Reallocation of foliicolous species of the genus Strigula into six genera (lichenized Ascomycota, Dothideomycetes, Strigulaceae)

Strigula sensu lato has been previously defined based on phenotype characters as a rather broad genus including tropical to temperate species growing on a wide array of substrata. In this study, based on a multilocus phylogenetic approach, we show that foliicolous species form six well-delimited clades that correlate with diagnostic phenotype features, including thallus morphology, carbonization of the involucrellum and excipulum, ascospore dimensions, and type of macroconidia. Given the topology, with five of the six clades emerging on long stem branches, and the strong phenotypical differentiation between the clades, we recognize these at the genus level, making mostly use of previously established genus names. Four genera, namely Phylloporis, Puiggariella, Raciborskiella, and Racoplaca, are resurrected for the S. phyllogena, S. nemathora, S. janeirensis, and S. subtilissima groups, respectively, whereas one new genus, Serusiauxiella gen. nov., is introduced for a novel lineage with peculiar macroconidia. The only sequenced non-foliicolous species, S. jamesii, is not closely related to these six foliicolous lineages but clusters with Flavobathelium and Phyllobathelium, revealing it as an additional undescribed genus-level lineage being treated elsewhere. Within the new genus Serusiauxiella, three new species are described: Serusiauxiella filifera sp. nov., S. flagellata sp. nov., and S. sinensis sp. nov. In addition, ten new combinations are proposed: Phylloporis austropunctata comb. nov., P. radiata comb. nov., P. vulgaris comb. nov., Puiggariella confluens comb. et stat. nov., P. nemathora comb. nov., P. nigrocincta comb. nov., Racoplaca maculata comb. nov., R. melanobapha comb. nov., R. transversoundulata, and R. tremens comb. nov. We also report on a peculiar, previously unrecognized growth behaviour of the macroconidial appendages in Strigula s.lat.


Introduction
The genus Strigula s.lat. in the family Strigulaceae (Order Strigulales) comprises about 70 species (Hyde et al. 2013;Lücking et al. 2017). Strigula was first mentioned as a nomen nudum in a summary of the lichen system by Fries (1821), with reference to his Systema Mycologicum 2 years later, where Fries gave a short description (Fries 1823). It is generally assumed that Fries at that time did not include a species, and therefore Santesson (1952) selected S. smaragdula Fr. as type, a name established by Fries 7 years later (Fries 1830). In the protologue, Fries (1823: 535) explicitly cited the name Endocarpon smaragdulum ("… sed ob thallum lobatulum Endocarpi smaragduli facie inter Lichenes locatum"). The latter name had been previously introduced as E. smaragdulum Wahlenb. in Acharius (1803) and corresponds to an unrelated, saxicolous lichen currently known as Myriospora smaragdula (Wahlenb.) Nägeli ex Uloth. It is unlikely that Fries identified his material with that species, and the least nomenclaturally disruptive interpretation is that Fries made a simply comparison and did not give the foliicolous taxon a name. Hence the later introduced name Strigula smaragdula (Fries 1830) indeed is to be considered the first established name for this species.
Numerous species currently included in Strigula s.lat. were described posterior to Fries (e.g. Fée 1825;Montagne 1845;Santesson 1952;Hawksworth et al. 1980;Harris 1995;Aptroot et al. 2003;Roux and Sérusiaux 2004;Lücking 2008;McCarthy 2009a). A detailed modern account of most of the foliicolous species was given in the monograph for Flora Neotropica (Lücking 2008). The separate family Strigulaceae was introduced by Zahlbruckner (1898) for an array of partly unrelated pyrenocarpus lichens growing on leaves. Eriksson (1982) and Vézda (1984) restricted the family to three genera, Strigula, Phylloporis and Raciborskiella, a classification followed in some other works (e.g. Llarens 2003). Species assigned to Raciborskiella grow exclusively on the leaf underside, whereas species assigned to Phylloporis (and previously to Porina) are supracuticular (Santesson 1952;Vézda 1984). The subcuticular foliicolous species classified in Raciborskiella and Strigula thereby associate with photobionts in the genus Cephaleuros (Trentepohliaceae), a semiparasitic alga which obtains carbohydrates through photosynthesis but other nutrients from the host leaves (Nelson 2008;Brooks et al. 2015). While Cephaleuros appears to represent a phylogenetically coherent entity, it is nested within a broadly defined genus Trentepohlia s.lat. whose generic delimitation remains unresolved (Nelsen et al. 2011a;Zhu et al. 2014). However, these ecological differences were not accepted as being sufficient for generic delimitation, and both Phylloporis and Raciborskiella were included in Strigula (Harris 1975(Harris , 1995, a classification accepted in major works (Roux and Sérusiaux 2004;Lücking 2008;McCarthy 2009a;Lücking et al. 2017).
With this broadened generic concept, Strigula s.lat. was further emended by adding numerous corticolous and saxicolous species (Hawksworth et al. 1980;Etayo 1993;Harris 1995;McCarthy 1995McCarthy , 1997McCarthy , 2001McCarthy , 2009aMcCarthy and Malcolm 1996;Roux and Sérusiaux 2004). Notably, the only non-foliicolous species thus far sequenced, the saxicolous Strigula jamesii (Swinsc.) R.C. Harris, was first described in the genus Geisleria and only later transferred to Strigula (Harris in Hawksworth et al. 1980). The type of Geisleria, G. sychnogonoides Nitschke, has been shown through molecular and morphological study to be an apotheciate taxon close to Stictidaceae and entirely unrelated to Strigulaceae (Aptroot et al. 2014). Hence the name Geisleria is not available as genus name for nonfoliicolous species of Strigula s.lat. including S. jamesii.
Strigulaceae have been analysed molecularly in various studies, including their higher level phylogenetic relationships (Nelsen et al. 2009(Nelsen et al. , 2011b and species delimitation in foliicolous representatives (Jayalal et al. 2013;Krishnamurthy and Subramanya 2016;Jiang et al. 2016Jiang et al. , 2017aKrishnamurthy and Kumar 2017). However, to date no broader systematic study of the genus based on molecular data has been carried out. Adding to the somewhat diffuse delimitation of the genus, Harris (1995) even suggested merging Phyllobathelium with Strigula, while other workers kept this genus separate Lücking 2008;McCarthy 2009a) or placed it in a separate family, Phyllobatheliaceae (Eriksson et al. 2004;Eriksson 2005). Molecular data (Nelsen et al. 2009) supported the inclusion of Phyllobatheliaceae within Strigulaceae, but clearly separate from foliicolous lineages of Strigula. Two new genera, Phyllocratera and Flavobathelium, were also added to the family Lücking et al. 1997;Nelsen et al. 2009). Further, the non-lichenized Oletheriostrigula (Huhndorf and Harris 1996) was transferred to Strigulaceae (Lücking et al. 2017), but based on macroconidial morphology, this placement is not justified. A recent systematic and phylogenetic analysis of some Porina-like foliicolous lichens resulted in the recognition of a novel lineage forming a new family, Tenuitholiascaceae, sister to Strigulaceae in Strigulales (Jiang et al. 2020).
In this study, we provide the first broad-scale multilocus phylogeny of Strigula s.lat. focusing on foliicolous taxa, employing markers from the rDNA (ITS, nuSSU, nuLSU) and protein-coding loci (TEF1-α, RPB2). The results prompted us to reclassify foliicolous taxa into six well-delimited genera.

Material examined
A total of 65 specimens of Strigulales belonging to 27 species from included for the ingroup taxon set, representing material from the Americas (Brazil, Cuba, Guatemala) and Asia (China). Additional material from other regions (Costa Rica, Australia) was included for morphological comparison. Specimens examined (and duplicates) are preserved in the Fungarium-Lichenarium of the Institute of Microbiology, Chinese Academy of Sciences (HMAS-L), the herbarium of the Botanischer Garten und Botanisches Museum und Botanischer Garten Berlin (B), the Australian National Herbarium, Centre for Australian National Biodiversity Research (CANB), and the herbarium of the Jardín Botánico Nacional de Cuba in Havanna (HAJB).

Morphological observations
For morphological and anatomical studies, a LEICA M125 dissecting microscope (Leica Microsystems, Singapore) and a Zeiss Axioscope2 compound microscope (Carl Zeiss, Göttingen) were used. Photographs were taken with an AxioCam MRc5 connected to a Zeiss Imager A2-M2 microscope (Carl Zeiss, Göttingen) for microscopic features. Thin-layer chromatography (TLC) (Orange et al. 2001) was employed for the detection of lichen substances in selected specimens. Features of the ascus apex were examined using Lugol's solution, without pre-treatment by KOH (Baral 1987).
PCR reactions were carried out in 25 μl reaction volume and the components used were 2 μl total DNA, 1 μl each primer (10 μM), 12.5 μl 2 × Taq MasterMix, 8.5 μl ddH 2 O. Amplification was performed using a Biometra T-Gradient thermal cycler. Cycling parameters of LSU, ITS and SSU 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 (Rehner and Buckley 2005). 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 (Liu et al. 1999). The PCR products were checked on 0.8% agarose electrophoresis gels stained with ethidium bromide and then sent to the sequencing facilities of Majorbiology Cooperation, China, for sequencing. Sequences were deposited in Gen-Bank (Table 2).
New sequences generated for each locus were first analyzed separately with other sequences obtained from Gen-Bank ( Table 2). Sequences of each marker were aligned using MAFFT 7.402 (Katoh and Toh 2010) and analysed separately (see below) to evaluate potential conflicts. Conflicts were evaluated manually by comparing the topology and position of each of the major clades in relation to their topology and position in the concatenated analysis (see Results).
ML trees were reconstructed for each individual marker and for a concatenated data set including the nuSSU, nuLSU,  SR3  GCT GCG TTC TTC ATC GTT GC  LR71  TAC TAC CAC CAA GAT CTG C  28SR4R  TTT CTG GCA CCT CTA GCC TC  28BYF1  ATG AAG AAC GCG GCA AAA TG  NEWR2 CTA AAC CCA GCT CAC GTT CC  TEF-α, and RPB2. For the individual markers and the concatenated data set, we analysed two taxon sets, a broader one encompassing related Dothideomycetes (Table 3) and a narrower one focusing on Strigulales (Table 2). For this, we used IQ-TREE 1.6.6 (Nguyen et al. 2015) with 1000 bootstrap pseudoreplicates. The best-fit substitution model was selected using ModelFinder (Kalyaanamoorthy et al. 2017): TIM2 + F + I + G4 was selected as the best model for the Dothideomycetes taxon set, and TNe + I + G4 for the Strigulales taxon set. For the ITS alignment, TN + F + G4 was selected as the best model. Bayesian analysis was performed with MrBAYES assuming the general time reversible model including estimation of invariant sites and a discrete gamma distribution with six rate categories (GTR + I + G), for the single-genes and the combined analyses (Ronquist et al. 2012). A run with 5,000,000 generations and employing 20 simultaneous chains was executed. Posterior probabilities above 90% and bootstrap support above 50% are considered significant supports. Analyses of the single markers and the combined data set resulted in basically the same topology.

Molecular phylogeny of the photobiont
Photosymbionts of selected specimens were also analyzed phenotypically and molecularly. ITS nrDNA sequences of the algal partners were amplified using the primers nr-SSU-1780-59 and nr-LSU-0012-39 (Piercey-Normore and Depriest 2001). The newly generated sequences were submitted to GenBank and aligned with fourteen samples of Trentepohliaceae from GenBank representing the three major lichenized genera currently distinguished in that family (Table 4).
The resulting alignment was subjected to randomized accelerated maximum likelihood in RAxML 8 with 1000 pseudoreplicates (Stamatakis 2014), using the GTR-GAMMA model chosen by JModeltest (Felsenstein 1985;Posada 2008) and otherwise with default parameters.

Molecular phylogeny of the mycobiont
Comparison between the individual markers and the concatenated data set revealed topological variation ( Table 5). The Phylloporis and Raciborskiella clades were monophyletic and strongly supported in all instances (no RPB2 data for the latter); its position varied according to marker but this variation was not supported. The Puiggariella and Racoplaca clades were also consistently recovered as monophyletic with strong support (somewhat lower support for the first in TEF1-α). In these cases, however, nuLSU showed a supported conflict for Puiggariella with RPB2 and the concatenated data set, whereas Racoplaca exhibited a supported conflict in the nuSSU with the nuLSU and the concatenated data set. The Serusiauxiella clade was also consistently monophyletic with strong support, but supported conflict was detected between TEF1-α, RPB2 and the concatenated data set. Finally, the Strigula s.str. clade was recovered as monophyletic with strong support in the nuSSU, RPB2 and the concatenated data set, but paraphyletic in the nuLSU and polyphyletic in TEF1-α (Table 5). In the latter case, the subclades behaving aberrantly were the S. nitidula and the S. prasina clades. Given that in 22 of 24 instances, including the concatenated data set, the six clades resulted monophyletic and with strong support, we use the concatenated tree as phylogenetic hypothesis for this group and interpret the few supported conflicts as the result of conflicting individual gene histories in the cases of the protein-coding markers. The few observed conflicts between the nuSSU and nuLSU cannot be attributed to individual gene histories, as both form part of the rDNA cistron that is expected to evolve in tandem. While these conflicts, and the resulting short branches in the concatenated backbone, make it challenging to evaluate the exact relationships between the six clades, they do not affect their interpretation as distinct phylogenetic entities.
For the Dothideomycetes taxon set, the concatenated alignment contained 4371 characters (1272 for nuLSU, 1176 for nuSSU, 886 for TEF1-α, and 1037 for RPB2. The resulting topology was similar for maximum likelihood (ML) and Bayesian analysis (Fig. 1). The analysis confirmed the placement of Strigulales with related orders in Dothideomycetes and the separation of the two families Strigulaceae and Tenuitholiascaceae (see Jiang et al. 2020). In this broad-scale analysis, foliicolous representatives of Strigula formed six clades (A-F) with strong support (between 95% and 100% BS and 100% PP).
For the Strigulales taxon set, with Tenuitholiascus porinoides S.H. Jiang, Lücking & J.C. Wei as outgroup, the concatenated alignment had the same length parameters. In the resulting topology (Fig. 2), the six clades of foliicolous Strigula s.lat. were again supported (between 82 and 100% BS and 100% PP), with the same topology. The six clades were composed as follows ( Fig. 1): clade A included the type species, S. smaradgula, and corresponds to Strigula s.str.; clade B represented a newly recognized group of species with peculiar macroconidia (see below), for which the name Serusiauxiella is introduced below; clade C included two hypophyllous species with large ascospores centered around Raciborskiella janeirensis (Müll. Arg.) R. Sant., and the name Raciborskiella is available for this clade; clade D featured species with non-carbonized perithecial wall, with the name Puiggariella available; clade E comprised the species of the S. subtilissima (Fée) Müll. Arg. aggregate, for which the name Racoplaca exists; and clade F contained the supracuticular species formerly separated in the genus Phylloporis. The non-foliicolous Strigula jamesii clustered in a separate clade, sister to foliicolous Strigula s.lat., together with Flavobathelium and Phyllobathelium.
The ITS-based tree for the genus Serusiauxiella ( Fig. 3) revealed three species-level lineages, with two species being closely related on shallow stem branches and one species being phylogenetically quite distinct on a long stem branch. These are formally introduced below as Ser. filifera sp. nov., Ser. flagellata sp. nov., and Ser. sinensis sp. nov. The latter two are very closely related but differ in 18 substitutions and two indels in the ITS, for a an overall similarity of 95.6%, distinctly below the standard species threshold level of 98.5% (Suppl. File S1). The rather shallow topology between these two species is therefore due to the substantial differences compared to S. filifera (95 substitutions, 26 indels; 73.5% similarity) and the outgroup taxon, S. smaragdula. This indicates an unusually high interspecific and intergeneric variation in the ITS, compared to notably uniformity within species, making the ITS barcoding marker an ideal locus to delimit species in foliicolous Strigula s.lat.

Morphology and anatomy of foliicolous Strigula s.lat
Specimens of Strigula s.lat. examined in this study are all characterized by a Strigula-type ascus apex, i.e. bitunicate, with a short tholus or ocular chamber at the tip, by which this clade can be distinguished from the sister group Tenuitholiascaceae (Jiang et al. 2020). As outlined above, the six foliicolous Strigula clades correlate with various phenotype characters, including photobiont type, thallus disposition, perithecial wall structure, ascospores, and macroconidida, particularly the nature of their appendages (Fig. 4). Thallus morphology was found to be strongly correlated with phylogeny. Apart from the supracuticular species in clade F (Phylloporis), the subcuticular species in the remaining five clades exhibit four distinctive thallus types: clades A (Strigula s.str.) and B (Serusiauxiella) feature mostly thickened, bright green to grey-green thalli, whereas clade C (Raciborskiella) has very thin, bluish grey thalli composed of dispersed patches and clade E (Racoplaca) includes species with olive-brown to dark olive-green, somewhat metallic thalli forming delicate lobes delimited by thin black, continuous or interrupted lines. Finally, clade D (Puiggariella) includes closely related species with a continuous to lobed, pale greenish thallus with somewhat folded surface and abundant white papillae.
The involucrellum of the perithecia is mostly carbonized to various degrees but consistently uncarbonized in clade D (Puiggariella). Ascospores show similar variation in most clades, ranging from small to medium-sized (less than 30 μm long) and in shape from oblong to fusiform. Exceptions are clade C (Raciborskiella), with large ascospores (30-70 μm long), and clade F (Phylloporis), with consistently small ascospores (up to 12 μm). The ascospores of Raciborskiella also display terminal gelatinous appendages. Likewise, macroconidial appendages vary from short to medium-sized in clade A, C, D, and E, whereas clade B (Serusiauxiella) is characterized by a peculiarity of its macronidial appendages. To elucidate this, we examined appendage length for the various lineages. Apparently, fresh mounts of any of the species in Strigula s.lat. feature macroconidia with rather short appendages only (15-25 μm). Depending on the lineage, after a given time period, these appendages may grow  to considerable lengths. In all genera except Serusiauxiella, little change in appendage length was observed after 1 h, but after 1 day, appendages would often have extended to 50 μm or more. In contrast, in all species of Serusiauxiella, the appendages extended to up to 70 μm already after 1 h, much faster and to greater lengths than in the other genera.
The functional role of this behaviour is unclear; since the appendages supposedly not contain plasma, they cannot be considered germ tubes, but their slow to rapid growth may have to do with fixation of the macroconidia on the leaf surface. Notably, the ascospores do not exhibit such a behaviour.

Photobiont of selected foliicolous Strigula s.lat
Most foliicolous Strigula s.lat. clades (Puiggariella, Raciborskiella, Racoplaca, Strigula) feature a Cephaleuroslike photobiont (Fig. 5a), with anastomosing filaments or groups of cells usually arranged in one to several layers; these lichens grow subcuticularly and are not separable from the leaf. The Phylloporis clade features Phycopeltis as photobiont, which anastoming filaments in one layer and forming regular radiating plates or irregular nets (Fig. 5b); these lichens grow supracuticularly and can be easily separated from the leaf surface. A different, more Trentepohlia-like photobiont (Fig. 5c) was found in Serusiauxiella, which is remarkable as these species exhibit a similar thallus morphology and subcuticular growth as in Strigula s.str. Notably, thalli of Serusiauxiella can be removed from the leaf surface more easily than Strigula s.str. and the other subcuticular genera. ITS sequence data confirmed our morpho-anatomical assessment of the photobionts, placing the photobionts of the studied lichens variously in proximity to lineages classified as Cephaleuros, Phycopeltis, or Trentepohlia (Fig. 6). However, these generic assignments are provisional, given that broader studies on trentepohlioid algae indicate that of the morphologically delimited genera within Trentepohliaceae, only Cephaleuros and Stromatochroon are monophyletic and the others (Phycopeltis, Printzina, Trentepohlia) are polyphyletic (Nelsen et al. 2011a;Zhu et al. 2017).

Assessment of non-foliicolous Strigula s.lat
The only non-foliicolous species of Strigula s.lat. sequenced thus far, Strigula jamesii, grows on rocks and is characterized by having 3-septate ascospores, with the second cell often the widest and 3-septate macroconidia (Swinscow 1967;Roux and Sérusiaux 2004). This species does not cluster with the foliicolous taxa and the underlying topology suggests that non-foliicolous taxa do not represent Strigula s.str. or related foliicolous genera. No other generic name exists for these taxa and the name Geisleria cannot be used (see above). Given that non-foliicolous species may represent more than one entity (Roux and Sérusiaux 2004;Lücking et al. in Hyde et al. 2013), and only one species has been sequenced so far, we refrain from proposing a new genus for the single sequenced species here. The non-foliicolous species currently classified in Strigula s.lat. are being treated in a separate paper (Jiang et al., in prep.).

Discussion
Our analysis revealed the division of foliicolous Strigula s.lat. into six well-supported clades. Each clade is characterized by one or several unique synapomorphies. Clade A includes the type species Strigula smaragdula and is characterized by usually thickened, bright green to greygreen thalli with a Cephaleuros photobiont, with carbonized perithecial walls, small to medium-sized ascospores and short to medium-sized macroconidial appendages without rapid growth in microscopic mounts. Most foliicolous species of Strigula belong in this clade. Clade B is an entirely novel lineage recognized here, under the name Fig. 3 Phylogenetic tree showing species delimitation in Serusiauxiella, constructed through ML analysis based on ITS sequences with an alignment length of 447 bp. Maximum likelihood bootstrap probabilities above 50% (left) and Bayesian inference posterior probabilities above 90% (right) are shown at nodes (ML-BP/B-PP). The tree was rooted with Strigula smaragdula Serusiauxiella, on a very long stem branch, with three species new to science. The three species are morphologically rather uniform, resembling e.g. Strigula microspora Lücking in habit and ascospore size. However, the appendages of the macroconidia in this clade are unique, in microscopic mounts quickly growing to up to 70 μm and making the macroconidia appear as if suspended on long, gelatinous strands. In addition, the photobiont in this clade appears to be different from other subcuticular Strigula s.lat., clustering with Trentepohlia and also morphologically resembling that genus rather than Cephaleuros. Clade C includes two species previously classified under the name Strigula janeirensis, but representing two lineages separated by ascospore size (see below). Both agree in the hypophyllous growth, forming very thin, dark bluish grey thalli with Cephaleuros photobiont, carbonized perithecial walls, large ascospores (30-70 × 5-7 μm), with appendages at both ends (see also Roux and Sérusiaux 1995), and macroconidia with short to medium-sized appendages. In addition, the paraphyses in this clade are slightly branched and anastomosing (Santesson 1952). These differences were not accepted for generic delimitation and therefore Raciborskiella was subsumed under Strigula (Harris 1975). However, this group is welldelimited both phylogenetically (longest branch in the tree) and phenotypically and hence the genus Raciborskiella is reinstated here. Another hypophyllous species previously included in Raciborskiella, S. prasina (Santesson 1952), has strongly deviating phenotype features fitting with Strigula s.str. (clade A) and is also included there phylogenetically.
Clade D is a small clade characterized chiefly by thallus morphology (with somewhat folded surface and abundant papillae) and pale, non-carbonized perithecial walls. The principal species in this clade, S. nemathora, is indeed quite distinct from all other foliicolous Strigula species, and our data suggest that it is a collective taxon, comprising several closely related species. The name Puiggariella is taken up for this genus. Clade E comprises the species of the Strigula subtilissima group, with thin, finely lobed thalli of olivebrown to dark (olive-)green color which feature a characteristic, thin black line along the lobe margins. Ascospores, macroconidia and other characters are otherwise similar to clade A (Strigula s.str.). The only superficially similar species in clade A is S. nitidula, which differs in the bright green thallus and the completely exposed, black perithecia,  Lücking (1992) whereas those of the S. subtilissima group are covered by thallus up to the ostiole. We resurrect the name Racoplaca for this clade below. Clade F comprises the small Strigula phyllogena group, with thin, supracuticular thalli featuring a Phycopeltis photobiont, carbonized perithecial walls, and comparatively short asci and small ascospores. The genus Phylloporis was established for this group by Clements (1909), but the species were treated under Porina by Santesson (1952) and eventually included in Strigula by Harris (1995). Phylogenetically, the group is also quite distinctive. It is therefore justified to take up the genus Phylloporis for this clade.
Overall, all foliicolous lineages of Strigula s.lat. form a monophyletic clade, separate from the only non-foliicolous species currently placed in Strigula, S. jamesii. This indicates that most non-foliicolous species are not actually part of Strigula s.lat. but have to be redispositioned otherwise (Jiang et al., in prep.). For the foliicolous lineages, two alternatives exist to reflect the division into six clades taxonomically: (1) recognizing the six lineages at genus level or (2) separating them at subgeneric level. The phylogeny itself does not provide conclusive data to favor one or the other solution, although the long branches of three of the six lineages (Phylloporis, Raciborskiella, Serusiauxiella) indicate a substantial period of separate evolutionary history that would justify generic status, which in turn would then require to also recognize the other three lineages as separate genera.
The concept of the so-called "evolutionary significant unit" (ESU), in tandem with the notion that the binomial offers the inclusion of phylogenetic affinities, provides another approach to solve this matter. Originally, the term was coined to denote species-or population-level lineages that due to unique genetic makeup deserve conservation status (Moritz 1994;Casacci et al. 2014;Cornejo et al. 2017). However, the concept can be expanded to genus-level taxonomy, with the argument that ESUs above species level should be reflected in their nomenclature. The standard nomenclature for species is the binomial, including the name of the genus and the specific epithet. Infrageneric subdivisions are not reflected in the binomial and there are not established standard citation forms of infrageneric classifications in botanical or mycological nomenclature, in contrast to zoological nomenclature. Therefore, if the six lineages above would be recognized at infrageneric level (e.g. as subgenera), standard listings or name citations would not reflect this classification, as the species would continue to be listed and cited under the genus Strigula. Since these lineages strongly correlate with phenotype features and are "recognizable" as taxonomic units, we consider them Scale bars: a = 20 μm, b, c = 10 μm to represent ESUs above species level and the only way to recognize this properly in their nomenclature is to establish generic rank for each lineage. We therefore introduce a revised generic concept below for foliicolous representatives of Strigula s.lat.

Key to genera of Strigulaceae containing foliicolous species
In the key below, genera not explicitly treated in this study are given in brackets; the remaining genera are treated in the order of the phylogeny (Fig. 1).

Strigula smaragdula
The original material is from Nepal and India; Santesson (1952) only cited the material from Nepal, as "holotype", but technically it constitures a lectotypification (ICN Art. 9.10), which means that the bulk of tropical material may not represent that species.
Distribution and ecology: Strigula smaragdula in its current circumscription is presumed to be pantropical (Santesson 1952;Lücking 2008). However, molecular studies indicate that it might deal with a collective taxon (Jiang et al. 2017a;Woo et al. 2020). For instance, the European material is now being recognized as separate species, S. buxi (Roux and Sérusiaux 2004). Numerous synonyms exist to potentially cover segregate species in various geographic regions (Santesson 1952;Lücking 2008 Diagnosis: Differing from Strigula s.str. in the Trentepohlia photobiont and the macroconidial appendages quickly growing to substantial length (up to 70-100 μm) in squash mounts.
Etymology: It is with great pleasure that we dedicate this new genus to Dr. Emmanuël Sérusiaux, in recognition of his important contributions to lichenology, foliicolous lichens and the genus Strigula.
Chemistry: No substances detected by TLC. Notes: As outlined above, the species of this new genus superficially resemble Strigula s.str., in particular S. microspora and similar taxa. Its most distinctive features are the macroconidial appendages becoming extremely long and the photobiont morphologically and phylogenetically representing a species of Trentepohlia rather than Cephaleuros. Thus far the genus includes three species, all new to science and so far only known from (sub-)tropical Asia.
Etymology: The epithet refers to the macroconidal appendages becoming long and filiform in squash mounts.
Chemistry: No substances detected by TLC.

Ecology and distribution:
The new species grows on leaves in wet tropical forest in southern China. At present, it is known only from the type locality, but documented by numerous specimens.
Etymology: The epithet refers to the macroconidial appendages resembling flagellae.
Chemistry: No substances detected by TLC. Ecology and distribution: Serusiauxiella flagellata is known through a number of collections from around the type locality in South China. It grows on leaves in wet tropic forest.
Notes: Serusiauxiella flagellata differs from Ser. filifera in the distinctly more exposed, hemispherical to wart-shaped perithecia and the ascospores with more distinctly tapering ends. Phylogenetically, the two species differ in numerous substitutions in the ITS (see above).
Etymology: The epithet refers to the discovery of this new species in China.
Chemistry: No substances detected by TLC. Ecology and distribution: This species is known from several collections from the type locality and its vicinity in South China. It grows on leaves in wet tropic forest.

3
Notes: This new species is very similar to Serusiauxiella filifera and the only tangible difference appears to be the shorter asci. The differences in the ITS are, however, substantial, with an overall similarity of only 95.6% (see above).
Chemistry: No substances detected by TLC.

Key to species of Raciborskiella
Chemistry: No substances detected by TLC. Distribution and ecology: This species is apparently eastern paleotropical, known from the type from Indonesia (Java) and from mainland China, there found in humid, semiexposed forest habitats.

Key to species of Puiggariella
Chemistry: No substances detected by TLC. Distribution and ecology: The distribution of Strigula nemathora is difficult to ascertain, but it may deal with a neotropical taxon, as this morphotype is mostly known from that region.
Notes: The name Strigula nemathora is here restricted to a taxon with distinctly laciniate, often widely spaced lobes, a frequent morphotype commonly found in the Neotropics (Lücking 2008). More data are required to disentangle this species complex and evaluate the status of the various synonyms of S. nemathora proposed by Santesson (1952  Type: VIETNAM. "Tonkin", Mont Ba Vi; 1889, B. Balansa 3 (holotype, G-G00292245!).
Chemistry: No substances detected by TLC. Distribution and ecology: This is likely an eastern pantropical species, thus far known from Vietnam and (sub-) tropical southern China.
Notes: Based on our molecular data, Strigula nemathora s.lat. as defined by Santesson (1952) comprises more than one species. The two sequenced clades differ in thallus morphology and distribution. We have therefore restricted the name S. nemathora to the neotropical material with typically separate, somewhat irregular laciniae and take up the name S. nigrocincta for the eastern paleotropical material with radiating, more contiguous lobes. The material from China is quite uniform in this respect and agrees well with the type from China. The epithet nigrocincta is a bit misleading, as it refers to the blackened area of the leaf around the thalli. This effect is likely phorophyte-specific and not a feature of the lichen.
Chemistry: No substances detected by TLC. Distribution and ecology: This is presumed to be a pantropical species; it has been reported from Guiana, Peru, French Guiana, Brazil, Sierra Leone, and Borneo (Santesson 1952;Lücking 2008).
Notes: Racoplaca melanobapha is characterized by thallus composed of dichotomously branched lobes bordered by a thin, black, interrupted line that form an attractively patterned reticulum. Racoplaca subtilissima differs in the