Twelve New Species Reveal Cryptic Diversification in Foliicolous Lichens of Strigula s.lat. (Strigulales, Ascomycota)

We employed a molecular phylogenetic approach using five markers (ITS, nuSSU, nuLSU, TEF1-α, and RPB2) to assess potential cryptic speciation in foliicolous members of Strigula s.lat. (Strigulaceae), including the recently segregated genera Phylloporis, Puiggariella, Raciborskiella, Racoplaca, and Serusiauxiella, from tropical areas in Asia, with selected materials from the Neotropics as reference. On the basis of combined molecular and phenotypic datasets, two new species of Racoplaca and 10 new species of Strigula s.str. are described: Racoplaca macrospora sp. nov., R. maculatoides sp. nov., Strigula guangdongensis sp. nov., S. intermedia sp. nov., S. laevis sp. nov., S. microcarpa sp. nov., S. pseudoantillarum sp. nov., S. pseudosubtilissima sp. nov., S. pycnoradians sp. nov., S. sinoconcreta sp. nov., S. stenoloba sp. nov., and S. subtilissimoides sp. nov. In addition, we propose the new combination Phylloporis palmae comb. nov. (≡ =Manaustrum palmae) and we validate the earlier combination Racoplaca melanobapha comb. nov. (≡ Verrucaria melanobapha; Strigula melanobapha). Our data clearly indicate a considerable degree of cryptic diversification in foliicolous representatives of Strigula s.lat., particularly in the presumably widespread taxa Strigula antillarum, S. concreta, S. nitidula, and S. smaragdula. Given that these phylogenetic revisions are thus far limited to few regions, we predict that our findings only represent the proverbial tip of the iceberg in this group of lichenized fungi.


Introduction
Understanding species delimitation is crucial for biological and applied sciences and for conservation assessments [1,2]. In lichenised fungi, alpha-taxonomy has traditionally been based on morphological, anatomical, and/or chemical characters [3,4]. However, ontogenetic, epigenetic, and environmental factors may influence phenotypic features and, as a result, it is often difficult to distinguish phylogenetically informative characters from intraspecific variability [5][6][7][8]. In addition, evolutionary homoplasy is frequent in fungi, including lichens, often leading to phenotypically cryptic, albeit not directly related species [9,10].
PCR reactions were carried out in 25 µL reaction volumes and the components used were: 2 µL total DNA, 1 µL each primer (10 µM), 12.5 µL 2 × Taq MasterMix, and 8.5 µL ddH 2 O. Amplification was performed using a Biometra T-Gradient thermal cycler. Cycling parameters for nuLSU, ITS, and nuSSU were set to an initial denaturation at 95 • C for 5 min, followed by 35 cycles of denaturation at 94 • C for 30 s, annealing at 54 • C for 30 s, extension at 72 • C for 1 min, and a final extension at 72 • C for 10 min. PCR amplifications of TEF1-α were initiated with a 2 min denaturation at 94 • C. The annealing temperature in the first amplification cycle was 66 • C, which was subsequently incrementally reduced by 1 • C per cycle over the next 9 cycles. An additional 30 amplification cycles were then performed, each consisting of 30 s denaturation at 94 • C, a 30 s annealing step at 56 • C, and a 1 min extension at 72 • C, concluding with a 10 min incubation at 72 • C [57]. The PCR conditions of RPB2 included: initial denaturation at 95 • C for 5 min; 35 cycles of 1 min at 95 • C, 2 min at 50 • C, an increase of 1 • C/5 s to 72 • C, and 2 min at 72 • C; and a 10-min incubation at 72 • C [58]. PCR products were checked on 0.8% agarose electrophoresis gels stained with ethidium bromide, and then sent to Majorbiology (Changping District, Beijing, China) for sequencing.

Sequence Alignment and Phylogenetic Analyses
Sequences for each marker were joined with others obtained from GenBank (Table S1), generating a separate ITS and a concatenated four-locus (nuSSU, nuLSU, TEF-α, and RPB2) dataset. Each marker was firstly aligned independently with MAFFT v. 7 [59], and the combinability was tested [60]. Each partition was firstly analysed separately using MrBayes. The Markov chain Monte Carlo algorithm of MrBayes ran with 20 chains simultaneously, each initiated with a random tree, for 2 M generations, sampling every 20th generation for a total of 100,000 trees sampled. The first 4500 sampled trees were discarded before calculating the majority-rule consensus tree to ensure that all chains had converged at a single level. A majority-rule consensus tree was calculated with DendroPy v. 4.5.2 for the remaining 95500 B/MCMC sampled trees [61]. The majority-rule consensus tree of each gene was constructed through SumTrees with the parameters "summary-target = consensus -burnin = 4500 -support-as-labels -min-clade-freq = 0.1". The conflict was assumed to be significant if different relationships for the same set of taxa (one being monophyletic and the other being nonmonophyletic), all with posterior probabilities (PP) 99%, were observed on the majority-rule consensus trees. Only if no significant conflict was detected throughout the majority-rule consensus trees, using this criterion, would the four partitions be combined [60].
An ML tree involving 1000 pseudoreplicates was generated by IQ-TREE v1.6.6 [62]. For this analysis, the best-fit substitution model was selected using ModelFinder [63]. GTR + F + I + G4 was selected as the best model for the ITS, and TIM2 + F + I + G4 for the four-locus dataset.
Bayesian analysis was performed in MrBAYES [64] under the general time reversible model, including estimation of invariant sites and a discrete gamma distribution with six rate categories (GTR + I + G), both for the single marker and for the combined analyses. A run with 5 M generations and employing 20 simultaneous chains was executed. Poste-rior probabilities above 95% and a bootstrap support value above 70% were considered threshold values.

Species Tree Assessment
Firstly, we utilised IQ-TREE v1.6.6 [62] to construct an ML phylogenetic gene tree for each fragment (nuSSU, ITS, nuLSU, TEF-α, and RPB2), with 1000 pseudoreplicates, respectively. ASTRAL finds the species tree that has the maximum number of shared induced quartet trees with the set of gene trees, subject to the constraint that the set of bipartitions in the species tree comes from a predefined one. We conducted the analysis of the best supported ML gene trees in the multi-individual version of ASTRAL v5.7.7 [66,67] to estimate a species tree annotated with posterior probabilities, as nodes support.

Phylogenetic Analyses
The dataset includes 169 ITS sequences, 20 nuLSU sequences, 17 nuSSU sequences, 19 TEF1-α sequences, and 19 RPB2 sequences newly generated for this study. Since we obtained a much larger number of ITS sequences, this gene locus was analysed separately ( Figure 1) and subsequently compared with the concatenated tree based on the other four gene loci.
The ITS tree contained a total of 225 terminals, based on an alignment with a length of 448 bp. The topology (ML tree shown) was consistent with the earlier findings by Jiang et al. [47], recovering the six genera now distinguished within foliicolous Strigula s.lat. (Figure 1). Both the genera and almost all species had strong support through ML and Bayesian analyses. The ITS phylogeny revealed 11 new lineages, here introduced as new species, two in Racoplaca (R. maculatoides, R. macrospora) and nine in Strigula s.str. (S. guangdongensis, S. intermedia, S. laevis, S. microcarpa, S. pseudoantillarum, S. pseudosubtilissima, S. pycnoradians, S. sinoconcreta, S. subtilissimoides). In addition, the ITS-based phylogeny supported the separation of Brazilian material of Phylloporis with a punctate thallus from P. phyllogena s.str. (Figure 1); for this material, the name P. palmae is taken up below. A further new species of Strigula, S. stenoloba, is described based on phenotype only, as no sequence data could be generated for this material.
The majority-rule consensus tree sampled with B/MCMC for the LSU, SSU, TEF1-α, and RPB2 datasets, respectively, exhibited-although similar in their overall topologyvarious differences. However, none of the different relationships revealed by the separate analyses received reciprocal posterior probabilities (PP) of 99%, and, therefore, combining these four datasets was not considered to have any detrimental effect in estimating phylogenetic relationships among these taxa [60,68]. The Serusiauxiella, Phylloporis, Puiggariella, and Raciborskiella clades were monophyletic and strongly supported in all instances (no RPB2 data for the last). Based on nuSSU evidence, Racoplaca was nested within Strigula prasina with a posterior probability of 86%, whereas it was monophyletic with the LSU, TEF1-α, RPB2, and in the combined analysis (all 100% posterior probability). Considering that our selection of 99% as the threshold to determine if datasets should be combined, this has no detrimental effect. Strigula s.str. clade was recovered as monophyletic in the RPB2 and the concatenated dataset, but paraphyletic in the nuLSU and polyphyletic in nuSSU due to the invasion of Racoplaca. In the concatenated dataset, the six clades were resolved as monophyletic and with strong support, so we use this to reflect the most likely natural evolutionary relationship ( Figure 2). The majority-rule consensus tree sampled with B/MCMC for the LSU, SSU, TEF1-α, and RPB2 datasets, respectively, exhibited-although similar in their overall topology-various differences. However, none of the different relationships revealed by the separate analyses received reciprocal posterior probabilities (PP) of 99%, and, therefore, combining these four datasets was not considered to have any detrimental effect in estimating phylogenetic relationships among these taxa [60,68]. The Serusiauxiella, Phylloporis, Puiggariella, and Raciborskiella clades were monophyletic and strongly supported in all instances (no RPB2 data for the last). Based on nuSSU evidence, Racoplaca was nested within Strigula prasina with a posterior probability of 86%, whereas it was monophyletic with the LSU, TEF1-α, RPB2, and in the combined analysis (all 100% posterior probability). Considering that our selection of 99% as the threshold to determine if datasets should be combined, this has no detrimental effect. Strigula s.str. clade was recovered as monophyletic in the RPB2 and the concatenated dataset, but paraphyletic in the The four-marker tree, based on a concatenated alignment with 4373 bp (nuSSU: 1176 bp; nuLSU: 1274 bp; TEF1-α: 886 bp; RPB2: 1037 bp) also recovered the six genera with strong support (ML tree shown), and likewise recovered the genera Puiggariella and Racoplaca, and Raciborskiella and Serusiauxiella as sister clades, respectively ( Figure 2). The clade including Puiggariella and Racoplaca had moderate support in the ITS ( Figure 1) and no support in the four-marker tree (Figure 2), whereas the clade containing Raciborskiella and Serusiauxiella had no support in the ITS (Figure 1), but strong support in the fourmarker tree ( Figure 2). Stem branches were rather long for Phylloporis, Raciborskiella, and Serusiauxiella, moderately long for Puiggariella and Racoplaca, and comparatively short for Strigula s.str. (Figures 1 and 2), agreeing with earlier findings [47]. Within the latter, both the ITS and the four-marker tree resolved the S. nitidula group on a long, strongly supported branch (Figures 1 and 2). J. Fungi 2022, 7, x FOR PEER REVIEW 6 of 31 nuLSU and polyphyletic in nuSSU due to the invasion of Racoplaca. In the concatenated dataset, the six clades were resolved as monophyletic and with strong support, so we use this to reflect the most likely natural evolutionary relationship ( Figure 2).

Figure 2.
Phylogenetic tree constructed through ML analyses based on four markers (nuSSU, nuLSU, TEF1-α, and RPB2), with a total alignment length of 4373 bp. Maximum likelihood bootstrap support value above 70% (left) and Bayesian inference posterior probabilities above 95% (right) are shown above branches (ML-BS/B-PP). Terminals in boldface indicate newly generated sequences for this study.
The four-marker tree included data for one of the two new species in Racoplaca (R. maculatoides), and also resolved all nine new species in Strigula s.str. separately delimited based on the ITS tree (S. guangdongensis, S. intermedia, S. laevis, S. microcarpa, S. pseudoantillarum, S. pseudosubtilissima, S. pycnoradians, S. sinoconcreta, and S. subtilissimoides).

Species Tree Assessment
Phylogenetic analysis using concatenated datasets does not take into account the stochasticity of the coalescent process, and thus may fail to recover the true species tree [69]. Therefore, we also used a coalescent-based method to infer the species tree from a set of gene trees by explicitly taking into account the inherent stochasticity associated with the coalescent process [66].
The resulting ASTRAL tree ( Figure 3) was quite similar to the single-marker ITS tree (Figure 1), except for the branch length patterns characteristic of the coalescent approach, supporting the usefulness of the ITS to delimit species in this clade and the congruence between the various markers. The genera also received good to strong support in the ASTRAL tree (87-99%), with the exception of Phylloporis (Figure 3). We further compared the interspecific topologies of each gene tree to the coalescent species tree topology and found that three of the five individual gene trees had partially different topologies, whereas the other two were congruent with the coalescent species tree topology.     Notes: The sequenced material corresponds morphologically to Phylloporis multipunctata (G. Merr. ex R. Sant.) Vězda, described from Indonesia and recently synonymised under P. cinefaciens (Nyl.) S.H. Jiang, Lücking & Sérus. [41]. However, microscopic examination revealed that the foliicolous specimens from Brazil have much smaller ascospores than the type material of P. cinefaciens and P. multipunctata (12-16 × 3.5-4.5 μm in the latter two [41]). We, therefore, take up the epithet introduced by Cavalcante et al. [70], originally synonymised with P. multipunctata by Lücking (1998), for this material [71]. Since the holotype only bears pycnidia, we designate one of the sequenced specimens with numerous perithecia as epitype.    [41]. However, microscopic examination revealed that the foliicolous specimens from Brazil have much smaller ascospores than the type material of P. cinefaciens and P. multipunctata (12-16 × 3.5-4.5 µm in the latter two [41]). We, therefore, take up the epithet introduced by Cavalcante et al. [70], originally synonymised with P. multipunctata by Lücking (1998), for this material [71]. Since the holotype only bears pycnidia, we designate one of the sequenced specimens with numerous perithecia as epitype.  Diagnosis: The new species externally resembles Racoplaca maculata but differs by the longer asci, larger, fusiform ascospores, and larger, one-septate macroconidia.
Chemistry: No substances detected by TLC.
Based on the ITS data, Racoplaca macrospora was found nested within another newly recognised species, R. maculatoides (see below). However, there are between 18 consistent base call differences between the two species (substitutions and indels), corresponding to a mean identity value of 96% (Table S2), substantially below what could be accepted for within-species variation. Given that R. maculatoides has distinctly shorter (15-25 µm vs. 22.5-27.5 µm long) ascospores and macroconidia with shorter appendages (10-18 µm vs. 17-35 µm), we consider R. macrospora a recently emerging, yet distinct species. Species evolving from paraphyletic residuals are now broadly accepted [36], but an artifactually paraphyletic topology could also result from the most closely related species not having been sequenced yet. Another possible explanation is that R. maculatoides itself represents a species complex. Indeed, separate analysis of the ITS only for Racoplaca revealed three subclades in this clade: one supported subclade (75%), including the type of R. maculatoides, with a basally emerging individual with 63% support; another shallow subclade with four individuals currently assigned to R. maculatoides; and a strongly supported subclade (100%) on a long branch, representing R. macrospora ( Figure S1). macroconidia (asci 40-60 × 5-8 μm, ascospores 12-18 × 2.5-3.5 μm, macroconidia aseptate, 4-6 × 1.5-2 μm in R. maculata [45]). Racoplaca maculatoides is most similar and most closely related to R. macrospora ( Figure 1). It differs by the shorter ascospores and the narrower macroconidia with shorter appendages. Racoplaca melanobapha resembles R. maculatoides in ascospores and macroconidia, but differs in the strongly and regularly laciniate, more brownish thallus [45]. Both are also phylogenetically distinct (Figures 1  and 2). Diagnosis: The new species externally resembles Racoplaca maculata but differs by the longer asci, larger, fusiform ascospores, and larger, one-septate macroconidia.
Chemistry: No substances detected by TLC.
Habitat and distribution: The new species was found in humid, semi-exposed forest habitats in southern China. Notes: Racoplaca maculatoides strongly resembles R. maculata in morphology but can be distinguished by the longer asci; larger, fusiform ascospores; and larger, one-septate macroconidia (asci 40-60 × 5-8 µm, ascospores 12-18 × 2.5-3.5 µm, macroconidia aseptate, 4-6 × 1.5-2 µm in R. maculata [45]). Racoplaca maculatoides is most similar and most closely related to R. macrospora (Figure 1). It differs by the shorter ascospores and the narrower macroconidia with shorter appendages. Racoplaca melanobapha resembles R. maculatoides in ascospores and macroconidia, but differs in the strongly and regularly laciniate, more brownish thallus [45]. Both are also phylogenetically distinct (Figures 1 and 2). Diagnosis: The new species can be distinguished by the beaked, more elongate pycnidia than in other species of the genus. The phylogenetic trees also indicate it forms an independent clade.
Chemistry: No substances detected by TLC.
Habitat and distribution: The new species grows on living leaves in wet tropical forest in China. Notes: This is another newly recognised species that belongs in the morphologically defined Strigula smaragdula complex. The latter has been described from Nepal and we have currently marked one phylogenetically distinct clade from China as a candidate for this species (S. cf. smaragdula; Figures 1 and 2). Strigula guangdongensis differs from this clade phylogenetically and is also set apart by its beaked pycnidia. Lücking also mentioned specimens of S. smaragdula s.lat. with shortly beaked pycnidia from the Neotropics, but these likely represent another unrecognised taxon [45]. A corticolous species from Madagascar, S. rostrata R.C. Harris & Aptroot, has more strongly beaked pycnidia and submuriform macroconidia [43]; that species is now placed in a different genus, as Swinscowia rostrata  the Philippines (S. sulcata Vain., Porina crenulata Vain. [42,45]). These names are, therefore, potentially available for cryptic or near-cryptic lineages in the corresponding regions. However, considering the striking diversification of the S. smaragdula complex in Asia alone, with no apparent overlap in species distributions between (sub)tropical China and Korea or Thailand [48-50; this paper], it seems unlikely that the Philippine material of Strigula sulcata or Porina crenulata is conspecific with any of the two Chinese lineages detected here, and, therefore, we introduce new species for the latter.  Description: Thallus subcuticular, 1.5-4.5 mm across and 7.5-22.5 µm thick, with entire to crenulate margins, bright green mottled with white, surface smooth, sometimes with black dots or lines. Photobiont a species of Cephaleuros, cells angular-rounded, 8-12 × 4-6 µm. Perithecia exposed, conical to wart-shaped, 0.25-0.5 mm in diameter and 100-190 µm high, black. Excipulum prosoplectenchymatous, 12.5-25 µm thick, brown. Involucrellum 25-62.5 µm thick, carbonaceous, black. Paraphyses unbranched. Asci cylindrical, 40-60 × 3-4 µm. Ascospores eight per ascus, uniseriate, ellipsoid, one-septate, sometimes with one to two oil droplets per cell when fresh, with distinct constriction at the septum and sometimes broken into parts outside the asci, 7.5-10 × 2-2.5 µm. Pycnidia wart-shaped, those producing macroconidia 0.1-0.15 mm, those producing microconidia 0.05-0.1 mm diameter, black. Macroconidia bacillar, zero-to one-septate, 4-6 × 1.5-2.5 µm, with appendage at one end or both ends c. 5-10 µm long. Microconidia fusoid, aseptate, 4-5 × 1.5-2 µm.
Chemistry Notes: Strigula intermedia conforms to the morphology of S. concreta, in particular the rather thin thallus with crenulate margins, the exposed, black perithecia, and the uniseriate asci with short ascospores sometimes breaking into halves [45]. There is at least one other new species belonging to this morphodeme, S. sinoconcreta (see below). Given that both are phylogenetically distinct in both the ITS and the four-marker phylogeny (Figures 1 and 2), we have to assume that S. concreta s.lat. forms a species complex, similar to S. smaragdula and intermingled with lineages representing the S. nitidula morphodeme (see below). The latter agrees with S. concreta anatomically, including the uniseriate asci with short ascospores often breaking into halves, but has a very thin thallus often bordered by a black line, as in Racoplaca [42,45]. The two thallus morphologies are sometimes difficult to distinguish, a feature attributed to leaf characteristics [45], but it appears that numerous lineages are involved. Strigula intermedia represents an intermediate thallus type, thinner than in typical S. concreta but thicker than in typical S. nitidula, and with short, irregular black lines not as distinct as in S. nitidula. Strigula concreta has been described from the Caribbean, with one current synonym from Brazil (S. rugulosa Müll. Arg.), one from Africa (S. atrocarpa Vain.), and two from the Philippines (S. sulcata Vain., Porina crenulata Vain. [42,45]). These names are, therefore, potentially available for cryptic or near-cryptic lineages in the corresponding regions. However, considering the striking diversification of the S. smaragdula complex in Asia alone, with no apparent overlap in species distributions between (sub)tropical China and Korea or Thailand ( [48][49][50], this paper), it seems unlikely that the Philippine material of Strigula sulcata or Porina crenulata is conspecific with any of the two Chinese lineages detected here, and, therefore, we introduce new species for the latter. Diagnosis: The new species was similar with Strigula microspora Lücking in the thallus, but can be distinguished by the shorter ascus and the biseriate ascospores.
Chemistry Notes: Strigula laevis belongs in the S. smaragdula complex and is somewhat similar to the neotropical S. nigrocarpa, due to the blackish perithecia. However, in S. nigrocarpa, the perithecia are sharply delimited from the thallus, and the ascospores are uniseriate in longer, cylindrical asci, and the ascospores are somewhat larger [45]. It is also similar to the neotropical S. minuta Lücking, but that species has much smaller ascospores Diagnosis: The new species was similar with Strigula microspora Lücking in the thallus, but can be distinguished by the shorter ascus and the biseriate ascospores.
Chemistry: No substances detected by TLC.
Habitat and distribution: Collected on living leaves in humid, semi-exposed forests of southern China. Notes: Strigula microcarpa bears some resemblance to the neotropical species S. microspora, but the latter has longer asci (50-70 × 5-6 μm) and uniseriate ascospores [45]. Strigula wandae M. Cáceres & Lücking from the Valdivian Forest in Chile also resembles S. microcarpa in morphology, but has larger ascospores (15-23 × 4-5 μm [45]). While the ITS resolved the new species as an early diverging lineage within Strigula s.str. (Figure  1), the four-marker tree revealed a strongly supported relationship with the S. prasina Müll. Arg. complex (Figure 2).  Notes: Strigula microcarpa bears some resemblance to the neotropical species S. microspora, but the latter has longer asci (50-70 × 5-6 µm) and uniseriate ascospores [45]. Strigula wandae M. Cáceres & Lücking from the Valdivian Forest in Chile also resembles S. microcarpa in morphology, but has larger ascospores (15-23 × 4-5 µm [45]). While the ITS resolved the new species as an early diverging lineage within Strigula s.str. (Figure 1  Diagnosis: Strigula pseudosubtilissima is similar to Racoplaca subtilissima Fée in the thin, laciniate thallus bordered by a black line, but can be distinguished by the uniseriate, short ascospores, as well as the green thallus colour and fully exposed, black perithecia. It differs from the closely related S. nitidula in the rather long, free lobes.
Chemistry: No substances detected by TLC. Diagnosis: Strigula pseudoantillarum is similar to S. antillarum (Fée) Müll. Arg. regarding the ascospores and aggregated pycnidia producing macroconidia, but it has a thin thallus and is also phylogenetically different from the latter.
Chemistry: No substances detected by TLC. Distribution and ecology: The new species was found on living leaves in humid, semiexposed forest habitats of southern China. Notes: Strigula pseudoantillarum is characterised by its aggregated, confluent pycnidia in the centre of the thallus. Together with the similar asci and ascospores, it thus resembles S. antillarum, and the material was originally reported under this name from China [48]. However, inclusion of authentic material of S. antillarum from the Caribbean, the type region, showed that both taxa are phylogenetically distinct and not even closely related (Figures 1 and 2). A consistent morphological difference is found in the thin thallus in the new species, compared to the thicker, somewhat bulging thallus patches in typical S. antillarum [48]. Strigula lacericola P.M. McCarthy, described from Australia, can be distinguished by the pseudostromatic groups of pycnidia and by the smaller ascospores and macroconidia [72]. Notes: At first, this new species resembles Racoplaca subtilissima, but it differs in the green thallus, the fully exposed, black perithecia, and the uniseriate, short ascospores breaking into halves. These latter features point to a close relationship with S. nitidula, which is supported by the molecular data. Strigula nitidula s.lat. often forms marginal black lines, akin to R. subtilissima, but differs in the above features from the latter [45]. Strigula nitidula was described from the Caribbean (Cuba), and the ITS data suggest that Cuban and Chinese material form a homogeneous clade representing a single species (Figure 1), thus far the only species within the family demonstrated to be genuinely pantropical. Both the ITS data and the four-marker tree show that S. pseudosubtilissima is phylogenetically distinct (Figures 1 and 2), also differing from typical S. nitidula morphologically in the rather long, partly free marginal laciniae. Diagnosis: Strigula pseudosubtilissima is similar to Racoplaca subtilissima Fée in the thin, laciniate thallus bordered by a black line, but can be distinguished by the uniseriate, short ascospores, as well as the green thallus colour and fully exposed, black perithecia. It differs from the closely related S. nitidula in the rather long, free lobes.
Chemistry: No substances detected by TLC.
Habitat and distribution: The new species was found on living leaves in humid, semiexposed forest in southern China. Notes: At first, this new species resembles Racoplaca subtilissima, but it differs in the green thallus, the fully exposed, black perithecia, and the uniseriate, short ascospores breaking into halves. These latter features point to a close relationship with S. nitidula, which is supported by the molecular data. Strigula nitidula s.lat. often forms marginal black lines, akin to R. subtilissima, but differs in the above features from the latter [45]. Strigula nitidula was described from the Caribbean (Cuba), and the ITS data suggest that Cuban and Chinese material form a homogeneous clade representing a single species (Figure 1), thus far the only species within the family demonstrated to be genuinely pantropical. Both the ITS data and the four-marker tree show that S. pseudosubtilissima is phylogenetically distinct (Figures 1 and 2), also differing from typical S. nitidula morphologically in the rather long, partly free marginal laciniae.
Chemistry  Diagnosis: Strigula pycnoradians is closely related and morphologically similar to S. acuticonidiarum, but can be distinguished by the shorter asci and smaller ascospores.
Chemistry: No substances detected by TLC.
Habitat and distribution: The new species was found in humid, semi-exposed tropical forest in Thailand.
Additional specimens examined: THAILAND, Nakorn Nayok, Wang Takrai [49]. Both share the general morphology of somewhat grouped or confluent pycnidia also seen in the neotropical S. antillarum, which is, however, phylogenetically separate ( Figure 1). Strigula pycnoradians differs from S. acuticonidiarum in the often radiating rows of pycnidia and the larger mucilaginous cap of the macroconidia. In addition, the ascospores are much smaller (12.5-20 × 3.7-5 µm in S. acuticonidiarum), leading also to much shorter asci (50-65 × 8-12 µm in S. acuticonidiarum [49]). The radiating pycnidia are reminiscent of those of S. novae-zelandiae (Nag Raj) Sérus. and S. oleistrata M. Ford, D.J. Blanchon & de Lange, both known from New Zealand [73]. However, in these two species, the macroconidia are shorter (7-15 µm)  Notes: Strigula sinoconcreta agrees with S. concreta in general habit and in the uniseriate, short ascospores partly breaking into halves, but can be recognized by the bright green thallus (grey-green in S. concreta [45]) and the larger, flatter, basally somewhat spreading perithecia. The new species was phylogenetically distinct from S. intermedia (Figures 1 and 2) and differs from the latter in the thicker thallus not bordered by black lines. Diagnosis: Strigula stenoloba differs from other species in the genus by its very narrow, irregular lobes.
Chemistry: No substances detected by TLC.
Habitat and distribution: The new species was discovered in humid forest in southern China. Individuals mostly grew on living leaves with a rigid texture.
Additional Notes: Strigula sinoconcreta agrees with S. concreta in general habit and in the uniseriate, short ascospores partly breaking into halves, but can be recognized by the bright green thallus (grey-green in S. concreta [45]) and the larger, flatter, basally somewhat spreading perithecia. The new species was phylogenetically distinct from S. intermedia (Figures 1 and 2) and differs from the latter in the thicker thallus not bordered by black lines.
Habitat and distribution: The new species was found on living leaves in humid, semi-exposed forest habitats in southern China.
Notes: Unfortunately, we were unable to obtain molecular data for this material. Yet, the morphological features are so unique that we decided to describe it formally as a new species. At first glance, one could consider this a barely lichenised form or an early ontogenetic stage, but the morphology was consistent across all individuals seen in the type collection and different from anything we have observed in other species. The formation of fully mature perithecia also contradicts the interpretation of this material as an aberrant form of a known species. The new species is characterised by its very thin, widely separated, partly anastomosing lobes. It should not be confused with S. delicata Sérus., a species described from New Zealand [74]. The latter has a much better developed thallus with narrow but regular lobes, although it agrees in ascospore and macroconidial size. Species of Racoplaca also differ in the regular thallus lobes, which are flatter and typically olive-green to brownish [45]. Somewhat similar is R. tremens (Müll. Arg.) S.H. Jiang, Lücking & J.C. Wei, known from Brazil, which agrees in the somewhat irregular, narrow, more or less free laciniae, but the laciniae are completely flat and bordered by a thin, black line, the ascospores are longer (17-23 × 3-5 μm), and the involucrellum is thinner (15-25 μm thick); macroconidia are unfortunately not known [45].  Diagnosis: Strigula stenoloba differs from other species in the genus by its very narrow, irregular lobes.
Chemistry: No substances detected by TLC.
Habitat and distribution: The new species was found on living leaves in humid, semiexposed forest habitats in southern China. Notes: Unfortunately, we were unable to obtain molecular data for this material. Yet, the morphological features are so unique that we decided to describe it formally as a new species. At first glance, one could consider this a barely lichenised form or an early ontogenetic stage, but the morphology was consistent across all individuals seen in the type collection and different from anything we have observed in other species. The formation of fully mature perithecia also contradicts the interpretation of this material as an aberrant form of a known species. The new species is characterised by its very thin, widely separated, partly anastomosing lobes. It should not be confused with S. delicata Sérus., a species described from New Zealand [74]. The latter has a much better developed thallus with narrow but regular lobes, although it agrees in ascospore and macroconidial size. Species of Racoplaca also differ in the regular thallus lobes, which are flatter and typically olive-green to brownish [45]. Somewhat similar is R. tremens (Müll. Arg.) S.H. Jiang, Lücking & J.C. Wei, known from Brazil, which agrees in the somewhat irregular, narrow, more or less free laciniae, but the laciniae are completely flat and bordered by a thin, black line, the ascospores are longer (17-23 × 3-5 µm), and the involucrellum is thinner (15-25 µm thick); macroconidia are unfortunately not known [45].

Discussion
At first glance, the discovery of so many new species in the foliicolous genera Strig ula and Racoplaca may seem surprising. However, based on Santesson's (1952) mono graph, the species concept in foliicolous lichens has been comparatively broad, wit many species presumably widespread, pantropical, or even subcosmopolitan [75]. Thi phylogenetic studies, including the present one, have revealed a great deal of cryptic speciation in these lichen-forming fungi [48][49][50]73]. Given that these studies are partly based on multiple markers (ITS, nuSSU, nuLSU, TEF1-α, RPB2), such revised, much finer species concepts reflect the reality much better than previous, morphology-based concepts, as well as because the newly recognized cryptic lineages often emerge on longstem branches and are not even directly related. Our data also show that ITS alone provides high resolution in Strigulaceae at the species level and the corresponding topology is highly congruent with that of other markers, making the ITS a promising single marker for a broad molecular screening in this group.
The notion that the currently available molecular data for this enigmatic group of lichen-forming fungi are almost exclusively from continental Southeast Asia, at the northern border of the tropical belt, and very few or no such data exist for the Neotropics, the African Paleotropics, the Indian subcontinent, the Indopacific, and Australasia, suggests that species richness in foliicolous Strigulaceae, in particular Strigula s.str., is grossly underestimated. Taking into account the presently available data, with 19 new species recognised based on molecular phylogenies from China, South Korea, Thailand, and New Zealand, and several other species reinstated from previous synonymy [48][49][50]73,76], we anticipate that additional cryptic species will be discovered in these widespread collective taxa in other regions, particularly in Phylloporis obducta, P. phyllogena, Puiggariella nemathora, Racoplaca maculata, R. subtillissima, Strigula antillarum, S. concreta, S. nitidula, S. smaragdula, and S. subelegans. In the hitherto studied species, based on the aforementioned and the present study, the factor of hidden diversity in presumably known species ranges from twofold (e.g., Racoplaca maculata, R. melanobapha; Strigula microspora, S. nitidula) to threefold (e.g., Puiggariella nemathora, R. subtilissima, Strigula concreta) to up to fivefold (S. antillarum, S. smaragdula), for a weighted mean of a threefold increase. These findings align well with Lücking et al. [36], who also showed multiplication factors between two and five in studied species complexes. Given that the data on Strigulaceae are mostly from Southeast Asia, even higher multiplication factors (a weighted mean of up to five) could be assumed when expanding such studies to tropical America and Africa. If we consider the currently known 67 species in this clade of six genera, assume that about one third of these represent species complexes with hidden diversity, and apply the factor five to them, the total species richness in this group can be extrapolated at close to 150 species. This estimate may still be conservative, given that the high levels of cryptic or near-cryptic speciation thus far revealed are concentrated within a narrow region in Southeast Asia. Our current findings, thus, appear to be just the proverbial tip of the iceberg, the actual size of which is difficult to assess.
Our results also put into perspective the recently proposed split of Strigula s.lat. into seven genera, namely Phyllocharis, Phylloporis, Puiggariella, Raciborskiella, Racoplaca, Serusiauxiella, and Strigula [41,47]. One may consider this an attempt at oversplitting. However, as explained by Jiang et al. [41,47], the newly recognised or reinstated genera not only emerge mostly on long-stem branches, indicating a longer time of separate evolution, but also correlate well with various phenotype characteristics. With an estimated total of at least 150 species in these genera overall, the species:genus ratio would result in a little over 20:1, which corresponds well to the overall ratio in lichenised fungi, and fungi and plants in general [46,77], while genera in animals are, on average, less than half this size [77].
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/jof8010002/s1, Table S1: Specimens and sequences for phylogenetic analysis; Table S2: Comparison between Racoplaca macrospora and Racoplaca maculatoides; Figure S1: Phylogenetic tree constructed through ML analyses based on the ITS for Racoplaca.