Understanding the evolution of phenotypical characters in the Micarea prasina group (Pilocarpaceae) and descriptions of six new species within the group

Abstract Six new Micarea species are described from Europe. Phylogenetic analyses, based on three loci, i.e. mtSSU rDNA, Mcm7 and ITS rDNA and ancestral state reconstructions, were used to evaluate infra-group divisions and the role of secondary metabolites and selected morphological characters on the taxonomy in the M. prasina group. Two main lineages were found within the group. The Micarea micrococca clade consists of twelve species, including the long-known M. micrococca and the newly described M. microsorediata, M. nigra and M. pauli. Within this clade, most species produce methoxymicareic acid, with the exceptions of M. levicula and M. viridileprosa producing gyrophoric acid. The M. prasina clade includes the newly described M. azorica closely related to M. prasina s.str., M. aeruginoprasina sp. nov. and M. isidioprasina sp. nov. The species within this clade are characterised by the production of micareic acid, with the exception of M. herbarum which lacks any detectable substances and M. subviridescens that produces prasinic acid. Based on our reconstructions, it was concluded that the ancestor of the M. prasina group probably had a thallus consisting of goniocysts, which were lost several times during evolution, while isidia and soredia evolved independently at multiple times. Our research supported the view that the ancestor of M. prasina group did not produce any secondary substances, but they were gained independently in different lineages, such as methoxymicareic acid which is restricted to M. micrococca and allied species or micareic acid present in the M. prasina clade.


Introduction
Traditionally, morpho-anatomical characters, together with secondary metabolites, have played an important role in the lichen classification (e.g. Brodo 1978Brodo , 1986Lumbsch 1998). With the introduction of molecular data, powerful tools for reconstructing phylogenetic relationships have become available. Furthermore, molecular phylogenies can serve as a backbone for tracing the evolution of morphological and chemical characters by reconstructing their ancestral states. Such interpretations of character evolution usually open new perspectives to the evolutionary history .
Secondary metabolites have been traditionally used in the taxonomy of lichens at different taxonomic levels, although their values have been questioned by many authors Leavitt et al. 2011;Lutsak et al. 2017). In many cases, molecular data do not correspond with the chemical variation and, therefore, the correlation between them has to be evaluated for each taxonomic group de novo (e.g. Goffinet and Miadlikowska 1999;Kroken and Taylor 2001;Molina et al. 2004;Divakar et al. 2005Divakar et al. , 2006Elix et al. 2009;Buschbom and Mueller 2006;Nelsen and Gargas 2008;Lendemer et al. 2015;Ossowska et al. 2018). Moreover, the production of certain secondary metabolites might be triggered by the environment (e.g. climate, edaphic factors, associated symbionts) (Spribille et al. 2016;Lutsak et al. 2017).
The genus Micarea Fr., comprising ca. 100 species, is a cosmopolitan group of lichens which has been extensively studied in Europe by Coppins (1983) and Czarnota (2007). Phenotypical diversity in this group of lichens is not limited to morphological characters, but also includes diverse secondary metabolites and, hence, chemical variation plays an important role in their taxonomy. Recently, Micarea has received more attention and numerous species have been described based on anatomical, morphological and chemical characters and, in some cases, also molecular data (e.g. Czarnota and Guzow-Krzemińska 2010;Svensson and Thor 2011;Cáceres et al. 2013;Aptroot and Cáceres 2014;Brand et al. 2014;van den Boom and Ertz 2014;Guzow-Krzemińska et al. 2016;McCarthy and Elix 2016;van den Boom et al. 2017;Kantvilas 2018;Launis et al. 2019a, b).
Species delimitation within Micarea has been especially difficult in the M. prasina group which was first characterised by Coppins (1983) based on morphological, anatomical and chemical features. At first, the group included M. prasina Fr., the type species of the genus, as well as M. hedlundii Coppins and M. levicula (Nyl.). Coppins (1983) also suggested that M. misella (Nyl.) Hedl., M. melanobola (Nyl.) Coppins and M. synotheoides (Nyl.) Coppins might be related to M. prasina; however, as supported by recent molecular studies, M. misella and M. synotheoides do not belong to this group (Czarnota and Guzow-Krzemińska 2010;van den Boom et al. 2017;Launis et al. 2019a). Micarea melanobola was synonymised with M. prasina (Czarnota 2007), but recently found to be a distinct species (Launis et al. 2019b). the role of secondary metabolites for species taxonomy within the M. prasina group were also evaluated. The production of selected secondary metabolites is further analysed (i.e. gyrophoric, methoxymicareic, micareic, prasinic and thiophanic acids), as well as the presence of several pigments in the apothecia commonly used in lichen taxonomy (Meyer and Printzen 2000) (i.e. Sedifolia-grey, Elachista-brown, Cinereorufagreen and Superba-brown). Ancestral state reconstruction of morphological characters i.e. goniocysts, isidia and soredia is also performed.

Materials
Material of the new species, including samples used for DNA analyses, is deposited in KTC, UGDA and LG, with additional specimens stored in private herbaria of van den Boom and Brand.

Morphology and chemistry
Apothecial sections and squashed thallus preparations were studied in tap water with or without the addition of C (commercial bleach) and K (water solution of potassium hydroxide) (Orange et al. 2001). Dimensions of all anatomical features were measured in water. Thin layer chromatography (TLC) was used for the determination of lichen substances according to the standard methods (Orange et al. 2001). All samples were studied in solvent C. The nomenclature of apothecial pigments follows Meyer and Printzen (2000). Crystalline granules were studied in polarised light (see Launis et al. 2019a, b).

Taxon sampling for DNA
A total of 63 new sequences were generated for this study (Suppl. material 2, Table S1). Additional sequences of mtSSU, Mcm7 and ITS rDNA from specimens of the Micarea prasina group were obtained from GenBank (Suppl. material 2, Table S1). Moreover, sequences of the above-mentioned markers from specimens of M. adnata Coppins, M. elachista Table S1), which were shown to be outside the group (e.g. Launis et al. 2019a) were also obtained from GenBank. In total, sequences of 119 specimens were subjected to analyses. Micarea peliocarpa (Anzi) Coppins & R. Sant. was chosen as the outgroup, based on the study of Launis et al. (2019a).

DNA extraction, PCR amplification and DNA sequencing
DNA was extracted directly from pieces of thalli using a modified CTAB method (Guzow-Krzemińska and Węgrzyn 2000). DNA extracts were used for PCR amplification and 25 μl of PCR mix contained 1U of Taq polymerase (Thermo Scientific) or 1U of DreamTaq polymerase (Thermo Scientific) and appropriate buffer, 0.2 mM of each of the four dNTPs, 0.5 μM of each primer and 10-50 ng of genomic DNA. PCR amplifications were performed using a Mastercycler (Eppendorf ).
Amplifications of mtSSU rDNA, employing mrSSU1 and mrSSU3R primers (Zoller et al. 1999), were performed using the following conditions: initial denaturation at 95 °C for 10 min followed by 6 cycles at 95 °C for 1 min, 62 °C for 1 min and 72 °C for 105 s and then 30 cycles at 95 °C for 1 min, 56 °C for 1 min and 72 °C for 1 min, with a final extension step at 72 °C for 10 min.
Amplifications of the Mcm7 region employing Mcm7_AL1r and Mcm7_AL2f primers (Launis et al. 2019a) were performed using the following conditions: initial denaturation at 94 °C for 5 min, followed by 38 cycles at 94 °C for 45 s (denaturation), 56 °C for 50 s (annealing) and 72 °C for 1 min (extension), with the final extension at 72 °C for 5 min.
Amplifications of the ITS region employed the following primer pairs: ITS1F (Gardes and Bruns 1993) and ITS4 (White et al. 1990) or ITS 5 and ITS4A (Kroken and Taylor 2001) or nu-SSU-1626-5' (Gargas and DePriest 1996) and nu-LSU-136-3' (Döring et al. 2000). The following PCR cycling parameters were applied to amplify nuclear ITS region: an initial denaturation at 94 °C for 3 min, followed by 35 cycles at 94 °C for 30 s, 54 °C for 30 s (for ITS1F and ITS4 or nu-SSU-1626-5' and nu-LSU-136-3' primers) or 62 °C for 30 s (for ITS5 and ITS4A primers) and 72 °C for 1 min, with a final extension at 72 °C for 7 min. PCR products were visualised on agarose gels in order to determine DNA fragment lengths. Subsequently, PCR products were purified using Clean-up Concentrator (A&A Biotechnology) following the manufacturer's protocol or 10 μl of PCR products were treated with a mixture of 20 units of Exonuclease I and 2 units of FastAP Thermosensitive Alkaline Phosphatase enzymes (Thermo Scientific) to remove unincorporated primers and nucleotides. Treatment with those enzymes was carried out at 37 °C for 15 min, followed by incubation at 85 °C for 15 min to completely inactivate both enzymes. Sequencing of each PCR product was performed in Macrogen (www.macrogen.com) using the PCR primers.

Sequence alignment and phylogenetic analysis
The newly generated sequences (GenBank accession numbers are given in Suppl. material 2, Table S1) were compared to the sequences available in the GenBank database (http://www.ncbi.nlm.nih.gov/BLAST/) using BLASTn search (Altschul et al. 1990) in order to confirm their identity. The sequences of each marker were aligned with se-quences of selected representatives of the genus Micarea obtained from GenBank (list of specimens and GenBank Accession Numbers are given in Suppl. material 2, Table  S1). Alignment was performed using Seaview software (Galtier et al. 1996;Gouy et al. 2010) employing the Muscle option, followed by manual optimisation. Portions of the alignment with ambiguous positions that might not have been homologous and terminal ends were excluded from the analyses. As the gene trees for each marker did not show any strongly supported conflicts, three datasets were combined into a concatenated matrix in the Seaview software (Galtier et al. 1996;Gouy et al. 2010) and the final alignment was deposited in Treebase (Accession No. S24731).
Partition Finder 2 (Lanfear et al. 2016), implemented at CIPRES Science Gateway (Miller et al. 2010), was used to determine the best substitution model for each partition under Akaike Information Criterion (AIC) and greedy search algorithm (Lanfear et al. 2012). The following models were found: TVM+I+G+X for mtSSU, TRN+I+G+X for Mcm7 and GTR+I+G+X for ITS regions.
The data were analysed using a Bayesian approach (MCMC) in MrBayes 3.2 (Huelsenbeck and Ronquist 2001;Ronquist and Huelsenbeck 2003) and best models determined by Partition Finder 2 were employed. Two parallel runs were performed, each using four independent chains and 10 million generations, sampling trees every 1,000 th generation. Tracer v. 1.5 (Rambaut and Drummond 2007) was used to ensure that stationarity was reached by plotting the log-likelihood values of the sample points against generation time. Posterior probabilities (PP) were determined by calculating a majority-rule consensus tree generated from the 15,002 post-burn-in trees of the 20,002 trees sampled by the two MCMC runs, using the sumt option of MrBayes.
Maximum likelihood analyses were performed using RaxML HPC v.8 on XSEDE (Stamatakis 2014) under the GTRGAMMAI model on CIPRES Science Gateway (Miller et al. 2010). Rapid bootstrap analyses were performed with 1,000 bootstrap replicates (BS). The RAxML tree did not contradict the Bayesian tree topology for the strongly supported branches. Therefore, only the maximum likelihood tree is shown with the posterior probabilities (PP) of the Bayesian analysis and the bootstrap support values added near the internal branches. BS ≥ 70 and PP ≥ 0.95 were considered significant. Phylogenetic trees were visualised using FigTree v. 1.4.2, in which the clades for previously described taxa are collapsed (Rambaut 2012).

Ancestral character state reconstruction
Morphological and chemical characters from taxa of the Micarea prasina group and selected outgroup taxa were obtained from herbarium material and complemented with data from literature. In order to reduce the number of missing data in our dataset, we did not include M. pycnidiophora, M. stipitata and M. synotheoides, which do not belong to the M. prasina group and for which mtSSU sequences were only available and Micarea sp. lineage A, which represents a single specimen that has not been formally described. The following secondary metabolites were analysed: gyrophoric, methoxymicareic, micareic, prasinic and thiophanic acids. The presence of apothecial pigmentation was also analysed and the following pigments were noted: Sedifoliagrey, Elachista-brown, Cinereorufa-green and Superba-brown. The presence of selected morphological characters was also analysed, i.e. goniocysts, isidia and soredia. The morphological and chemical characters were coded as a multistate data matrix (Suppl. material 2, Table S2) and a binary dataset (Suppl. material 2, Table S3) and subjected to ancestral character state reconstruction using the parsimony model with characters treated as unordered and the likelihood method (Mk1 model) in Mesquite v.3.5 (Maddison and Maddison 2018). Ancestral state reconstructions were based on the topology of the consensus tree obtained using Mr Bayes 3.2 (Huelsenbeck and Ronquist 2001;Ronquist and Huelsenbeck 2003)

Results
The final DNA alignment consisted of sequences obtained from 119 individual specimens and three markers, i.e. mtSSU, Mcm7 and ITS rDNA, with a total of 1784 characters. Since the topologies from the maximum likelihood and Bayesian analyses did not show any strongly supported conflict, the maximum likelihood tree (RaxML Optimisation Likelihood was -14426.795913) is presented in Figure 1 with added posteriori probabilities from Bayesian analysis (Harmonic mean was -13101.16). In order to reduce the size of the tree, highly supported clades were collapsed for previously described taxa.
The phylogenetic reconstruction ( Fig. 1) shows that the M. prasina group is highly supported and monophyletic (100 BS and 1 PP) and it agrees with previous phylogenies based on a mtSSU marker (e.g. Czarnota and Guzow-Krzemińska 2010;Guzow-Krzemińska et al. 2016) or three loci (Launis et al. 2019a). Two main lineages are further distinguished, i.e. the M. micrococca clade and the M. prasina clade with sequences of M. tomentosa forming a highly supported lineage, basal to the two clades ( Fig. 1). Moreover, M. hedlundii and M. xanthonica are closely related (82 BS) and sister to the M. micrococca clade (Fig. 1).
The Micarea micrococca clade in Figure 1   supported group with the type species of M. prasina s.str. (100 BS and 1 PP), whereas specimens of M. prasina form a well-supported group (87/0.99). Furthermore, they are sister to M. nowakii and M. herbarum, which are the only species within the M. prasina group developing almost entirely an endosubstratal thallus with only a few areoles. With the exception of M. herbarum and M. nowakii, this lineage (97 BS and 1 PP) also includes a sequence which seems to be different from both species (EF453665) and may indicate the existence of an undescribed taxon. Specimens of the newly described M. isidioprasina form a highly supported group (100 BS and 1 PP) with a single sequence from North America originally assigned to M. prasina (AY756452; see Andersen and Ekman 2005), but genetically more similar to M. isidioprasina. This sample is also morphologically similar to M. isidioprasina due to the isidioid thallus and pale apothecia (Czarnota and Guzow-Krzemińska 2010) and, therefore, is named here as M. cf. isidioprasina in Figure 1. Micarea meridionalis, M. soralifera and M. subviridescens form a highly supported group (80 BS and 1PP).
To investigate the diagnostic traits traditionally used for the taxonomic classification within the M. prasina group, we focused both on the M. micrococca and the M. prasina clades separately and the whole M. prasina together ( Fig. 1) and employed both maximum parsimony and Mk1 models, based on the multistate and binary datasets (Suppl. material 2, Tables S2-S3 and Suppl. material 1, Figs S1-S15). The likelihoods for each set of characters are given in Suppl. material 2, Table S4. Our analyses found that the presence of methoxymicareic acid is restricted to the M. micrococca clade that accommodates several species containing this substance. However, M. levicula and M. viridileprosa are exceptions by producing gyrophoric acid (Suppl. material 1, Fig. S13). The ancestral state reconstructions show that the presence of methoxymicareic acid is the most parsimonious and the most likely ancestral state for the M. micrococca clade (Fig. 1, Tables 1-3 and Suppl. material 1, Figs S3, S13). On the other hand, micareic acid is the ancestral state for M. prasina clade in all analyses ( Fig. 1, Tables 1-3, Suppl. material 1, Figs S2, S13). However, the reconstructions of ancestral state for the whole M. prasina group show the lack of any secondary metabolites in their ancestors in most of the analyses. However, the maximum likelihood analysis, based on the multistate dataset, suggests uncertainty as both the lack of any secondary metabolites and the presence of micareic acid are more likely than other states ( Fig. 1, Tables 1-3 and Suppl. material 1, Figs S2, S13).
The evolution of pigments, present in the apothecia, was also analysed, but some of the results remain uncertain in our analyses. Parsimony reconstructions, based on the binary dataset, suggest the lack of any pigment in the apothecia, while other analyses do not exclude the possibility that Sedifolia-grey pigment was present in the ancestor of M. prasina group (Fig. 1, Tables 1-3, Suppl. material 1, Figs S9-S12, S14). Moreover, the results obtained for the M. micrococca clade, using two different methods, are not fully consistent. Maximum parsimony analyses suggest a lack of pigments in their ancestors therefore resulting in multiple gains of Sedifolia-grey pigment and a single gain of Cinereorufa-green pigments in this lineage. However, maximum likelihood analyses show that both the lack of pigments in apothecia and the presence of Sedifolia-grey pigment may have occurred in their ancestor ( Fig. 1, Tables 1-3, Suppl.  material 1, Figs S9-S12, S14). In case of the M. prasina clade, maximum likelihood analyses, based on the binary dataset, give uncertain results as both presence and absence of Sedifolia-grey are equally likely; however parsimony analysis for the binary dataset and both analyses for the multistate dataset show the presence of Sedifolia-grey pigment in apothecia of their ancestor.

Discussion
Challenges in species delimitation within M. prasina group were already mentioned by Coppins (1983) and other authors (e.g. Czarnota 2007;Czarnota and Guzow-Krzemińska 2010;van den Boom et al. 2017;Launis et al. 2019a, b). Since Coppins (1983), who treated M. prasina in a wide sense with morphologically variable chemical races which were further recognised as distinct species, the introduction of molecular data revealed even greater variability within this group and numerous other species were described based on phenotypic and molecular data (e.g. Czarnota and Guzow-Krzemińska 2010;Guzow-Krzemińska et al. 2016;van den Boom et al. 2017;Launis et al. 2019a, b). Many species within this group have goniocystoid thallus, micareoid photobiont and Sedifolia-grey pigment in the apothecia, however a high variation in secondary metabolites production, which are treated as diagnostic characters, is observed within the M. prasina group. In the phylogenetic tree ( Fig. 1) two main clades were distinguished; M. micrococca clade which groups mainly taxa containing methoxymicareic acid and M. prasina clade which mainly comprises species containing micareic acid. However, there are some exceptions as other substances may be produced by selected representatives of the group, e.g. gyrophoric, prasinic or thiophanic acids or some taxa do not produce any secondary metabolites. Within this group, numerous phenotypic differences are applied to distinguish species, e.g. size and shape of apothecia, size and type of paraphyses, size of ascospores, thallus structure including the vegetative diaspores and presence of pigments. Recently introduced crystalline granules showed to be valuable traits in the taxonomy of the group (Launis et al. 2019a, b). However, the application of molecular data seems to be essential to support delimitation of species within this group (e.g. Launis et al. 2019a, b; this study). The evolution of new morphological characters involves multiple subsequent evolutionary steps. In our study, ancestral state reconstructions showed that the presence of goniocysts is the most parsimonious and most likely state for the ancestor of the M. prasina group (Fig. 1, Tables 1-3 and Suppl. material 1, Figs S6, S15). However, the development of goniocysts was apparently lost in some lineages during evolution as several species within the group do not develop such structures but produce other vegetative diaspores (soredia and/or isidia). Whether the structures from which soredia and isidia develop are goniocysts or areoles is not easy to assign. Based on literature, goniocysts are more or less round vegetative diaspores (therefore similar to soredia) and are produced from the endosubstratal parts of thalli multiple times to form a layer as in M. prasina s.str. (Coppins 1983;Barton and Lendemer 2014). As the thallus parts developing isidioid or soredioid diaspores did not resemble goniocysts as defined in previous works, we determined all these structures as areoles, as already proposed by Guzow-Krzemińska et al. (2016). Although soredia in the newly described M. microsorediata and recently recognised M. soralifera (Guzow-Krzemińska et al. 2016) may resemble goniocysts, they are at least at the beginning produced in delimited soralia over the thallus and differ in the structure and colour from the non-sorediate parts of thalli.
In our study, ancestral state reconstructions suggest that isidia evolved independently multiple times in this group of lichens resulting in the formation of almost entirely isidiate thalli in four species, i.e. M. aeruginoprasina, M. isidioprasina, M. nigra and M. pauli (Suppl. material 1, Figs S6-S8, S15). Prieto et al. (2013) suggested that losing an existing character could be expected to occur much more rapidly and in fewer steps than gaining a new character. A similar case is represented by sorediate species and the production of soredia developed in unrelated lineages. Only one lineage lost the ability to produce goniocysts or any other lichenised vegetative diaspores (i.e. M. herbarum and M. nowakii). Species belonging to this clade develop thin episubstratal thalli with few areoles or merely an endosubstratal layer (Czarnota 2007;van den Boom et al. 2017). The acquisition of different thallus organisation may have resulted from adaptation to drier ecological niches. Many collections of the species from this clade were found in drier and open habitats (Czarnota 2007;van den Boom et al. 2017). In comparison, taxa developing distinct episubstratal thalli seem to be confined to more humid and shaded localities (Czarnota 2007). However, this hypothesis needs further ecological studies.
Secondary metabolites have been extensively used in the chemotaxonomy of lichens. The Micarea prasina group shows a high variation in chemistry even in closely related species (e.g. Czarnota 2007;Czarnota and Guzow-Krzemińska 2010). Species belonging to this group produce gyrophoric, micareic, methoxymicareic and prasinic acids, as well as xanthones (Elix et al. 1984;Coppins and Tønsberg 2001;van den Boom and Coppins 2001). Gyrophoric acid is the simplest tridepside comprising three orsellinic units which originate from condensation of one acetyl-CoA and three malonyl-CoA units as shown by Mosbach (1964). Although gyrophoric acid is commonly produced in the genus Micarea (e.g. Coppins 1983;Czarnota 2007) Micareic and methoxymicareic acids are the most common secondary metabolites produced by species of the M. prasina group. They are structurally related diphenyl ethers ('pseudodepsidones') (Huneck and Yoshimura 1996), but they have a distincly different substitution pattern and probably also biosynthetic origin (Elix et al. 1984). As numerous diphenyl ethers co-occur with structurally related depsidones, it was hypothesised that they are biosynthesis precursors or catabolites of similarly substituted depsidones (Huneck and Yoshimura 1996). In the work on the secondary metabolites of chemical races of the M. prasina s.l., Elix et al. (1984) suggested that enzymatically induced Smiles rearrangement of para-depside prasinic acid might lead to the formation of micareic acid, a very likely biosynthetic pathway for this metabolite. They also pointed out that other rearrangements, such as nuclear hydroxylation followed by O-methylation, are necessary for the formation of methoxymicareic acid, but the actual order of those processes remain unknown. However, the chemical races of M. prasina s.l. they studied actually represent several species which were later distinguished as M. micrococca (methoxymicareic acid chemotype), M. prasina s.str. (micareic acid chemotype) and M. subviridescens (prasinic acid chemotype) (Coppins 2009); furthermore, other new species have also been recognised within the M. prasina group. Both micareic and methoxymicareic acids are produced by several species within the M. prasina group, while prasinic acid has only been reported from M. subviridescens. So far, no co-occurrence of any of those substances has been observed in any species within the M. prasina group.
Reconstructions of the ancestral state for the whole M. prasina group suggest that the most recent common ancestor did not produce any secondary metabolites. This may suggest that the production of a wide range of secondary metabolites in this group of lichens could have resulted from independent gains of ability to biosynthesise various substances during evolution. The scenario, in which the ability to produce micareic acid in the ancestor of M. prasina clade or methoxymicareic acid in the ancestor of M. micrococca clade being gained only once during evolution, seems to be reasonable since losing an existing character could be expected to occur more rapidly and in fewer steps than gaining a new character (e.g. Prieto et al. 2013). Those evolutionary events could have been followed with the loss of those traits in some lineages and successive independent gains of ability to biosynthesise prasinic (M. subviridescens) or gyrophoric acids (M. levicula and M. viridileprosa) in some species.
To summarise, our study showed that phenotypical variation within the Micarea prasina group has been previously underestimated and, based on field work and laboratory studies, six new species within this group are described (see Taxonomy).

Habitat and distribution.
In the type locality Micarea aeruginoprasina grows abundantly on trunks of Juniperus brevifolia, in a subnatural degradated forest, dominated by J. brevifolia shrubs and trees. In other localities, it was found on Cryptomeria and Erica trunks, also in forested areas.
The new species is only known from the island Terceira in the Azores, where it is known from several localities.
Etymology. The epithet refers to the often aeruginose colour of the apothecia and the resemblance in secondary chemistry to M. prasina.
Habitat and distribution. To date, known only from the Azores archipelago (Terceira island) from three localities where it was found on bark of trees.
Etymology Notes. The new species is resolved as sister to M. prasina s.str. with strong support, being morphologically and chemically similar to that species, but differing in the absence of the Sedifolia-grey pigment, responsible for the typical reaction K+ violet in M. prasina s.str (Coppins 1983;Czarnota 2007;Launis et al. 2019a). Instead of Sedifoliagrey pigment, Superba-brown is present in M. azorica.
The identity of M. prasina s.str. has been recently solved by Launis et al. (2019a, b) and its occurrence is confirmed from boreal and temperate Europe (Finland, Germany, Poland) and Eastern North America (Canada: New Brunswick and USA: Maine) (Launis et al. 2019b;this paper). Other records need confirmation as, previously, other species have been included in the variation of M. prasina.
Habitat and distribution. The species grows on wood (decomposing logs) and acidic bark of trees in various forest communities in well preserved forest.
To date, it is known from Belgium, Germany, France, Poland and Romania. Notes. Micarea isidioprasina is an isidiate species of the M. prasina group containing micareic acid as the main secondary metabolite. It is usually sterile and in Poland often grows in similar habitats with M. pauli, a species described in this paper, from which it can be separated with certainty by analyses of secondary metabolites, as the latter contains methoxymicareic acid.
Micarea aeruginoprasina and M. nigra also develop similar isidiate thalli, but M. aeruginoprasina has pale cream to pale brown or aeruginose apothecia (often mottled with all colours in the same apothecium) and M. nigra develops dark greyish to black apothecia. When sterile, all three species may be more difficult to separate, especially M. aeruginoprasina which also produces micareic acid (M. nigra contains methoxymicareic acid), but that species has pale brown isidia. Additionally, the so far known distributions of all three species do not overlap and M. aeruginoprasina and M. nigra are known from the Azores and continental Portugal, respectively.
Habitat and distribution. The new species occurs on acidic bark of various trees such as Alnus, Betula, Fagus and Quercus, usually in humid forests, also on decaying wood (logs and stumps) and rarely on terrestrial decaying mosses in, for example, steep slopes in heath and dunes. It is a very common species in the south of the Netherlands and some areas in Poland and is mostly found on microhabitats where only few other lichens species co-occur. On several occasions, Normandina pulchella (Borrer) Nyl. and squamules of Cladonia spp. are the only accompanying lichens.
To date, the species has been found in Belgium, Germany, the Netherlands, Poland and Portugal.
One specimen of Micarea microsorediata was invaded by Nectriopsis micareae Diederich, van den Boom & Ernst (see below additional specimens examined).
Habitat and distribution. Abundantly present on a trunk of a fern tree in a parkland where many tropical and exotic fern and tree species have been introduced.
To date, it is only known from the type locality in Portugal (Sintra). Etymology. The epithet chosen for this species refers to its very dark appearance, the thallus being dark greenish and the apothecia mostly blackish.
Notes. This species is resolved in the M. micrococca group (Fig. 1) and is unique because of its dark grey to almost black apothecia and the presence of Cinereorufagreen pigment in epihymenium.
Micarea nigra resembles M. aeruginoprasina, M. isidioprasina and M. pauli. Micarea aeruginoprasina and M. isidioprasina differ in the presence of micareic acid instead of methoxymicareic acid and paler apothecia. In addition, M. aeruginoprasina produces different pigment in the apothecia (Sedifolia-grey). Micarea pauli differs in the production of methoxymicareic acid, Sedifolia-grey pigment in the apothecia and different distribution (see under that species).
Habitat and distribution. This species is so far known only in Poland from Białowieża Forest, where it grows in deciduous forests on bark of Alnus glutinosa (5 specimens), Tilia cordata (1 specimen) and on wood (2 specimens).
Etymology. The species is named after our friend, Paweł Czarnota, specialist in the genus who monographed it in Poland. Notes. Micarea pauli is an isidiate species with Sedifolia-grey pigment in its apothecia. It can be separated from the similar M. isidioprasina, with which it grows in Białowieża Forest, by the presence of methoxymicareic acid.
Micarea aeruginoprasina and M. nigra are also similar in thallus morphology, but they differ in the pigmentation of apothecia. Micarea aeruginoprasina develops pale cream to pale brown or aeruginose apothecia, which are often mottled in colour in one apothecium, whereas in M. nigra the apothecia are dark greyish to black. Without apothecia, they can be difficult to separate from M. pauli, especially M. nigra which also contains methoxymicareic acid (M. aeruginoprasina produces micareic acid), but so far, M. aeruginoprasina and M. nigra are only known from the Azores and continental Portugal, respectively.