Platevindex burnupi ( Collinge, 1902 ) Goulding & Bourke & Comendador & Khalil & Quang & Tan & Tan & Dayrat, 2021

Goulding, Tricia C., Bourke, Adam J., Comendador, Joseph, Khalil, Munawar, Quang, Ngo Xuan, Tan, Shau Hwai, Tan, Siong Kiat & Dayrat, Benoît, 2021, Systematic revision of Platevindex Baker, 1938 (Gastropoda: Euthyneura: Onchidiidae), European Journal of Taxonomy 737 (1), pp. 1-133 : 88-96

Platevindex burnupi ( Collinge, 1902) comb. nov.

Figs 46–53

Onchidium burnupi Collinge, 1902: 17 , figs 1–2.

Onchidium burnupi – Collinge 1910: 171–172. — Connolly 1912: 224; 1939: 456. — Hoffmann 1928: 84.

Material examined

Lectotype SOUTH AFRICA • lectotype (here designated; 21/ 16 mm); Natal , Umlaas Lagoon [now the Umlazi River]; NHMUK 1905.4.12.33 .

Paralectotype SOUTH AFRICA • 1 paralectotype (18/ 12 mm); NHMUK 1905.4.12.34 .

Notes on type material The notum of the lectotype was cut open for the present study, with all internal organs remaining. The paralectotype was not dissected.

Other material MADAGASCAR • 1 spec. (10/6 [3599] mm); Mangaoka area ; 12°15.984′ S, 49°06.103′ E; 18 May 2014; station 160; mangrove of Avicennia and Rhizophora , many small trees in center of mangrove had been cut; MNHN-IM-2019-1396 • 1 spec. (16/9 [3150] mm); Madagascar, Ambaro Bay; 13°24.769′ S, 48°47.341′ E; 23 May 2014; Station 164; large, very dry mangrove of mostly Avicennia ; MNHN- IM-2019-1398 • 1 spec. (14/9 [3151] mm);same collection data as for preceding; MNHN-IM-2019-1399 • 1 spec. (12/6 [3602] mm); same collection data as for preceding; MNHN-IM-2019-1397 • 1 spec. (16/10 [3603] mm); same collection data as for preceding; MNHN-IM-2019-1395 .

MOZAMBIQUE • 1 spec. (13/10 [729] mm); Mecufi , 50 km S of Pemba, Cabo Delgado Province [about 13° S]; 13 Jul. 2006; D.G. Reid leg.; on base of Ceriops bush in landward zone of mangrove belt, 30 cm above firm mud; NHMUK 20060274 .

SOUTH AFRICA • 1 spec. (32/11 [6146] mm); East Cape , Kwelera River estuary ; 32°54.169′ S, 28°03.954′ E; 23 Jun. 2017; M. Cole leg.; mud amongst plants in supratidal fringe; ELM W04105 • 1 spec. (23/9 [6154] mm); same collection data as for preceding; ELM W04107 • 1 spec. (22/10 [6153] mm); same collection data as for preceding; ELM W04106 • 1 spec. (20/10 [6149] mm); same collection data as for preceding; ELM W04108 • 1 spec. (19/7 [6148] mm); same collection data as for preceding; ELM W04109 .

Description

Color and morphology of live animals ( Fig. 46 ) The dorsal notum of live animals can be covered with mud, in which case their color may be obscured until they are washed. The dorsal notum is brown with dark brown or black markings. The hyponotum is light blue-grey and the foot is yellow. Papillae with dorsal eyes are present. Their exact number is variable, approximately from 15 to 38 were observed in specimens from South Africa and between 12 and 20 were observed in specimens from Madagascar. Each papilla bears one dorsal eye. Eyes are distributed across the notum but are absent from the margin (i.e., eyes are never <2 mm from the notum edge).

Digestive system ( Figs 2G , 47–48B , 53 ) Radulae measure up to 3.3 mm in length. Examples of radular formulae are presented in Table 5 . The intestinal loops are of type II, with a transitional loop oriented at approximately 7 o’clock ( Figs 2G , 48A , 53 ).

Reproductive system ( Figs 49–51 ) In the posterior (female) reproductive system, the oviduct is short, wider than the deferent duct (approximately twice as wide). The length of its distal section (distal to the spermatheca) is shorter than or equal to that of its proximal section. The deferent duct is slightly longer than the oviduct, is only loosely attached to it and varies from loosely to tightly coiled ( Fig. 49 ). The female pore is approximately 0.5 to 2 mm from the anus. The distal, flexible region of the penis with hooks is 1 to 1.5 mm long. Penial hooks are large, between 25 to 70 µm long ( Fig. 51A ), and can be seen inside the semi-transparent penis. The retractor muscle of the penis inserts on the right side at the posterior end of the visceral cavity (from the posterior end of the heart, up to the posterior end of the visceral cavity). The length of the retractor muscle varies from half the length of the penial sheath to approximately two times its length ( Fig. 50B ). The deferent duct is highly convoluted.

Distinctive diagnostic features ( Table 4 )

Platevindex burnupi is the only known species of Platevindex from the western Indian Ocean. It can be distinguished from P. martensi and P. aptei sp. nov. by its grey hyponotum and yellow foot, and from P. amboinae and P. latus by the presence of dorsal eyes. Platevindex burnupi is similar in appearance to P. luteus and P. applanatus , but lacks distinctly raised dorsal papillae. Its dorsal surface can even be smooth. It is difficult to distinguish P. burnupi from P. coriaceus and P. tigrinus due to their similar coloration and granular dorsal appearance. Although individuals of P. burnupi tend to be shorter than those of P. coriaceus and P. tigrinus , animal sizes overlap and cannot be used for identification.

Internally, the intestinal loops of type II differentiate P. burnupi from P. amboinae , P. latus , P. applanatus and some specimens (with intestinal loops of type I) of P. luteus . The reproductive system can be used to distinguish P.burnupi from all other species of Platevindex . The deferent duct in the posterior reproductive system is loosely attached to the oviduct in P. burnupi , which distinguishes it from P. coriaceus and P. tigrinus (deferent duct closely attached to the oviduct). The distal region of the penis with penial hooks is also much shorter in P. burnupi (1–1.2 mm) than in P. coriaceus (2–8 mm). Platevindex burnupi is very difficult to differentiate from P. luteus internally. They differ only slightly with respect to the insertion of the penial retractor muscle (near the heart in P. luteus and in the last posterior quarter of the visceral cavity in P. burnupi ). DNA sequences are the most reliable way to distinguish Platevindex burnupi and P. luteus , which are more than 10.5% divergent based on COI sequences.

Distribution ( Fig. 10A )

Madagascar, Mozambique and South Africa (type locality and newly-collected material).

Habitat ( Figs 45 A–B, 52C–D)

Like other species of Platevindex , P. burnupi is found on mangrove trees but, unlike other species, it may also be found on mud. Specimens in Madagascar were found high in the intertidal on moist mud not saturated in water, and one specimen was found on a piece of wood. A specimen from Mozambique was collected on the base of a Ceriops tree, 30 cm above ground. All specimens in South Africa were on mud, in the supra-tidal zone of a salt marsh, a few meters away from the river.

Remarks

Onchidium burnupi must be transferred to Platevindex because of the following combination of characters in the lectotype: flattened body, rectal gland present and no penial gland. The type locality (Umlaas Lagoon, Natal, South Africa) was visited twice by our team and several times by experts from the Natal Museum (Dai Herbert, pers. com.), but no onchidiid species were ever found. This is very likely due to the fact that the Umlaas Lagoon has been developed and natural vegetation destroyed. However, onchidiids similar to the lectotype of O. burnupi were recently collected by Mary Cole and Philippe Bouchet in the estuary of the Kwelera River, East Cape, South Africa (approximately 400 km south of the type locality). As of today, no other live population of P. burnupi is known in South Africa (assuming that there is no live population at the type locality). Note that several authors have mentioned the existence of Onchidium burnupi in South Africa but exclusively based on the original

description, i.e., without providing new specimens or records ( Connolly 1912, 1939; Collinge 1910; Hoffmann 1928).

Platevindex burnupi is known from two localities in South Africa, three localities in Madagascar and one locality in Mozambique. High genetic divergences between populations from Madagascar and Mozambique versus South Africa (5.4–6.0%) as well as the fact that they are consistently recovered as reciprocally-monophyletic units ( Figs 4–8 ) suggest that there could be more than one species. However, there is not enough evidence to justify the description of a new species in Madagascar. The anatomy of populations from South Africa and Madagascar are similar, and little is known about intraspecific variation across its distribution. The genetic divergences between Madagascar and East Africa ( Mozambique and South Africa) could be the result of the large geographic distance between localities. However, there are some noticeable differences between the populations in the two regions.

The most noticeable difference between regions is animal length: individuals collected from South Africa (up to 32 mm) are significantly larger than those from Madagascar (up to 16 mm) and Mozambique (13 mm). However, intermediate sizes may be found when additional populations are discovered, or this could be due to a difference in climate. The deferent duct in the posterior (female) reproductive system is slightly longer in specimens from Madagascar than in specimens from South Africa (even though animal size is larger in South Africa), but this could also be the result of population structure (this trait is known to vary between individuals). In addition, the insertion of the retractor muscle is more posterior in specimens from Madagascar than in South Africa. Finally, Madagascan individuals were found in a different habitat (mangrove) than in South Africa (salt marsh). Salt marshes in Madagascar were not visited for this study and may need to be explored in the future.

Figures

Fig. 46. Platevindex burnupi (Collinge, 1902) comb. nov., live animals.A–B. Dorsal view [unnumbered], South Africa, East London. C. Dorsal view, 16 mm long [3150], Madagascar, Ambaro Bay (MNHN- IM-2019-1398). D. Dorsal view, 10 mm long [3599], Madagascar, near Mangaoka (MNHN- IM-2019-1396). E. Dorsal view, 14 mm long [3151], Ambaro Bay (MNHN-IM-2019-1399). F. Dorsal view, 22 mm long [6153], South Africa, East London (ELM W04108). G. Ventral view, 20 mm long [6149], South Africa, East London (ELM W04107). H. Ventral view, same as F. I. Ventral view, same as C. J. Ventral view, same as D.

Fig. 47. Platevindex burnupi (Collinge, 1902) comb. nov., radula, South Africa. A. Left, lateral teeth [6153]. B. Rachidian tooth and innermost lateral teeth, same as A. C. Right, lateral teeth [6149]. D. Right, outermost, lateral teeth [6148]. Abbreviation: lp = lateral protuberance. Scale bars: A = 30 μm; B = 10 μm; C–D = 20 μm.

Fig. 48. Platevindex burnupi (Collinge, 1902) comb. nov., digestive system, dorsal view, intestinal loops of type II, with a transitional loop oriented at 7 o’clock. A. South Africa, lectotype (NHMUK 1905.4.12.33). B. Madagascar [3150] (MNHN-IM-2019-1398). Abbreviations: ddg = dorsal digestive gland; i = intestine; pdg = posterior digestive gland; rg = rectal gland; st = stomach. Scale bars = 2 mm.

Fig. 49. Platevindex burnupi (Collinge, 1902) comb. nov., posterior reproductive system.A. SouthAfrica, lectotype (NHMUK 1905.4.12.33). B. Madagascar [3603] (MNHN-IM-2019-1395). Abbreviations: dd = deferent duct; fgm = female gland mass; hd = hermaphroditic duct; hg = hermaphroditic gland; ov = oviduct; rs = receptaculum seminis; sp = spermatheca. Scale bars: A = 2 mm; B = 1 mm.

Fig. 50. Platevindex burnupi (Collinge, 1902) comb. nov., anterior, male copulatory apparatus. A. South Africa [6148] (ELM W04109). B. South Africa, lectotype (NHMUK 1905.4.12.33). C. Madagascar [3150] (MNHN-IM-2019-1398). Abbreviations: dd = deferent duct; ps = penial sheath; rm = retractor muscle; v = vestibule. Scale bars: A = 1 mm; B–C = 2 mm.

Fig. 51. Platevindex burnupi (Collinge, 1902) comb. nov., penial hooks A. South Africa [6148] (ELM W04109). B. Madagascar, Ambaro Bay [3603] (MNHN-IM-2019-1395). Scale bars: A = 30 μm; B = 20 μm.

Fig. 52. Platevindex burnupi (Collinge, 1902) comb. nov., radula, Madagascar, Ambaro Bay [3603] (MNHN-IM-2019-1395). A. Rachidian and innermost lateral teeth. B. Lateral teeth. C–D. Outer lateral teeth. Scale bars: A, C–D = 20 μm; B = 30 μm.

Fig. 53. Habitats of Platevindex burnupi (Collinge, 1902) comb. nov. A. Madagascar, near Mangaoka (station 160). B–D. Madagascar, Ambaro Bay, vegetation and close view of the high intertidal mud (station 164). E–F. South Africa, East Cape, Kwelera Estuary, estuarine plants in the supra-tidal fringe of salt marsh.

Fig. 2. Intestinal loops of type II. Small black arrows show the direction of the intestinal transport. The clockwise loop is in blue. The counterclockwise loop is in yellow. The transitional loop (between clockwise and counterclockwise loops) is in green. Red arrows indicate the orientation of the transitional loop. A. Type II (as defined by Plate 1893), with a transitional loop oriented at 9 o’clock, redrawn from Plate (1893: pl. 8, fig. 31). B. Type II, with a transitional loop oriented at 9 o’clock, Platevindex coriaceus coriaceus (Semper, 1880) unit #2 [3170] (PNM 041236), Luzon, Philippines. C. Type II, with a transitional loop oriented at 9 o’clock, P. martensi (Plate, 1893) [2333] (UMIZ 00111), Sulawesi, Indonesia. D. Type II, with a transitional loop oriented at 6 o’clock, P. martensi [964] (USMMC 00029), Peninsular Malaysia. E. Type II, with a transitional loop oriented at 8 o’clock, P. luteus (Semper, 1880) [3367] (PNM 041246), Bohol, Philippines. F. Type II, with a transitional loop oriented at 8 o’clock, P. aptei Goulding & Dayrat sp. nov., [5955] (USMMC 00078), Peninsular Malaysia. G. Type II, with a transitional loop oriented at 7 o’clock, P. burnupi (Collinge, 1902) [3150] (MNHN-IM-2019-1398), Madagascar. H. Type II, with a transitional loop oriented at 7 o’clock, P. coriaceus coriaceus unit #2 [3256] (PNM 041238), Bohol, Philippines. Scale bars: B–H = 2 mm.

Fig. 10. Geographic distributions of species of Platevindex. Colored areas correspond to hypothetical ranges proposed based on all known records, and the colors are the same as those used for each species in the phylogenetic trees (Figs 1–5). Distinct colors are used for each subspecies and mitochondrial unit in P. coriaceus (Semper, 1880) as well as for each subspecies of P. burnupi (Collinge, 1902). Black dots show sites sampled for this study and red dots (China and Japan) show localities of previously published genetic sequences included in the phylogenetic analyses. A. Platevindex luteus (Semper, 1880), P. applanatus (Simroth, 1920), P. burnupi. B. Platevindex tigrinus (Stoliczka, 1869), P. coriaceus coriaceus (units #1, #2, and #3) and P. coriaceus darwinensis subsp. nov. C. Platevindex martensi (Plate, 1893), P. aptei Goulding & Dayrat sp. nov., P. amboinae (Plate, 1893) and P. latus (Plate, 1893).

Fig. 45. Habitats of Platevindex applanatus (Simroth, 1920) comb. nov., Indonesia (A–C), Papua New Guinea (D–E) and Mauritius (F). A. Ambon, log behind a rocky mangrove in the very high intertidal, shaded by trees (station 127). B. Kei Islands, rocks with algae at the edge of the mangrove (station 137). C. Kei Islands, Un, rocks in the high intertidal behind a mangrove, surrounded by dry vegetation, algae growing on the rocks (station 137). D. Madang, Hargun Island, night tide (station PM24). E. New Ireland, Povalval, east coast of New Ireland, mixed hard platform and seagrass bed at outlet of rivulet (station KM05). F. Mahebourg, close up of rotting petiole of palm frond at margin of mangrove where juvenile onchidiids were found.

Fig. 4. Phylogenetic relationships of species of Platevindex based on partitioned COI, 16S, ITS2 and 28S sequences. Support values are reported above each branch with bootstrap values (maximum likelihood) listed first, followed by posterior probabilities (Bayesian analysis).

Fig. 5. Maximum parsimony consensus tree of species of Platevindex based on unique ITS2 haplotypes. Numbers above the branches are bootstrap values.

Fig. 6. Maximum parsimony consensus tree of species of Platevindex based on ITS2 and 28S sequences. Numbers above the branches are bootstrap values.

Fig. 7 (part 1 of 3). Phylogenetic tree based on COI sequences for 195 individuals (plus outgroups). Only bootstrap values> 60% (maximum likelihood) and posterior probabilities>0.9 (Bayesian analysis) are indicated. Some outgroups sequences are from previous studies (Dayrat et al. 2011a, 2016, 2017, 2018, 2019a, 2019b, 2019c, 2019d; Goulding et al. 2018a, 2018b, 2018c). Information on specimens can be found in the lists of material examined and in Table 1. All sequences of specimens of Platevindex are new, except sequences obtained from GenBank for specimens from China.

Fig. 8 (part 1 of 2). Phylogenetic tree of species of Platevindex based on mitochondrial COI and 16S sequences. Support values are reported above each branch with bootstrap values (maximum likelihood analysis) followed by posterior probabilities (Bayesian analysis); only values>60% and> 0.9, respectively, are indicated.