Quantitative integrative taxonomy informs species delimitation in Teloschistaceae (lichenized Ascomycota): the genus Wetmoreana as a case study

The genus Wetmoreana was studied using quantitative integrative taxonomy methods to resolve the genus delimitation and explore its taxonomy diversity at the species level. As a result, the genus Fulgogasparrea is synonymized with Wetmoreana, and the latter includes 15 formally described species, one subspecies, and three further, thus far undescribed species: W. appressa, W. awasthii comb. nov., W. bahiensis sp. nov., W. brachyloba comb. nov., W. brouardii, W. chapadensis comb. nov., W. circumlobata sp. nov., W. decipioides, W. intensa comb. nov., W. ochraceofulva comb. nov., W. rubra sp. nov., W. sliwae sp. nov., W. sliwae ssp. subparviloba subsp. nov., W. subnitida comb. nov., W. texana, and W. variegata sp. nov. Eleven of 19 examined taxa are newly placed within this genus or confirmed to belong to it. Two species, W. awasthii and W. intensa, are transferred to Wetmoreana without additional analysis but based on previous studies. The W. brouardii and W. ochraceofulva species complexes are discussed in detail. Additionally, Caloplaca muelleri and C. rubina var. evolutior are transferred to Squamulea, and the latter is elevated to the species rank. Supplementary Information The online version contains supplementary material available at 10.1186/s43008-024-00140-1.


Introduction
Teloschistaceae is one of the families of lichenized Ascomycota where internal genus-level classification is most contentious, although, with the exception of some questionable studies based on demonstrably chimeric data, subfamily divisions are stable (Arup et al. 2013;Bungartz et al. 2020;Wilk et al. 2021).Many new genera have been described in recent years, with the total reaching more than 90 (71 listed in Wijayawardene et al. 2022, with a further 23 treated as synonyms or possible synonyms of accepted genera), including about one thousand known species (Arup et al. 2013), resulting in a ratio of 11 species per genus.In more speciose families, the number of accepted genera is comparatively lower: 2765 species and 77 genera in Parmeliaceae (36 species per genus), 2161 species and 79 genera in Graphidaceae (27 species per genus), and 943 species and 43 genera in Verrucariaceae (22 species per genus) (Lücking et al. 2017).
Genus-level classification in Teloschistaceae is particularly challenging due to the lack of clear correlations of potentially diagnostic characters with phylogenetically defined clades, along with a high level of morphological homoplasy, particularly across lineages containing crustose taxa formerly placed in the collective genus Caloplaca (Gaya et al. 2008;Arup et al. 2013;Bungartz et al. 2020).Consequently, accurate placement of taxa at generic level without molecular data is difficult or often impossible (e.g., taxa with the 'Flavoplaca citrina morphology'; Vondrák et al. 2009;Wilk et al. 2021).It is, however, unclear whether such diagnostic characters are indeed absent or have not been properly assessed using quantitative approaches (Vondrák et al. 2013).Although morphology is often integrated into phylogenetic revisions at the family level, e.g., in Collemataceae (Otálora et al. 2013), Graphidaceae (Rivas Plata et al. 2012a), Pannariaceae (Ekman et al. 2014), Parmeliaceae (Thell et al. 2012), or Ramalinaceae (Kistenich et al. 2018), morphological data are less frequently quantitatively assessed in the form of a matrix or using statistical methods (e.g., Kistenich et al. 2018).
Some exceptions are found in cladistic approaches to assess morphological data, e.g., in the order Arthoniales (Tehler 1990), the families Gomphillaceae (Lücking et al. 2005) or Parmeliaceae (Saag and Randlane 1995;Saag et al. 2002), or the genus Diploschistes Norman in Graphidaceae (Lumbsch and Tehler 1998).When complementing molecular studies, the power of quantitative approaches to properly delimit and define taxa becomes apparent, e.g., in the Graphidaceae (Parnmen et al. 2012;Rivas Plata et al. 2012a;Lücking et al. 2015;Lücking and Kalb 2018) or in the Arthoniales (Perlmutter et al. 2020).Crustose Teloschistaceae have been subjected to quantitative phenotypic analysis within a phylogenetic framework (Frolov et al. 2016), using a comprehensive character and data matrix established by Vondrák et al. (2013); however, the applied technique has been limited to morphometric (continuous) data.
Here, we use the recently established genus Wetmoreana as an example to test the combination of phenotype-based phylogenetic binning (PBPB) and multiple response permutation procedure (MRPP) to assess species and genus boundaries and to clarify the nomenclature within this complex.Wetmoreana belongs to subfamily Teloschistoideae, with W. texana as its type (Arup et al. 2013).Almost at the same time when the genus was established, Kondratyuk et al. (2013) considered it polyphyletic and proposed its division into Fulgogasparrea and Wetmoreana s.str.However, that division has not been broadly accepted (Lücking et al. 2017;Bungartz et al. 2020;Wilk et al. 2021;Wijayawardene et al. 2022).Wetmoreana sensu Arup et al. (2013) is a relatively small group consisting of four accepted species, besides W. texana also W. appressa, W. brouardii, and W. decipioides, plus several additions proposed in the present study.The relocation of Xanthoria tenax within Wetmoreana by Kondratyuk et al. (2013) was based on misidentified material representing an undescribed species of Wetmoreana (Kondratyuk et al. 2020).Kondratyuk et al. (2013Kondratyuk et al. ( , 2015) ) circumscribed Fulgogasparrea as a genus including three of the four species originally assigned to Wetmoreana, keeping the latter restricted to its type, W. texana, and later adding also F. awasthii (Mishra et al. 2020).A further species, F. intensa, was recently described from Brazil (Aptroot et al. 2021).Species of Wetmoreana (including Fulgogasparrea) form marginally lobate (placodioid) or squamulose thalli with anthraquinones and often produce vegetative propagules, such as isidia, papillae, schizidia or soredia.Apothecia are less common or unknown, e.g. in W. decipioides (Arup et al. 2013).The species are exclusively saxicolous, occurring mostly on siliceous rocks.Most Wetmoreana species, including those added in the present study, occur in South America, a few in North America (U.S.A., Mexico), and a one each in Africa and Asia (Arabian Peninsula and South Korea) (Arup et al. 2013;this study).The origin of the genus is unknown, but the highest diversity occurs in South America and may indicate this continent as the centre of its diversification.
The objective of the present study included: (i) assess specific and intraspecific delimitation of Wetmoreana species based on phenotype and DNA-based phylogenetic reconstructions and multi-response permutation procedure (MRPP), (ii) evaluate potentially diagnostic characters for Wetmoreana at the genus level using phenotype-based phylogenetic binning (PBPB), (iii) assess the validity of the separation of Fulgogasparrea through molecular data and MRPP, (iv) assess the potential phylogenetic position of several South American putative Wetmoreana species for which no sequence data are unavailable by PBPB, and (v) analyze phenotypic variation within the newly recognized W. ochraceofulva/ W. variegata complex in Africa, the Arabian Peninsula and South America using MRPP.Based on our results, the genus Fulgogasparrea is synonymized with Wetmoreana, six new taxa are established, namely W. bahiensis sp.nov., W. circumlobata sp.nov., W. rubra sp.nov., W. sliwae sp.nov., W. sliwae subsp.subparviloba subsp.nov., and W. variegata sp.nov., and further six species are transferred to Wetmoreana: Caloplaca brachyloba, C. chapadensis, C. ochraceofulva, C. subnitida, F. awasthii, and F. intensa.We further transfer Caloplaca muelleri and C. rubina var.evolutior to Squamulea, and C. rubina var.evolutior is elevated to species rank.Two larger subclades within Wetmoreana are recognized: the W. brouardii clade, including W. appressa, W. bahiensis sp.nov., W. brouardii, W. intensa comb. nov. and W. subnitida comb. nov., and the W. ochraceofulva clade, including W. ochraceofulva comb. nov. and W. variegata sp. nov.

Taxon sampling
The study was based on the herbarium material from B, BCN, BM, C, E, F, G, GZU, KoLRI, KRAM, LPB, LD, M, MIN, S, UPS, and the private herbarium of U. Arup.The material originates mainly from South America, in some cases from other continents, including also types and original collections.The studied samples were represented by marginally lobate, squamulose or rarely areolate species of Wetmoreana and the following morphologically similar genera: Aridoplaca, Calogaya, Cinnabaria, Gyalolechia, Squamulea, Teuvoahtiana, and selected representatives of Caloplaca s. lat.representing all three subfamilies within Teloschistaceae.Among the studied material, we included poorly known neotropical Caloplaca species (e.g., C. brachyloba, C. chapadensis, C. muelleri and C. rubina), which were revised using type or original material.A large collection of W. ochraceofulva from Africa, the Arabian Peninsula and South America (BM, E, G, KRAM, LD, and MIN) was studied in detail.

Phenotype assessment
Morphological characters were studied on herbarium material using a Nikon SMZ 1270 dissecting microscope (Tokyo, Japan).Anatomical details of thalli, apothecia and pycnidia were examined using hand-cut sections mounted in water under a Nikon Eclipse 50i compound microscope.Ascospore sizes are given as an average with standard deviation and extremes are given in brackets, with the number of measurements (n) and specimens examined (N) also indicated in brackets.Spores were measured without preheating the slides (a method used in Wetmore 2003).Conidia were measured based on calibrated photographs of preparations.In case of lobate species, only terminal lobes were measured, and lobe length refers to the distance from the tip of the lobe back to the first branch (Wetmore and Kärnefelt 1998).The widths of lobes were measured below their tips (cf.Brodo et al. 2001: Fig. 3).Apothecial margins were measured according to Vondrák et al. (2013: Fig. 3).Apothecium types follow Clauzade and Roux (1985: Fig. 61), where 1-2 refer to immersed apothecia, 3 to erumpent apothecia, and 4-5 to sessile apothecia.The thalline margin in zeorine apothecia was classified as ± persistent, partly reduced, or much reduced (Fig. 1).The presence of crystals on anatomical structures was observed under polarized light and the solubility of crystals were determined using 25% KOH (K), 65% nitric acid (N), 5% acetic acid (abbreviated as AA), and 2% hydrochloric acid (HCl).Calcium oxalate crystals (CaOx crystals) are defined as insoluble in K, soluble in N, insoluble in AA and soluble in HCl without giving off gas (Yasue 1969;Wilk et al., in prep.).To study the occurrence of crystals within the thallus, especially in W. ochraceofulva and similar species, cross sections of thalline lobes were prepared by selecting those which overlap other lichen thalli where possible, which allowed to observe the delimited layer of CaOx crystals.Hydrochloric acid was used to test for the presence of calcium carbonate (CaCO 3 ) in rock substrata.Terminology of thallus and apothecial structures follows Bungartz (2002) and Ryan et al. (2002Ryan et al. ( , 2012)).Photographic documentation was made with a Nikon DS-Fi2 digital camera combined with the imaging software NIS-Elements D 4.30 (Nikon Corporation, Japan).All specimens examined were photographed and the photo documentation was used to compare species and obtain dimensions of structures that were difficult to measure under the microscope, such as conidia.
In total, 112 specimens were included in the phenotype data matrix, representing 31 taxa (Additional file 1: Table S1).For 39 specimens the ITS sequences are available, which constitute 35% of all studied specimens.These specimens provide the basis for the identification of the reference taxa and were used to generate the 'molecular reference tree' for the phenotype-based phylogenetic binning analysis.The remaining 73 specimens (65%), for which ITS sequences are unknown, constitute 'taxon queries' .The latter group includes among others the poorly known, mostly neotropical species, such as Blastenia fernandeziana f. validior, Caloplaca brachyloba, C. chapadensis, C. muelleri, C. rubina var. evolutior, and C. xanthobola.They were chosen for comparison with representatives of Wetmoreana due to their morphological similarities to Wetmoreana spp., and were considered to probably belong to this genus.The 73 phenotypic characters, which comprehensively characterizes the studied group, were used in analyses, including 59 qualitative and 14 morphometric characters.Among them, 32 characters belong to thallus morphology; 2 to thallus anatomy; 13 to apothecia morphology; 9 to apothecia anatomy; 2 to pycnidia morphology; and 6 to conidia anatomy; 6 to CaOx crystals occurrences; and 1 to habitat preferences.The binary and ordered multistate characters were evaluated, measured and coded (Additional file 2: Table S2).
The "query taxa" represented Teloschistaceae which morphologically, anatomically, and geographically relate to the study fresh material from South America and include putative undescribed Wetmoreana species.They also include some poorly examined tropical Teloschistaceae for which DNA sequences are unavailable.The chosen reference taxa encompass species for which DNA sequences are available and which have morphological and anatomical resemblance with Wetmoreana, e.g.Aridoplaca, Calogaya, Cinnabaria, Gyalolechia, Squamulea, and Teuvoahtiana.Other placodioid or squamulose genera, such as Austroplaca, Follmannia, Gondwania, Orientophila, Polycauliona, Scutaria, Sirenophila, Xanthopeltis, where not considered, as they deviate in important features from Wetmoreana and geographic distribution (e.g., endemic to Australasia) and/or ecology (e.g., restricted to seashore rocks).

Molecular phylogenetic analyses
The monophyly of the Wetmoreana genus was demonstrated in previous papers (Arup et al. 2013;Søchting et al. 2014;Wilk et al. 2021).To assess phylogenetic relationships between of all available Wetmoreana spp., we first used a three-marker data set (ITS, nuLSU, mtSSU), for which additional DNA sequences were downloaded from GenBank (Additional file 1: Table S1).Teloschistes flavicans was chosen as an outgroup because it is not closely related to Wetmoreana and is located in a sister clade to that genus (Arup et al. 2013: Fig. 1).Sequences of two specimens of W. ochraceofulva from Africa and Saudi Arabia were newly generated following the methods outlined in Wilk et al. (2021) (Additional file 1: Table S1), assembled and edited using CodonCode Aligner 6.0.2, and subjected to BLAST queries for an initial verification of their identities.Alignments of the three different markers were assembled using MAFFT v. 7 (Katoh et al. 2019), and then corrected manually in AliView 1.17.1 (Larsson 2014).They were first analysed separately to check for potential incongruence.The one incongruence was present in the ITS analysis in relation to the other two (see chapter Results for details).The concatenated alignment of the three markers was assembled, with 34 terminals and a length of 2980 bases (ITS: 623, nuLSU: 1476, mtSSU: 881).The alignment was subjected to maximum likelihood (ML) analyses in RAxML v. 8.2.0 (Stamatakis 2014) under the GTR-Gamma model.Given the low number of terminals and the high number of characters, no partitioning scheme was applied, as it would not result in any notable differences.Branch support was assessed by non-parametric bootstrapping with 100 replicates.Phylogenetic trees were visualized and edited in FigTree v. 1.4.4 (Rambaut 2018).Nodes with bootstrap values ≥ 70 were considered as supported.
As a second step, phylogenetic relationships of Wetmoreana with the morphologically similar genera Aridoplaca, Calogaya, Cinnabaria, Gyalolechia, Squamulea, Teuvoahtiana, and the species Caloplaca fernandeziana, were assessed based on the ITS marker, following the methods outlined above, with the corresponding additional ITS sequences downloaded from GenBank (Additional file 1: Table S1).The resulting ITS alignment included 39 terminals and was 650 bases long.Given previously published phylogenies, the topology was internally rooted with subfamily Xanthorioideae, defining the clade formed by Teuvoahtiana altoandina, Squamulea spp., and Calogaya spp. as outgroup.The resulting phylogenetic tree was consistent with previously published phylogenies (e.g., Arup et al. 2013) and was used as the molecular reference tree for the downstream PBPB analysis.
The alignments for all phylogenetic analyses performed are available in Additional files 3-6: FILES S1-S4.

Phenotype-based phylogenetic binning (PBPB)
PBPB was carried out following Berger et al. (2011) and Lücking et al. (2015).The procedure was performed in three steps: (i) computing the molecular reference trees (ITS rDNA marker) for the terminals for which sequence data were available; (ii) performing weight calibration for the phenotype characters of these same terminals, using maximum likelihood (ML) and maximum parsimony (MP); and (iii) performing the binning analysis using the Evolutionary Placement Alogarithm (EPA), under both parsimony and maximum likelihood weights for the remaining terminals for which no DNA sequences where available.All steps were performed in RAxML.During the binning procedure, alternative placement of each terminal based on its phenotype characters were being assessed through non-parametric bootstrapping using 100 replicates.The resulting classification trees were visualized using FigTree, with some file modifications following Lücking et al. (2015), and the classification tables were edited for display in Microsoft Excel.Results were summarized as cartoon trees.
The final phenotype data matrix was used to run PCA ordinations (Principal Component Analysis) to test how well the separation of clades defined by molecular phylogeny and morphological binning corresponded to discrimination of the included species by means of morphological data.PCA analysis was performed in PC-ORD v. 5.03 (McCune and Mefford 2011;McCune et al. 2002) using 108 specimens of all studied taxa of Wetmoreana and other genera, and separately using 73 specimens of Wetmoreana exclusively.The morphological data included 71 and 63 variables respectively; missing data, mostly of apothecia, were interpolated, and Callopisma subnitidum (syntype, PBPB no.49), Calogaya biatorina (PBPB no.127) and W. decipioides (PBPB no.70 and 122), were not included in the PCA analyses because of missing data of apothecia/pycnidia.PCA was done using correlation to compute the cross-products matrix, and a randomization test was performed using 999 replicates.

Multi-response permutation procedure (MRPP)
MRPP was performed in PC-Ord 5.03 following Parnmen et al. (2012), with the objective to test for the statistically significant phenotypic differences between predefined groups (clades).The groups constituted our initial taxon delimitation hypotheses from the previous binning analysis.The clades tested were of two types, namely the clades defined by molecular data only, and the extended clades produced by adding the terminals binned to each molecular clade with support, according to ML and MP weighting techniques separately.At first, we focused on the phylogenetically defined complexes comparison, to decide how many species to distinguish within.Two complexes: ochraceofulva/ variegata (= W. ochraceofulva clade) and sliwae/ subparviloba/ brachyloba (= W. sliwae clade), corresponding to ML and MP classification trees, were analysed (Fig. 2, Additional file 7: Fig. S1).The test was performed for complete and incomplete datasets; the latter were devoid the apothecial characters which were missing in some specimens, especially in case of W. ochraceofulva clade, but also in the case of W. texana and W. brouardii.For the complete dataset, missing data were interpolated from other specimens, the type material when possible.However, in case of W. decipioides and Callopisma subnitidum (syntype, PBPB no.49), none of the specimens studied produced apothecia, and these species were analysed only in case of the incomplete dataset.
In case of the cross-species comparison three grouping scenarios were employed: (i) W. ochraceofulva vs. W. variegata, (ii) W. sliwae vs. W. sliwae ssp.subparviloba, and (iii) W. sliwae vs. W. brachyloba.In case of the W. ochraceofulva clade, both dataset types (with and without apothecia) were used.In the last comparison, W. sliwae s. lat.and W. sliwae s.str.[without W. sliwae ssp.subparviloba and C. rubina var.evolutior no.96 (syntype)] were compared with W. brachyloba separately.The additional MRPP analysis was made to assess Squamulea subsoluta vs. S. evolutior comb.nov (PBPB no.107-108), as they are similar and closely related taxa on the classification trees.The PCA analyses were done in all cases to check the phenotypic variation between the studied taxa.
Finally, we tested the separation of Fulgogasparrea from Wetmoreana as proposed by Kondratyuk et al. (2013).We defined the clades as follows: 1) W. decipioides (type species for Fulgogasparrea) and W. appressa, and 2) W. brouardii, W. ochraceofulva/ W. variegata/ W. rubra, and W. texana (type species for Wetmoreana)/ W. sliwae/ W. sliwae ssp.subparviloba, based on Fig. 3.The test was Fig. 2 Summary tree showing placements of 73 queried Teloschistaceae specimens for which the DNA sequences are unknown, using 73 phenotypic traits for 112 specimens in total, based on PBPB and ML weighting technique.Numbers in square brackets indicate node numbers and numbers after the query name indicate bootstrap support for a particular placement of the query taxon based on its morphological features.The numbers following the species names are a unique PBPB number.The colors used on the figure indicate subfamilies as follows: Xanthorioideae and Caloplacoideae in blue, Teloschistoideae including Wetmoreana in orange and the Cinnabaria-Aridoplaca complex in green.Species names in black without the prefix "QUERY" are the reference taxa for which DNA sequences were available.The remaining terminals, all with the prefix "QUERY", represent the query taxa, the gray-colored terminals indicating bootstrap support (in cases where the same individual was binned into several, alternative clades).Clade designations are summarized on the right side of the tree, with newly recognized species in bold  performed based on specimens for which both molecular and phenotypic datasets were present, with the exception of W. brouardii which was supplemented by an additional sample lacking sequence data.Additionally, to increase sample size, the analysis was repeated on an expanded dataset with additional binned specimens of each species.Because W. decipioides is sterile, the incomplete datasets were used in both analyses.

Molecular phylogenetic analysis
The Wetmoreana clade consisted of 33 terminals, clustering into five main, strongly supported subclades and two singleton lineages: the W. ochraceofulva/ W. variegata, W. brouardii/ W. intensa, W. texana/ W. sliwae, W. decipioides and an unnamed subclade, with two additional singleton lineages representing W. 'appressa' and W. rubra.The relationships among the five subclades and the singleton lineages were not supported in the three-locus phylogeny (Additional file 8: Fig. S2).Wetmoreana texana was fully supported as sister to the newly described W. sliwae, which in turn included the newly recognized W. sliwae subsp.subparviloba in a nested position.Wetmoreana ochraceofulva and W. variegata were recovered as sister on short stem branches, and each of the species-level clades was only moderately supported.The recently described W. intensa was recovered as sister to W. brouardii.Finally, W. rubra was recovered on a long, separate branch, and the species is clearly divergent from other members of the genus.Its phylogenetic position was not congruent in the individual gene trees, forming a supported sister group relationship with the W. ochraceofulva/ W. variegata clade in the ITS tree, but with the W. texana/ W. sliwae clade in the other trees (Additional files 9-11: Figs.S3-S5).Fortunately, this supported conflict did not interfere with the objectives of the study regarding the potential delimitation of Fulgogasparrea vs. Wetmoreana or species delimitation in this group.Wetmoreana 'appressa' lies within the genus clade on a very long branch.Depending on the analyses it was recovered as sister to W. decipioides (Fig. 3) or was not grouped with any other clade (Additional files 8-9: Figs.S2-S3).
Phylogenetic binning of the chosen twelve Caloplaca s.lat.species for which no molecular data are available, based on ML weighting, suggested placement of seven species in Wetmoreana, whereas the remaining five species were suggested to not form part of Wetmoreana, nested instead within Squamulea or clustering with Caloplaca fernandeziana (one species, i.e.B. fernandeziana f. validior).MP weighting placed the same seven species in Wetmoreana, whereas three of the remaining species were placed outside Wetmoreana, in the same positions as with ML weighting; the positions of the remaining two query samples were not supported when using MP weighting (Table 1, Additional file 12: Table S3, Fig. 2, Additional file 7: Fig. S1).
Additional 14 query specimens of other Teloschistaceae for which molecular data at the species level are generally available, and which included in the reference tree, mostly grouped with the corresponding species.Two exceptions were found for W. appressa and Teuvoahtiana altoandina.In the case of W. appressa, the non-sequenced type specimen did not cluster with the sequenced specimen bearing that identification but grouped with W. brouardii instead under ML weighting, BS = 74 (MP weighting did not give support).This implies that the sequenced specimen labelled W. appressa in GenBank may not represent that species and is more likely an undescribed species within Wetmoreana.In the case of Teuvoahtiana altoandina, only MP weighting grouped together all specimens of this species (Additional file 7: Fig. S1), the alternative placement suggesting that T. altoandina (PBPB no.92) might represent an undescribed species probably belonging in the genus Calogaya (Fig. 2).Notably, the thirteen W. ochraceofulva query specimens, including the type, grouped only with one of the two sequenced specimens of that species, i.e.W. ochraceofulva PBPB no.55 from Saudi Arabia, but not with W. ochraceofulva PBPB no.60 from Namibia.Even if quite similar, none of the W. ochraceofulva query specimens grouped with W. variegata, suggesting that the subtle differences between the two taxa are diagnostic.The type specimen of W. ochraceofulva from Africa and syntype of Callopisma subnitidum from South America were shown to be conspecific, both grouping with the sequenced W. ochraceofulva from Saudi Arabia (Additional file 12: Table S3, Fig. 2, Additional file 7: Fig. S1).
The phylogenetic binning method also allows assessment of clade-based correlations, and hence the potentially diagnostic importance of morphological characters through the ML and MP weighting process.Of the 73 characters, 34 received more-or-less significant weighting scores (from 70 to 100) in both ML and MP analyses.The remaining 30 characters received higher weights only in one analysis.Among them, 23 were weighted higher by the ML and seven by the MP analysis.Nine characters received low weights (< 70) in both analyses (Additional file 2: Table S2).Characters with high weight in both analyses included the type of thallus morphology (e.g., areolate, squamulose), the morphology of the squamules (their margins and their spatial growth), the presence of propagules, their type and way of formation, the presence and color of the prothallus, the type of apothecia, the apothecial disc color and pruinosity, the structure of the parathecium and the ratio of parathecium and amphithecium, the ratio of length and width of ascospores, the shape of the conidia, and the presence of CaOx crystals in the thalline cortex and in the inner part of the apothecia (parathecium and basal part).Characters with higher weights in only one of the two approaches encompassed, e.g., the type of substrate, the shape of areoles, the size and shape of marginal lobes, the presence of the necral layer and the continuity of the algal layer in the thallus, the frequency of apothecia, their diameter and degree of emergence, the visibility of both apothecial margins, the development of thalline margin of apothecia, the length of ascospores, the length of conidia, and the ratio of length and width of conidia.Characters with low weights in both analyses included the shape of apothecia, the continuity of the apothecial disc surface, the color of the apothecial margin, the width of parathecium, the width of ascospores, the thickness of the acospore septum, the ratio of ascospore length and septum width, the frequency of pycnidia, and the width of conidia (Additional file 2: Table S2).In summary, 57 out of 73 characters received significant weight scores in ML.Among them, qualitative traits accounted for 50 (85% of all qualitative traits), and morphometric traits accounted for 7 (50% of all morphometric traits).In contrast, 41 characters received support in MP and 37 were qualitative traits (63% of all qualitative traits) and 4 were morphometric traits (29% of all morphometric traits).

Ordination of morphological dataset
In the extended PCA including 27 species representing all study genera, the lobate members of Wetmoreana occupied the upper left quadrant of the diagram and, surprisingly, did not group with other lobate taxa such as Calogaya, Gyalolechia and Teuvoahtiana.In contrast, the squamulose W. sliwae fell together with squamulose representatives of Squamulea (Fig. 4).The sublobate W. rubra and Cinnabaria boliviana, characterized by the same ecology, took isolated positions in the diagram (Fig. 4), supporting the molecular data indicating these taxa as more divergent lineages (Fig. 3 The restricted PCA focusing on the 14 Wetmoreana species recognized in this study (Additional file 13: Fig. S6) showed some differences compared to the more extended PCA.Above all, in this analysis, W. ochraceofulva and W. variegata no longer overlapped morphologically.In contrast, W. appressa, W. bahiensis, W. brouardii, W. subnitida, and W. texana more-or-less overlapped with W. variegata, corresponds to the high variation of the latter taxon.Wetmoreana 'appressa', W. circumlobata and especially W. chapadensis remained as clearly morphologically distinct from the other lobate Wetmoreana species.
The MRPP gave the highest A value (between 0.35 and 0.4) when all clades were treated as separate taxa and the lowest (between 0.2 to 0.3) when the two larger complexes, the ochraceofulva/ variegata clade and the sliwae/ subparviloba/ brachyloba clade, were treated as two more broadly defined taxa (Fig. 5, Additional file 17: Fig. S10a, b, c).The cross-species MRPP indicated the absence of significant differences between the W. ochraceofulva and the W. variegata clades, suggesting that these species should be treated as a single taxon (Table 2).The analysis gave weak support for distinguishing both taxa when apothecia and more samples were included (A = 0.1), but only two samples were found fertile out of 15 specimens of W. ochraceofulva, and seven out of 19 specimens in W. variegata, leading to the observed uncertainty of the results.PCA ordination supported the separation of both taxa also when apothecia were included (Additional file 14: Fig. S7), although this effect was not seen when PCA was performed on all genera studied (Fig. 4).

Discussion
Wetmoreana is placed in Teloschistoideae, where it constitutes a well-separated lineage.Its phylogenetic relationship with other closely related genera is resolved without support (Arup et al. 2013;Wilk et al. 2021).The delimitation of Wetmoreana as proposed by Arup et al. (2013) is confirmed in this study, in contrast to the delimitations of this group presented by Kondratyuk et al. (2013Kondratyuk et al. ( , 2015)).According to the latter authors, Wetmoreana is polyphyletic, comprising two generic clades, namely Fulgogasparrea and Wetmoreana s.str.Following this concept, the first taxon includes W. decipioides (type), W. 'appressa', W. awasthii, W. brouardii, and W. intensa, while the genus Wetmoreana includes only the type species, W. texana.Our phylogenetic analyses based on the extended datasets of the Wetmoreana clade show that Fulgogasparrea sensu Kondratyuk et al. cannot be molecularly defined because it would be highly paraphyletic (Fig. 3, Additional file 8: Fig. S2).We performed, the MRPP analyses to test a possible separation of Fulgogasparrea (including W. decipioides and W. 'appressa' only) from the rest of the molecularly defined clades (the W. brouardii, W. ochraceofulva, and W. texana clades) based on their morphological variation and using the one-loci molecular phylogeny as framework (Fig. 3).The MRPP results did not show significant phenotypic differences between Fulgogasparrea and the remaining clades (A = 0.03435718, p = 0.06735338).Therefore, we synonymize Fulgogasparrea with Wetmoreana, and transfer F. awasthii and F. intensa to Wetmoreana.Unfortunately, we could not resolve the placement of W. awasthii within Wetmoreana, as the authors of the combination of that species into Fulgogasparrea (Mishra et al. 2020) failed to deposit the underlying sequence data, ignoring agreed standards of open and fair access to data.
Currently, Wetmoreana includes 15 formally described species, one subspecies, and three further, yet undescribed species.These are W. appressa, W. awasthii, W. bahiensis, W. brachyloba, W. brouardii, W. chapadensis, W. circumlobata, W. decipioides, W. intensa, W. ochraceofulva, W. rubra, W. sliwae, W. sliwae ssp. subparviloba, W. subnitida, W. texana, and W. variegata.Eleven of the 19 tested taxa are newly placed within this genus or confirmed to belong to it based on quantitative integrative taxonomy methods.Two species (W.awasthii and W. intensa) are transferred to Wetmoreana without additional analysis, based on previous studies only (Mishra et al. 2020;Aptroot et al. 2021).Wetmoreana species Table 2 MRPP results for pairwise species comparisons using clades defined from molecular data or extended clades including terminals binned to each molecular clade with support by ML and MP weighting.Chance-corrected within-group agreement (A) describes within-group homogeneity, compared to the random expectation.When all items are identical within groups A = 1 (the highest possible value for A).If heterogeneity within groups equals expectation by chance, then A = 0.If there is less agreement within groups than expected by chance, then A < 0   4).Several examined phenotypic characters in the examined lobate species of Calogaya, Gyalolechia and Teuvoahtiana are responsible for this division, for example, the absence of vegetative diaspores, frequent and abundant apothecia, which are mostly sessile and larger up to 1.5 mm in diam.(vs.0.9 mm in Wetmoreana spp.), the amphithecium being dominant over the parathecium, mostly widely ellipsoid ascospores (vs.regularly ellipsoid), and the frequent presence of CaOx crystals in the thalline cortex (vs. in the thalline medulla).However, to confirm the indicated discriminatory phenotypic characters between those two groups of lichens, it is necessary to consider an expanded dataset with more representatives of lobate species from different genera.
The newly recognized Wetmoreana ochraceofulva/ W. variegata complex in South America, Africa, and the Arabian Peninsula was analyzed by us in terms of molecular and phenotypic variation using PBPB and MRPP.Molecularly, the two species are separated, but with only moderate support and with low genetic divergence (Fig. 3, Additional file 8: Fig. S2).Wetmoreana ochraceofulva was described from Somaliland (East Africa) at the end of nineteenth century (Müller 1885).According to Dodge (1971), there are four other taxa with W. ochraceofulva-morphology in Africa, namely Gasparrinia gracilescens, G. sympageella, G. granulifera, and G. granulifera var.subvitellina.Also, Müller indicated in the protologue the similarity of W. ochraceofulva with G. granulifera (Müller 1885, p. 504).Therefore, more studies are needed on African material of W. ochraceofulva to detect potential (semi-)cryptic diversity.Kärnefelt (1988Kärnefelt ( , 1990) ) synonymized the South American Callopisma subnitidum with W. ochraceofulva and discussed the disjunctive geographic range of this species.This was also supported by our PBPB analyses, since Callopisma subnitidum (syntype, PBPB no.49) grouped with full support with African W. ochraceofulva (Fig. 2, Additional file 12: Table S3).However, the typification of the name Callopisma subnitidum, and consequently its placement as a synonym of C. ochraceofulva, is to be questioned and this topic is discussed under the species W. subnitida below.The distribution of W. ochraceofulva on the South American continent appears narrow, while the species is dominant on the African continent (Fig. 6).
The other species similar to W. ochraceofulva that is common on the South American continent is the newly proposed W. variegata.Our molecular studies indicate that both are closely related, and we suppose that the two taxa have recently diverged, given the short stem branches leading to each (Fig. 3).In comparison to W. ochraceofulva, W. variegata is a highly variable species which may resemble, besides W. ochraceofulva, also other species, such as W. appressa and W. texana.Wetmoreana variegata differs from W. appressa and W. texana in some morphological aspects (e.g., distinctly thinner ascospore septum compared to W. appressa, and distinctly shorter conidia and a limited layer of CaOx crystals in the medulla compared to W. texana), and by their phylogenetic positions (Figs.2-3).Separating W. variegata from W. ochraceofulva can be problematic when a specimen of the former may produce isidia, as does W. ochraceofulva.Still, some other distinguishing features may discriminate between the two taxa (Table 3).In the PBPB analyses, both taxa were placed on the phylogenetic tree separately (Fig. 2, Additional file 7: Fig. S1).Also, the MRPP comparison between the ochraceofulva/ variegata and sliwae/ subparviloba/ brachyloba complexes shows the largest A value when all taxa are treated separately (GR06, all datasets), or at least W. ochraceofulva and W. variegata (GR05 and GR04, only complete datasets) (Fig. 5, Additional file 17: Fig. S10).When defining the clades only molecularly, these being less well sampled, all solutions have similar support (Additional file 17: Fig. S10c,  d).However, MRPP analyses focusing on species pairwise comparisons show some support for distinguishing the two taxa after accounting for apothecial traits and expanding sampling to include additional binned specimens (Table 2).The PCA ordination of Wetmoreana supports the separation of W. ochraceofulva and W. variegata, but only when apothecial characters are included.Wetmoreana ochraceofulva forms one uniform group, whereas W. variegata, with its high morphological variation, overlaps with other Wetmoreana species on this diagram (Additional file 13: Fig. S6).Overall, the separation of the two species is not supported in the case of an extended dataset including additional genera (Fig. 4).
The newly proposed Wetmoreana rubra is a highly divergent species, confirmed besides molecular data also by phenotypic characters .The morphology of this species is rather unique within the genus, characterized by a sublobate thallus, ascospores with very thin septa, partly immersed apothecia, and occurrence on strongly calcareous rocks.The PCA ordination diagram shows that Cinnabaria boliviana forms a group together with W. rubra.Although both species are separated molecularly within Teloschistoideae, they are superficially similar and may grow side by side on   Another new species recognized in this study is Wetmoreana sliwae.It has a strongly supported sister group relationship with W. texana (Fig. 3), although both species are completely different in morphology, squamulose versus lobate, respectively.Instead, the samples of W. sliwae overlap morphologically with the unrelated genus Squamulea, rather than with other representatives of Wetmoreana, in the PCA ordination diagram (Fig. 4).Wetmoreana sliwae is a variable species.Some specimens producing small lobules on the margins of squamules and slightly thinner ascospores are treated here as a separate taxon at the subspecies level, namely W. sliwae ssp.subparviloba.PCA ordination analyses partially support their separation, depending on the data set analyzed, i.e., regarding only the W. sliwae clade or all Wetmoreana spp.(Additional files 13 and 16: Figs.S6 and S9).PBPB (only ML) and MRPP analyses support the separation of both infraspecific taxa (Fig. 2, Table 2).
Among the twelve query Caloplaca s. lat.species, for which molecular data are currently unavailable, seven were placed within Wetmoreana based on ML and MP weighting methods.Those are Callopisma subnitidum (syntype), Caloplaca bahiensis, C. brachyloba, C. chapadensis, C. rubina var.evolutior (syntype), C. subnitida, and one unnamed species, Caloplaca sp. 1.The query species C. brachyloba was placed within Wetmoreana in a sister group relationship with W. sliwae with full support, based on both weighting techniques in PBPB (Fig. 2, Additional file 7: Fig. S1).The species is quite similar to W. sliwae and differs from the latter in having a thinner areolate to subsquamulose thallus, paraplectenchymatous and more-or-less even thalline cortex, ± continuous algal layer, smaller apothecia, orange ostiole of pycnidia, and CaOx crystals within medulla of thallus.The separation of both taxa was supported by all analyses (Table 2, Fig. 4, Additional files 13 and 16: Figs.S6 and S9).
The query species Caloplaca chapadesis was located within the W. rubra clade with strong support in both ML and MP analyses, suggesting a close affinity with the latter (Fig. 2, Additional file 7: Fig. S1).The obscurely lobate W. chapadensis has a rather unique morphology, characterized by the presence of a prothallus and by ascospores with very thick septa, clearly different from W. rubra.Moreover, both species grow on different rock substrates, W. chapadensis on siliceous and W. rubra on calcareous rocks.The PCA diagram clearly shows the uniqueness of W. chapadesis within Wetmoreana (Additional file 13: Fig. S6), supporting its recognition as a species distinct from W. rubra.
The placement of Caloplaca bahiensis within Wetmoreana is not resolved in the same way by the two analyses used (Fig. 2, Additional file 7: Fig. S1).It is either nested within the W. brouardii (ML, both studied samples) or the W. variegata clade (MP, only one studied sample; the other one has different unsupported placement).The partially recovered affinity of W. bahiensis with W. brouardii is considered as more accurate due to morphological characters such as the thickness and color of the thallus and the length of the marginal lobes.Besides W. brouardii, W. bahiensis grouped with W. appressa and W. subnitida, taxa of the same clade (Fig. 2).This group of four species (W.brouardii, together with W. appressa, W. bahiensis and W. subnitida) is recognized here as the W. brouardii complex.Wetmoreana bahiensis is a poorly known species, the epithet not having been validly published (without description) by Zahlbruckner (Luetzelburg 1923;Mackenzie-Lamb 1963).The name was detected by us thanks to Clifford Wetmore's note on one of the specimens of this taxon examined and the epithet is taken up below for this taxon.Based on our study, W. bahiensis is superficially similar to the recently described W. intensa (Aptroot et al. 2021).Both taxa are reported from Brazil from the same or close to the same region.Interestingly, W. intensa is sister to W. brouardii in our phylogenetic analysis (Additional file 8: Fig. S2), just as W. bahiensis in the PBPB analysis (ML only, Fig. 2).Wetmoreana bahiensis and W. intensa may perhaps be conspecific, but more studies are needed to compare both taxa, especially in their anatomical structure.Aptroot et al. (2021) reported that in W. intensa the thallus is quite thick, with the marginal lobes up to 1 mm long, resembling Rusavskia elegans.In contrast, W. bahiensis has longer lobes reaching 2 mm in length, the thallus is thin, and the marginal lobes are 100-200 µm thick.Also, the thalline medulla of W. bahiensis is filled with crystals of the secondary metabolites, mixed with some smaller clusters of CaOx crystals, all prominent in polarized light.From W. brouardii, W. bahiensis differs in lacking vegetative propagules, containing abundant crystals in the medulla of the thallus and apothecia, rather common apothecia with coarsely orange pruinose discs and ascospores with thicker septum up to 5 µm (vs.3.5 µm).Wetmoreana bahiensis differs from W. subnitida in having coarsely orange pruinose discs of apothecia, slightly narrower ascospores (4-5.5 µm vs. 4-7 µm) with slightly thicker septa (mean thickness 4.6 µm vs. 3.2 µm), and abundant crystals in the whole medulla and the apothecia.It differs from W. appressa in having reddish colored thallus, coarsely orange pruinose apothecial discs, distinctly smaller ascospores [(9-)10.8(-13)× (4-)4.6(-5.5)µm vs. (10-)12.6(-15)× (5.5-)6.8(-8)µm], with thinner septa (4-5 µm vs. 4.5-9 µm), and distinctly thinner conidia [(0.5-)0.7(-1.1)µm thick vs. (0.9-)1.1(-1.3)µm].
Unexpectedly, the type of Wetmoreana appressa, for which molecular data are unavailable, grouped with the W. brouardii complex in the PBPB analysis (only ML, Fig. 2).The separation of W. appressa from W. brouardii is clear and without a doubt.It differs from W. brouardii in having abundant apothecia, wider ascospores with a noticeably thicker septum up to 9 µm (vs.3.5 µm), and lacking vegetative diaspores and by containing CaOx crystals in the medulla.Wetmoreana appressa is also different from W. bahiensis and W. subnitida, the taxa grouping with W. brouardii, by having a distinctly thicker thallus with CaOx crystals (absent in W. subnitida), and by larger ascospores with a much wider septum.Our analyses indicate that the sequenced specimen of W. appressa (available in GenBank) is not conspecific with the studied type of this taxon.Both examined specimens differ in morphology, and the sequenced W. 'appressa' may belong to an undescribed species.The sequenced W. 'appressa' differs from the type material of W. appressa in having a thallus with bullate centre, shorter and broader marginal lobes, and a thinner ascospore septum (see notes under W. 'appressa').In the PBPB analyses, the sequenced W. 'appressa' has sister position to W. decipioides (Fig. 2, Additional file 7: Fig. S1).
Four query taxa, Caloplaca muelleri, C. rubina var.evolutior, C. cf.tucumanensis, and C. xanthobola, grouped with Squamulea in the PBPB (Fig. 2, Additional file 7: Fig. S1).In the case of C. cf.tucumanensis and C. xanthobola, the results only indicate that these species are similar to Squamulea, but they likely do not belong to that genus because the species produce a clearly prosoplectenchymatic exciple.An expanded dataset, including members of more Teloschistaceae genera, is needed for more correct binning and to elucidate the possible systematic positions of C. cf.tucumanensis and C. xanthobola.In contrast, C. muelleri and C. rubina var.evolutior are very similar to Squamulea spp.and both may even constitute synonyms of S. subsoluta.However, we prefer to keep them separate and proposed new combinations, i.e., Squamulea muelleri and Squamulea evolutior (see notes under these taxa below).The latter species differs from S. subsoluta in some aspects, e.g., the squamules in S. evolutior are strongly reduced in the central part and only marginal incised squamules are present.Although S. subsoluta may also have a reduced thallus, the squamules are dispersed and present in the central portion between apothecia.Both species, S. evolutior and S. subsoluta, are likely present on the Juan Fernández Islands (Chile).Caloplaca rubina appears to be a tiny species, producing very small, vivid red apothecia and small, scattered areoles.The superficial morphology of the species (we could not study its anatomy due to the small amount of material, and thus this taxon was not included in our PBPB analyses) agrees with Squamulea, and C. rubina can be considered to belong to this genus.An anatomical study should be done especially concerning the apothecial exciple, or molecular analysis based on fresh material, to possibly transfer this taxon to Squamulea.
The Caloplaca fernandeziana clade represents an undescribed, molecularly clearly separate genus within Caloplacoideae (Wilk et al. 2021: Fig. 1).The taxon is quite divergent morphologically from the rest samples studied due to producing typical areolate thallus and biatorine or pseudolecanorine apothecia.
Distribution and ecology: The species grow mainly on siliceous rocks, rarely calcareous (W.rubra), or on both types of rocks (W.ochraceofulva and W. variegata).They are reported from the Americas, Africa, and Asia, where they were collected in exposed or shaded habitats, at different elevations, from 20 to 4500 m.Among the species, W. sliwae occurred at the highest-altitude sites, i.e., 3500-4500 m above sea level.Wetmoreana appressa s. lat., W. awasthii, and W. decipioides occurred at the lowest elevations up to 1000 m (Wetmore and Kärnefelt 1998;Joshi and Upreti 2007;Lumbsch et al. 2011).In contrast, W. brouardii occurred in the widest ranges of altitude between 10 to 2800 m above sea level (Wilk 2021).
The lobate members of Wetmoreana differ from the studied lobate species of Calogaya, Gyalolechia, and Teuvoahtiana by the frequent occurrence of the vegetative diaspores, often rare and not so abundant apothecia, which are mostly erumpent and clearly smaller up to 0.9 in diameter (vs.1.5 mm), ellipsoid ascospores (vs.often broadly ellipsoid), frequent occurrence of CaOx crystals and the location of these in the thalline medulla (vs. in the thalline cortex when present).
In contrast, the occurrence of CaOx crystals in the thallus and apothecia is a good discriminating character for this group of lichens.The traits characterizing CaOx crystals (six in number) received significant weighting scores in both or at least one PBPB analysis.60% of the examined Wetmoreana spp.have such crystals in the thallus medulla, sometimes also in apothecia, and very rarely in the thalline cortex (only in W. rubra and in small amounts).In some species, the crystals form a relatively thin and distinct crystalline layer that separates the algal layer from the lower part of the crystal-free medulla.Such a limited crystalline layer occurs in W. appressa, W. circumlobata, W. ochraceofulva, and W. variegata.Taxa in which CaOx crystals have not been observed are W. brouardii, W. chapadensis, W. decipioides, W. sliwae, W. sliwae ssp.subparviloba, and W. subnitida.CaOx crystals were also found in some other studied species unrelated to Wetmoreana, such as Calogaya pusilla, C. biatorina, Gyalolechia gomerana, Teuvoahtiana altoandina, or closely related Cinnabaria boliviana.Compared to Wetmoreana, the crystals are not always present in those taxa, and if they are present, they often occur in the thalline cortex.CaOx crystals were used as a diagnostic feature by Wetmore and Kärnefelt (1998) in the case of W. appressa and W. texana.The authors studied nineteen species of "Gasparrinia", and in three of them [including also Caloplaca eugyra], the crystals were present in the thalline medulla.Other authors also examined the occurrence of CaOx crystals within Teloschistaceae.For example, within the Caloplaca saxicola group, the crystals are commonly present in the thalline cortex, but only in one species, Calogaya pseudofulgensia, they occur abundantly also in the medulla of the thallus (Gaya 2009).In the case of Variospora aurantia and V. flavescens, the presence of CaOx crystals in the thalline cortex is treated as the most helpful feature to distinguish the two similar species (e.g., Śliwa and Wilk 2008).The occurrence of the CaOx crystals in the thalline cortex is manifested as pruina on the thallus surface, which protects lichens against excessive insolation and pollution (Modenesi 1993;Modenesi et al. 1998).While data on CaOx crystals found in the inner parts of the thallus, such as the medulla, are rather sparse, and their role is unknown.7a-g)

Key to
Etymology: The epithet is derived from the name of the area where the species was collected, an epithet previously proposed by Zahlbruckner but not validly published (Luetzelburg 1923).

MycoBank: MB851727.
Etymology: The epithet is derived from clearly placodioid thallus in this species.
Distribution and ecology: The species was collected in Paraguay, in the open pasture.It grows on siliceous sandstone rocks.
Notes: The examined specimen of W. circumlobata was previously reported as Callopisma brachylobum by Malme (1926) (see the commentary under this species).Wetmoreana circumlobata differs from W. brachyloba mainly in having a placodioid thallus (vs.areolate-squamulose in W. brachyloba), convex central areoles, polygonal to granular (vs.plane to slightly convex, polygonal to sublobate mostly on margins), and apothecia sessile (vs.erumpent).The ascospores and conidia are the same in both species, as is the ecology.Both species were collected in Paraguay on similar loose sandstone rocks.There is more material of Callopisma brachylobum collected by Malme in Paraguay (e.g., G. O. Malme 1355Malme , 1596Malme , 1468b, all in S) , all in S) and Brazil (G.O. Malme 1134, S) (Malme 1926), which needs to be studied to determine its correct taxonomic identity.
Distribution and ecology: Wetmoreana rubra is known only from Bolivia from dry Inter-Andean Valleys, at an altitude of ca.3000 m.It grows on calcareous rocks, in well-lit conditions together with Cinnabaria boliviana.
Notes: Wetmoreana rubra is one of the most distinctive species in the genus, both molecularly and morphologically (Figs.3-4).It produces a sublobate thallus and ascospores with very thin septa, slightly constricted at the central part, and occurs on strongly calcareous rocks.Morphologically, it is most similar to the distantly related, but still within the Teloschistoideae, Cinnabaria boliviana.Both species grow side by side in the field, and the most apparent superficial difference between them is the color of the thallus, which is yellowish orange in C. boliviana.Moreover, the apothecia in C. boliviana are ± persistently immersed and produce widely ellipsoid ascospores with slightly thicker septa (2-3 µm vs. 1-2 µm wide in W. rubra), and distinctly lower length/septum width ratio (4.6 vs. 10.9 times).Cinnabaria boliviana also produces different, shorter conidia, ovoid to short bacilliform in shape (length/width ratio 2.6 vs. 4 times), and has a discontinuous algal layer with algae arranged in the distinct groups.While the anatomical structure of the apothecia are very similar in both species, and the apothecia contain CaOx crystals that are characteristically located along the parathecium (but crystals are not always present in C. boliviana).
Etymology: Named after Lucyna Śliwa, Polish lichenologist, in appreciation of her lichen taxonomic knowledge and her influence on the first author taxonomic experience.
Distribution and ecology: Wetmoreana sliwae is known from Bolivia and Peru, where it occurs in semi-desert, high mountain regions at an altitude between 3462 and 4437 m.It grows on siliceous rocks, in well-lit conditions.
Calogaya miniata differs from the less well-developed specimens of W. sliwae by having a crustose, lobate thallus with areoles in the central portion, continuous algal layer in the thallus, smaller ascospores (7-15 µm length vs. 12-22 µm), lack of pycnidia, and different ecology, occurring on calcareous rocks.
A specimen from Chile from the Juan Fernandez Islands collected by C. & I. Skottsberg (S L2589) was grouped within the W. sliwae clade in the PBPB analysis (ML, BS = 100; MP, PB = 93) (Fig. 2, Additional file 7: Fig. S1).It was considered to represent W. sliwae, but there are several phenotypic differences suggesting that the Chilean specimen should be regarded as a separate undescribed species.The Chilean specimen differs from W. sliwae in having a relatively thinner thallus (150-170 µm thick) without any cracks on its surface, paraplectenchymatous thalline cortex, the algal layer more-or-less continuous (almost all specimens of W. sliwae have algae in distinct groups or columns), sessile and slightly smaller apothecia (up to 0.8 mm), thinner apothecial margins, pycnidia with orange ostioles, and shorter conidia (mean length 3.1 µm).This specimen is a syntype of Caloplaca rubina var.evolutior but differs from the holotype of Squamulea evolutior by its undulate or verruculose surface of the squamules, absence of a prothallus, prosoplectenchymateous exciple, and larger ascospores Etymology: The epithet is derived from similarities to Squamulea parviloba.
Distribution and ecology: Wetmoreana sliwae ssp.subparviloba is only known from Peru, where it occurs together with W. sliwae in semi-desert habitats in high mountain regions at an altitude of 3500 m.It grows on siliceous rocks in well-lit conditions.
Notes: Wetmoreana sliwae ssp.subparviloba differs from W. sliwae s.str.by having an irregularly growing squamulose thallus, consisting of strongly convex, shiny squamules with finger-like lobules on their edges, smaller, erumpent to sessile apothecia (0.3-0.6 mm vs. 0.2-1.2mm in diameter), and distinctly smaller ascospores (9-13 µm length vs. 12-22 µm) with thinner septa (2-3 µm vs. 2-6 µm).Even though the two taxa are not separated in our phylogenetic tree (Fig. 3), we decided to treat them as distinct at the subspecific level.We are aware that the differences in the thallus morphology may be the result of the habitat conditions, but the taxa also differ in other features that are statistically significant, such as size of ascospores.Therefore, we felt that it would be better to show these differences for further verification by specialists.
Additional Etymology: The epithet is derived from the high phenotypic variability of the species.
Distribution and ecology: Wetmoreana variegata is known from Argentina, Bolivia, Ecuador and Peru, where it occurs in semi-desert, high mountain regions at an altitude between 2774 and 3349 m.It grows on siliceous, rarely calcareous rocks, in well-lit conditions, and it is often accompanied by Squamulea subsoluta.
Notes: Wetmoreana variegata is a highly variable species that, together with W. ochraceofulva forms the newly recognized W. ochraceofulva complex (Fig. 3, Table 3).Wetmoreana variegata is characterized by a rather large, orange yellow, lobate thallus, the presence of apothecia or schizidia or rarely isidia in the central portion of thallus; both reproductive structures, such as apothecia and propagules, are rarely present on the same specimen.The algal layer in the thallus is always discontinuous and consists of algae gathered in distinct columns, and CaOx crystals are always present as a limited crystalline layer at the base of the algal layer and in the apothecial medulla.
Fertile specimens of W. variegata are superficially very similar to W. appressa.In contrast, sterile ones resemble W. ochraceofulva or W. texana.There is one atypical specimen from Ecuador (R. Davis 34, BM; not sequenced), representing very well-developed thallus in which apothecia, and isidia and schizidia grow together.The type of W. appressa differs from W. variegata in having a more yellowish thallus, continuous algal layer, entire apothecial disc surface, and ascospores with distinctly thicker septa (4.5-9.0 µm vs. 3-5 µm in W. variegata).Wetmoreana texana has a thallus tightly to loosely adhering to the substrate, more yellowish in color and partly white pruinose, without any cracks on its surface, more convex marginal lobes, fewer apothecia when present, and the joint occurrence of apothecia and vegetative propagules.Moreover, W. texana differs from W. variegata by having ± widely ellipsoid ascospores with thicker septa (4.4 µm vs. 3.6 µm on average), distinctly longer conidia (3.8-4.4-4.9 µm length vs. 2.8-3.2-3.6 µm on average), and CaOx crystals that are not always present but if so, then not forming a distinct limited layer at the base of the algal layer.The two species are not closely related, and W. texana is sister to W. sliwae (Fig. 3).Wetmoreana ochraceofulva is distinguished from W. variegata by consistently forming isidia which occur together with apothecia, if the latter are present (they are extremely scarce).In contrast to W. variegata, the isidia of W. ochraceofulva are usually rounded, distinctly smaller, almost indistinguishable from each other at low magnification, and formed at the margins of areoles in a regular pattern.The ascospores of W. ochraceofulva have septa slightly thicker than in W. variegata, and the length/septum width ratio is higher (Table 3).Finally, W. ochraceofulva is a morphologically uniform species found mainly on the African continent, while W. variegata is highly variable and only occurs in South America (Fig. 6).Wetmoreana variegata may also be confused with W. brouardii.The latter species, however, differs from W. variegata in having smaller marginal lobes [0.4-2(-2.2) mm long and 0.2-1 mm wide at tips vs. 0.2-4.5 mm long and 0.2-1.6(-2)mm wide] that are thinner (90-250 µm vs. 125-600 µm), and more closely adhering to the substrate.Moreover, apothecia in W. brouardii are extremely rare, and when present the ascospores are smaller ascopores (10-13 µm length vs. 10-17 µm) with thinner septa (2-3.5 vs. 3-5 µm); also, abundant distinctive small papillae always cover the thallus surface.Finally, the algae form a continuous layer or are clustered but not in such distinct groups as in W. variegata, and CaOx crystals are not present in the medulla.Wetmoreana circumlobata differs from W. variegata in having slightly shorter marginal lobes (0.3-1.4 mm vs. 0.2-4.5 mm length), strongly convex central areoles, a continuous to discontinuous algal layer in the thallus (algae not forming distinct columns), sessile apothecia with entire apothecial discs, more reduced thalline apothecial margin, thicker parathecium (up to 190 µm vs. up to 119 µm), larger and wider ascospores [(13-)16.2(-19)× (7-)7.5(-9)µm vs. (10-)13.1(-17)× (5-)6.3(-9)µm] with thicker septa (mean thickness 4.6 µm vs. 3.6 µm), long bacilliform conidia (length/width ratio 3.8 vs. 2.7), and lack of vegetative propagules.
Although W. variegata and W. ochraceofulva are not well separated in our molecular and integrative approaches, the emerging patterns, especially considering vegetative propagules and ascospore morphology, led us to recognize W. variegata as a distinct taxon.Its recognition will also help to test the taxonomy of the two species with further data in the future, rather than encourage premature lumping.
Notes: The placement of W. awasthii within Wetmoreana (= Fulgogasparrea) is uncertain.The sequences of this species were published by Mishra et al. (2020) are not present in GenBank.Therefore, the molecular study must confirm W. awasthii placement within the genus Wetmoreana.
Distribution and ecology: The species is known only from Paraguay, where it grows on siliceous, sandstone rocks.
Notes: The studied material of W. brachyloba is morphologically uniform.The species has some similarities with W. sliwae, as both have a ± squamulose thallus and produce similar ascospores.For a detailed comparison, see the discussion of W. sliwae.Wetmoreana brachyloba is also similar to W. circumlobata in anatomical aspects of thallus and apothecia, including the size of ascospores and conidia (see comments under W. circumlobata).Malme (1926) proposed the new combination Callopisma brachylobum for A. brachylobum.Notable, the material studied by Malme (1926) is not conspecific with A. brachylobum (G, BM, and M) and represents an undescribed species.The latter is described above as new species, W. circumlobata (see also the commentary under this species).Nevertheless, since nomenclatural acts follow types, the combination of A. brachlobum into Callopisma by Malme is valid.
Distribution and ecology: Wetmoreana chapadensis is known only from Brazil (Matto Grosso), where it was collected on siliceous stones in a sunny field covered by grasses and herbs (Malme 1926).
Notes: Wetmoreana chapadensis is a rather distinct species, clearly separated from other Wetmoreana spp.The PBPB analysis indicates that W. chapadensis is closest to W. rubra, but it differs from the latter in having an orange prothallus, mostly sublobate central areoles (vs.polygonal), sessile apothecia (vs.erumpent), apothecial discs ± concolorous with the thallus, and lack of a clear distinction between proper and thallus margin of the apothecia; in addition, it differs by the indistinct pycnidia, ascospores with distinctly thicker septa (5-7 µm vs. 1-2 µm), thin thalline cortex not covered by a necral layer, and lack of CaOx crystals in the thallus and apothecia.Wetmoreana chapadensis can be confused with W. bahiensis and W. subnitida.Both of those species have a more-or-less reddish thallus.However, W. chapadensis differs from the latter two species in having an obscurely lobate thallus, sublobate central areoles, the presence of prothallus, and larger ascospores with distinctly thicker septa.
Notes: Wetmoreana intensa is closely related to W. brouardii, hence belongs to the newly recognized W. brouadrii complex.Wetmoreana intensa is very similar to W. bahiensis and W. subnitida.For comparisons with these species, see notes under the corresponding taxa.
Distribution and ecology: The species is reported mainly from Africa, with single records from the Arabian Peninsula and South America (Argentina and Uruguay) (Kärnefelt 1990;Wilk 2022).It grows on calcareous and siliceous rocks.
Notes: The lectotype of W. ochraceofulva is incomplete, the specimen is represented only by the central part of thallus, and it entirely lacks the thallus margin, which should produce marginal lobes.This was considered during the PBPB analyses, and the character concerning marginal lobes was coded as unknown (not as
Distribution and ecology: Wetmoreana subnitida is known from Argentina and Peru.It grows on siliceous rocks.
Notes: This species was described by Malme (1926) from South America (Argentina) under the name Callopsima subnitidum.The original material of this species is heterogeneous, including two different lobate species, i.e., one fertile and the second sterile forming isidia.Kärnefelt (1990) proposed lectotypification of Callopisma subnitidum based on the sterile specimen with isidia, and consequently synonymized this species with Caloplaca ochraceofulva.However, our investigation revealed that the original description of Callopsima subnitidum (Malme 1926) is clearly based on the fertile material producing apothecia.The description includes details about apothecia and ascospores but there is no mention of vegetative propagules.Furthermore, according to the protologue, the algal layer is more or less continuous, and the tips of marginal lobes are usually about 0.5 mm wide, rarely up to 1 mm.These features better fit the fertile specimen (R.E.Fries 15), because the sterile specimen (R.E.Fries 50) has a discontinuous algal layer where algae are arranged in distinct groups, and the marginal lobes are wider on average.Malme also highlighted the character of the ascospores of Callopsima subnitidum when comparing this species with another taxon, Variospora aurantia (= Lecanora callopisma).All this clearly shows that the lectotypification by Kärnefelt (1990) is not in accordance with the protologue and has to be superseded [ICNafp Art. 9.19(c)].In doing so, an existing name becomes available for a taxon that would otherwise have to be described as new.Therefore, we superseed the previous lectotypification of C. subnitidum with the fertile specimen R. E. Fries 15, treating its combination into Wetmoreana as a taxon different from W. ochraceofulva.The sterile specimen originally given as a syntype of C. subnitidum (R. E. Fries 50, LD 1076640) is confirmed as representing Wetmoreana ochraceofulva.Wetmoreana subnitida differs from W. ochraceofulva mainly in anatomical features, in addition to the absence of vegetative propagules and a slightly white-pruinose thallus.Wetmoreana subnitida is further distinguished by the presence of a necral layer, a continuous algal layer, abundant apothecia, noticeably smaller ascospores (8-12 µm vs. 10-16 µm long) with slightly thicker septa (2-5 µm vs. 2-4 µm wide), a lower length/septum width ratio (3 vs. 4.3-4.4),and the absence of crystals in the thallus and apothecia.
The additional examined W. subnitida specimen collected by W. J. Eyerdam (26,062, F; PBPB no.65) is somewhat problematic, as it shows some similarities to W. bahiensis as well.The PCA diagram has not resolved its affiliation, and W. subnitida (PBPB no.65) is close to both the type of W. subnitida and W. bahiensis (Additional file 18: Fig. S11).However, W. subnitida (PBPB no.65), like the lectotype of this species, lacks crystals luminescent under polarized light in both the medulla of thallus and apothecia.While in W. bahiensis such crystals are found in the whole medulla of the thallus (these are crystals of unknown origin, and some CaOx crystals).This chemical evidence supports the placement of the W. J. Eyerdam specimen within W. subnitida.However, considering some uncertainty, the description of W. subnitida is based only on type specimen.
Distribution and ecology: Squamulea evolutior is known only from Chile, the Juan Fernandez Islands.It was collected on siliceous rocks.
Notes: Squamulea evolutior is placed within Squamulea with strong support in the PBPB analyses (ML, Fig. 2) as sister to S. subsoluta.The species fits very well with the morphology of Squamulea, and we proposed a new combination for it.Originally the taxon was described as Caloplaca rubina var.evolutior (Zahlbruckner 1924), but our study indicates that it is not conspecific with C. rubina s.str.(see below).Squamulea evolutior differs from C. rubina by having a well developed squamulose thallus, with clearly zeorine, larger apothecia (0.1-0.8 mm vs. 0.2-0.5 mm in diameter), and slightly larger ascospores (8-13 × 5-8 µm vs. 9-11 × 5-6 µm) (Zahlbrucner 1924).Squamulea evolutior is similar to S. subsoluta but differs from the latter in having a distinctly squamulose thallus and a much reduced thalline margin visible at the base of mature apothecia (see also Discussion).Squamulea evolutior may be confused with W. sliwae.Both species were discussed under the latter taxon.Notes: Caloplaca rubina is distinctive species producing reddish, tiny thalli consisting of dispersed, small and tightly adhered to the substrate areoles, surrounded by a black prothallus, and very small concolorous with thallus apothecia.It grows on volcanic rock.The possible affiliation of this species to Squamulea must be confirmed by molecular analysis or anatomical studies of the apothecial margin.The specimen collected by C. & I. Skottsberg (UPS 204817), labelled originally as Caloplaca rubina, appears to represent Caloplaca rubina var.evolutior.The latter produces rather well-developed squamulose thalli and is conspecific with the lectotype of Squamulea evolutior.
Notes: Squamulea muelleri is a historically described species described by Wainio in 1890 from Brazil and has been known only from South America till present (e.g.,  Malme 1926;Osorio 1972;Spielmann 2006).However, the studied material of this species is not homogenous and requires further study to resolve its taxonomy.Nevertheless, the type material of C. muelleri used in this study clearly clustered with Squamulea based on PBPB analysis (only ML, BS = 100, Fig. 2).This affiliation was also confirmed by the morphological studies, which showed the presence of a paraplectenchymatous exciple of apothecia and thalline cortex, and similar, rather small ascospores characteristic for Squamulea spp.(Fig. 13im).Caloplaca muelleri is located in the S. subsoluta clade and was considered by us a synonym of the latter species since the morphology of both taxa is quite similar.However, examined type of C. muelleri has ascospores quite smaller than in S. subsoluta (10-12 × 5-6.5 µm vs. 9.5-15 × 5-9 µm), and such small ascospores are diagnostic for C. muelleri according to Wainio (1890) and Malme (1926).Therefore, we decided to treat this species as a separate taxon pending more study material and results of molecular studies providing additional evidence, especially since the entire S. squamulosa/subsoluta group is still largely unresolved taxonomically, and the species S. subsoluta, appears to include many cryptic taxa (Bungartz et al. 2021).
The additional examined specimens of C. muelleri collected by Lorentz (KRAM) [treated as C. muelleri by Malme (1926)] and H. S. Osorio 2342 (F), both from Uruguay, have similar ascospores to the type of C. muelleri, but superficially they are rather similar to S. evolutior producing distinctly squamulose and slightly stipitate thalli.The C. muelleri taxonomic description was based on an isolectotype, because the studied material of C. muelleri appears to be not homogenous.

Taxon with unresolved taxonomy
Notes: Wetmoreana 'appressa' is characterized by bullate central areoles, rather short marginal lobes, closely or loosely adherent to the substrate, and apothecia with distinctly swollen margins (Fig. 14a-g).
The sequenced specimen of Wetmoreana 'appressa' (GenBank No. KC179332) differs phenotypically from the holotype of W. appressa, for which DNA sequences are unavailable.The two taxa are also located in different clades in our PBPB analyses (Fig. 2).These results indicate that the two species are not conspecific.The high phenotypic varability of W. appressa was also discussed by the author of the species (Wetmore and Kärnefelt 1998).Therefore, this species, W. appressa s. lat., requires further investigation, including molecular data, to see if the observed variability has support at the molecular level.

Conclusion
Our study provides a first attempt at an integrative taxonomic approach to solve the delimitation of the genus Wetmoreana and its species.PBPB is so far the only approach known that allows to obtain predictive placements of specimens for which no sequence data are available and which provides a measure of statistical confidence for such placements.The obtained solutions are therefore testable, in that way differing from ad hoc taxonomy which weights individual characters subjectively in the decision-making process.The Wetmoreana case provides a good case study, as the results would not have been expected using a traditional ad hoc approach.A shortcoming of this method is that it relies on existing molecular reference trees, and the possibility that some query taxa for which DNA is unavailable may represent genera not yet known to science cannot be excluded.Also, the morphological taxon sampling is not always complete and here we might have missed some additional, similar and poorly understood South American taxa.On the other hand, this approach helps to precisely identify those taxa that require targeted sampling for molecular data, and additional phenotypes can be easily added later to undertake a broader analysis.As such, we consider our results as a testable hypothesis that serves as reference for further, more detailed studies in this group.An obvious challenge to this approach is the time-consuming assessment of a large number of characters and their assembly in a comprehensive data matrix.However, once assembled, the matrix allows a reproducible approach and opens avenues to various downstream analyses regarding character evolution.

Fig. 1
Fig. 1 Stage of development of the thalline margin in the zeorine apothecia: A ± persistent, B partly reduced, C much reduced, and two other types of apothecia D pseudolecanorine and E biatorine not producing thalline margins

Fig. 3
Fig. 3 Overall molecular phylogeny of the members of Wetmoreana and additional genera Aridoplaca, Calogaya, Cinnabaria, Gyalolechia, Squamulea, Teuvoahtiana, and the species Caloplaca fernandeziana which members are similar to Wetmoreana.Phylogenetic ITS tree was performed in RAxML.The support values associated with branches indicate maximum likelihood bootstrap values.The values ≥ 70 are considered as significant support.Newly proposed taxa are marked.The numbers after species names corresponds to PBPB numbers

Fig. 6
Fig. 6 Map of world distribution of Wetmoreana ochraceofulva (black triangle) and W. variegata (black circle).The numbers in parentheses refer to additional specimens cited in the literature but not examined by us

Fig. 7
Fig. 7 Species of the Wetmoreana brouardii clade.A-G Wetmoreana bahiensis (lectotype, A thallus habit, B apothecia with coarse orange pruinose discs, C section of thallus in polarized light showing medulla filled with crystals of unknown origin and some calcium oxalate, D section of apothecium showing paraplectenchymateous parathecium, E section of apothecium showing crystals prominent in polarized light, F ascospores, G conidia).H-M W. subnitida (lectotype, H thallus habit, I erumpent to sessile apothecia, J section of thallus showing paraplectenchymatous upper cortex, K section of apothecium showing prosoplectenchymatous parathecium, L ascus with spores, M ascospore, N conidia).Scale in A, B, H, I = 2 mm, in C, D, E, J, K = 100 µm, in F, L, M = 20 µm, in G, N = 10 µm

(
See figure on next page.)

Table 1 Placements of twelve query Caloplaca s.lat. species within studied clades based on phenotype-based phylogenetic binning method under ML and MP weighting. The numbers in brackets indicate the number of specimens per species studied. The proposed
placement of the query taxa means that species is the closet species with respect to morphological characters but not necessarily that it belongs to that species are exclusively saxicolous, crustose, more-or-less lobate at the margins (placodioid), or rarely squamulose.The squamulose representatives, namely W. brachyloba, W. sliwae, and W. sliwae ssp.subparviloba, are similar to the unrelated genus Squamulea (subfamily Xanthorioideae), which differs from Wetmoreana mainly in the paraplectenchymatous apothecial exciple, the frequent presence of a prothallus, and shorter ascospores [8-15 µm vs. (9)12-22 µm; based on examined samples].In contrast, the crustose-lobate (placodioid) Wetmoreana species are similar to many other lobate taxa of different genera widely dispersed in all three Teloschistaceae subfamilies.Surprisingly, the studied lobate members of Calogaya, Gyalolechia, and Teuvoahtiana are clearly separated from the lobate members of Wetmoreana in the PCA ordination diagram (Fig.