Afrocyclus oxygala, Cole, 2019

Cole, Mary L., 2019, Revision of Chondrocyclus s. l. (Mollusca: Cyclophoridae), with description of a new genus and twelve new species, European Journal of Taxonomy 569, pp. 1-92 : 67-68

Afrocyclus oxygala gen. et sp. nov.

urn:lsid:zoobank.org:act:7B7B852C-8B49-494A-A652-B313EBDC6372

Figs 11 A–B, 28B, 29, 30

Chondrocyclus isipingoensis – Connolly 1939: 540 . — Herbert & Kilburn 2004: 91.

Diagnosis

Shell very small, depressed, discoidal; periostracum with axial costae producing spiral rows of simple hairs; operculum very fragile and duplex, exterior portion very shallowly concave, with low multispiral lamella terminating in a solid fringe, radula with two large cusps on second lateral tooth and rachidian tooth without serrated upper edge.

Etymology

The specific name is derived from the Greek ‘ oxygala ’, meaning ‘sour milk’, a translation of the name of the type locality, Maasstrom.

Type material examined

Holotype

SOUTH AFRICA – Eastern Cape • Bedford, Maasstrom Farm , forest in kloof; 32.6830°S, 26.0830°E; 24 Jan. 2011; M. Cole, R. Daniels and V. Ndibo leg.; NMSA P0638 /T4153. ( Fig. 11 A–B)

Paratypes

SOUTH AFRICA – Eastern Cape • 9 specimens; same collection data as for holotype; ELM D16895/ T151 • 3 specimens; same collection data as for holotype; ELM W03646/T172 • 1 specimen; same collection data as for holotype; NHMUK 20160078 • 1 specimen; Bedford, Maasstrom Farm , forest in kloof; 32.6830°S, 26.0830°E; 1 Jan. 2007; M. Bursey leg.; ELM D15108/T153 • 1 specimen; same collection data as for preceding; ELM W3122 /T154 • 5 specimens; Bedford, Maasstrom Forest ; J. Farquhar leg.; NMSA V6684 /T5154 • 40 specimens; Somerset East, Glen Avon Falls , forest beside base of falls; 32.6770°S, 25.6368°E; 19 Jan. 2002; D. Herbert leg.; in leaf-litter; NMSA W0754 /T4155 • 72 specimens; Somerset East, Glen Avon , downstream of falls; 32.6770°S, 25.6367° E; 25 Jan. 2011; M. Cole, V. Ndibo and R. Daniels leg.; ELM D16896/T155 • 50 specimens; same collection data as for preceding; ELM W03647/T156 • 3 specimens; same collection data as for preceding; NHMUK 20160076 • 2 specimens; same collection data as for preceding; NHMUK 20160077 • 3 specimens; same collection data as for preceding; RMNH.MOL.338285 • 4 specimens; same collection data as for preceding; NMW.Z.2016.003.00003 .

Description

SHELL ( Fig. 30 A–C). Very small, depressed, discoidal, adult diameter 2.19–3.0 mm, height 1.12–1.68 mm, diameter:height 1.61–2.12 (n = 29). Spire low, each whorl just rising above the next, apex acutely mammillate and tilted ( Fig. 30A ). Embryonic shell ( Fig. 28B ) just over 2.25 whorls, microscopically malleate, junction between embryonic shell and teleoconch evident with termination of malleation. Teleoconch comprising just over two whorls, very rapidly increasing, convex, suture impressed. Aperture circular, last whorl descending steeply nearing aperture, peristome simple, continuous and free. Umbilicus very wide, exposing all the whorls ( Fig. 30C ). Periostracum glossy and lacquer-like with lamellate axial costae at regular intervals ( Fig. 30D ), the average number on last whorl varying between 63 and 95 ( Table 5), which produce three spiral rows of long, simple hairs at periphery; intervals between costae usually with approx. five microscopic axial threads. Shell translucent when fresh.

OPERCULUM ( Fig. 30 E–F). Very fragile and duplex, outer portion consists of multispiral lamella with five–six whorls, height of lamellar blade very low thus operculum is very shallowly concave to almost flat, thickened horizontal ridge on lamellar blade just above disc surface; long fringe of fused bristles emanates from this ridge, fused to blade and then curving outwards, leaving no furrow between fringe and vertical portion of blade, fringe of each whorl does not appear to be fused to lamella of following whorl, fringe of outer whorl fairly long and overlaps disc slightly; inner portion of operculum is a thin disc, without a prominent tubercle in centre.

RADULA ( Fig. 30G ). Rachidian with five cusps, central one very long, outermost two relatively short, first lateral tooth with four cusps and a vestigial fifth, third cusp (from centre) very long; second lateral tooth with two large cusps, second cusp (from centre) larger, a third small cusp and a vestigial fourth.

PENIS ( Fig. 30 H–I). Relatively short, shaft more or less cylindrical.

Distribution and habitat

Eastern Cape, southern end of Great Escarpment, in Eastern Cape Escarpment Thicket ( Hoare et al. 2006) associated with river valleys ( Fig. 29 ).

Remarks

The molecular phylogeny suggests that A. oxygala gen. et sp. nov. belongs to a lineage distinct from all other populations of Afrocyclus gen. nov. sampled ( Fig. 1 ) but further study is required to resolve phylogenetic relationships between A. oxygala gen. et sp. nov. and other species of Afrocyclus gen. nov.

The species is readily distinguished by having only three rows of long hairs on lamellate costae round the periphery (although in adults the hairs were usually worn off), while A. isipingoensis gen. et comb. nov. has four rows around the periphery. A. oxygala gen. et sp. nov. lacks the additional spiral rows of hairs present in A. isipingoensis gen. et comb. nov.: one row of very short hairs below the suture and approx. five-seven rows of shorter hairs between the periphery and umbilicus. A. oxygala gen. et sp. nov. has a higher spire (and consequently deeper umbilicus) and a more strongly sculptured protoconch ( Fig. 28B ). The operculum has a smooth inner disc without a tubercle in its centre. The radula more closely resembles those of species in the Southern-Eastern Cape clade (smooth upper edge of rachidian and two large cusps on second lateral tooth) than those of other specimens of Afrocyclus gen. nov. examined. The penis of A. oxygala gen. et sp. nov. is relatively short.

This lineage occurs at the western extremity of the distribution of gen. nov. and the populations at Somerset East and Bedford are on opposite sides of the upper Great Fish River Valley ( Fig. 29 ). Although considered an arid barrier for dispersal of montane taxa ( Weimarck 1941; Stuckenberg 1962), this valley has not consistently been a major hindrance to connectivity along this region of the Great Escarpment ( Clark et al. 2010) and may have been an area of persistence of thicket during the Last Glacial Maximum (24000 to 18000 years ago) ( Potts et al. 2012) and perhaps prior maxima.

Figures

Fig. 11. Holotypes of species of Afrocyclus gen. nov. A–B. A. oxygala gen. et sp. nov., diameter 2.86 mm (NMSA P0638/T4153). C–D. A. potteri gen. et sp. nov., diameter 2.86 mm (NMSA P1125/T4283). E–F. A. bhaca gen. et sp. nov., diameter 2.87 mm (NMSA P1124/T4282).

Fig. 30. Afrocyclus oxygala gen. et sp. nov. A. Aperture view (ELM D16896). B. Dorsal view (ELM W3647). C. Ventral view (ELM W3647). D. Portion of juvenile shell, showing detail of periostracum (ELM W3647). E. Portion of radula (ELM W3646). F–G. Operculum (ELM W3647). H–I. Penis, showing position, dorsal and lateral views (ELM W3647). Scale bars: A–D, H = 1 mm; E = 20 µm; F–G = 200 µm; I = 500 µm.

Fig. 28. Protoconch detail of species of Afrocyclus gen. nov. A. Afrocyclus isipingoensis (Sturany, 1898) gen. et comb. nov. from Durban (near type locality) (ELM W3648). B. A. oxygala gen. et sp. nov. (ELM W3647). C. A. potteri gen. et sp. nov. (ELM W4052). D. A. bhaca gen. et sp. nov. (NMSA V4889). Scale bars: A–D = 200 µm.

Fig. 29. Distribution map of Afrocyclus gen. nov. T indicates type localities. Provinces and major cities are labelled.

Fig. 1. Bayesian Inference majority consensus tree of the “taxa complete” concatenated CO1-16S sequence dataset for Chondrocyclus s.l. Ancey, 1898 (84 sequences) and seven other caenogastropods with the tree rooted on Pomacea insularum (d’Orbigny, 1837) (modified from Cole et al. 2019). Colours indicate major clades of Chondrocyclus Ancey, 1898 and correspond with colours used in maps. Support values are given as posterior probabilities for Bayesian analysis above nodes (only values ≥ 0.95 are shown) and as bootstrap percentages for ML analysis below nodes (only values ≥ 65% are shown); maximal support (Bayesian = 1, ML = 100%) is indicated by an asterisk. Scale bar indicates substitutions per site.