Three samples of moss growing on sand were collected from the type locality of X. inermis, the fossil sandy dunes of Gela (Sicily, Italy). Two additional moss samples containing Xerobiotus euxinus were collected in Viagrande (Sicily, Italy) and at Serra La Nave on Mount Etna (Sicily, Italy), respectively. Photos of the sampling site and substrates are provided in Fig. S1A. After collection, samples were brought to the laboratory and air-dried. To extract tardigrade animals and eggs, each sample was rehydrated for 2 hours and sieved through two mesh sizes (250 µm and 37 µm). The material retained by the finer mesh was then examined under an EZ4D stereomicroscope, and animals and eggs were isolated using a glass pipette. The moss samples from which the tardigrades were extracted were stored at the University of Catania for future identification. Isolated material was split into groups destined for different analyses. Detailed information about the samples and isolated material is provided in Table 1.

The animals destined for DNA analysis were first observed in vivo under phase contrast microscopy (PCM) at magnifications up to 1000× with oil immersion to confirm identification. Hologenophores were prepared through photovouchering following Cesari et al. (2011) (photographs available at https://doi.org/10.6084/m9.figshare.31370419), and, when possible, by recovering exoskeletons after DNA extraction. Recovered exoskeletons were mounted on permanent slides using Hoyer’s medium. Detailed information about vouchers is provided in Table SS1. Total genomic DNA was extracted from individual specimens using the Chelex® 100 resin (Bio-Rad) protocol described by Casquet et al. (2012), with modifications detailed in Stec et al. (2020). Molecular analyses targeted fragments of the small and large nuclear ribosomal subunits (18S and 28S rDNA), the mitochondrial cytochrome c oxidase subunit I (COI) gene, and the nuclear internal transcribed spacer 2 (ITS-2). All fragments were amplified and sequenced according to the protocols described in Stec et al. (2020). The primers used for each marker are listed in Table SS2. Sequencing was performed on an ABI 3130xl Genetic Analyzer by Genomed (Warsaw, Poland). Sequence editing was conducted in MEGA11 (Tamura et al. 2021). The COI sequences were also translated into amino acid sequences in MEGA11 to check against pseudogenes. All sequences were submitted to GenBank and the accession number for each voucher is provided in Table SS1.

To compute phylogenetic trees, we used a dataset from Stec et al. (2024), comprising all the sequences of the investigated genes (18S, 28S, ITS2 and COI) of the family Macrobiotidae belonging to clades A, B, C (sensu Stec et al. 2021). We added to this data set sequences of Xerobiotus and Macrobiotus clade B published in the meanwhile (Vincenzi et al. 2023; Polishchuk et al. 2024; Rocha et al. 2024) as well as the DNA sequences newly obtained in this study. The second dataset included only COI sequences of Xerobiotus, also incorporating the new sequences. The dataset with accession numbers of downloaded sequences used in the phylogenetic analysis is provided in Table SS3. Sequences of 18S rRNA and 28S rRNA were aligned with MAFFT v.7 (Katoh et al. 2002; Katoh et al. 2017), using the Q-INS-i algorithm; for COI and ITS-2 the G-INS-i algorithm was used. Alignments were inspected and their borders trimmed to the first position with less than 50% missing nucleotides. The alignments used for phylogenetic analyses were concatenated with the R package ‘concatipede’ v1.0.0 (Vecchi and Bruneaux 2021).

The sequences listed in Table SS3 were used for Maximum Likelihood (ML) and Bayesian inference (BI) analyses with MrBayes (v3.2.7), while the COI-only dataset was used for ML and species delimitation analyses. Model selection and Maximum Likelihood phylogenetic reconstruction was conducted on both the concatenated alignment and a COI-only alignment (the latter comprising only members of the genus Xerobiotus) using the IQtree online software (Trifinopoulos et al. 2016). One thousand ultrafast bootstrap (UFBoot) replicates were applied to provide support values for branches (Hoang et al. 2018). For the concatenated dataset, sequences of Minibiotus, Paramacrobiotus and Tenuibiotus were used as outgroup. For the COI-only dataset, the outgroup comprised species of the Macrobiotus pallari group. For BI (Bayesian) phylogenetic reconstruction, the best partitioning scheme was obtained with PartitionFinder2 (Lanfear et al. 2016). Phylogenetic BI inference was done with the software MrBayes (v3.2.7). Two runs with one cold chain and three heated chains were run for 50 million generations, sampling a tree every 1000 generations. An average standard deviation of split frequencies of < 0.01 was used as a guide to ensure the two independent analyses had converged. Posterior distribution sanity was checked with Tracer v1.7 (Rambaut et al. 2018). The effective sample size (ESS) values were greater than 200 and the consensus tree was obtained after summarizing the resulting topologies and discarding the first 10% of generations as burn-in. The phylogenetic trees were visualized with FigTree v1.4.4 (Rambaut 2007) and the final image was edited with Inkscape 0.92.3 (Bah 2011). Raw maximum likelihood (ML) and Bayesian inference (BI) trees for concatenated and cytochrome c oxidase subunit I (COI) only datasets are provided in File S1. The COI-only dataset was also used for distance-based species delimitation analyses, including Assemble Species by Automatic Partitioning (ASAP) and Automatic Barcode Gap Discovery (ABGD), using Jukes-Cantor (JC69) distances (Puillandre et al. 2012, 2021). Tree-based species delimitation analyses were performed using the Poisson Tree Processes (PTP) model and its Bayesian implementation (bPTP). The COI-only dataset, after deleting the outgroup and aligned, was also used for distance-based species delimitation analyses such as ASAP and ABGD utilizing JC69 distances. The PTP (Zhang et al. 2013) was run using the ML tree and bPTP was run on both the BI consensus tree and a sample of 200 trees from the BI posterior trees distribution (to account for uncertainty in phylogenetic reconstruction). If not otherwise specified, species delimitation analyses parameters were the default ones. For PTP and bPTP, phylogenetic reconstruction was performed as above. For PTP and bPTP, phylogenetic reconstruction was performed as above. Species delimitation analyses were run on the iTaxoTools softwares (Vences et al. 2021). Species delimitations results are available in File S2. For comparative purposes, we calculated p-distances between all COI and ITS-2 sequences of Xerobiotus and Pseudohexapodibius with MEGA11 (File S2).

The mitochondrial genome of X. inermis was sequenced (Fig. S1B). One individual from sample DFG4 was subjected to Whole Genome Amplification (WGA) following the protocol of Vecchi and Stec (2024), but with the REPLI-g Advanced DNA Single Cell Kit (Qiagen) instead than the REPLI-g Mini Kit (Qiagen). CeleroTM DNA-Seq Library Preparation kit (Tecan Genomics, Redwood City, CA) was used for library preparation following the manufacturer’s instructions. Both input and final library were quantified by Qubit 2.0 Fluorometer (Invitrogen, Carlsbad, CA) and quality tested by Agilent 2100 Bioanalyzer High Sensitivity DNA assay (Agilent technologies, Santa Clara, CA). Libraries were then prepared for sequencing and sequenced on NovaSeq X in paired-end 150 bp mode. Library preparation and sequencing were performed by a commercial provider (IGA Technology, Udine, Italy). Reads were trimmed and quality filtered with the software fastp (options: -q 15 -u 50 -l 100 –correction --detect_adapter_for_pe; Chen et al. 2018), then the mitogenome was assembled with NOVOPlasty v.4.3.5 with k-mer size 33 (Dierckxsens et al. 2016) using a COI sequence from an individual of the same population as bait. Mitogenome annotation and visualization was performed as in Vecchi and Stec (2025). Raw reads are deposited in NCBI SRA under Bioproject PRJNA1287536.

A dataset with raw morphometric data of Xerobiotus and Macrobiotus Clade B taxa (sensu Stec et al. 2021a) and Pseudohexapodibius was assembled from publicly available data and the data produced in this study (File S3A). Traits were normalized to the buccal tube length (Pilato 1981), and only individuals with less than 25% of missing data were retained. A second dataset was assembled with only with raw morphometric data of Xerobiotus extracted from the first dataset (File S3B). The R language was used for data analysis (R Core Team 2013; v3.0.2). Missing data were imputed with a PCA-based imputation approach using the R package “missMDA” (Josse and Husson 2016). PCA was conducted on the scaled data using the R package “FactoMineR” (Josse and Husson 2008). PCA results were visualized with the R packages “ggplot2”, “ggforce”, “gghighlight” and “patchwork” (MacLean 2023). The R scripts used for PCA analyses are available in File S3C, D.

Animals and eggs used for light microscopy analyses were mounted on permanent slides using Polyvinyl-Lactophenol (PVLF) as the mounting medium. Slides were examined under a Leica DM1000 phase contrast microscope (PCM), and photographs were taken with a Leica Flexacam C3 digital camera. Topotypic specimens of X. euxinus were examined and photographed with a Zeiss AX10 phase contrast and differential interference contrast microscope (PCM, DIC) equipped with a DLT-Cam PRO digital camera. For scanning electron microscopy (SEM) specimens were processed following the protocol described as “A2” in Camarda et al. (2024) and sputter-coated with gold. The imaging was carried out using a Phenom XL G2 SEM at the University of Catania. As the comparative material we used permanent microscope slides containing animals and eggs belonging to the type series of Xerobiotus inermis, and holotype and paratypes of Xerobiotus euxinus. Specimens of the topotypic population studied recently by Polishchuk et al. (2024) of Xerobiotus euxinus were also examined.

Measurements (in µm) were taken using Leica Enersight software on a Leica DM1000 Phase Contrast Microscope (PCM) equipped with a Leica Flexacam C3 digital camera. Morphometric data were collected only when structures were undamaged and properly oriented. Body length was measured from the anterior extremity to the end of the body, excluding the hind legs. Buccal apparatus and claw types were classified following Pilato and Binda (2010) and Kaczmarek et al. (2014). Animal and egg measurements and terminology follow Pilato and Binda (2010), Kaczmarek and Michalczyk (2017), Vincenzi et al. (2024) and Stec (2024). The pt ratio, defined as the ratio of the length of a given structure to the length of the buccal tube, is expressed as a percentage (Pilato 1981). Morphometric data were processed using the ‘Parachela’ v.1.8 template available from the Tardigrada Register (Michalczyk and Kaczmarek 2013). Raw morphometric data for X. euxinus (two Sicilian populations) and X. inermis (type series and topotypic population) are provided in File S4. Tardigrade taxonomy follows Bertolani et al. (2014), Stec et al. (2021a) and Vincenzi et al. (2024).

The phylogenetic reconstruction based on 4 concatenated markers (Fig. 1) provided comparable topology based on ML and BI methods. The relationships within Macrobiotidae Superclade I (sensu Stec et al. 2021) and intergeneric topology were recovered as (Sisubiotus + Mesobiotus) (Macrobiotus [including Xerobiotus and Pseudohexapodibius]). The genera Xerobiotus and Pseudohexapodibius form a clade nested within a paraphyletic Macrobiotus. More specifically, the clade Xerobiotus + Pseudohexapodibius is placed within Macrobiotus clade B (sensu Stec et al. 2021).

Depending on the analyses, molecular species delimitations recovered 3–10 putative species within the Xerobiotus + Pseudohexapodibius clade (Fig. 2). Most conservative solutions with a lower number of species were provided by distance-based methods such as ABGD and ASAP (3–6 putative species). The tree-based methods like PTP and bPTP always recovered more putative species (7–10). In the ABGD analyses, the initial partition results were stable (i.e., the same) across a broad range of prior intraspecific divergence while the recursive partitions results were not stable (File S2). Thus, for ABGD the first initial partition results were chosen as representative, and it identified 3 putative species (Fig. 2). ASAP delimitation identified two partitions which showed higher ASAP scores compared to all the others (partition #9 with ASAP score 2.0, and partition #6 with ASAP score 2.5; File S2). Partition #9 found the same 3 putative species as ABGD, whereas partition #6 identified 6 putative species (Fig. 2). The PTP analysis conducted on the ML tree recovered the highest number of 10 putative species, while bPTP results both on the BI consensus tree and on the posterior trees provided identical results, indicating 7 putative species in the dataset (Fig. 2).

Despite the considerable discrepancies between outcomes of different delimitation methods, three clades appear to be more distinct and evident with genetic divergence in COI between them, having a mean group p-distance of 17.4 to18.6%. The first clade was recovered as one putative species across all the delimitation approaches and contains specimens of X. euxinus sequenced in recent studies (including sequences from topotypic population) and two sequences of X. pseudohufelandi (AY598776, AY598777) published by Guidetti et al. (2005). Intra-clade divergence in COI ranged in the first clade from 0 to 5.1%. The second clade exhibited a wider range of intra-clade divergence in COI (0 to 11.8%) resulting in mixed outcomes from different delimitation approaches. Two results were recovered twice by different delimitation approaches: one putative species found by ABGD and ASAP #9 and three putative species found by both bPTP alternatives. The first putative species comprises mostly sequences of P. degenerans, the second putative species comprises mostly sequences of X. naginae and X. reductus, while unidentified Xerobiotus specimen from Poland (MN888325) constitute a third putative species. The p-distance range between these three putative species is as follows: [1–2] 5.3 to 6.7%; [1–3] 10.7 to 11.1%; [2–3] 10.8 to 11.8%. The third clade was recovered as one species by the ABGD and ASAP #9, but by all other four approaches as three putative species; its intra-clade divergence in COI ranges from 0 to 11.6%. The first putative species comprises specimens of X. inermis, X. gretae, X. arenosum, and unidentified Xerobiotus individuals from Australia (OR397025 –29) and Italy (OR397009 –12, OR397013 –18), with an internal clade divergence ranging from 0 to 3.3%. The second putative species includes exclusively specimens of X. litus while the third putative species is represented by only one unidentified Xerobiotus individual from Australia (OR397030). The p-distances range between these three putative species is as follows: [1–2] 9.3 to 11.4%, [1–3] 8.9 to 11.6%; [2–3] 10.3 to 11%. The genetic divergence in the case of ITS-2 dataset was lower compared to the COI dataset. The three big clades showed inter-clade p-distances as follows: [1–3] 2.9 to 6.3%, [1–2] 5.1 to 9.6%, and [2–3] 5.1 to 8.2%. The intra-clade divergence was also lower compared to COI, with ranges for the first, second and third clade: [1] 0 to 2.54%, [2] 0 to 1.19%, and [3] 0 to 3.52%. All detailed results of the species delimitation analyses, including tables of p-distance values and outputs from ABGD and ASAP, are provided in File S2.

We performed PCA on two datasets: one combining morphometric data from Pseudohexapodibius, Xerobiotus and Macrobiotus Clade B taxa, and a second including only raw measurements from Xerobiotus and Pseudohexapodibius. In both datasets, the first principal component (PC1) showed strong positive loadings for claw lengths and structures associated with the buccal apparatus, suggesting that PC1 primarily reflects overall size variation and morphological scaling in these structures. The second principal component (PC2), by contrast, was characterized by generally small or negative loadings for claw lengths and positive loadings for buccal apparatus traits, indicating a potential shape-related axis of variation less influenced by size. Notably, the position of the stylet support insertion point contributed negatively to PC1, while ventral lamina length had a negative loading in PC2, highlighting these characters as potential exceptions to the broader patterns of trait covariation.

As regards the first dataset, the PCA analysis summarized the variation in morphometric traits (relative values, pt) variability in two Principal Components (PCs) which together explained 78.12% of the total variance. The taxa of the genera Xerobiotus and Pseudohexapodibius clustered together and separated from the other Macrobiotus clade B species (Fig. 3A–C). The separation occurs mostly along the PC1, indicating smaller claws and buccal apparatus structures size in the formers. However, the separation between these two species groups is not complete, as some taxa like X. euxinus and Macrobiotus margoae Stec, Vecchi & Bartels, 2021 occupy intermediate positions and show significant overlap.

In the second dataset (PCA based on the Xerobiotus/Pseudohexapodibius-only morphometric measurements), the three clades identified within Xerobiotus/Pseudohexapodibius occupy separate regions of the morphometric space, even though with some overlap. The first two PC components together explained 64.98% of the total variance. In clade I (Fig. 3A), which includes only Xerobiotus euxinus, individuals from Sicily (SN2 and V1) and from the type locality in Ukraine displayed almost complete overlap, even if a separation is present between SN2 and the topotypic population. Clade II (Fig. 3B), comprising P. degenerans, X. naginae, and X. reductus, showed only partial separation among the three taxa. Clade III (Fig. 3C) included X. inermis, X. arenosum, X. litus, and X. gretae. The last two species exhibited almost complete morphometric overlap while X. inermis and X. arenosum overlapped only partially, what indicates limited morphometric distinctiveness between these populations. The loadings regarding the morphological characters included in the second PCA analysis are provided in Fig. S1C.

Phylum: Tardigrada Doyère, 1840

Class: Eutardigrada Richters, 1926

Order: Parachela Schuster et al., 1980

Superfamily: Macrobiotoidea Thulin, 1928 (in Marley et al. 2011)

Family: Macrobiotidae Thulin, 1928

Genus: Xerobiotus Bertolani & Biserov, 1996

The integrative redescription of X. inermis and the analysis of two Sicilian populations of X. euxinus provided new morphological and genetic data that enabled an updated phylogenetic reconstruction of the family Macrobiotidae, with a focus on the genera Xerobiotus and Pseudohexapodibius. This reconstruction recovered three well-supported clades: (1) a clade comprising all X. euxinus sequences, (2) a clade comprising P. degenerans, X. reductus, and X. naginae, and (3) a clade including X. inermis, X. litus, X. arenosum, and X. gretae. These groupings, combined with integrated analysis of genetic data and a re-evaluation of morphological characters, demonstrate that Pseudohexapodibius is nested within Xerobiotus, and Xerobiotus is nested within Macrobiotus, rendering the latter paraphyletic. To restore monophyly and resolve this conflict, both Xerobiotus and Pseudohexapodibius should be suppressed as valid genera. Additionally, our results support several taxonomic changes at the species level, including two synonymities and three status revisions. The detailed nomenclatural decisions and their justifications, along with the composition of the three major clades and the diagnostic reliability of key morphological traits, are discussed in the following sections.

The findings of our study provide additional evidence for abolishing the genera Xerobiotus and Pseudohexapodibius and transferring their species into Macrobiotus. This conclusion is supported by integrative analyses of genetic, morphological, and morphometric data, which collectively demonstrate that Xerobiotus lacks both phylogenetic independence and clear morphological distinctiveness from Macrobiotus (in particular, claw reduction, which is advocated as main trait separating these genera, is demonstrated to be a gradient rather than clear cut morphological states; see Fig. 2 and paragraph below). Phylogenetic reconstructions based on concatenated markers (18S, 28S, ITS-2, COI) place Xerobiotus and Pseudohexapodibius within Macrobiotus, firmly within Macrobiotus clade B sensu Stec et al. (2021), forming a lineage internal to Macrobiotus rather than a divergent sister group (Fig. 1). This phylogenetic pattern is consistent with previous studies that questioned the validity of Xerobiotus due to its low genetic divergence from Macrobiotus (Stec et al. 2021; Stec et al. 2022; Stec 2024; Vecchi et al. 2022).

Historically, Xerobiotus was separated from Macrobiotus based on characters thought to be apomorphic, such as a short basal tract of the claws lacking a distinct peduncle, direct insertion of the secondary branch, reduced lunulae on legs I–III, relatively shorter legs (Bertolani and Biserov 1996). Massa et al. (2021) added further putative synapomorphies, including smaller legs I, garter-like structures with microdigitations, claws IV with a relatively long common tract, and minute dorsolateral cuticular pores. However, our comparative studies on these characters (and those reported in recent studies) demonstrate that none are exclusive or consistent across Xerobiotus. For example, a basal peduncle is present but only visible under SEM; lunulae are present but reduced to so called claws’ cuticular plates; garter-like structures and small cuticular pores occur in Macrobiotus species such as M. mileri Stec, 2024 and M. paulinae Stec, Smolak, Kaczmarek & Michalczyk, 2015; and cuticular pores are not confined exclusively to the dorsolateral region. Thus, the only remaining consistent differences with Macrobiotus involve subtle variations in claw IV morphology and lunules reduction. Morphometric data further undermine the distinctiveness of Xerobiotus. Our Principal Component Analysis (PCA) shows a continuum of pt ratios for claw and buccal apparatus dimensions across Xerobiotus and other species of the Macrobiotus clade B (Fig. 3A, B), with no clear separation. This overlap, along with the clear phylogenetic position of Xerobiotus, strengthens the argument that the genera cannot be reliably separated, and that the differences observed may instead reflect a gradient of morphological adaptations to a xeric and sandy environment.

The inclusion of P. degenerans, a species lacking claws IV, within the Xerobiotus further challenges the morphological boundaries of the group. Despite its morphological deviation, P. degenerans clusters phylogenetically within Macrobiotus, and its morphological traits fall within the range observed in Xerobiotus species. Given the lack of support for the distinctiveness of P. degenerans, X. naginae, and X. reductus in the presented phylogeny, and the seamless morphological gradient among them, maintaining Pseudohexapodibius as a separate genus introduces unnecessary taxonomic complexity. Taken together, these findings provide robust justification for the suppression of Xerobiotus and Pseudohexapodibius, and their incorporation into a broader, morphologically cohesive Macrobiotus. Such reclassification restores monophyly and eliminates an artificial division unsupported by phylogenetic or morphological evidence.

This clade, consistently recovered as a single species by all species delimitation methods (Fig. 2), includes X. euxinus specimens from multiple localities: Italy (7 populations), Ukraine (2 populations), Georgia (1 population) (Vincenzi et al. 2024; Polishchuk et al. 2024; this study) as well as two sequenced individuals originally identified as X. pseudohufelandi (Guidetti et al. 2005). These latter specimens may represent misidentifications of X. euxinus, or alternatively, they could belong to the true X. pseudohufelandi, in which case a redescription of the latter will be essential to assess the potential synonymy between the two taxa. Given the considerable intraspecific morphological variability demonstrated within X. euxinus, both hypotheses remain plausible. The mentioned considerable variability includes the presence of two morphotypes (forma aporata and forma porata), as well as variation in claw size, with larger claws observed in individuals from Sicilian moss samples collected from rock. Observations on the newly investigated populations and the topotypic population revealed that both morphologies (i.e., forma porata and forma aporata) are present. The same applies to the separation of the dorsal crests of the OCA, which does not always appear as a single continuous crest. This may depend on the degree of compression caused by the coverslip, as more flattened specimens allow a clearer visualization of the buccal armature, or it may reflect intraspecific variability. Moreover, the species was reported to be dioecious (Vincenzi et al. 2024) and appears to be relatively widespread, which is noteworthy given that widely distributed tardigrade species are typically parthenogenetic, whereas sexual species tend to have more restricted ranges (Guidetti et al. 2019; Stec et al. 2020; Stec et al. 2021; Bertolani et al. 2023; Kayashta et al. 2023a, b).

The second big Xerobiotus clade yielded inconsistent results across species delimitation methods (Fig. 2). As also noted in earlier phylogenetic studies (Vincenzi et al. 2024), Pseudohexapodibius is nested within Xerobiotus and appears more closely related to X. reductus than X. reductus is to other Xerobiotus species. This pattern further blurs the boundaries among Pseudohexapodibius, Xerobiotus, and Macrobiotus, which likely form a morphological continuum shaped by adaptation to edaphic conditions.

Notably, in our study X. reductus, X. naginae, and P. degenerans are grouped as a single species by ABGD, ASAP_6, and ASAP_9, while other methods (PTP, bPTP_cons, bPTP_post) suggest the presence of two or more distinct species. Importantly, X. naginae and X. reductus (which show COI p-distances ranging from 1.1% to 2.6%) are consistently recovered as a single species by all delimitation methods. The species were primarily differentiated based on the morphology of the third band of teeth (transversal crests) which appears as a continuous dorsal ridge in X. reductus and as three separate teeth in X. naginae. However, since a similar variation is also observed in X. euxinus, this difference is more likely attributable to the compression of the animal under the coverslip or to intraspecific variability. Therefore, given the minimal morphological differences and the broader context of observed intraspecific variation, we consider their synonymization to be justified.

All species delimitation analyses conducted in our study consistently recover X inermis, X. arenosum, and X. gretae as belonging to the same species. The morphological traits previously used to distinguish these taxa are minimal and fall within the range of intraspecific variability observed in other species, such as X. euxinus demonstrated in this study. In fact, no diagnostic morphological differences (whether in adult or egg characters) can be identified between X. inermis and X. arenosum based on their descriptions. This conclusion about them being the same species is further supported by low COI p-distance values between populations (0.7–2.7%), further justifying their synonymization. Importantly, the case of X. gretae is more complex. While genetic distances between X gretae and X. inermis are similarly low (1.7–2.4%), X. gretae was described as having a smooth egg chorion (Massa et al. 2021) and such morphological character would be consider as having potential phylogenetic relevance and constituting a diagnostic character. Although this trait could be bona fide species-specific, (i) the variability observed even within single populations raises the possibility that chorion sculpturing may be a plastic or variable character, as reported for Paramacrobiotus bifrons (Brandoli et al. 2024), or (ii) the reticulation may be present but difficult to see under LM, and the SEM preparation may have generated an artifact (a similar case occurred in our study, where a X. euxinus egg appeared to have a smooth chorion under SEM; see Fig. S1H). However, this hypothesis remains untested. Until additional data becomes available, we propose that X. gretae be treated as a species inquirenda.

Our study further supports X. inermis as a valid species. Moreover, the sequencing of its complete mitochondrial genome provides additional information and contributes valuable data to a still limited dataset, as relatively few tardigrade mitogenomes are currently available. This new resource will be useful for future phylogenomic analyses, which are expected to improve the resolution of intra- and interspecific variability within the Macrobiotus pseudohufelandi complex.

The species is morphologically distinct from the extant species, but critical gaps remain, and further analyses are needed on the relation with X. pseudohufelandi. In particular, the incomplete description of X. pseudohufelandi highlights the need for a thorough re-investigation and redescription using modern techniques. The species was originally described as lacking cuticular pores, but it is plausible that pores are present and detectable only under SEM. Additionally, data on the oral cavity armature (OCA) are limited, making X. pseudohufelandi morphologically similar to both X. inermis and the forma aporata of X. euxinus.

However, based on an interpretation of the drawing presented by Iharos (1966) in the original description, the egg processes of X. pseudohufelandi appear more similar to those of X. euxinus than to those of X. inermis (Iharos 1966; Pilato et al. 2011; Polishchuk et al. 2024; present contribution). Furthermore, X. pseudohufelandi and X. euxinus were described from geographically close regions (Austria and Ukraine, respectively), and the latter is a widespread species, further supporting the possibility that these two taxa may in fact be synonyms.

Given the arguments presented and discussed in the previous sections, the following nomenclatural acts are proposed:

(1)Xerobiotus Bertolani & Biserov, 1996 syn. nov. and Pseudohexapodibius Bertolani & Biserov, 1996 syn. nov. are hereby synonymized with Macrobiotus C.A.S. Schultze, 1834.

(2)Xerobiotus reductus Vincenzi et al., 2024 syn. nov. is from now treated as junior synonym of Xerobiotus naginae (Vecchi et al., 2022).

(3)Xerobiotus arenosum Vincenzi et al., 2024 syn. nov. is from now treated as junior synonym of Xerobiotus inermis (Binda & Pilato, 1971).

(4)Xerobiotus gretae Massa et al., 2021 is from now treated as species inquirenda.

The first action requires a transfer of nominal species from the synonymized genera to the genus Macrobiotus with the following designations:

Macrobiotus euxinus (Pilato, Kiosya, Lisi, Inshina & Biserov, 2011) comb. nov.

= Xerobiotus euxinus Pilato, Kiosya, Lisi, Inshina & Biserov, 2011.

Originally described as Xerobiotus euxinus, transferred by Stec et al. (2021a) to Macrobiotus, and re-transferred to Xerobiotus by Vincenzi et al. (2024).

Macrobiotus inermis Binda & Pilato, 1971 stat. rev.

= Xerobiotus inermis (Binda & Pilato, 1971): Vincenzi et al. (2024).

= Xerobiotus arenosum Vincenzi, Cesari, Kaczmarek, Roszkowska, Mioduchowska, Rebecchi, Kiosya & Guidetti, 2024 syn. nov.

Originally described as Macrobiotus inermis, later synonymized with M. pseudohufelandi by Pilato (1973), re-validated and transferred to Xerobiotus by Vincenzi et al. (2024).

Macrobiotus litus (Vincenzi, Cesari, Kaczmarek, Roszkowska, Mioduchowska, Rebecchi, Kiosya & Guidetti, 2024) comb. nov.

= Xerobiotus litus Vincenzi, Cesari, Kaczmarek, Roszkowska, Mioduchowska, Rebecchi, Kiosya & Guidetti, 2024.

Macrobiotus pseudohufelandi Iharos, 1966 stat. rev.

= Xerobiotus pseudohufelandi (Iharos, 1966): Bertolani and Biserov (1996).

Originally described as Macrobiotus pseudohufelandi, later moved to Xerobiotus (Bertolani and Biserov 1996), re-transferred by Stec et al. (2021a) to Macrobiotus, and retransferred to Xerobiotus by Vincenzi et al. (2024).

Macrobiotus xerophilus (Dastych, 1978) comb. nov.

= Parhexapodibius xerophilus Dastych, 1978.

= Xerobiotus xerophilus (Dastych, 1978): Bertolani and Biserov (1996).

Originally described as Parhexapodibius xerophilus, transferred to Xerobiotus by Bertolani & Biserov (1996), then transferred to Macrobiotus by Stec et al. (2021a) and re-transfer back to Xerobiotus by Vincenzi et al. (2024).

Macrobiotus degenerans (Biserov, 1990) comb. nov.

= Parhexapodibius degenerans Biserov, 1990

= Pseudohexapodibius degenerans (Biserov, 1990): Bertolani and Biserov (1996).

Originally described as Parhexapodibius degenerans, later accommodated within a monotypic genus Pseudohexapodibius (Bertolani & Biserov, 1996).

Macrobiotus naginae Vecchi, Stec, Vuori, Ryndov, Chartrain & Calhim, 2022 stat. rev.

= Xerobiotus reductus Vincenzi, Cesari, Kaczmarek, Roszkowska, Mioduchowska, Rebecchi, Kiosya & Guidetti, 2024 syn. nov.

= Xerobiotus naginae (Vecchi, Stec, Vuori, Ryndov, Chartrain & Calhim, 2022): Vincenzi et al. (2024)

Originally described as Macrobiotus naginae, transferred to Xerobiotus by Vincenzi et al. (2024).

Macrobiotus gretae (Massa, Guidetti, Cesari, Rebecchi & Jönsson, 2021) comb. nov.

= Xerobiotus gretae Massa, Guidetti, Cesari, Rebecchi & Jönsson, 2021.

Originally described as Xerobiotus gretae, transferred to Macrobiotus by Stec et al. (2022), re-transferred to Xerobiotus by Vincenzi et al. (2024). Given the uncertain status of this taxon, which combines unique egg morphology with close genetic affinity to M. inermis, we treat it as a species inquirenda pending further investigation.

Since the nominal genus Xerobiotus is now synonymized, to aid communication between researchers we propose to group all the species listed above within a Macrobiotus pseudohufelandi species complex as has been already proposed by Stec et al. (2021a). The complex groups macrobiotid taxa characterized by claws with strongly reduced or absent lunules and a peduncle not visible under LM on the first three pairs of legs, whereas the claws of leg IV exhibit an elongated common tract and short primary branches; the latter claws may be strongly reduced or absent.

Following the synonymization of Xerobiotus and Pseudohexapodibius with Macrobiotus, the diagnosis of the latter should also be revised. We propose the following amended diagnosis: Macrobiotidae characterized by: (i) a porous cuticle; (ii) a mouth opening surrounded by ten peribuccal lamellae; (iii) a rigid buccal tube strengthened by a ventral lamina and lacking a ventral hook; (iv) two elongated macroplacoids and a microplacoid positioned in close proximity; (v) Y-shaped claws of the hufelandi type with lunulae on each leg, or claws with reduced lunulae (restricted to the Macrobiotus pseudohufelandi complex); (vi) claws IV sometimes strongly reduced or absent (restricted to the M. pseudohufelandi complex); and (vii) eggs with an ornamented shell laid freely in the environment.