Taxonomic revision of the Plagiothecium curvifolium complex

Supported by the examination of specimens from the entire range and by the analysis of type specimens and the diagnosis of individual names, morphological and genetic studies of the Plagiothecium curvifolium complex resulted in the conclusion that this taxon should be recognized as four separate taxa. In addition to P. curvifolium s.str., there is a variety that is proposed as a new combination–P. curvifolium var. recurvum; resurrection of the forgotten P. decursivifolium; and the description of a new species–P. imbricatum. The features that distinguish individual taxa focus primarily on: plant size; arrangement of leaves on the stem; the symmetry, dimensions, shape, concavity and folding of leaves; cell length; serration of the leaf apex; the shape of the decurrencies; the length of the sporophyte and the shape of the operculum. For all described taxa, the distribution, ecological preferences, key to their identification and detailed photographic documentation have been provided.


Introduction
Plagiothecium curvifolium Schlieph. ex Limpr., a fairly widespread species in the Northern Hemisphere, is common in Europe, less frequentin North America and Asia, and considered doubtful in North Africa [1][2][3].
This species was described 125 years ago in Die Laubmoose Deutschlands, Oesterreichs und der Schweiz [4]. Limpricht, in the diagnosis, does not indicate any specimen as a holotype. However, in the protologue (Fig 1) he indicated a Karl Schliephacke collection as the one on which this taxon was described: "Schliephacke sammelte die Exemplare, die er 1880 vertheilte, im Thüringerwalde bei der Schmücke in feuchten Nadelwäldern am 29. Juli 1880." Limpricht [4] listed specimens from this collection in the diagnosis, along with other analyzed materials (Fig 1).
After publication, at the end of the 19 th and the beginning of the 20 th century, this species was not distinguished by researchers [e.g., [5][6][7][8]. On the other hand Meylan [9], shortly after its publication, proposed to change its status, recognizing it as a variety of P. denticulatum-P. denticulatum var. curvifolium (Schlieph. ex Limpr.) Meyl., additionally he proposed a distinctive form-P. denticulatum var. curvifolium fo. albescens Meyl. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 At the turn of the 19 th and 20 th centuries, several varieties of P. denticulatum were published, which are considered as synonyms of P. curvifolium sensu lato [2,4,10,11]. For example, Dixon [12,13], citing Spruce [14], wrote about plants (at the rank of variety) characterized by downward curved leaves. He undoubtedly made a mistake, because Spruce [14] described this as a subspecies-P. denticulatum subsp. aptychus Spruce. Despite this error, the morphological features caused Dixon [10] to consider this taxon a synonym of P. curvifolium. On the other hand, Grout [15] proposed to change the status of this taxon to a variety-P. denticulatum var. aptychus (Spruce) Grout, stating that these plants, e.g., have"leaves smooth and shining, not shrunken (. . .) apices usually more or less curved downwards", which strongly suggests plants currently understood as P. curvifolium sensu lato.
In the second half of the 20 th century, the understanding of P. curvifolium changed significantly, which was related to three important taxonomic revisions [27][28][29]. These studies strongly influenced the way this taxon was perceived by successive generations of bryologists, because it was proposed to abandon the recognition of any subspecies, varieties, forms and subforms of this taxon, or even its synonymy with P. laetum [27][28][29].
Ireland [27], in his revision of specimens originating in North America (Canada and U.S. A.) and subsequent articles [30,31], treated species of Plagiothecium very broadly. In addition, he proposed many synonyms, including P. curvifolium with P. laetum, which has led to a significant reduction in the number of taxa recognized in this genus from North America. This point of view was adopted and maintained in this area over the next decades [32,33], and it did not change until the beginning of the 21 st century [2,34].
Another revision based on material from Japan was done by Iwatsuki [28]. This author proposed a synonymization of the specimens described as P. laetum by Sakurai [35] with P. curvifolium and the exclusion of the former from the Japanese bryoflora. This point of view was adopted in subsequent studies by Japanese scientists [36,37]. But, on the other hand, Iwatsuki [28] mentioned that "P. curvifolium may be a variety of P. laetum" which is reflected in another Japanese moss checklist, where Iwatsuki [38] proposed to recognize P. curvifolium as a synonym P. laetum. So far, this approach has been recognized and accepted in Japan [39].
The third revision, proposed by Lewinsky [29], was of specimens from Denmark and has been widely adopted in Europe. In this publication, Lewinsky [29] disagreed with Ireland [27,30,31] and Iwatsuki [28,36,38] and distinguished both of the above-mentioned species. However, she pointed out that P. curvifolium is very variable and can sometimes be mistaken with P. laetum or even P. denticulatum [29].
Only the end of the 20 th century brought a taxonomic article proposing a new variation of the described species-P. curvifolium var. hypnophyllum Ukrainskaya [40]. However, this variety is currently not accepted [41][42][43], and recently only one form of this taxon is recognized-P. curvifolium fo. julaceum [2,3].
In the diagnosis of P. curvifolium, Limpricht [4] described in detail the new species, writing, e.g., that its turf is creeping, loose and clearly glossy; color is yellowish green to light green; leaves 1.6-2.0 × 0.75-0.9 mm, asymmetrical, ovate, overlapping, tightly arranged on the stem and more or less downwards; margin is entire; cells 9 μm wide and they are 8-16 times as long as they are wide; seta up to 2 cm; capsule horizontal or inclined.
Taking into account the above facts, research was undertaken aimed at a taxonomic revision of Plagiothecium curvifolium sensu lato throughout its entire geographical range.
Additionally, other specimens of P. curvifolium that had been previously genetically tested (2) were borrowed and examined (CP10515, CP10621) from the herbarium of C.

DNA isolation, amplification and sequencing
Leafy stems of mosses were cut from dried material. Approximately 20 mg of dry tissue from each specimen in duplicates was placed in a 1.5 ml Eppendorf Safe-Lock tube and frozen (-20˚C) for homogenization. Tissue homogenization was performed using a hand-held stainless steel homogenizer (Schlüter Biologie, Eutin, Germany). Total DNA was extracted using the GeneMATRIX Plant & Fungi DNA Purification Kit (Eurx, Gdansk, Poland) following the manufacturer's protocol. DNA extracts were quantified with a BioDrop DUO Spectrophotometer (BioDrop Ltd, Cambridge, UK). From the duplicates, the sample with higher quality DNA (1.7-1.9 OD 260 /OD 280 ) was selected for further analysis.
For each sample, all markers were amplified by PCR in a few replicates to obtain high quality amplicons for sequencing. PCR for ITS and rpl16 was performed using primers and reaction conditions as described in Wolski, Nowicka-Krawczyk [52]. To obtain the best results in trnK-psbA (matK) amplification, three parallel reactions were performed. Two reactions amplified the region in overlapping fragments using following set of primers: 1) trnK-F/matK-1307 and 2) trnk-450F/psbARbryo; while for some cases third reaction had to be performed to obtain high quality of a amplicon containing the trnK-psbA spacer using 3) trnK-2284F/ psbARbryo set of primers. The reaction conditions for first two amplifications PCR products were visualized on an agarose gel (1.5%, 90 V, 40 minutes) stained with GelRED™ fluorescent dye (Biotum, Fremont, CA, USA) and two replicates of each marker per sample were chosen for sequencing. Amplicons after PCR reaction were cleaned using Syngen Gel/PCR Mini Kit (Syngen Biotech, Wroclaw, Poland) according to the manufacturer's protocol. Samples were sequenced with Sanger sequencing using primers from amplification by SEQme s.r.o. company (Dobris, Czech Republic). The obtained sequences were assembled in Geneious 11.1.5 (Biomatters Aps, Aarhus, Denmark) (http://www.geneious.com). The sequences were submitted to the NCBI GenBank database (www.ncbi.nlm.nih.gov) under the accession numbers ON202485-ON202493 for ITS, ON228316-ON228324 for trnK-psbA (matK), and ON228307-ON228315 for rpl16.

Phylogenetic analyses
Phylogenetic analyses of studied specimens and other species in the Plagiothecium group were performed based on concatenated ITS-matK-rpl16 sequences matix (4127 bp). Voucher information for the specimens included in this study, with corresponding GenBank accession numbers, are presented Table 1. Sequences were aligned using the MAFFT v. 7 web server [53] (http://mafft.cbrc.jp/alignment/server/) where the auto strategy was applied, the scoring matrix of 200PAM with Gap opening penalty of 1.53, UniREf50 for Maft-homologs and Plot and alignment with threshold of 39 score were set. The obtained alignments were checked for poorly and ambiguously aligned regions and small corrections were made by eye. The evolutionary models were calculated using PartitionFinder 2 software [54] chosen according to the Akaike Information Criterion. Summary of partitions for ITS-matK-rpl16 matrix evolutionary model selection and phylogenetic interference were submitted to figshare online database (10.6084/m9.figshare.16570353.v1).
Phylogenetic calculations were performed using maximum likelihood analysis (ML) in the IQ-TREE web server [55] (http://iqtree.cibiv.univie.ac.at/) with the ultrafast bootstrap (UFBoot) pseudolikelyhood algoritm [56] and 10000 replicates; and Bayesian inference (BI) in MrBayes 3.2.2 [57] where two parallel Markov chain Monte Carlo (MCMC) runs for four million generations each, with trees sampled every 1000 generations were performed. The average standard deviation of split frequencies in both cases remained below 0.01 for the last 1000 generations and posterior probabilities were estimated from the 50% majority-rule consensus tree after elimination of the first 25% of samples as burn-in. Raw data sequences, the alignment file, evolutionary model set for partitions and tree files were submitted to figshare online database (10.6084/m9.figshare.16570353).

Results
Studied specimens (P. curvifolium JE04004091, HBG02115, PC01322640, WRSL; P. curvifolium fo. julaceum C-M-9120, MO3974490; P. denticulatum var. recurvum JE04004201), all analyzed names [4,9,14,15,18,21,23,24,40,49] as well as test specimens from the entire range of the studied taxon (C, F, HBG, JE, LOD, MO, PC, SZUB-B, UBC, VLA, WRSL, YU) are different from each other in terms of macro-and microscopic qualitative and quantitative characteristics. The most important of them differentiating individual taxa include, e.g., the size and color of the turf; arrangement of leaves on the stem; symmetry, dimensions and shape of the leaves; cell length; serration of the apex; the shape of the decurrencies; the length of the sporophyte and the shape of the operculum. This changeability is reflected in the genetic variability of the studied complex.

Genetic analyses
Phylogenetic analyses (Fig 2) based on the concatenated ITS-matK-rpl16 placed studied specimens within the branch of a Leptophyllum sect. clade, next to "P. curvifolium" analyzed by Wynns [2], with a very high branch support by Bayesian inference (PP = 1) and maximum likelihood (B = 100). Moreover, with high branch support from both analyses (PP = 1; B = 99) the topology of the tree revealed division of the Wolski specimen clade into two subclades containing different morphotypes. As the first subclade was monospecific, the second possessed internal division highly supported by BI (PP = 0.98) (Fig 2).  Wolski 424 (LOD 15015) were outside the clade "P. curvifolium" (Fig 2) analyzed by Wynns [2]. During this revision the type of P. decursivifolium (PC0132686), deposited at Herbarium PC, was tested. Its quite modest description on the envelope in part matches the diagnosis (Fig 3). The revision of this material showed that it clearly belongs to the P. curvifolium complex, also this is confirmed by the diagnosis [58]. Morphologically, it fits to the specimens (CP0010515, CP0010621) genetically examined by Wynns [2] and also is identical to the P. curvifolium fo. julaceum type (C-M-9120, MO3974490). All the above-mentioned specimens are characterized by, e.g., asymmetric, lanceolate, longitudinally folded, clearly concave, with a wide base and often torn leaves; decurencies usually forming very distinct auricles; capsules inclined to horizontal; and operculum rostrate.
Taking into account the above facts and accordance to Art. 9.3 of the Shenzhen Code [59], "A lectotype is one specimen or illustration designated from the original material (Art. 9.4) as the nomenclatural type, in conformity with Art. 9.11 and 9.12, (. . .)" specimen PC0132686 from Herbarium PC should be designated as the lectotype of P. decursivifolium.
Whereas, in the P. curvifolium fo. julaceum case taking into account the existence of two original materials of this taxon and on the basis of Art. 9.6 of the Shenzhen Code [59] "A syntype is any specimen cited in the protologue when there is no holotype, or any one of two or more specimens simultaneously designated in the protologue as types" all the above-mentioned specimens should be regarded as syntypes. Additionally, accordance to Art. 9.3 of the Shenzhen Code [59] quoted above specimen C-M-9120 from Herbarium C should be designated as the lectotype of P. curvifolium fo. julaceum.  Wolski 109,and Wolski 199 represent the material with a fairly flat turf with leaves slightly curved towards the ground; leaves that are symmetrical or almost symmetrical, lanceolate, concave and very often incurved; decurrencies not forming distinct auricles; capsules inclined to horizontal; and an operculum conical, obtuse. These specimens morphologically correspond to the type of P. curvifolium (JE04004091, HBG02115, PC01322640, WRSL). While, the specimens: Wolski 358, Wolski 39, and Wolski 145 represent the material, among others, with a turf with leaves strongly curved towards the ground; leaves that are asymmetrical, hooked, concave or slightly concave, very often folded; decurrencies not forming distinct auricles; capsules inclined to horizontal, and operculum rostellate, and these specimens fit perfectly to the P. denticulatum var. recurvum type (JE04004201) and other specimens like P. curvifolium var. hypnophyllum (VLA) type.
Warnstorf [60], when proposing a new variety, P. denticulatum var. recurvum, described it as being distinguished by its hooked leaves. However, he did not indicate any original materials or types. He only wrote that "Auf nacktem Boden in Kiefernschonungen vor Altruppin!!" (Fig 4). During this revision, a specimen (JE04004201) was found in Herbarium JE, which, as stated on the envelope, was collected in September 1880 and signed by C. Warnstorf as "P. denticulatum var. recurvum" (Fig 5). Taking into account the above facts and that the location and habitat description is consistent with the diagnosis, specimen JE04004201 should be considered as original material on the basis of which Warnstorf [60] described the new variety. Taking into account the above facts and in accordance to Art. 9.3 of the Shenzhen Code [59], quoted above specimen JE04004201 form Herbarium JE should be designated as the lectotype of P. denticulatum var. recurvum.
When Limpricht [4] described P. curvifolium he proposed the synonymization of P. denticulatum var. recurvum with this species. In the diagnosis, the author did not indicate any specimen or figures as a type, but stated that the new species was described on the basis of materials  [4]. In the same year, according to the author, these specimens were sent to many herbaria (Fig 4).
During this revision, in Herbarium JE, a specimen (JE04004091) was found which was collected July 29, 1880 by C. Schliephacke and signed as "Plagiothecium curvifolium Schlieph." (Fig 6). At the same time, in many European herbaria (e.g., HBG02115, PC01322640, WRSL) materials originating from Herbarium Europeanum Dr Karl Baenitz are deposited and labeled as P. denticulatum var. recurvum = P. curvifolium (Fig 7). Data on the envelopes of all the above-mentioned specimens indicating, e.g., Karl Schliephacke as their collector; date (July 29, 1880) and place of collection are consistent with the diagnosis of P. curvifolium [4], which indicates that they came from one collection and are the original materials used by Limpricht to describe a new taxon. Taking into account the above facts and on the basis of Art. 9.6 of the Shenzhen Code [59] quoted above all the above-mentioned Karl Schliephacke specimens should be regarded as syntypes.
All mentioned specimens are described quite similarly and contain data characterizing the Schliephacke collections [4], however with the difference that specimens JE04004091 (as diagnosis) is described in German, it contains the seal 'Herbarium Karl Schliephacke Osterfeld' and is not as detailed as the other materials, but at the same time it more accurately reflects the data contained in the diagnosis. Whereas, the remaining materials are described in identical handwriting in Latin. Considering the above facts, and that specimen (JE04004091) has a large well-preserved turf of material, and according to Art. 9.3 of the Shenzhen Code [59] cited above specimen JE04004091 form Herbarium JE should be designated as the lectotype of P. curvifolium.
Genetic analyzes also show the third clade, which is composed of two specimens: Wolski 175 (LOD 15014) and Wolski 424 (LOD 15015) (Fig 2), these materials represent a new species-P. imbricatum, which, compared to the whole P. curvifolium complex, is characterized by a unique combination of gametophytic features.
Thus, the revision of herbarium materials, supported by the examination of types and original collections; a detailed analysis of the diagnoses of individual names; and the history of the described taxon supported by DNA research allowed four separate taxa within P. curvifolium sensu lato to be distinguished. Two of them are P. curvifolium and a variety of it which is proposed as a new combination; one is the resurrected P. decursivifolium; and the fourth is a newly described species-P. imbricatum. Description: Plants medium-sized, yellow-green to green, with a metallic luster (Fig 8); stems complanate-foliate, gently imbricate, 1.5-2.5 cm long; leaves lanceolate to ovate-lanceolate, concave, slightly curved towards the ground. Symmetrical or almost symmetrical leaves dominating, those from the middle of the stem 1.7-2.7 (M 2.2) mm long, and the width measured at the widest point 0.7-1.5 (M 1.0) mm; margin incurved, delicately on both sides or strongly on one side; the apex acuminate, usually not denticulate; costae two, variable, but usually strong, usually to 1/3 leaf length, reaching 0.  Distribution and ecology: the presently known range of Plagiothecium curvifolium var. curvifolium is Europe, Asia and North America. In this area, it is recorded mainly from coniferous forests (dominated by Pinus, Picea or Abies), less often from deciduous forests (dominated by Fagus or Quercus). This taxon most often is found on epigeic habitats (on soil, humus), less often is it epiphytic (bark of Pinus, Picea, Fagus, Quercus, Alnus), epixylic (logs) or epilithic (sandstones) (S1 Text).
Distribution and ecology: the currently known range of this taxon are Europe, Asia and North America. In this area it is mainly recorded from coniferous forests (mainly dominated by Picea, less often by Pinus, Abies or Pseudotsuga). Also, most often P. curvifolium var. recurvum is found in epigeic habitats (on mineral soil, humus, litter), less often is it epiphytic (bark of Picea or Quercus, Fagus), epixylic (log, stump) or epilithic (rock) (S1 Text).
Distribution and ecology: the presently known range of P. decursivifolium are Europe, Asia and North America. In this area, it is mainly recorded from mixed and coniferous forests (dominated by Picea), less often from Pinus (including also Pinus monocultures), Abies, Fagus or Alnus dominated forests. This species most often was found in epigeic habitats (on humus, mineral soil), less often was it epiphytic (Betula pendula, Quercus sp., Pinus sylvestris, Picea abietis, Abies alba), epixylic (trunk, log), or epilithic (rock) or in anthropogenic habitats such

PLOS ONE
as old railroads (S1 Text). Additional specimens examined: as a supplementary materials (S1 Text).
Description: Plants small, bright-green to green, with a metallic luster (Fig 14); stems clearly julaceous and imbricate, 0.7-1.5 cm long, densely foliate; two types of leaves: symmetrical and asymmetrical. The symmetrical ones folded, lanceolate, concave, sometimes strongly cracked at the base, asymmetrical ones ovate, slightly concave or flat, both types of leaves identical in size, those from the middle of the stem 1.2-2.3 (M 1.7) mm long, and the width measured at the widest point 0.7-1.0 (M 0.8) mm; margin plane; the apex acuminate, not denticulate; costae two, short and thin, extending from 1/5 to 1/3 leaf length, reaching 120-700 (M 400) μm; cells from the midleaf linear-vermicular, 80-190 (M 140) × 5-9 μm, cell areolation very tight; decurrencies 250-270 × 95-100 μm, forming distinct auricles, created by 3-4 rows of rectangular, quadrate, quite often inflated cells (Fig 15), 65-70 × 20-25 μm; sporophytes unknown so far. Distribution and ecology: currently known range of P. imbricatum is mainly central Europe, single positions are given from northern and western Europe and North America. From this area this species mainly is recorded from mixed and Fagus forests, it also is noted mainly in epigeic habitats (on humus and mineral soil, slopes), less often on anthropogenic habitats (such as drainage ditches), and on the bark of living trees (S1 Text).
Additional specimens examined: as a supplementary materials (S1 Text).
For decades, the genus Plagiothecium has hardly been the subject of any research. It is changing very intensively now, and molecular research, which are now commonly used in

PLOS ONE
taxonomy, along with other methods help not only explain the intraspecific variability of problematic taxa, but also shed new light on the relationship between closely related species [2,3].
We agree with what is stated by Wynns [2] that among P. curvifolium sensu lato other taxa can be distinguished, however for a taxon with a feature P. curvifolium fo. julaceum there is another previously published name representing the same characteristics-P. decursivifolium.
Our research proposes a resurrection of P. decursivifolium [58]. In the diagnosis, Macoun & Kindberg [58] stated that it is intermediate between P. latebricola Wilson ex Schimp. and P. pseudo-latebricola Kindb. in Macoun. Despite the similarity of the name (P. decursivifolium and P. curvifolium) in both cases indicating curved leaves, lack of comparison P. decursivifolium to P. curvifolium is related to the fact that it was not distinguished by bryologists at the end of 19 th and at the beginning of 20 th centuries [e.g., [5][6][7][8].
Macoun & Kindberg [58] in the diagnosis wrote that P. decursivifolium is characterized, e.g., by flattened foliage; inclined capsules; and a curved operculum. Grout [15], who quoted the type of P. decursivifolium, indicated that it is a form of P. latebricola "with narrower leaf cells, about 5-6 μ wide". While Ireland [27] synonymized this taxon with P. laetum, he wrote "P. laetum (. . .) has two forms (. . .). The most common form has (. . .) capsules that are Whereas, Wynns [2] reported that specimens of this taxon erroneously described as "types" are deposited in NY Herbarium (NY164182, NY164138), and he added that these materials represent P. latebricola. Indeed, these specimens appear to be the mentioned taxa. However, a detailed analysis of the description of the envelopes of these materials (NY164182!, NY164138!, both available online) showed that, habitat "on earth", location "Ottawa", as well as the date of collection "Oct. 5, 1907", as well as features of these specimens are inconsistent with the diagnosis. Therefore these specimens cannot be types of P. decursivifolium, which confirms the premises given by Wynns [2]. Additionally, in the Natural History Museum, Herbarium BM a specimen (BM13777462) signed as "probable original material Hypnum (Plagiothecium) latebricola Lindb." has been found. Based on description of habitat, location and a reference to H. Passaicense Austin, it reflects the data contained in the aforementioned diagnosis quite well. However, the features of this specimen are also inconsistent with the characteristics stated in diagnosis of P. decursivifolium as in the case of specimens from NY Herbarium.
Detailed diagnosis analysis of all the species mentioned above, in particular diagnosis of P. decursivifolium and the characteristics given for this taxon [58] exclude not only P. latebricola, but also P. laetum, because none of them is characterized by, e.g., inclined capsules and a curved operculum [62]. Taking into account the above facts, above-mentioned specimens cannot be considered as original collection on which this species was described.
Features characterizing the resurrected species, e.g., quite wide cells 6-10 (M 8) μm; decurrencies forming distinct auricles, created by 3-5 rows of rectangular, quadrate, quite often inflated cells could cause many researchers to mention the possibility of confusion of P. curvifolium sensu lato (including this taxon) with the P. denticulatum complex [e.g., 29,37]. However, its genetic distinctiveness and unambiguous morphological features support its restoration and treatment as a separate species.
In the 20 th and 21 st centuries, bryologists not only pointed to the problematic aspect of P. curvifolium sensu lato, but also reported a large range of many diagnostic features of this taxon. Mentioned, e.g., symmetrical or nearly symmetrical to asymmetrical leaves; margin entire or with denticulations at apex; cells from the middle part of the leaf 80-160 × 6-10 μm [28][29]37,[44][45][46][47], at present explain the intraspecific variability of the described complex and help to separate individual taxa from each other.
As reported by Wynns et al. [50] in genetic analyzes, P. curvifolium forms a clade with the "widely distributed austral species P. lucidum". Our research shows similar relationships and places P. curvifolium as more closely related to P. lucidum (Southern Hemisphere) than to P. laetum (Northern Hemisphere), which is usually considered as closely related with P. curvifolium. Obviously, these results confirm the legitimacy of the discussed distinction of P. curvifolium and P. laetum as separate [27][28][36][37]39], however, they pose a number of new questions about the relationship between the species of both hemispheres, as P. curvifolium and P. lucidum or P. schofieldii G.J.Wolski & W.R.Buck and P. lamprostachys (Hampe) A.Jaeger and others.
Despite the fact that over the last decades the genus Plagiothecium has not been the subject of detailed studies, recent years indicate that it is changing intensively [2][3][50][51][52]63,64]. Extensive research focused on the taxonomic revision of many problematic taxa not only allows to describe their intraspecific variability, but also allows the description of new species. Plagiothecium imbricatum is another species of this genus described over several years from a well-studied lowland area of central Europe [3,52]. These studies, as well as other reviews published in the past few years [50,51], indicate that the whole genus Plagiothecium, as well as many similar plagiotorphic genera, still requires detailed taxonomic studies.

Conclusion
1. So far, Plagiothecium curvifolium sensu lato was considered to be one widespread species.
2. The conducted research shows that the described taxon is a complex consisting of four separate taxa-P. curvifolium var. curvifolium, P. curvifolium var. recurvum, P. decursivifolium and P. imbricatum.
3. The most important features distinguishing the studied taxa are related to plant size; arrangement of leaves on the stem; the symmetry, dimensions, shape, concavity and folding of leaves; cell length; serration of the leaf apex; the shape of the decurrencies; the length of the sporophyte and the shape of the operculum.