Echinoderes cyaneafictus, Cepeda & González-Casarrubios & Sánchez & Spedicato & Michaud & Zeppilli, 2022

Cepeda, Diego, González-Casarrubios, Alberto, Sánchez, Nuria, Spedicato, Adriana, Michaud, Emma & Zeppilli, Daniela, 2022, Two new species of mud dragons (Scalidophora: Kinorhyncha) inhabiting a human-impacted mangrove from Mayotte (Southwestern Indian Ocean), Zoologischer Anzeiger 301, pp. 23-41 : 33-38

4.1. Designation of Echinoderes cyaneafictus sp. nov. and E. parthenope sp.nov. to a putative species group

One of the best supported putative groups within Echinoderes is the E. coulli -group ( Sørensen 2014; Randsø et al., 2019). The group is mainly composed of intertidal, brackish species with enlarged nephridiopores (oval or triangular, convex plate with multiple pores plus a posterior, oval, concave area with a single pore), a feature that supposedly confers an adaptive advantage to cope with extreme salinity fluctuations ( Lundbye et al., 2011; Randsø et al., 2019). However, there are some exceptions to this. Echinoderes regina Yamasaki, 2016 , Echinoderes rex Lundbye et al., 2011 and Echinoderes serratulus Yamasaki, 2016 are strictly subtidal (up to 13 m depth), marine species ( Lundbye et al., 2011; Yamasaki 2016). In addition, there is a still undescribed species from the New Zealand deep-sea (up to 1013 m depth) (Sørensen, personal communication). These species unequivocally belong to the group since they possess the enlarged nephridiopore, suggesting its adaptation to salinity fluctuations ( Randsø et al., 2019). This E. coulli -group is furthermore characterized by sharing the following combination of morphological features: middorsal spines absent or poorly developed on segment 4 only, lateroventral spines absent or poorly developed on segments 6–7 only, female lateral terminal accessory spines absent or poorly developed, and lateral tubes on segments 5 and 8 ( Sørensen 2014; Randsø et al., 2019). However, as it can be seen, the presence of enlarged nephridiopores remains as the only consistent apomorphy present throughout the group ( Randsø et al., 2019).

E. cyaneafictus sp. nov. seems to fit well within the E. coulli -group since it has all the morphological characteristics shared by the members of the group ( Randsø et al., 2019). E. cyaneafictus sp. nov. shows closest resemblance to E. regina , E. rex and E. serratulus , of which E. rex is part of the Clade C/III ( Randsø et al., 2019). The phylogenetic position of E. regina still remains problematic, as it is very sensitive to the different inference approaches and choice of molecular markers ( Randsø et al., 2019). E. regina was never recovered in the molecular and total-evidence phylogenies together with E. rex or E. serratulus , and the support value of Clade A ( E. regina + Echinoderes annae Sørensen et al., 2016 and E. serratulus ) in the morphological phylogenies was low ( Randsø et al., 2019). On the other hand, E. serratulus seems to be more closely related to E. annae and Echinoderes hwiizaa Yamasaki & Fujimoto, 2014 according to both molecular and total-evidence phylogenies (Randsø et al.,

2019) and, as above mentioned, its occurrence together with E. regina in Clade A in the morphological phylogenies cannot be trusted. Thus, this apparent morphological resemblance of E. regina , E. rex , E. serratulus and E. cyaneafictus sp. nov. is not indicative of a close relationship.

We also consider E. parthenope sp. nov. part of the E. coulli -group since it agrees with the only identified morphological apomorphy: the presence of enlarged nephridiopores composed of an anterior, oval or triangular, convex plate with multiple pores and a posterior, oval, concave area with a single, large pore ( Randsø et al., 2019). Actually, the shape of its nephridiopores, as a triangle instead of an oval area,

(caption on next page)

resembles that of Echinoderes maxwelli Ostmann et al., 2012 and E. komatsui ( Lundbye et al., 2011; Yamasaki & Fujimoto 2014). However, Randsø et al. (2019) did not find support for a homology between the triangular sieve plates.

4.2. Remarks on diagnostic and differential taxonomic features of Echinoderes cyaneafictus sp. nov. and Echinoderes parthenope sp.nov

The E. coulli -group currently comprises 13 species: E. annae , E. applicitus , E. coulli Higgins, 1977 , E. hwiizaa , E. komatsui , Echinoderes marthae Sørensen, 2014 , E. maxwelli , Echinoderes ohtsukai Yamasaki & Kajihara, 2012 , E. regina , E. rex , E. serratulus , E. strii , and E. teretis ( Randsø et al., 2019) . Furthermore, Echinoderes aspinosus Sørensen et al., 2012 , E. bengalensis , and E. caribiensis could also be part of the group, although the presence of the characteristic sieve plate needs to be confirmed ( Randsø et al., 2019).

E. cyaneafictus sp. nov. shares a similar spine and tube pattern with

E. maxwelli , E. regina and E. serratulus , with slight differences. Spines of segment 6 are in lateroventral position in E. maxwelli (sublateral in the new species), and tubes of segment 8 are in sublateral position in E. regina and lateroventral position in E. serratulus (lateral accessory in the new species) ( Yamasaki 2016; Randsø et al., 2019). Furthermore, E. serratulus possesses tubes in midlateral position on segment 9 ( Yamasaki 2016), which are absent in E. cyaneafictus sp. nov. The three congeners may also be distinguished from the new species by the presence of minute type 2 glandular cell outlets throughout segments 1–8 ( Yamasaki 2016; Randsø et al., 2019), structures otherwise missing in E. cyaneafictus sp. nov. Finally, the new species is characterized by three pairs of large, rounded sensory spots on segment 1 with a traversal row of conspicuously elongated hairs at the posterior part of the papillae area. Although the other three species also have sensory spots on segment 1, they are morphologically different and lack the significant, posterior row of elongated hairs. These sensory spots have been observed in the same positions of segment 1 of another still undescribed congener of the E. coulli -group with scanning electron microscopy (Cepeda et al. unpublished). E. cyaneafictus sp. nov. may be distinguished from this other congener by the absence of type 2 glandular cell outlets, apart from other minor differences concerning the sensory spot arrangement (Cepeda et al. unpublished).

E. parthenope sp. nov. is unique within the E. coulli -group due to the presence of spines on segment 8. Only E. caribiensis was originally described with lateroventral spines on this segment ( Kirsteuer 1964), but the description of this species is quite incomplete, type material is missing and it has been never found again (despite the multiple taxonomic efforts done at the Caribbean area, e.g. Higgins 1983; Sørensen 2006; Pardos et al., 2016b; Cepeda et al., 2019a; 2019c). Thus, we can question the actual validity of the morphological data concerning E. caribiensis . Members of the E. coulli -group may have spines in middorsal position on segment 4 and in lateral position on segments 6–7, but these spines are conspicuously small, poorly sclerotized and weakly articulated. In contrast, tubes, which may be present at different positions throughout the trunk, are large and well developed ( Sørensen 2014; Randsø et al., 2019). Thus, the lateroventral spines on segments 8–9 of E. caribiensis could be actually tubes, as they were described and illustrated as large, conspicuous, well sclerotized cuticular appendages ( Kirsteuer 1964). Nevertheless, the spine and tube pattern of E. caribiensis is completely different from that of E. parthenope sp. nov., including the absence of middorsal spine and lateral spines on segments 6–7 ( Kirsteuer 1964), structures otherwise present in the new species. The characteristic triangular sieve plate of E. parthenope sp. nov. is also found in E. komatsui and E. maxwelli , but both congeners may be easily distinguished from the new species. E. komatsui lacks a middorsal spine on segment 4 and lateral spines on segments 6–7 ( Yamasaki & Fujimoto 2014), which are present in E. parthenope sp. nov. E. maxwelli also lacks spines on segment 8, having tubes instead in lateral accessory position ( Ostmann et al., 2012; Randsø et al., 2019) ( E. parthenope with lateral accessory spines and sublateral tubes on this segment), but otherwise is quite similar to the new species in terms of spine and tube arrangement. However, both species differ in the type 2 glandular cell outlet arrangement as well as some sexually dimorphic features, including the presence of laterodorsal tubes on segment 10 only in males in E. parthenope sp. nov (both sexes in E. maxwelli ) and the shape of the male tergal plate of segment 11 ( Randsø et al., 2019), which is unique of the new species.