the current study. Melanconis taxa occurring on Alnus are marked blue, those on Betula in green. The broken branches to the outgroup were scaled to 10%. Notes. Tulasne (1856) had already mentioned Melanconis, but did not give a generic diagnosis. Hence, the species he newly described were invalid, but became validated by reference in Tulasne and Tulasne (1863) (Paul Kirk, pers. comm.).
In contrast to Diaporthe, species of Melanconis always develop in bark, never in wood and lack stromatic zones. Pseudostromata are pulvinate to conical, circular to elliptic in outline and usually slightly project beyond the bark surface with perithecial contours remaining indistinct. Ectostromatic discs usually project distinctly from the surface of the pseudostromata and are bright, white to yellowish, to brown when old.
Nomenclaturally, the older genus Melanconium potentially competes with the younger genus Melanconis. However, as outlined in Rossman et al. (2015), the generic concept of Melanconium and the true identity of its generic type, M. atrum, are obscure and they therefore recommended to protect the well-defined Melanconis over Melanconium, which was formally adopted in the last ICN (Turland et al. 2018, Appendix III). Diagnosis. Melanconis alni is recognised by ascospores having filiform, tapering appendages and dark brown α-conidia with a pale to subhyaline median area.
Culture: Colony on CMD at 16 °C first hyaline, turning yellowish-brown from the centre, becoming covered by flocks of white aerial hyphae and conidiomata forming around the centre or colony irregular, with limited growth, turning green to black due to conidiomata; on MEA first hyaline, circular, with short aerial hyphae, forming concentric zones, the outer white, the inner turning brown, black conidiomata forming between the zones, margin becoming diffuse and the entire colony turning brown. Odour indistinct.
Distribution and ecology. Melanconis alni occurs in Europe on dead twigs and branches of Alnus glutinosa and A. incana, mainly at lower elevations.  Notes. Melanconis alni was described by Tulasne from Alnus glutinosa in 1856 after a presentation of the topic in April 1856. Tulasne and Tulasne (1863) validated the name in Melanconis, illustrated ascospores with typical long acute appendages and mentioned material from Meudon and Chaville. In PC, nine specimens of Tulasne are extant in the Melanconis alni folder; three of them were collected after its description in 1856 and, for one, no collection data are available. PC 0723590, PC 0723591, PC 0723593, PC 0723594 and PC 0723595 were collected after the publication date. PC 0723588 (no data) and PC 0723589, PC 0723596 from 1852 only contain asexual morph, but in the protologue, the sexual morph is also described. Therefore, we select PC 0723592, which also contains few pseudostromata of the sexual morph, as the lectotype. In PC 0723592 and PC 0723595, both α-and β-conidia are present. Generally, β-conidia are inconspicuous and produced in small numbers, i.e. they are easily overlooked. Asci in old herbarium material are shrunk and difficult to rehydrate, therefore significantly smaller than those of fresh material. In KOH, the ascus apex becomes very thick and the ring disappears; also ascospore appendages disappear in KOH. Tulasne and Tulasne (1863) and Wehmeyer (1941) listed the following asexual morph names, amongst others, as linked to M. alni: Stilbospora microsperma Pers. Material with this name is not accessible in L; Melanconium sphaeroideum Link (1825) is more generally given as the name of the asexual morph. Sieber et al. (1991) used another name described by Link (1825), Melanconium apiocarpum, for the asexual morph of Melanconis alni. As Link´s type material of these taxa is not extant in B, we are unable to draw a conclusion about their identity; in addition, the descriptions in Link (1825) are vague and he gave no hosts. Therefore, we continue to use the name M. alni, which is generally well-known. Type material of Melanconium atrum Link, the generic type of Melanconium, described from Germany (K(M) 171588, slide from Melanconium atrum type material from Persoon´s herbarium) has conidia of the same shape, size and lighter median band (Fig. 4p) and may thus be conspecific with M. alni, but it was described from Fagus sylvatica. According to Sutton (1964), Link had sent his material to Persoon, because in the herbarium of the latter 3 specimens labelled M. atrum were extant. The host of one of these materials was identified as Fagus, based on bark structure. This specimen was selected as lectotype. The slide K(M) 171588 (= IMI 102914) was prepared from the lectotype and is thus an isotype. Accordingly, Melanconium atrum is a different species, despite its morphological similarity with M. alni, because the latter only occurs on Alnus spp. We have not seen any Melanconium on Fagus, but Petrini and Fisher (1988), Sieber et al. (1991) and Kowalski and Kehr (1992) reported and isolated M. atrum as an endophyte of Fagus. For α-conidia of isolates from Fagus sylvatica and Quercus robur, Sieber et al. (1991) reported mean sizes of 11.7-12 × 8.5-8.9 µm, which were similar to those from Alnus glutinosa (on average, 10.1-12.3 × 5.9-7.4 µm). However, the protein profiles revealed by isozyme electrophoresis differed markedly between the isolates from Alnus glutinosa and those from Fagus/Quercus, confirming them to represent distinct species that may not even be congeneric. Another fact may support the presence of morphologically similar but rare taxa on Fagaceae, as, for example, Melanconium gourdaeforme with similar conidia was described by Kobayashi (1968) from Castanea. A narrow light band is also characteristic for conidia of Melanconiella ostryae .
Ascospore appendages of Melanconis alni may sometimes be similar to those of M. marginalis, at least in fractions, although truncate appendages in M. alni are rather a consequence of microscopic mount preparation. On Alnus incana both species occur, therefore the asexual morph should be sought for to reliably identify the species.  Description (after Bohn 1993): Colonies on PDA and MEA 30-33 mm after 10 d (52-62 mm after 20 d), appearing leathery, at first whitish to greyish, later becoming greyish-orange, particularly on MEA; margin superficial, entire on MEA but fimbriate to lobate on PDA; exudate and diffusible pigment absent; reverse greyish-orange, especially at the margin; brownish, thick-walled, chlamydospore-like swollen portions 6-18 µm diam. present. Conidiomata appearing after ca. 14 d as dark green pustules of various sizes, irregularly scattered over the colony surface, but sometimes arranged in concentric rings, particularly in old cultures, initially covered by mycelium but becoming almost black and shiny at later stages due to the mass of conidia; conidiomata sporodochial (acervular?), irregular, dark green, up to 2 mm diam., scattered, gregarious or coalescent, composed of a 50-70 µm high stroma of textura intricata and conidiophores. Marginal hyphae and setae absent. Conidiophores arising from the stroma, branched, septate, yellowish to brownish, ca. 40-75 µm × 2-4 µm. Conidiogenous cells cylindrical to subulate, 15-25 × 2-3 µm, arranged in verticils of 2-4 at the top of the conidiophore, sometimes also intercalary, provided with conspicuous, pigmented collarettes and producing conidia by percurrent growth. Conidia black and shiny in mass, olivaceous to brownish under the microscope, straight, cylindrical with rounded ends, sometimes slightly narrowing towards the base or apiculate, (9-)10-12(-15) × (5-)6(-7) µm, with smooth wall. Teleomorph not formed after 3 months incubation.

Melanconis betulae
Culture (own observations): Colony on MEA circular, first hyaline, turning and long remaining whitish, with age forming narrow concentric zones with tooth-like margins and turning pale brownish. Odour indistinct to unpleasant.
Distribution and ecology. Melanconis groenlandica is known from North America Note. This species was isolated as a putative endophyte from Betula nana and described from MEA and potato dextrose agar as a species of Myrothecium. In our phylogenetic analyses, three isolates from North America and one from Japan grouped with the ex-type isolate of M. groenlandica with high support. Tak Description. See  and Fan et al. (2016). Culture: Colony on MEA circular, first hyaline, forming a white outer and brown inner zone, with radial stripes; conidiomata forming mostly in the inner zone. Odour indistinct.

Melanconis larissae
Culture: Colony on MEA at room temperature circular, dense, first hyaline, turning rosy. Odour indistinct to musty.
Distribution and ecology. Melanconis larissae is known from a single specimen collected in New York State from an unidentified species of Betula.
Notes. The description of this taxon is based on a single specimen with overmature sexual morph and well-developed asexual morph with thick masses of conidia. Melanconis larissae differs from M. stilbostoma by the broad light-coloured zone of its conidia. No β-conidia have been detected in this specimen, but oblong to ellipsoid, hyaline to dilute brownish conidia 5-9 × 1.7-5 µm, which we interpret as immature α-conidia. Notes. This species is here subdivided into four subspecies below. See under subsp. marginalis for the original species.

Melanconis marginalis
Culture: Colony on CMD at 16 °C first hyaline, partly or entirely turning brownish or ochre, either covered by a dense white mat of aerial hyphae or not, sometimes becoming indistinctly zonate, sometimes forming irregularly disposed conidiomata; on MEA at room temperature, first hyaline to whitish, soon forming a few broad zones with uneven margins forming teeth, the latter partly turning brown.
Distribution and ecology. Notes. This subspecies differs mainly in its occurrence in (Central) Europe and by forming a clade of its own in phylogenetic analyses (Fig. 1). While the differences of the European accessions in each marker included are few, they are consistent, resulting in a well-delimited clade in the multigene analyses. As the morphological differences from M. marginalis subsp. marginalis are only small, we prefer to classify the European taxon as a subspecies rather than a separate species.
Under the name Melanconis alni, Podlahová (1973) described both sexual and asexual morphs of a Czech collection from Alnus alnobetula which clearly represents M. marginalis, and Szász (1966) listed and described the species (as Melanconium dimorphum) from Romania, again from Alnus alnobetula. In his isozyme studies of Melanconium, Sieber et al. (1991) included a Swiss isolate from Alnus alnobetula (as Melanconium sp. 1). This isolate showed a distinct but similar isozyme pattern to North American collections of Melanconis marginalis and had a mean conidial size of 11.7 × 4.3 µm, indicating that this isolate also represents Melanconis marginalis subsp. europaea. Notes. It is presently unclear, whether this poorly described and illustrated taxon that is only known from a single collection is simply Melanconis marginalis subsp. europaea or merits a subspecies name of its own. First, the host given by the authors, Alnus cordata, naturally occurs in southern Italy and Corsica and, thus, may be correct only if planted in the collection area, which is not given by the authors. Secondly, the ascospores are in the range of other subspecies and appendages are neither mentioned nor illustrated, although a few are visible in their ascus images. Apparently, ascospores were mounted in KOH, where appendages are invisible. Thirdly, they describe the asexual morph from culture and include only a poor image of conidia without giving any measurements. Last but not least, only LSU, ITS and rpb2 are available, which are insufficient to reliably resolve its true phylogenetic position. In addition, instead of comparing their taxon with M. marginalis, they compare it with M. alnicola (Jaap 1917), which is a synonym of Alnecium auctum. (from ascospores), D321a (from α-conidia), D321b (from β-conidia); MBT390382).

Distribution and ecology. Widespread in North America and also occurring in
Japan and eastern Russia on various subspecies of Alnus alnobetula and A. incana; recorded also from A. rubra (Sieber et al. 1991; see also material cited below).
Sizes of asci depend on the age of the material. They shrink with time and in specimens, which are 20 or more years old, they are smaller and do not obtain the original size even in KOH; also, it is very difficult to release ascospores from asci. In fresher specimens, asci are easily separable and ascospores are readily released. Vital asci open readily in mounts. Nonetheless, fresh asci of the epitype of subsp. marginalis were distinctly smaller than fresh asci of subsp. europaea.
Poor representation of the asexual morph in fungarium specimens may be due to the fact that the sexual morph is usually abundant, with numerous white ectostromatic discs; thus, the asexual morph may have been neglected during collecting or even discarded. β-conidia are often absent or scant and old amongst α-conidia in dark conidial deposits, hence they are either not formed or produced before α-conidia.
Culture: Colony on MEA dense, first hyaline to white, with restricted growth, forming brown radial portions mostly submerged in the agar. Odour unpleasant.
Distribution and ecology. Co-occurring with Melanconis marginalis subsp. europaea in a subalpine area of eastern Tyrol, Austria, Europe, on Alnus alnobetula.

Notes.
As this subspecies differs morphologically only subtly from the other varieties of M. marginalis, we prefer to classify it as a subspecies rather than a separate species. While the ITS sequences of Melanconis marginalis subsp. tirolensis differs from Melanconis marginalis subsp. europaea in only a single base pair, the differences are substantial in all other markers included, particularly tef1 and tub2. Jaklitsch & Voglmayr,sp. nov. MycoBank No: 834112 Fig. 12 Diagnosis. This species is characterised by its occurrence on Alnus rubra and α-conidia, which are wider and darker than those of M. marginalis and differ by a different shape and absence of a light band from those of M. alni.

Additional materials examined
Asexual morph acervular, intermingled with pseudostromata of the sexual morph or developing separately, conspicuous. First white tissue (central column) forming within the bark, becoming surrounded by sterile yellow margin and narrow discs rupturing bark epidermis, followed by the production of conidia in olivaceous to black chambers containing black conidial masses translucent though bark. Conidiomata 0.9-3.2 mm diam., subconical or pulvinate, more or less circular in outline, scattered or crowded.
Culture: Colony on CMD at 16 °C forming irregular white and brown to ochre zones partly covered by aerial hyphae or hyaline, undifferentiated, forming brown spots and irregularly disposed conidiomata; on MEA at room temperature first white, later with broad white and brown zones with undulating margin and conidiomata forming mostly on the outer margin. Odour indistinct to fruity.

Validation of neotypification
Here we also validate the neotypification of Melanconium pterocaryae, the basionym of Juglanconis pterocaryae by Voglmayr et al. (2019), where the new requirement to explicitly state the MBT number in the typification proposal was missing:
All melanconis-like species form their fructifications in bark and lack black zones, which delimit the pseudostromata from surrounding bark tissue in genera like Diaporthe. The sexual morph in Melanconis sensu stricto is characterised by distinctly projecting white to yellowish ectostromatic discs, which continue as stromatic central columns downwards, by entostroma, which is optically scarcely different from internal bark tissue, by long cylindrical ostiolar necks, which converge in the disc, by hyaline bicellular ascospores with or without appendages, by absence of paraphyses at maturity and asci, which have an apical ring and are released from the subhymenium at maturity. Conidiomata of the asexual morph are acervular. They commonly produce two types of conidia, melanconium-like brown α-conidia and narrow hyaline to brownish β-conidia. Species of Dendrostoma in the Erythrogloeaceae Jiang et al. 2019) also produce two types of conidia on the same conidiophores, but both are hyaline. Acervuli of Melanconis, however, particularly in M. marginalis, form chambers, in which first β-conidia are produced. Such chambers are still present when α-conidia are produced, but in the latest stages of maturation, the entire fertile region around the central column is filled with α-conidia and appears as a single locule. In species of the morphologically most similar genera Melanconiella  and Juglanconis , pseudostromata are less conspicuous and project to a lesser degree from the bark surface than in Melanconis. The central column in Melanconiella is usually grey, dull yellow to greenish, only rarely white and often poorly developed and ascospores may be hyaline or brown. The most striking difference between Melanconis and Melanconiella lies in the asexual morph. In Melanconis, each species produces α-and β-conidia in the same conidiomata, whereas each species of Melanconiella only produces a single type of conidia, either brown melanconiumlike (corresponding to α-conidia) or hyaline discosporina-like conidia (corresponding to β-conidia). Species of Juglanconis only produce melanconium-like conidia, which have a gelatinous sheath (also present in a few Melanconiella spp.) and differ from the other genera by the presence of verrucae on the inner surface of the conidial wall.

Molecular phylogeny, species numbers, concept and delimitation
In Melanconiella, 15 species have been recognised Fan et al. 2018) and five in Juglanconis . Fan et al. (2016Fan et al. ( , 2018 included five species of Melanconis sensu stricto in their phylogenetic trees. Here we add three species, of which two are new. While all betulicolous species, except for the basal M. betulae, formed a highly supported clade, those on Alnus were scattered in between, so no general evolutionary pattern in host association could be revealed. Remarkably, within species, a commonly high genetic divergence and variability was observed (e.g. within M. groenlandica, M. itoana, M. marginalis and M. stilbostoma; see Fig. 1), contrary to Melanconiella and Juglanconis, where the species clades were genetically rather homogeneous Fan et al. 2018). This may, in part, be attributed to the wider geographic distribution and host range of these Melanconis species, but it may also indicate that they are within the process of evolutionary radiation and speciation. Although the species concept in Melanconis is primarily based on phylogenetic analyses, we consider morphological and ecological evidence as important criteria for taxonomic conclusions. The taxa on Betula spp. may be more or less easily distinguished by differences in the morphology of α-conidia and by ecology: α-conidia of M. larissae have a large light-coloured zone, those of M. itoana have a l/w ratio of > 3 and those of M. betulae and M. groenlandica, as given by the respective authors, are shorter than those of the other species, albeit similar. However, the latter two species occur on different host species: M. betulae on Betula albosinensis, M. groenlandica on Betula maximowicziana, B. nana and B. papyrifera.
Taxa on Alnus spp. may pose difficulties in differentiation. Ascospores of M. alni and M. marginalis differ in shape, size and particularly in appendages from each other. Nonetheless, all features are overlapping and, for example, ascospore appendages of M. alni are not always long and pointed, particularly in old fungarium specimens, but show some similarities with those of M. marginalis. In such cases, it is important to have the asexual morph in order to study its conidia, which are strikingly different from those of M. marginalis. The same applies to Melanconis accessions from the western North American Alnus rubra, where the co-occurring M. pacifica and M. marginalis can be reliably distinguished by their conidia (see, for example, also fig. 2 in Sieber et al. 1991).
The situation is particularly complex within M. marginalis, which splits up into four subclades in our phylogenetic analyses. Morphology amongst those subclades is very similar, measurements are heavily overlapping and only subtle differences or tendencies are recognisable. In addition to the lack of distinctive morphological characters, there is also a substantial amount of genetic variation within the two of the four subclades, for which several accessions are available, particularly within M. marginalis sensu stricto, which will certainly increase if more accessions from additional geographic areas and Alnus species and subspecies are added. Only a small part of the distribution area of M. marginalis is yet sampled. We, therefore, do not think that these subclades should be interpreted as different species, but as a single variable species. Acknowledging the geographical and genetic differentiaton, we decided to classify them as subspecies that may be within the process of speciation. Vicariant speciation may be the reason for splitting of the M. marginalis clade into two main clades, but the residual two clades that are only based on a single and two specimens, were gathered within a small restricted region in Austria and northern Italy. The internal structure of the whole clade may therefore change, in particular, if isolates from additional specimens collected in western and central Russia were added to the phylogenetic analyses and if sequences of all phylogenetic markers of Melanconis marginalis subsp. italica were included.
Misidentification of M. alni and M. marginalis is also prominent in GenBank sequences that were used in all published phylogenetic analyses including these species, resulting in an interchanged application of the names. Based on, as we now know, incorrect assumptions purported in the literature (e.g. Wehmeyer 1941) that M. marginalis is a North American and M. alni a European species, Central European accessions of M. marginalis were misidentified as M. alni. Vice versa, M.E. Barr misidentified her Canadian isolate from Alnus rubra, that is closely related to M. alni and here described as M. pacifica, as M. marginalis. Therefore, all sequences currently deposited in GenBank as M. alni actually represent M. marginalis, while those of M. marginalis belong to M. pacifica.

Hosts
While Juglanconis is confined to the Juglandaceae, subtribus Juglandinae , both Melanconiella and Melanconis occur on the Betulaceae. So far, species of Melanconiella primarily occur on the subfamily Coryloideae with the exception of M. betulae and M. decorahensis, which inhabit Betula Fan et al. 2018). In contrast, Melanconis is confined to Alnus and Betula, the sole genera of the subfamily Betuloideae. While all known Melanconis species are highly host specific on the generic level (i.e. no Melanconis species occurs on Alnus as well as Betula hosts), host specificity is less expressed and variable concerning their host species range. In addition, the same host species is commonly used by more than one Melanconis species. For instance, the widely distributed M. stilbostoma has been recorded from various species of Betula, which is likewise true for M. groenlandica (for confirmed hosts, see Table 1). Conversely, M. betulae is so far only known from a single host, B. albosinensis, which, however, is also host for M. itoana . For Melanconis species on Alnus, M. alni and M. marginalis show some host specificity but are not strictly host specific; while A. glutinosa and A. alnobetula are apparently only colonised by M. alni and M. marginalis, respectively, both species occur on A. incana. Melanconis pacifica, here described as a new species, seems to be host specific on A. rubra, which, however, also harbours M. marginalis. Therefore, the host species are of limited use for species identification and additional investigations are required to elucidate the host range of the various Melanconis species.