Green armoured tardigrades (Echiniscidae: Viridiscus), including a new species from the Southern Nearctic, exemplify problems with tardigrade variability research

Ranges of tardigrade intraspecific and interspecific variability are not precisely defined, both in terms of morphology and genetics, rendering descriptions of new taxa a cumbersome task. This contribution enhances the morphological and molecular dataset available for the heterotardigrade genus Viridiscus by supplying new information on Southern Nearctic populations of V. perviridis, V. viridianus, and a new species from Tennessee. We demonstrate that, putting aside already well-documented cases of significant variability in chaetotaxy, the dorsal plate sculpturing and other useful diagnostic characters, such as morphology of clavae and pedal platelets, may also be more phenotypically plastic characters at the species level than previously assumed. As a result of our integrative analyses, V. viridianus is redescribed, V. celatus sp. nov. described, and V. clavispinosus designated as nomen inquirendum, and its junior synonymy with regard to V. viridianus suggested. Morphs of three Viridiscus species (V. perviridis, V. viridianus, and V. viridissimus) are depicted, and the implications for general echiniscid taxonomy are drawn. We emphasise that taxonomic conclusions reached solely through morphological or molecular analyses lead to a distorted view on tardigrade α-diversity.


Green
Ranges of tardigrade intraspecific and interspecific variability are not precisely defined, both in terms of morphology and genetics, rendering descriptions of new taxa a cumbersome task.This contribution enhances the morphological and molecular dataset available for the heterotardigrade genus Viridiscus by supplying new information on Southern Nearctic populations of V. perviridis, V. viridianus, and a new species from Tennessee.We demonstrate that, putting aside already well-documented cases of significant variability in chaetotaxy, the dorsal plate sculpturing and other useful diagnostic characters, such as morphology of clavae and pedal platelets, may also be more phenotypically plastic characters at the species level than previously assumed.As a result of our integrative analyses, V. viridianus is redescribed, V. celatus sp.nov.described, and V. clavispinosus designated as nomen inquirendum, and its junior synonymy with regard to V. viridianus suggested.Morphs of three Viridiscus species (V.perviridis, V. viridianus, and V. viridissimus) are depicted, and the implications for general echiniscid taxonomy are drawn.We emphasise that taxonomic conclusions reached solely through morphological or molecular analyses lead to a distorted view on tardigrade α-diversity.
Tardigrades, also known as water bears or moss piglets, represent an invertebrate phylum closely related to onychophorans and arthropods within Panarthropoda 1 , which consists of ca.1500 described species to date 2 .These are grouped into two classes: Heterotardigrada and Eutardigrada.Heterotardigrades are characterised by the presence of cirrus A [3][4][5] , whereas eutardigrades are generally distinguishable by their simplified, vermiform external morphology 6 .Tardigrades inhabit limno-terrestrial and aquatic, both freshwater and marine, habitats, the former typically in association with bryophytes, algae, lichens, or leaf litter 5,7,8 .Although the number of documented tardigrade species has increased significantly in the past three decades, most regions globally remain sparsely sampled for these diminutive animals 9,10 .The distributions of a few terrestrial tardigrade species are well documented for some regions of North America, e.g.see 11 , but the general paucity of North American tardigrade specialists has left much of the continent unexplored in this regard (most recently reviewed in 12,13 ).
New material examined.Populations from Alabama and Florida, 346 specimens in total were processed for PCM, SEM and DNA analyses (Table 5); additional few hundred specimens were frozen for future analyses.
Type locality.North America, USA, Alabama, Auburn.Additional localities.North America, USA, New Mexico 23 ; the Azores, Ribeira Fria, Lages do Pico 23,29,30 ; North America, USA, New Jersey 28 ; Central America, the Lesser Antilles, Antigua 28 .Given the reported variability in the pattern of the dorsal armour, these additional localities should be verified.
Etymology.From Latin viridianus = greenish.An adjective in nominative singular.Animals.Females (i.e., from the third instar onwards; measurements and statistics in Table 1).Body cavity with yellowish pigments (typical for most echiniscids), whereas dorsal and pedal cuticular plates light to dark green (Figs.1C, 3E,F).Red eyes and yellow pigments present in live specimens, but dissolve after mounting in Hoyer's medium, thus only green pigmentation persists.Body bulky (Fig. 6), with a poorly delimited cephalic region (Fig. 7).The cervical (neck) plate is well-developed, but sculptureless (Fig. 7).Weakly developed lateralmost, rectangular portions of the scapular plate with a weak sculpturing (Fig. 1C).Dorsal plate sculpturing ordinarily comprising polygonal epicuticular granules with scarce micropores, barely identifiable, even with SEM (Fig. 8A).Lateral and ventral endocuticle with intracuticular pillars, visible in PCM as minute dark dots, but identifiable in SEM only when the thin epicuticle is ruptured (Fig. 8B-D).Pillars larger and more sclerotised in proximal and central limb portions, forming longitudinal, narrow pulvini, and pedal platelets, respectively (Fig. 9).Some specimens exhibit a differently formed central pedal portion, more convex than the remainder of each platelet (Figs. 9, 11E).Areas of more sclerotised pillars always form a pair of merged subcephalic plates (Fig. 10F).Cirrus A short (< 20% of body length) and thin.A remarkable diversity of primary clava shapes: some specimens within a population have both clavae pointed and clearly conoid (Figs.7A, 10A-C), some have both clavae dactyloid, i.e., elongated, but without a pointed tip (Fig. 10D), and in some individuals both clavae are tubby, i.e., of a typical, Echiniscus-type shape (Fig. 10E).Importantly, numerous specimens showed a mixture of these shapes, that is the clava of one specimen differed in morphology from the other one on the same specimen.Claws massive and isonych (Figs.10G-I, 11A-D).
Males (i.e., most probably from the third instar onwards; measurements and statistics in Table 2).No detectable sexual dimorphism besides the circular gonopore.
Juveniles (i.e., the second instar; measurements and statistics in Table 3).Smaller than adults, but qualitatively like them.Gonopore absent.
Larvae (i.e., the first instar; measurements and statistics in Table 4).Body size overlaps with juveniles.Anterior portions of paired segmental plates, and median plate 2 sculptureless.Large cuticular pores in the dorsal armour.No gonopore or anus.
Eggs.Up to five orange eggs per shed exuvia, but typically fewer (see 31 ).Remarks.Males were present in all examined populations of the species.5).
Etymology.From Latin celatus = hidden, concealed.The name refers to the fact that the new species was not identified as new taxon for a long time, although the locality has been extensively sampled for tardigrades ( 16,27 referred to the species as V. perviridis based on the identification by Maucci 32 , and morphological characters, including cirrus A length).An adjective in nominative singular.Animals.Females (i.e., from the third instar onwards; measurements and statistics in Table 6).Body mediumsized and bulky.Body cavity with yellowish pigments (typical for most echiniscids), whereas dorsal and pedal cuticular plates olive green (Figs.1C, 3E,F).Red eyes and yellow pigments present in live specimens, but dissolve after mounting in Hoyer's medium, thus only green pigmentation persists (Figs. 12, 13B).Except for cirrus A, with a tubby clava near the cirrophore (Fig. 12), other dorsal and lateral trunk appendages are absent.Cephalic   appendages include internal and external peribuccal cirri with tubby cephalic papillae between them (Fig. 12).Dorsal plate sculpturing comprises large epicuticular granules (Fig. 12), which may be poorly developed in central plate portions (Fig. 13B).Sponge layer identifiable beneath granules.Granules appear more convex in anterior portions of paired segmental plates than in the remainder of the armour in PCM.Micropores not visible in PCM and their presence or absence remains to be confirmed in SEM.
All plates strongly sclerotised and with clear edges.The cephalic plate with a well-marked anterior chaliceshaped incision, the cervical plate and lateral sections of the body lack dense granulation and are covered with fine regular punctuation.The scapular plate contains three portions.Only the central part is visible in the dorsal view, and two small, weakly delineated, trapezoidal sections are present on the lateral portions of the body, with intracuticular pillars visible (Fig. 13B).The first median plate is triangular and unipartite, the second median plate is subdivided into two portions, and the anterior portion lacks the sponge layer.The third median plate is absent, but the area between the paired segmental plate II and the caudal plate is covered with large granules.Paired segmental plates I and II have two clearly delineated parts.Intersegmental plate is inserted between the posterolateral edge of the paired segmental plate I and anterior margin of paired segmental plate II.The caudal incisions are unsclerotised and weakly marked (Figs. 12, 13B).
Venter densely granulated in PCM (endocuticular pillars); a pair of subcephalic plates present (Fig. 14A).Gonopore hexapartite.Pulvini (= narrow proximal bands of intracuticular pillars) and pedal platelets (= broad central bands of pillars) are visible on all legs.Dentate collar with numerous irregular teeth (Fig. 14C).Sensory organs present on all legs: a tiny spine on leg I embedded at the edge of pedal platelet; hemispherical rudimentary papillae on legs II-III, embedded in the centre of pedal platelets (identifiable only when specimens are dorsolaterally oriented); and papilla IV on hind legs (Fig. 13B).Claws anisonych; primary spurs I-III tiny and thin, positioned slightly lower on branches compared to more massive spurs IV (Fig. 14B-C).
Males (i.e., most probably from the third instar onwards; measurements and statistics in Table 7).No sexual dimorphism observable in body size or qualitative traits (Fig. 13A).Gonopore circular.www.nature.com/scientificreports/Juveniles (i.e., the second instar; measurements and statistics in Table 7).Smaller than both females and males.Qualitatively like adults; gonopore absent.
Larvae.Not found.Eggs.Not found.
Remarks.Found only in association with large populations of V. viridissimus 17 .Differential diagnosis.The new species from Tennessee is differentiated from all Viridiscus spp.based on the presence of plesiomorphic papillae on legs II-III.These structures are, however, barely identifiable in specimens oriented dorsoventrally, hence we enumerate other criteria making V. celatus sp.nov.distinct from: • Viridiscus clavispinosus, by the relative length of cirrus A (34-49% vs < 15% of the body length), and primary spurs IV less divergent from claw branches; • Viridiscus perviridis, veritably reported from the Holarctic and Oriental regions 17 , by the length of cirrus A (34-49% vs typically ≫ 50% of the body length, see the subsection below that addresses this character), the weakly developed caudal incisions (strongly sclerotised and well-marked in all syntypes of V. perviridis), and the body colour (light to olive green vs usually dark green to almost black in V. perviridis, also in mounted specimens); • Viridiscus viridianus, reliably reported only from the USA, by the relative length of cirrus A (34-49% vs < 20% of the body length), the lack of pedal platelets with a distinctly formed central portion, and the more pronounced sculpturing of the anterior portions of paired segmental plates; • Viridiscus viridis, reliably reported only from the Hawaiian Archipelago 19,26 , by the relative length of cirrus A (34-49% vs < 10% of the body length), and a different pattern of dorsal sculpturing (in general V. viridis has noticeably fewer epicuticular granules on all plates, see fig. 1 in 26 ); • Viridiscus viridissimus, with a likely wide distribution in the Holarctic, Oriental, and Neotropical regions 17,33 , by the absence of pores in dorsal armour, and a better developed sponge layer of cuticle.

Viridiscus viridianus vs Viridiscus clavispinosus
Viridiscus viridianus was described by Pilato et al. 23 based on populations from Alabama (the origin of the holotype, thus locus typicus), New Mexico (the Nearctic region), and the Azores (Macaronesia, the westernmost part of the Palaearctic region).A few years later, Fontoura et al. 24 described V. clavispinosus (from the Archipelago of Cape Verde, which is also a part of Macaronesia, but located southwards of the Azores), using a seemingly sound autapomorphy, namely conoid primary clavae (with pointed apices).However, the data gathered in the present study undermine the validity of V. clavispinosus, as the shape of primary clava is evidently a variable character in some species of Viridiscus (Fig. 10A-E), rendering it unsuitable for species delineation.The surface between the paired segmental plate II and the caudal plate can be weakly sculptured (Figs.1C, 3E, 6A,C) or unsculptured (Fig. 3F) in a population of one species.Contrary to what was stated in 24 , neither V. viridianus nor V. clavispinosus exhibit well-marked epicuticular granules in the cephalic and cervical plates (compare figs.1b, e in 23 , fig.1a in 24 , and Fig. 7 herein).In echiniscid species, the dorsal plate sculpturing is less pronounced in these two anteriormost elements of armour.The sculpturing of the less sclerotised anterior portion of median plate 2 can also be more or less developed, and sometimes absent within a single population.The remaining characters referring to the dorsal sculpturing mentioned in 24 also are invalid in light of various atypical morphs presented for Viridiscus species in this paper, including V. viridianus.We refrained however, from synonymising V. clavispinosus with V. viridianus for two reasons: (1) as rightly stated by Fontoura et al. 24 , primary spurs IV of V. clavispinosus are more developed than those of V. viridianus (compare fig. 2c in 24 and Fig. 11D herein); (2) although V. viridianus potentially has a wide circum-Atlantic distribution (whilst molecular evidence is lacking), V. clavispinosus is known from a different part of Macaronesia.Therefore, V. clavispinosus is designated as nomen inquirendum, given the substantial doubts regarding its separateness from V. viridianus described above.www.nature.com/scientificreports/

Inter-vs intraspecific variability in Viridiscus and its consequences for echiniscid taxonomy
5][36][37] ).In the preliminary study that integratively addressed variability of Viridiscus 17 , we accentuated the role of combining molecular and morphological approaches in tardigrade taxonomy, as utilising only one line of evidence either leads to taxonomic inflation (e.g., the synonymy of V. miraviridis 16,17 ) or deflation (e.g., the Milnesium case 38 ).We have already underscored high variability of characters previously deemed as universally stable at the species level in the Echiniscidae (cirrus A length, details of cuticular ornamentation) in Echiniscus and Nebularmis spp. 39,40.This study adds further observations on atypical morphs within Viridiscus, which show disparities with typical morphs of a given species that are larger than interspecific variability in multiple new characters, such as the general phenotype of the dorsal sculpturing, the shape of primary clava, and morphology of pedal platelets.It is gradually becoming obvious that even within such a morphologically coherent family as Echiniscidae 41 , phenotypic plasticity of diagnostic characters varies between lineages at various taxonomic levels (genus, species, etc.), as it happens with the shape of primary clavae in Viridiscus, which is labile in contrast to other known echiniscid genera.Consequently, conservative characters in one genus should not simply be assumed to be conservative in other taxa.In parallel, analogous variability in characters such as the shape and presence of spine I and dorsal sculpturing, has been also reported recently in Claxtonia spp. 42.On the other hand, genetic data, when taken in isolation, can also be deceiving, and this pertains specifically to the COI marker, often uncritically utilised and termed as a universal tardigrade barcode as (I) it is often www.nature.com/scientificreports/difficult to amplify, and (II) may fail in delimiting species (Fig. 5; 43 ).Multiple markers are advisable 44 , and both ITS markers are good predictors of intra-and interspecific differences, as they are congruent with a spectrum of morphological variability analysed on many animals.Altogether, this reinforces the necessity of the integrative approach in tardigrade studies 17 .
In the light of our findings, some generalisations can be made regarding the diversity and classification of Viridiscus:    A characterises V. perviridis (most often ≫ 50% of bo), followed by cirri A longer than the echiniscid average (30-50% of bo) in V. celatus sp.nov.and V. viridissimus, and cirri A that can be classified as short (< 20% of bo) in V. clavispinosus nom.inq., V. viridianus, and V. viridis 23 .3. Atypical morphs are probably present in populations of all Viridiscus spp.We recorded them in V. perviridis, V. viridianus (this study), and V. viridissimus 17 .This means that morphology alone does not offer a fully credible dataset for establishing new species.4. The description of V. viridis by Pilato et al. 23 was based on only two individuals and likely does not reflect the spectrum of morphological variability of this species, but plainly indicates the specific nature of its dorsal sculpturing.The sculpturing embraces particularly widely spaced epicuticular granules, making it dissimilar to all other Viridiscus spp.Unless integratively redescribed from the Hawaiian Islands and later verified from other regions of the globe, it should be temporarily considered an endemic of this archipelago.

Conclusions
The extent of intraspecific morphological variability within Viridiscus is considerable and exceeds the usual variation encountered within echiniscid populations.Atypical morphs of Viridiscus species are not linked with sexual dimorphism, phenology, or ontogeny, and seem to represent idiopathic deviations from the most common morphs.Viridiscus celatus sp.nov., found only in association with large populations of V. viridissimus, is described integratively based on genetics and morphology.In addition, the validity of V. clavispinosus is questioned, and its synonymy with V. viridianus is implied.Males were reported in populations of Viridiscus celatus sp.nov., V. viridianus, and V. viridissimus so far.

Sample collection and processing
Previous sampling experience 31,45 showed that Viridiscus species are often found in mosses and lichens growing on rapidly draining vertical surfaces (see e.g. 46) such as the sides of rocks (Fig. 15), gravestones, and tree trunks in human-altered environment.Therefore, moss and lichen samples were collected from several localities in Alabama, Florida, and Tennessee from February 2020 to May 2022 (Table 5).All samples were stored in acid-free paper envelopes and allowed to dry overnight in an air-conditioned room at 25 °C and 50% relative humidity.The air-dried samples were suspended over deionised water using the custom-built Baermann pan described by Davison 47 for at least 12 h to separate the motile meiofauna from the moss and lichen particles.The deionised water was then filtered through 25 µm mesh and the sieve was rinsed into 30 mm Petri dishes with locally collected rainwater to search for tardigrades.The air-dried lichens were soaked in deionised water for at least 12 h, active tardigrades were extracted with an Irwin loop and transferred into 30 mm Petri dishes containing rainwater.

Microscopy and imaging
Individual tardigrades from the samples were visually inspected for diagnostic traits using an Olympus BX53 phase contrast microscope (PCM) associated with an Olympus DP74 digital camera.All specimens were divided into morphological and molecular analyses (details in Table 5).The excess of individuals was frozen to store for future analyses.For morphology and morphometry, individuals were permanently mounted on microscope slides in Hoyer's medium.Dried slides were sealed with nail polish and examined under the Olympus BX53 PCM.Specimens for imaging in the SEM were CO 2 critical point-dried, coated with gold and examined in the Versa 3D DualBeam SEM at the ATOMIN facility of the Jagiellonian University.All figures were assembled in Corel Photo-Paint X8.For deep structures that could not be fully focused on a single PCM photograph, a series of images was taken every ca.0.1 mm of vertical focusing and then assembled manually in Corel Photo-Paint into a single deep-focus image.

Morphometry
All individuals of the new species from Tennessee and selected individuals of V. viridianus from Alabama were chosen for morphometry.The measurements are in micrometres (µm), and only undamaged structures with the appropriate orientation were used.The body length measurement is from the anterior to the posterior ends of the body, excluding the hind limbs.The bo% was calculated for cirrus A, which is the ratio of cirrus A length to the body, and sp% is the ratio of the length of the given structure to the length of the scapular plate 48 .Morphometric data were handled using the Echiniscoidea ver.1.4 template available from the Tardigrada Register, http:// www.tardi grada.net/ regis ter 49 .

Genotyping
A Chelex ® 100 resin (Bio-Rad) extraction method was used for DNA extraction 50,51 .Hologenophores were recovered after the extraction and mounted on permanent slides in Hoyer's medium in most cases; in other cases, paragenophores were preserved 52 ; all are deposited in the Faculty of Biology, Jagiellonian University in Kraków.
Five DNA fragments (nuclear markers: 18S rRNA, 28S rRNA, ITS-1, ITS-2; mitochondrial marker: COI) were amplified and sequenced according to the protocols described in 51 ; primers and original references for specific PCR programmes are listed in Supplementary Material 3. GenBank accession numbers for the species used in calculating phylogenies and the genetic framework (mainly based on the dataset from Gąsiorek et al. 17 ) are provided in Table 8.ITS and COI sequences were separately aligned with sequences from Echiniscus succineus Gąsiorek & Vončina, 2019 53 as an outgroup using the ClustalW Multiple Alignment tool 54 implemented in BioEdit ver.7.2.5 55 .The remaining gaps were left intact in ITS alignments.The 18S and 28S rRNA gene fragments were not used for the species delimitation purposes but are provided for future broader-scale phylogenies.

Phylogeny
The sequences of the ITS fragments were concatenated to generate a matrix of 1058 bp in SequenceMatrix 56 .Using PartitionFinder version 2.1.1 57 with applied Bayesian Information Criterion (BIC) and the greedy algorithm 58 , the best substitution model (the lowest BIC) and partitioning scheme were chosen for posterior phylogenetic analysis.As the best-fit partitioning scheme, PartitionFinder indicated one partition with the best-fitted model GTR + G. Bayesian inference (BI) marginal posterior probabilities were calculated using MrBayes v.3.2 59 .Random starting trees were used, and the analysis was run for ten million generations, sampling the Markov chain every 1000 generations.An average standard deviation of split frequencies of < 0.01 was used as a guide to ensure that the two independent analyses had converged.Tracer v1.3 60 was then used to ensure Markov chains had reached stationarity and to determine the correct 'burn-in' for the analysis, in this case the first 10% of generations.The Effective Sample Size values were greater than 200, and the consensus tree was obtained after summarising the resulting topologies and discarding the 'burn-in' .
ModelFinder 61 was used to choose the best-fit models for two partitions in Maximum Likelihood (ML): TIM3e + G4 (ITS-1) and K2P + G4 (ITS-2), chosen according to the BIC.W-IQ-TREE was used for ML reconstruction 62,63 .One thousand ultrafast bootstrap (UFBoot) replicates were applied to provide support values for branches 64 .
Both ITS and COI alignments were uploaded separately to the Assemble Species by Automatic Partitioning (ASAP) webpage 65 , and automatic barcode gap discovery (ABGD) web 66 to obtain six independent marker-based primary species hypotheses using uncorrected pairwise distances.The partitions with the lowest ASAP and ABGD scores and p values < 0.05 were chosen as the best-fit hypotheses.Finally, Bayesian Poisson tree processes (bPTP 67 ) were applied to the ML phylogenetic trees run on three markers separately in W-IQ-TREE (COI bestfitted model: HKY + F + G4).The calculations were conducted with 100,000 MCMC generations, thinning the set to 100, with 10% burn-in, and with searches for maximum likelihood and Bayesian solutions.All final consensus trees were visualised by FigTree v.1.4.3 available from https:// tree.bio.ed.ac.uk/ softw are/ figtr ee.

Figure 2 .
Figure 2. Two main morphotypes of dorsal plate sculpturing present in Viridiscus (SEM): pores dominant, only V. viridissimus (the population from Vietnam): (A) a fragment of the scapular plate; (B) close up of the posterior portion of the second paired segmental plate; epicuticular granules dominant, micropores visible only in SEM, all remaining Viridiscus species (the population of V. perviridis from Vietnam shown): (C) a fragment of the scapular plate; (D) close up of the central portion of the scapular plate.Scale bars in μm.

2 AFigure 5 .
Figure 5. Phylogenetic relationships of the genus Viridiscus: Bayesian tree based on the concatenated ITS-1 + ITS-2 dataset (1058 bp); vertical bars denote different delineation methods used in the formulation of the primary molecular species hypotheses: ASAP, ABGD, and bPTP; (COI) refers to COI delimitation in all three methods).Asterisks indicate the maximal (1.00/100) posterior probability/bootstrap value.Echiniscus succineus was used as an outgroup.Scale bar represents substitutions per position.

Figure 9 .Figure 10 .
Figure 9. Leg morphology of V. viridianus (PCM).Arrows indicate pedal platelets in central leg portions, white asterisks indicate distinctly demarcated, central oval areas in pedal platelets, and black asterisks indicate pulvini in proximal leg portions.Scale bars = 20 μm.

Figure 11 .
Figure 11.Leg structures of V. viridianus (SEM): (A) claws I; (B) claws II; (C) claws III; (D) claws IV; (E) leg morphology.Arrows indicate pedal platelets in central leg portions, white asterisks indicate distinctly demarcated, central oval areas in pedal platelets, and black asterisks indicate pulvini in proximal leg portions.Scale bars in μm.

Figure 15 .
Figure 15.Habitats harbouring Viridiscus populations in Andreaea mosses growing on boulders by Lake Harris, Tuscaloosa, Alabama: (A) an overall view of the rock substrate; (B) close up of the moss growing on the vertical surfaces of the boulder (arrow indicates the moss matt); (C) close up of the pleurocarpous moss and leaf structure when dry; (D) moss leaves seen under a stereomicroscope (arrows indicate animals on moss leaves in the tun state).

Description of Viridiscus celatus sp. nov. Momeni, Gąsiorek, Nelson & Michalczyk
Material examined.Populations from Tennessee, 19 specimens in total processed for PCM and DNA analyses (Table

Table 1 .
Measurements (in µm) of selected morphological structures of adult females of Viridiscus viridianus mounted in Hoyer's medium.sp the proportion between the length of a given structure and the length of the scapular plate, ?unknown.

Table 3 .
Measurements (in µm) of selected morphological structures of juveniles of Viridiscus viridianus mounted in Hoyer's medium.sp the proportion between the length of a given structure and the length of the scapular plate, ?unknown.

Table 4 .
Measurements (in µm) of selected morphological structures of larvae of Viridiscus viridianus mounted in Hoyer's medium.sp the proportion between the length of a given structure and the length of the scapular plate, ?unknown.

Table 6 .
Measurements (in µm) of selected morphological structures of adult females of Viridiscus celatus sp.nov.mounted in Hoyer's medium.sp the proportion between the length of a given structure and the length of the scapular plate, ?unknown.

Table 7 .
Measurements (in µm) of selected morphological structures of adult males and juveniles of Viridiscus celatus sp.nov.mounted in Hoyer's medium.sp the proportion between the length of a given structure and the length of the scapular plate, ?unknown.

Table 8 .
GenBank accession numbers for the sequences analysed in this work.New sequences in bold.*Viridiscus aff.viridianus in 17 .