What If Multiple Claw Configurations Are Present in A Sample? A Case Study with the Description of Milnesium pseudotardigradum sp. nov. (Tardigrada) with Unique Developmental Variability

© 2019 Academia Sinica, Taiwan Open Access What If Multiple Claw Configurations Are Present in A Sample? A Case Study with the Description of Milnesium pseudotardigradum sp. nov. (Tardigrada) with Unique Developmental Variability Bartłomiej Surmacz, Witold Morek*, and Łukasz Michalczyk Institute of Zoology and Biomedical Research, Faculty of Biology, Jagiellonian University, Gronostajowa 9, 30-387 Kraków, Poland. *Correspondence: E-mail: [email protected] (Morek). E-mail: [email protected] (Surmacz); [email protected] (Michalczyk) Received 14 June 2019 / Accepted 3 September 2019 / Published 11 November 2019 Communicated by Benny K.K. Chan The tardigrade fauna of Iceland has been a subject of studies since mid XX century. So far, only a single species of the genus Milnesium has been reported from the island, M. tardigradum, which at the time was assumed to have a cosmopolitan distribution. However, the record comes from before the redescription of M. tardigradum, thus the validity of the Icelandic report is questionable. Some species of Milnesium are characterised by developmental variability, which is most pronounced in the morphology of secondary branches of claws, exhibited by shifts in the number of spurs. In this contribution, we present a case study in which multiple claw configurations (CC) were found in a single lichen sample from Iceland, indicating the presence of more than one species and/or ontogenetic variability. To elucidate this puzzle, we utilised a range of integrative tools, including detailed morphology, morphometry, barcoding and development tracking. We present the workflow, which enabled us to collect the data for each species/morphotype. In result, we revealed the presence of three species, two characterised by ontogenetic CC change (M. variefidum and a new species) and one with a stable CC (a new species). Here, we describe one of these new species, M. pseudotardigradum, which is extremely similar to M. tardigradum, but can be phenotypically differentiated by a unique, double CC change pattern. Key words: Cryptic species, Developmental variability, Ontogeny, Pseudocrypsis, Species distribution. Citation: Surmacz B, Morek W, Michalczyk Ł. 2019. What if multiple claw configurations are present in a sample? A case study with the description of Milnesium pseudotardigradum sp. nov. (Tardigrada) with unique developmental variability. Zool Stud 58:32. doi:10.6620/ZS.2019.58-32. BACKGROUND Tardigrades are a phylum of microscopic invertebrates, inhabiting both aquatic and terrestrial environments. The limno-terrestrial water bears are most often found in mosses and lichens all around the world, across diverse environments (Nelson et al. 2015). So far more than 1200 species have been recognised (Guidetti and Bertolani 2005; Degma et al. 2007–2019; Degma and Guidetti 2007). Even though the tardigrade fauna of Iceland was investigated by numerous researchers (e.g., de Coninck 1939; Tuxen 1941; Morgan 1980; Maucci 1996; Buda et al. 2018), all records of the genus Milnesium are represented by only a single species, Milnesium tardigradum Doyère, 1840. This is surprising given that the number of described Milnesium species in the genus is 37 (Degma et al. 2007 2007–2019) and recent research shows that species diversity in the genus is much greater (Morek and Michalczyk, in press). However, it is important to note that the Icelandic Zoological Studies 58: 32 (2019) doi:10.6620/ZS.2019.58-32 1 © 2019 Academia Sinica, Taiwan records come from before the redescription of M. tardigradum (Michalczyk et al. 2012a b) and therefore should be treated with caution. In fact, the verified range of this species is currently restricted to Europe (Morek et al. 2019a), with a single record from Asia, in the vicinity of Lake Baikal (Morek and Michalczyk, in press). The geographic ranges of the majority of Milnesium species are restricted to their type localities, with a few reported from multiple sites (e.g., Kaczmarek et al. 2014 2016). Almost all, however, lack molecular support. Given that pseudocryptic diversity has been recently documented in the genus Milnesium (Morek et al. 2019b), records based solely on phenotypic identification are of very limited reliance and usability. The developmental variability in tardigrades includes mainly differences between immature (first two instars) and mature life stages (third instar onwards; Morek et al. 2016a; Guidetti et al. 2016; Kosztyła et al. 2016). In Milnesium, the change in claw configuration (CC) was discovered for the first time in M. variefidum (Morek et al. 2016a) where extra points on claws develop between the last immature instar and adult. Later, the phenomenon was showed to be more common in the genus (Morek et al. 2019a; Morek and Michalczyk, in press), including examples of both increases and decreases in the number of spurs on the secondary branches that can occur during both the first and the second moulting. Apart from CC, also the appearance of a dorsal cuticle may differ between life stages (e.g., pseudopores may be present only in adult animals; Morek et al. 2016a). In this contribution, we present a case study where Milnesium specimens with four different CCs were found in a single lichen sample, indicating the presence of multiple species and/or multiple phenotypes representing different ontogenetic stages. We show how we dealt with this challenging material and obtained integrative data for each species. As a result, we present a description of a new, pseudocryptic species, which is most similar to M. tardigradum, but is characterised by an extraordinary ontogenetic CC change pattern. We also discuss the importance of our findings in relation to the assessment of geographic ranges of Milnesium species as well as to the taxonomy of the genus. MATERIALS AND METHODS Sampling and specimens handling A lichen sample collected from stone by Małgorzata Mitan and Małgorzata Osielczak on 07.10.2015 in Reykjavík Botanic Garden, Iceland, 64°08'23"N, 21°52'09"W, 14 m asl was examined according to the standard protocol described by Stec et al. (2015). The sample yielded multiple Milnesium specimens exhibiting four different claw configurations (CCs), which indicated the presence of more than a single species or/and multiple developmental phenotypes. In order to identify the species and instars exhibiting particular phenotypes, we followed the workflow scheme presented in figure 1 (the detailed descriptions of particular methods are presented below). Firstly, the dead animals were preserved on permanent microscope slides (Fig. 1. step 1) to examine general morphology of animals in phase contrast light microscopy (PCM) and to assess how many CCs and putative species are present in the sample. In the next step, the alive specimens and exuviae with eggs were individually examined on the temporary water slides (Fig. 1. step 2) to determine the CC of each specimen/exuvia. This laborious procedure allows to accurately assess how many specimens per CC (putative species/developmental stage) are available for further analysis and therefore enables a desired assignment of specimens to most important analysis. In this study, the animals representing each CC were destined for DNA sequencing (Fig. 1, step 4), whereas the remaining ones, together with the exuviae with eggs, were placed in 24-well plastic plates to establish a culture, where the animals were kept individually (Fig. 1, step 3). This kind of culture allows for developmental tracking (Fig. 1, step 6). Provided the cultures were thriving, additional animals can be used for slide preparation (Fig. 1, step 1), DNA sequencing (Fig. 1, step 4), imaging in scanning electron microscopy (SEM; Fig. 1, step 5) and developmental tracking (Fig. 1, step 6). For the exact number of specimens destined for each analyses in this study, see table 1. Microscopy, imaging and morphometry The specimens were mounted on permanent microscope slides in Hoyer’s medium according to the method by Morek et al. (2016b) to examine general morphology in PCM and provide morphometric data. The measurements follow Tumanov (2006), the buccal tube widths were measured according to Michalczyk et al. (2012a) and the body length was measured from the anterior to posterior margin of the body, excluding the hind legs. The pt index is a ratio of a given structure to the length of the buccal tube (Pilato 1981), and in the text is given in italics. The number of measured specimens follows the recommendation of Stec et al. (2016). The morphometric data were handled using the Apochela spreadsheet ver. 1.3. available from the Tardigrada Register (Michalczyk and Kaczmarek 2013), www.tardigrada.net. All the measurements and page 2 of 15Zoological Studies 58: 32 (2019) © 2019 Academia Sinica, Taiwan Fig. 1. The scheme illustrating the general workflow when dealing with samples containing Milnesium specimens characterised by multiple claw configurations (CCs). After the extraction of specimens and exuviae with eggs from the sample, the material is divided into the dead and alive specimens. The dead ones are mounted on permanent slides (1) and immediately examined under PCM to check for the potential presence of multiple species in the sample. Provided there is a large number of dead specimens, they can be also used for imaging in SEM (6) but note there can be multiple species in the sample. Next, the alive specimens and exuviae are examined on the temporary water slide to assign the CC to the specimen (2). This laborious step allows for the assessment of the exact number of specimens per CC available for further analysis. In this study we immediately set aside some specimens from each CC for DNA extraction and sequencing (4). In parallel, a separate culture for each CC is established (3) as when the culture is thriving additional specimens can be dedicated for required analysis i.e., permanent slides preparation (1), DNA sequencing (4), imaging in SEM (5) or developmental tracking (6). page 3 of 15 Zoological Studies 58: 32 (2019) © 2019 Academia Sinica, Taiwan photographs were taken with the Olympus BX53 PCM, associated with the Olympus DP74 digital camera (PCM). For deep-focus structures a series of up to 22 pictures were taken and merged into one image using Corel Photo-Paint X8. Some specimens were processed for SEM imaging according to the protocol by Stec et al. (2015) and examined under high vacuum with HITACHI S-4700 SEM at the Institute of Geological Sciences, Jagiellonian University. Culturing and developmental tracking Initially, the specimens isolated from the lichen sample were individually kept on 24-well plastic plates and reared under conditions described in Kosztyła et al. (2016), at 10°C. The ontogenetic variability was assessed by developmental tracking (Morek et al. 2016a) for each CC separately. In short, individuals representing consecutive instars were mounted on permanent microscope slides and later examined under PCM. At least three specimens per each of the three putative species and each of three stages were collected (minimum 27 specimens for the entire tracking). Once the developmental tracking was complete, the remaining specimens were transferred from the plates to plastic Petri dishes to facilitate further culture growth (Stec et al. 2015). Genotyping The DNA was extracted from individual specimens (for sample size see Table 1) following the Chelex® 100 resin (Bio-Rad) extraction method by Casquet et al. (2012), with modifications by Stec et al. (2015). Before the DNA extraction, the specimens were mounted on temporary water slides in order to identify the morphotype (CC) and associate it with the genotype. The four standard molecular markers were sequenced, three nuclear: the small ribosomal subunit (18S rRNA), large ribosomal subunit (28S rRNA), Internal Transcribed Spacer 2 (ITS-2); and one mitochondrial, Cytochrome Oxidase C subunit I (COI). The PCR protocols followed Stec et al. (2015), primers and PCR programmes with relevant references are listed in table 2. The obtained chromatograms were checked manually in BioEdit ver. 7.2.5 (Hall 1999). In addition, the COI sequences were translated into amino acids using MEGA 7 (Kumar et al. 2016) to ensure that no pseudogenes were amplified. All sequences are deposited in GenBank (accession numbers are listed in Table 1). RESULTS Morphological species delineation In the first step, permanent slides made from dead animals (Fig. 1, step 1) found in the sample were analysed, revealing the presence of four distinct CCs: [2-2]-[2-2], [2-3]-[2-2], [2-3]-[3-2] (including a single specimen with an additional spur on claws III), and [33]-[3-3]. Based on the latest key to the genus Milnesium (Morek et al. 2016a), the species were initially identified Table 1. The details of the Milnesium species found in the analysed sample, with indication of the claw configuration present in each species. Data/analysis types: PCM - imaging and morphometry in light microsopy, DNA - DNA extraction and sequencing, DEV - developmental tracking, SEM - imaging in scanning electron miscroscopy. The GenBank accession numbers are provided below the species names Claw configuration Species Number of specimens analysed PCM DNA DEV SEM [2-2]-[2-2]: hatchlings and juveniles [2-3]-[2-2]: adults Milnesium variefidum 18S rRNA: MK484080 28S rRNA: MK483988 ITS-2: MK484014 COI: MK492293 18 2 9 0 [3-3]-[3-3]: hatchlings [2-3]-[3-2]: juveniles and adults Milnesium pseudotardigradum sp. nov. 18S rRNA: MK484088 28S rRNA: MK483997 ITS-2: MK484022 COI: MK492297 108 5 93 20 [3-3]-[3-3]: all life stages Milnesium sp. nov. 18S rRNA: MK484089 ITS-2: MK484023 10 3 9 0 page 4 of 15Zoological Studies 58: 32 (2019) © 2019 Academia Sinica, Taiwan as M. cf. variefidum Morek et al. 2016a ([2-2]-[2-2] and [2-3]-[2-2]), M. cf. tardigradum ([3-3]-[3-3] and [2-3]- [3-2]), and a putative new species with a [3-3]-[3-3] CC. However, this morphological identification was only preliminary and served as a starting point for further genetic and developmental analyses, which allowed for an integrative species delineation and identification. The alive specimens were either destined for DNA extraction and sequencing (Fig. 1, step 4), with prior examination of the CC on temporary water slides, or used to establish isogenic cultures (Fig. 1, step 3). In the long term, only cultures founded with specimens with adult [2-3]-[3-2] CC were stable culture and provided sufficient numbers of animals for all planned analyses. Genetic delineation Individuals with a [2-2]-[2-2] CC and a [2-3]-[22] CC shared the same haplotypes in all four markers, unequivocally indicating they represent a single species. The comparison of these DNA sequences with all available Milnesium sequences showed the closest affinity to the type sequences for M. variefidum (p-distances: 0.0% in 18S rRNA and 28S rRNA, 2.6– 3.0% in ITS-2, and 3.6% in COI). Thus, genetic data confirmed the initial morphological identification of [22]-[2-2] CC individuals as hatchlings and juveniles, and [2-3]-[2-2] CC animals as adults of M. variefidum. Similarly, some small specimens with a [3-3]- [3-3] and all medium or large individuals with a [23]-[3-2] CC exhibited the same haplotypes in each of the four sequenced markers, signifying that they represent a single species. The comparisons with all available Milnesium sequences indicated that despite morphological and ontogenetic similarities to M. tardigradum, these individuals in fact represent a species new to science. The comparison of these DNA sequences with all available Milnesium sequences showed the closest affinity to M. tardigradum (p-distances: 1.6–2.1% in 18S rRNA; 3.8–4.1% in 28S rRNA, 12.5–22.1%; in ITS-2; and 12.6–14.0% in COI). Since we were able to obtain sufficient numbers of individuals for our integrative analyses, we formally describe this species as M. pseudotardigradum sp. nov. below. Finally, some small and all large specimens with a [3-3]-[3-3] CC shared the same haplotypes in both 18S rRNA and ITS-2 (28S rRNA and COI failed to sequence), indicating that they represent a single species. The comparisons with all available Milnesium sequences indicated that the individuals represent a species new to science. The comparison of these DNA sequences with all available Milnesium sequences showed the closest affinity to another undescribed species designated in Morek and Michalczyk (in press) as “Milnesium sp. nov. 5 GB.096” (p-distances: 1.0% in 18S rRNA; 3.9%; in ITS-2). Since we were not able to obtain sufficient numbers of individuals for our integrative analyses, we decided to postpone the description of this species until more specimens are available. Developmental variability Developmental tracking revealed that specimens with a [2-2]-[2-2] and a [2-3]-[2-2] CC represent the same species that exhibits late positive CC change. Some of the hatchlings with a [3-3]-[3-3] CC moulted into [2-3]-[3-2] juveniles, i.e., they exhibited early negative CC change. Finally, some specimens had a stable [3-3]-[3-3] CC from hatchlings through juveniles to adults. Thus, this analysis indicated the presence of three species in the sample, two with ontogenetic CC change and one with a stable CC. When mounting large specimens with a [2-3]-[32] CC from the culture (Fig. 1, step 1) on microscope slides, we spotted that in the largest animals (presumably after the 3rd moult), additional spurs may appear, especially on claws II and III. Thus, we extended the developmental tracking to older life stages in order Table 2. Primers and references for specific protocols for amplification of the four DNA fragments sequenced in the study DNA fragment Primer name Primer direction Primer sequence (5’-3’) Primer source PCR programme 18S rRNA 18S_Tar_Ff1 forward AGGCGAAACCGCGAATGGCTC Stec et al. (2017) Zeller (2010) 18S_Tar_Rr1 reverse GCCGCAGGCTCCACTCCTGG 28S rRNA 28S_Eutar_F forward ACCCGCTGAACTTAAGCATAT Gąsiorek et al.(2018) Mironov et al. (2012) 28SR0990 reverse CCTTGGTCCGTGTTTCAAGAC Mironov et al. (2012) ITS-2 ITS2_Eutar_Ff forward GCATCGATGAAGAACGCAGC Stec et al. (2018) Stec et al. (2018) ITS2_Eutar_Rr reverse TCCTCCGCTTATTGATATGC COI COI_Mil.tar_Ff forward TATTTTATTTTTGGTATTTGATGTGC Morek et al. (2019a) Morek et al. (2019a) COI_Mil.tar_Rr reverse CCTCCCCCTGCAGGATC page 5 of 15Zoological Studies 58: 32 (2019) © 2019 Academia Sinica, Taiwan to test whether the additional spurs were aberrations present only in some specimens or whether it was a stable pattern characterising older animals. The tracking revealed first that they are truly additional spurs, because they exhibit typical morphology, i.e., they are of a typical shape and similar size, being only slightly smaller than the regular ones, in contrast to aberrations, which are shorter and less curved or even straight (Kaczmarek et al. 2012; Morek et al. 2019a). Secondly, the additional spurs are very common in the 4th instar onwards, at least one additional spur is present in 86% specimens, mostly on claws II–IV (Fig. 2). Most commonly, specimens are equipped with two additional spurs (19%), although single specimens with zero and eight spurs (i.e., resulting in a [3-3]-[3-3] CC) were observed (see also Description of females below). TAXONOMIC ACCOUNT Phylum: Tardigrada Doyère, 1840 Class Apotardigrada Guil, Jørgensen and Kristensen, 2018 Order Apochela Schuster et al., 1980 Family Milnesiidae Ramazzotti, 1962 Genus Milnesium Doyère, 1840 Milnesium pseudotardigradum sp. nov. (Figs. 2-6, Table 3) urn:lsid:zoobank.org:act:0BE101DC-62F5-44FE-BEF106FDA9B4B157 Integrative description Females (morphometrics and holotype measurements in Table 3): Milnesium of moderate length, up to 862 µm (Fig. 3A, B), yellowish. Eyes present in alive specimens, but dissolved in 80% of specimens (86/108 of the type series) mounted in Hoyer’s medium. Cuticle smooth without pseudopores or pseudoplates both in PCM (Fig. 3E) and SEM (Fig. 3F). The six rather short peribuccal papillae present around the mouth opening, with the ventral one being the smallest. Mouth opening with six triangular peribuccal lamellae of unequal size, i.e., the two lateral lamellae are significantly smaller than the pair of dorsal and lateral lamellae, the 4+2 configuration (Fig. 3D), which is detectable only under SEM. Two short lateral papillae present. Buccal tube funnel shape and of moderate width (Fig. 3C). Claws typical for this genus, primary branches with tiny, but well-visible accessory points (Fig. 4C, D, H). Typically internal and anterior secondary branches equipped with a basal spur, i.e., with a [2-3]-[3-2] CC (Fig. 4C, D, G). In the majority of specimens from the 4th life stages onwards (86% of the type series) additional spurs on external and/or posterior secondary branches may be present (Fig. 4E, F). These additional spurs are always smaller than the regular spurs (Fig. 4H), but they are characterised by a regular, normal shape, indicating they are not aberrations. The number of extra spurs in a single animal ranges from one to eight resulting in such cases in specimens with a [3-3]-[3-3] CC. This means that the number of spurs can be uneven and different on left and right pairs of claws. Cuticular bars under claws I–III are always present. Fig. 2. The graph showing the frequencies of number of additional spurs in 4th+ specimens of Milnesium pseudotardigradum sp. nov. on internal/ posterior claws of each pair of legs. page 6 of 15 Zoological Studies 58: 32 (2019) © 2019 Academia Sinica, Taiwan Fig. 3. Milnesium pseudotardigradum sp. nov. A, habitus, ventral view (holotype, PCM). B, habitus, dorsal view (paratype, SEM). C, Buccal apparatus (holotype, PCM). D, six peribuccal lamellae of unequal size, i.e., 4 + 2 configuration (paratype, SEM). E, smooth dorsal cuticle (holotype, PCM). F, smooth dorsal cuticle (paratype, SEM). All the scale bars in µm. page 7 of 15 Zoological Studies 58: 32 (2019) © 2019 Academia Sinica, Taiwan Fig. 4. Milnesium pseudotardigradum sp. nov. A, claws III of hatchling with [3-3]-[3-3] CC (paratype, PCM), the empty arrow indicate the external spur, which disappear after the first moult. B, claws IV of hatchling with [3-3]-[3-3] CC (paratype, PCM) ), the empty arrow indicate the posterior spur, which disappear after the first moult. C, claws III of adult (paratype, PCM), the arrow indicates the accessory points. D, claws IV of adult (holotype, PCM), the arrow indicates the accessory points. E, claws III of adult (paratype, PCM) with additional spur indicated by empty arrow. F, claws IV of adult (paratype, PCM) with additional spur indicated by empty arrow. G, claws III of adult (paratype, SEM). H, claws IV of adult (paratype, SEM) with additional spur indicated by empty arrow. All the scale bars in µm. page 8 of 15Zoological Studies 58: 32 (2019) © 2019 Academia Sinica, Taiwan Table 3. Measurements (in μm) and the pt values of selected morphological structures of 108 specimens of Milnesium pseudotardigradum sp. nov. from type locality, Reykjavík Botanic Garden, Iceland, mounted in Hoyer’s medium. Individuals were chosen to represent the entire body length range, with as equal representation of all available life stages as possible CHARACTER N RANGE MEAN SD µm pt µm pt µm pt Body length 106 290–862 1312–2092 639 1791 155 166 Peribuccal papillae length 82 3.8–10.3 13.7–24.8 7.2 19.9 1.8 2.5 Lateral papillae length 94 3.3–7.9 11.0–21.8 5.5 15.6 1.2 1.8 Buccal tube Length 108 21.6–45.4 – 35.4 7.2 Stylet support insertion point 104 15.2–30.3 61.8–75.9 23.6 67.2 4.5 2.9 Anterior width 99 8.2–19.4 32.5–49.4 15.0 42.2 3.4 3.5 Standard width 98 7.0–20.3 25.5–50.8 13.4 37.5 3.9 5.7 Posterior width 99 6.5–20.6 26.5–52.0 13.2 37.0 3.8 6.1 Standard width/ length ratio 98 25%–51% – 38% 7% Posterior/anterior width ratio 98 65%–121% – 87% 14% Claw 1 heights External primary branch 87 10.0–20.3 36.6–53.7 15.6 44.1 3.1 3.0 External base + secondary branch 65 5.8–15.7 26.2–37.4 11.5 32.8 2.6 2.4 External spur 11 2.5–3.3 11.0–14.1 2.9 12.7 0.9 1.0 External branches length ratio 58 57%–80% – 75% 11% Internal primary branch 88 9.9–19.4 34.9–51.2 14.7 42.8 3.0 3.2 Internal base + secondary branch 92 7.2–15.6 26.0–37.2 11.3 32.6 2.4 2.6 Internal spur 91 2.6–9.0 10.7–22.0 6.3 17.4 1.6 2.1 Internal branches length ratio 77 61%–91% – 76% 10% Claw 2 heights External primary branch 89 10.5–21.3 40.1–58.2 17.1 48.8 3.3 3.4 External base + secondary branch 79 7.1–15.3 29.1–41.0 12.2 35.3 2.6 2.7 External spur 19 2.3–6.7 10.4–18.8 4.3 13.7 1.5 2.1 External branches length ratio 68 60%–81% – 72% 10% Internal primary branch 90 10.1–21.0 39.1–55.6 16.5 47.4 3.4 3.6 Internal base + secondary branch 82 7.1–15.5 25.8–51.9 11.6 34.1 2.6 3.5 Internal spur 99 3.2–9.6 14.1–23.4 7.0 19.5 1.7 1.9 Internal branches length ratio 68 59%–81% – 72% 10% Claw 3 heights External primary branch 93 10.8–23.0 40.1–57.5 17.6 49.8 3.4 3.3 External base + secondary branch 74 6.9–15.9 28.0–41.6 12.2 35.5 2.7 2.8 External spur 22 2.9–7.4 7.6–19.9 4.5 13.6 1.5 3.2 External branches length ratio 65 63%–78% – 71% 9% Internal primary branch 81 10.5–21.2 39.9–56.6 17.1 48.3 3.4 3.9 Internal base + secondary branch 63 6.2–14.4 28.1–42.0 11.6 34.1 2.5 3.0 Internal spur 93 3.4–9.4 15.0–23.5 6.9 19.5 1.5 1.8 Internal branches length ratio 48 61%–82% – 71% 11% Claw 4 heights Anterior primary branch 107 10.2–25.5 43.5–71.9 21.1 59.6 4.2 6.0 Anterior base + secondary branch 100 7.5–18.0 31.9–45.4 13.9 38.9 3.0 3.5 Anterior spur 91 3.1–9.3 12.9–23.1 6.8 18.9 1.6 2.1 Anterior branches length ratio 100 53%–83% – 66% 8% Posterior primary branch 103 11.1–27.2 46.4–76.0 22.5 63.7 4.5 5.9 Posterior base + secondary branch 102 7.0–20.4 31.7–51.0 14.9 41.8 3.3 4.1 Posterior spur 18 2.2–5.7 5.3–20.3 3.7 12.6 1.2 3.7 Posterior branches length ratio 98 55%–77% –66% 8% page 9 of 15Zoological Studies 58: 32 (2019)