A revision of the chilodontidae (Gastropoda: Vetigastropoda: Seguenzioidea) of southern Africa and the south-western Indian Ocean.
The taxa discussed in this revision have traditionally been referred to the trochid subfamilies Margaritinae and Monodontinae (see for example Keen 1960). McLean (1981, 1982), however, identified certain lineages within these subfamilies that shared distinctive features of the shell, radula and epipodium. He proposed (McLean 1981) that these represented Recent descendents of Mesozoic lineages (Amberleyidae) previously thought to have become extinct in the Oligocene or earlier. He later suggested (McLean 1982, 1984) that they be recognised as separate tribes within the Margaritinae, but this idea was subsequently revised in the publication of Hickman and McLean (1990) in which a much modified trochoidean subfamily, the Eucyclinae, was employed for the taxa concerned, based on the extinct families Eucyclidae, Cirridae and Amberleyidae. This subfamily was further divided into three tribes, the Eucyclini, Calliotropini and Chilodontini (the latter two including extant taxa) on the grounds of differing shell morphology, radula form and external anatomy. Waren and Bouchet (1993), on account of uncertainties regarding the affinities of the exclusively fossil Eucyclini, subsequently chose to regard the other two tribes as subfamilies in their own right, although this was not followed by Hickman (1998).
A radically different position was subsequently proposed by Bouchet et al. (2005) in which the Calliotropinae and Chilodontinae were removed from the Trochoidea and referred instead to the Seguenzioidea, largely on the basis of molecular data (Waren et al. 2003). More recently, Kano (2008) has analysed the relationships of the Seguenzioidea using sequence data from a wider range of taxa and has confirmed that calliotropine and chilodontine species cluster together with seguenziids in a well supported clade, although no support was evident for the monophyly of either the Calliotropinae or Chilodontinae, nor for the two treated together as the Chilodontidae. He also highlighted potentially significant anatomical similarities including a peduncle posterior to the right eyestalk (distinct from the subocular tentacle) and hooded rachidian teeth in the radula, but pointed out that these may be plesiomorphic character states. Subsequently, Williams et al. (2008) recovered a monophyletic Chilodontidae, with monophyletic subfamilies Chilodontinae and Calliotropinae, but their study did not include other seguenzioid taxa. Most recently Kano et al. (2009), including additional seguenzioid taxa, recovered well supported Chilodontinae and Calliotropinae clades, but found no support for the combined Chilodontinae + Calliotropinae clade. Consequently, they proposed that the two groups be recognised as separate families within the Seguenzioidea. Further support for the seguenzioid affinities of these families was recently provided by Aktipis and Giribet (2012). Kano et al. (2009) also demonstrated that Turcica H. & A. Adams, 1854, traditionally grouped with chilodontid taxa (Hickman & McLean 1990), belongs within the Calliotropidae. For the purposes of this revision, I follow the recommendations of Kano et al. (2009); however, it should be noted that Bandel (2010) has proposed a very different classification for the taxa discussed here, in which he recognised a superfamily Eucycloidea and proposed several new families.
Since chilodontids have only recently been recognised as a distinct entity, there has been no previous collective analysis of the group in this region. In many cases little information over and above that contained in the original descriptions is available. Hickman and McLean (1990) considered the genus Synaptocochlea Pilsbry, 1890 to belong to this group, but although it occurs in the south-western Indian Ocean, it is not treated here. Morphological and molecular evidence now indicates that Synaptocochlea belongs within the Trochidae (Herbert 1998; Williams et al. 2008, 2010), in the subfamily Fossarininae Bandel, 2009. Another western Indian Ocean genus which Bandel (2010) referred to the Eucycloidea is Pagodatrochus Herbert, 1989, creating for it a new family, the Pagodatrochidae. Whilst this taxon shares some features of early teleoconch ribbing and microsculpture with chilodontids and has a radula somewhat atypical of trochids, the proposal of a new family in the absence of corroborating evidence from soft-part anatomy and molecular data seems premature. In terms of its external morphology there are no cephalic lappets or laterally expanded snout flanges, nor is a right post-ocular peduncle evident, and the ctenidium is monopectinate (Herbert unpubl. observ. on the type species Pagodatrochus variabilis (H. Adams, 1873)). None of these character states suggests a closer relationship with the Chilodontidae than with the Trochidae. It is evidently one of many small, anomalous vetigastropod taxa for which molecular sequence data could provide valuable phylogenetic insights.
MATERIAL AND METHODS
The material studied is kept in the following institutions:
AMS--Australian Museum, Sydney, Australia;
MCSNG--Museo Civico di Storia Naturale "Giacomo Doria", Genova, Italy;
MHNB--Museum d'Histoire Naturelle, Bordeaux, France;
MHNG--Museum d'Histoire Naturelle, Geneva, Switzerland;
MMUM--Manchester Museum, University of Manchester, UK;
MNHN--Museum National d'Histoire Naturelle, Paris, France;
NHMUK--Natural History Museum, London, UK;
NMSA--KwaZulu-Natal Museum, Pietermaritzburg, South Africa;
NMW--National Museum of Wales, Cardiff, UK;
OXUM--Oxford University Museum, Oxford, UK;
SAMC--Iziko South African Museum, Cape Town, South Africa;
UCBL--Universite Claude Bernard (Centre des Sciences de la Terre), Lyon, France;
USNM--United States National Museum, Smithsonian Institution, Washington, USA;
WAM--Western Australian Museum, Perth, Australia;
ZMAN--Zoologisch Museum Amsterdam, The Netherlands.
The bulk of the material discussed in this work was obtained during the Natal Museum Dredging Programme (1981-1997) off the east coast of South Africa, on board the RV Meiring Naude and RV Sardinops. This has been augmented by additional material from Madagascar, Mozambique and Reunion obtained during the Benthedi (1977), Mainbaza (2009), Miriky (2009) and Atimo Vatae (2010) campaigns (MNHN) (see acknowledgements). Littoral and shallow water material has also been accumulated over many years and from throughout the region, during the course of on-going field research and through the collecting activities of amateur malacologists.
Shell length and diameter measurements were made with the shell held in apertural view with the axis of coiling vertical. Immature shells were excluded from the data used to calculate length:diameter ratios. The dimensions of the protoconch were measured according to the method set out previously (Herbert 1987). Photographs of shells were taken with a Nikon F4 or Nikon D70 camera. Radulae were extracted by maceration of the buccal mass in dilute NaOH and then thoroughly rinsed in distilled water. For light microscopy radulae were stained in Shirlastain A [SDL Atlas] and for SEM they were dehydrated in ethanol, placed on stubs with double-sided carbon tape and manipulated into position using fine entomological pins whilst air-drying. Shells and radulae for SEM examination were coated with gold-palladium and examined at low accelerating voltage (5-10 kv) in Jeol T/200, Hitachi S-570 SEM, Philips XL30 ESEM and Zeiss EVO 10LS scanning electron microscopes.
The following acronyms and abbreviations are used in this paper:
CSIR--Council for Scientific and Industrial Research, South Africa;
IIOE--International Indian Ocean Expedition;
L(S)T--low (spring) tide;
NMDP--Natal Museum Dredging Programme;
NPB--Natal Parks Board [now Ezemvelo KZN Wildlife];
New taxa have been registered with ZooBank (Appendix I, p. 502).
Protoconch (Fig. 1)
SEM illustrations of chilodontid protoconchs have been given by Marshall (1979), Bandel (1982, 2010), Guidastri et al. (1984), Moolenbeek & Faber (1989), Sasaki (1998) and Engl & Rolan (2009). These together with the illustrations given here reveal considerable variation in protoconch morphology and sculpture. Although the paucispiral form (ca 1.25 whorls) is typical of vetigastropods and the apical beak consistently absent or only weakly developed, features associated with the terminal lip and superficial sculpture vary significantly.
In some species the terminal lip is straight or very slightly curved, but a number have developed a strong beak or sinusigera-like projection (Fig. 1B, C). This is only visible in the most well preserved protoconchs, but the strong angulation in the somewhat worn protoconchs of a number of other taxa is perhaps evidence of the occurrence of a similar structure (e.g. Ascetostomaprovidentiae, Perrinia angulifera and Vaceuchelus cretaceus). I am not aware of any prior reports of such a projection on the protoconch lip in the Vetigastropoda, although some seguenziids have a protoconch with a strongly convex lip (Marshall 1991). A beak or sinusigera-lip is found in the pelagic larvae of a wide variety of caenogastropods and is thought to be indicative of relatively long-lived, planktotrophic larvae (Lima & Lutz 1990; Fretter & Graham 1994). Its presence in lecithotrophic vetigastropods is less easy to explain, but could perhaps indicate a larger velum and a longer, more active swimming phase than is typical for vetigastropods. None of the species examined show any kind of varix (terminal or subterminal) associated with the protoconch lip.
The different patterns evident in the superficial sculpture of the chilodontid protoconch can be grouped as follows:
--almost totally smooth, usually exsert and rather globose [Perrinia konos, Pholidotrope gloriosa] (Fig. 1A).
--finely granular, sometimes with traces of spiral lines, exsert and sometimes globose
[Clypeostoma salpinx, Granata sulcifera and Putzeysia spp. (Engl & Rolan 2009)] (Fig. 1B, C).
--widely spaced spiral threads with close-set axial or oblique lines in the intervals; rather globose [Ascetostomaprovidentiae] (Fig. 1D). Similar sculpture has been illustrated in some seguenziids (Marshall 1983, 1991).
--strong irregular axial ridges with little or no spiral sculpture; not obviously globose, more or less level with first teleoconch whorl or at most weakly exsert [Herpetopoma (s.s.), most species of Vaceuchelus and 'Euchelus7 guttarosea Dall, 1889 (Moolenbeek & Faber 1989)] (Fig. 1E, F).
This degree of variation is considerable when compared with that seen in some other vetigastropod groups, e.g. Calliostomatidae, Solariellidae, Trochinae, Umboniinae (Herbert 1987, 1992, 1993; Marshall 1995), but is no more diverse than that evident in others, e.g. Scissurellidae and Fissurellidae (Sasaki 1998; Geiger & Jansen 2004). Although care needs to be taken in drawing conclusions of a phylogenetic nature based on vetigastropod protoconch microsculptural patterns, due to frequent homoplasy (Sasaki 1998), it seems that in some instances the sculptural patterns evident in the taxa studied here are broadly consistent within some genera (e.g. Herpetopoma (s.s.) and Vaceuchelus), in others less so (e.g. Perrinia). For the majority of genera, however, data on protoconch microsculpture are available for too few species to assess its intrageneric consistency.
Similarly, there is also variation in the size of the protoconch and its elevation above the first teleoconch whorl. In most species protoconch diameter lies between 200-300 [micro]m; however, in Clypeostoma salpinx, Danilia textilis and Vaceuchelus gemmula, protoconch diameter may reach 340-380 um. This appears to be more of a species character than a generic one, as other species in these genera have protoconchs within the 200-300 um range. Mostly the protoconch projects a little above the first teleoconch whorl and is slightly down-tilted, but in some species it is strongly exsert (Clypeostoma salpinx, Pholidotrope gloriosa, Perrinia konos) or slightly sunken below the level of the first teleoconch whorl rendering the apex truncated (Danilia spp., Perrinia angulifera, Vaceuchelus cretaceus).
Microsculpture (Fig. 2)
Teleoconch microsculpture within the Chilodontidae appears to show distinct juvenile and adult facies, both of which reveal commonly occurring features that warrant further study as characters of potential phylogenetic significance. The spiral cords of most chilodontid taxa develop only during the second and third teleoconch whorls, and the first whorl is for the most part sculptured only by axial pliculae. Between these there is often an unusual vermiform microsculpture of fine, broken, wavy and frequently bifurcating, spiral threads which continues onto the second and sometimes subsequent whorls (Fig. 2A). Although this is cannot be said to be a ubiquitous feature in the Chilodontidae, it is nonetheless common within local representatives of the group, occurring in species belonging to Ascetostoma, Clypeostoma, Danilia and Herpetopoma and has been illustrated in other chilodontids by Marshall (1979: fig. 2h), Moolenbeek & Faber (1989: figs 12, 13) and Engl & Rolan (2009: figs 6, 11). Hickman (1998) indicted that this microsculpture is present from the first appearance of the group in the fossil record (Early Mesozoic).
The microsculpture of later teleoconch whorls is generally of a completely different nature, comprising densely crowded, collabral, scratch-like marks (Fig. 2B). In living or fresh shells these are filled with a chalky, white, dirty-brown or ash-grey intritacalx, deposited as elongate grains which fill the scratch-like marks. Such a sculpture is evident also in many members of the Calliotropidae (Marshall 1979: fig. 3g, k, o, and pers. observ.).
Although the two types of microsculpture described above are by far the most common, they are not the only ones evident in family. Marshall (1979) and Moolenbeek & Faber (1989) have illustrated a microsculpture of crisp granules in other chilodontid species. Similar sculpture is evident on the early teleoconch whorls in Perrinia spp., Herpetopoma helix, Pholidotrope gloriosa and Vaceuchelus jayorum, in place of the more common vermiform spiral threads (Fig. 2C).
Microsculptural details are not always evident and may be effected by a number of parameters. Firstly, wear, which removes fine sculptural details, although in live-collected shells this usually only effects the apical whorls (and thus the juvenile microsculpture), and some part of the adult shell generally remains unworn. Secondly, encrustations can also obscure superficial details and this is particularly so in chilodontids, which are frequently encrusted with sponges or coralline algal growths, in the case of sponges sometimes almost completely so (e.g. Ascetostoma providentiae and Clypeostoma salpinx). Thirdly, several taxa have a particularly well developed intritacalx, which instead of being grain-like, occurs as a thick chalky layer obscuring all microsculptural details (e.g. Herpetopoma helix and Vaceuchelus cretaceus). In such cases microsculpture is evident only after ultrasonic cleaning, although the extent to which such treatment might also effect the microsculpture itself is not clear.
Radula (Fig. 3)
The chilodontid radula is not simple to interpret. Unlike that of many other vetigastropod groups such as the Fissurellidae, Solariellidae and Trochidae, it is difficult to prepare for SEM. The marginal teeth are long and fine, and naturally tend to fold inwards, overlapping each other and the central radula field, and thus obscuring the rachidian and lateral teeth. Careful micro-manipulation of the radula is needed during the final drying stages in order to splay the marginal teeth outward. A further complication is that the differentiation between the lateral and marginal series is not always clear, the transition from one to the other being somewhat gradual.
Illustrations of chilodontid radulae were given by Troschel (1879, Euchelus), Hutton (1883, Herpetopoma), Odhner (1917, Hybochelus), Thiele (1924, Euchelus, Danilia and Granata), Cotton (1946, 1959, Granata and Herpetopoma), McLean (1970, Mirachelus), Olsson (1971, Turcica), Beu & Climo (1974, Danilia), Quinn (1979, Mirachelus) and Guidastri et al. (1984, Danilia and Putzeysia). More recently, radula morphology within the group as a whole was discussed and summarised by Hickman and McLean (1990) who used radula characters exclusively in their diagnosis of the subfamily Eucyclinae. Amongst the most significant features noted were the so-called 'hooded' rachidian, complexly interlocking lateral tooth bases, rectangular lateromarginal plate and broad, mitten-shaped outermost marginal tooth. The expanded lateral flanges on the rachidian shaft (the 'hood') and the interlocking of lateral tooth bases are clearly evident in the taxa examined here (Fig. 3A, D). A broad outermost marginal tooth is also present in many of the local species (Fig. 3B), but it seems not to be as large as that illustrated for Turcica (Olsson 1971; Hickman & McLean 1990) and Bathybembix (Hickman 1981). These teeth are generally difficult to see, often being obscured by the reflected underlying lingual membrane. In contrast, I have been unable to identify a rectangular latero-marginal plate in any of the species examined in this revision, and such has not been mentioned by the earlier authors cited above. Odhner (1917) considered the fourth lateral tooth of Hybochelus mysticus to be vestigial and transitional, but in his drawing this tooth is far from vestigial and retains a serrated cusp. In reality, Hickman and McLean (1990) illustrated latero-marginal plates in a species of Calliotropis and in view of our enhanced understanding of the classification of their 'Eucyclinae' it may be that latero-marginal plates are restricted to the Calliotropidae. Similarly, since Turcica is also now thought to belong to the Calliotropidae (Kano et al. 2009), the extremely large, mitten-shaped outermost marginal tooth may also be a character more distinctive of the Calliotropidae than the Chilodontidae, in which the distal portion of the outermost marginal, though broadened, is not as massively enlarged. A similar expansion of this tooth is evident in some Scissurelloidea (disaster & Geiger 2011; Luque et al. 2001), suggesting that it may be a more widespread, but seldom recorded feature in the Vetigastropoda.
Where possible I have examined the radula of at least one species in each genus to provide information on intergeneric variation in radula form. The rachidian cusp is generally well developed and the lateral flanges on its upper shaft expanded to create a hood of variable extent. In most genera there is also a conspicuous raised transverse ridge at the base of the rachidian cusp, giving the tooth a distinctly hunched appearance (Fig. 3A, D) (e.g. Ascetostoma, Clypeostoma, Danilia, Herpetopoma, Perrinia and Vaceuchelus). Guidastri et al. (1984) reported the same structure in Putzeysia Sulliotti, 1889, and it is also evident in some calliotropids (Hickman & McLean 1990). This articulates with the similarly shaped posterior border of the base-plate of the tooth in front. Commonly, the rachidian cusp is relatively coarsely dentate with a strong, lanceolate or acuminate central denticle and 3-5 smaller, also lanceolate, lateral denticles on each side (Fig. 3A, D, E) (e.g. Ascetostoma, Clypeostoma, Danilia, Herpetopoma, Perrinia, Vaceuchelus herein, and Mirachelus (McLean 1970; Quinn 1979)), but in Granata (Fig. 3C) the rachidian cusp is narrowly triangular and its lateral margins finely serrate toward the base (also illustrated by Thiele 1924 and Hickman & McLean 1990). A coarsely dentate rachidian cusp morphology is also found in some seguenziids (Marshall 1991, subfamily Asthelysinae and tribe Fluxinellini), suggesting that this commonly occurring character state in the Chilodontidae may be the ancestral condition.
Within row interaction of the tooth bases is complex in the central field, with extensive tooth overlap (Fig. 3A, C, D). The cusp of the innermost lateral is usually similar in size or slightly larger than that of the rachidian, with those of the remaining laterals being similar in size or progressively decreasing slightly in size (rarely the reverse). The shaft of the lateral teeth comprises a thickened median pillar, usually straight, but set at an angle to the long axis of the radula, with an alate flange on both margins, that on the inner margin slotting behind the median pillar of its inner neighbour, that on the outer margin lying behind the inner flange of its outer neighbour (Fig. 3A). The cusp of these teeth is generally obliquely trigonal and rather coarsely dentate, with the central denticle often spathulate, but in Granata the cusps are less oblique and, like the rachidian, finely serrate (Fig. 3C). In Danilia (Fig. 3D) the lateral tooth cusps are strongly asymmetrical, finely serrate on the inner margin and more coarsely dentate on the outer margin (also illustrated by Thiele 1924 and Beu & Climo 1974). The transition from the lateral series to the marginal series is moderately clear in some genera (e.g. Herpetopoma and Vaceuchelus, though not in all species), but in others (e.g. Clypeostoma and Granata) there is no clear-cut boundary and determining the number of lateral teeth is therefore difficult. Whether the fourth tooth is considered a lateral or a marginal is open to question in these genera. Many authors have cited the number of pairs of lateral teeth in chilodontid taxa as four (e.g. Odhner 1917; Thiele 1924, 1929; Cotton 1946; Beu & Climo 1974; Quinn 1979; Guidastri et al. 1984), but others have cited three (e.g. McLean 1970; Bandel 2010) and some Vaceuchelus species evidently have only two (see below). However, I am not inclined to put much emphasis on this character for several reasons. Firstly, because it is often somewhat subjective to determine; secondly, because it is likely to be influenced by the size of the species (hence some small Vaceuchelus species with only two pairs of laterals); and thirdly, in one species where the transition is clear (Herpetopoma scabriusculum), I observed one individual with three pairs of laterals and another with four. The character is thus also individually variable.
The marginal teeth are numerous, but the precise number is difficult to establish. In Herpetopoma and Vaceuchelus, though long and slender, they remain relatively robust (Fig. 3E), but in some other genera the marginals are even more elongate and fine, particularly so in Danilia in which all the marginal teeth are extremely slender and delicate (Fig. 3F), appearing as a mop-like mass along the radula margin (described as a 'hairy rope' by Beu & Climo 1974). The inner marginals are coarsely dentate or pectinate along their outer margin in most genera, particularly Herpetopoma and Vaceuchelus, and the denticles may extend some way down the shaft. However, in Clypeostoma and more so Danilia, the dentition on these teeth is finer. As mentioned above, the marginal teeth at the end of the series, particularly the outermost one, have a broader, more spathulate cusp with a finely dentate, almost feathered margin (Fig. 3B). Another unusual feature of the Danilia radula is the buttressed outer shaft base of the inner marginals and its raggedly dentate edge (Fig. 3F). There is also some evidence of this in Perrinia, and it may be a more widely occurring, but seldom visible feature.
In calliotropid taxa, judging by the illustrations given by Hickman and McLean (1990), the rachidian is narrower and has a smaller cusp relative to the laterals, but it retains a well-developed hood and has a particularly strong transverse ridge at the cusp base. Only three pairs of lateral teeth are present and they increase markedly in size from first to third (something not seen on chilodontids). The marginal series is clearly distinct from the lateral one and there is an intervening latero-marginal plate. By comparison with chilodontids, the marginals are considerably less numerous. However, Hickman and McLean (1990) only illustrated the radula of Calliotropis species and how reflective these are of other calliotropid taxa is unclear. If, as molecular data suggests, genera such as Turcica are calliotropid rather than chilodontid (Kano et al. 2009), then radula form in calliotropids is certainly more diverse than is evident from Calliotropis species alone.
Operculum (Fig. 4)
The chilodontid operculum is thin, corneous and generally multispiral. In Euchelus, however, Pilsbry (1890 in 1889-1890) noted that the operculum is of few whorls and he likened it to that of Littorina, using this as a character separating Euchelus from Herpetopoma, which he stated possessed a conventionally multispiral operculum. Beu and Climo (1974) discussed the matter further in relation to the genus Danilia and confirmed that the Euchelus operculum is indeed paucispiral like that of Littorina. This, however, is an oversimplification, for whilst the operculum of Euchelus has a long growing margin and thus relatively few whorls, it is not as paucispiral as that of littorinids and has a noticeably less eccentric nucleus (compare Figs 4A and 4B). In reality, the Euchelus operculum represents a well-developed example of a trend, common in chilodontids, towards the reduction in the number of opercular whorls, particularly in comparison to trochoidean taxa (Hickman & McLean 1990). This trend is also evident in the Seguenziidae (Marshall 1983; Quinn 1983).
The coiling of the operculum is usually tight initially, but the outermost whorls expand somewhat more rapidly and the growing margin is correspondingly broader. Ultimately, as in Euchelus, the angle of accretion becomes almost tangential to the preceding opercular whorl and the growing edge is long. Operculum form in the taxa examined herein ranges from relatively tightly multispiral (e.g. Herpetopoma helix, Perrinia stellata and Vaceuchelus semilugubris, Fig. 4I, L, O), to more openly multispiral (e.g. Ascetostoma, Clypeostoma and Danilia species, as well as Herpetopoma s.s., Fig. 4D-H), to oligospiral (Euchelus-like) in Granata species (Fig. 4C), but the gradations are relative rather than categorical. Of necessity, as the rate of whorl expansion increases, the nucleus of the operculum becomes increasingly eccentric.
Hickman and McLean (1990) observed that opercular development in chilodontids does not always keep pace with growth of the shell (aperture), noting that the operculum in Granata it is vestigial. My own observations on Granata sulcifera indicate that its operculum, though it does not completely close the shell aperture, is not truly vestigial, its diameter equalling approximately half that of the aperture. In G. imbricata it is considerably smaller, relative to the aperture, but still not vestigial when compared to the miniature operculum of the similarly shaped stomatelline trochid genus Pseudostomatella (Herbert 1998).
External anatomy (Figs 5, 6)
Hickman & McLean (1990) and Hickman (1998) have summarised the basic features of the external anatomy of chilodontid taxa, building upon earlier descriptions provided by Beu & Climo (1974) and Guidastri et al. (1984). Judging from the species examined in this study, although the basic ground plan is similar, there is considerable variation in detail. However, since these details are largely available for only one species in each genus, it is not possible to establish whether the features reported here are broadly characteristic for the respective genera as a whole.
Cephalic lappets are present, the free margin ranging from finely digitate to microscopically fimbriate. The cephalic tentacles are well developed and micropapillate. The eyes are prominent, on short stalks not fused to the cephalic tentacles; the eye itself is subterminal and black. Immediately behind the right eyestalk, sometimes appearing to arise from its base, is a post-ocular peduncle. This may be nearly as long or longer than the eyestalk and is not micropapillate, but it is often slightly expanded and flattened distally, and a dorsal longitudinal groove is frequently evident. It is present in both sexes. An additional very small process, the subocular tentacle is occasionally present projecting from beneath the base of the right eyestalk (e.g. in Granata, see below and Kano 2008). The snout is often pigmented, has extensive ventro-lateral flanges and the ventral lip is deeply split in the mid-line.
The neck lobes are not fused to the eyestalks, arising instead adjacent to and overlapping the snout flanges, below the cephalic tentacles. Both neck lobes are well developed, the free margin microscopically fimbriate and bearing a series of micropapillate tentacles often of two or three size ranks. There is considerable variation in the number and arrangement of the neck lobe tentacles between genera. Commonly these arise from just under the lobe margin. Similar tentacles occur posteriorly along the length of the epipodial fold, the number depending on the size of the species and the individual. Again there are two or three size ranks. In Ascetostoma, Clypeostoma, Granata and Perrinia there are minute tentacles between the larger tentacles, but these are absent in Danilia and Vaceuchelus. Free-standing, mushroom-like epipodial sense organs such as occur in many trochiodean genera are not evident, even beneath neck lobes, but most of the larger epipodial tentacles have a basal swelling on the ventral side which probably represents an epipodial sense organ. Similar swellings are sometimes evident at the base of the larger neck lobe tentacles and Danilia possesses several distinct sense organs on the underside of its neck lobes.
The ctenidium is bipectinate with a short afferent membrane and thus the free portion is long. The right hypobranchial gland is well developed and trigonal in shape, surrounded on two sides by the rectum which loops around its left margin. In Granata the gland is particularly large, ovate to kidney-shaped, with the rectum curving around the left side. It is responsible for producing the foul-smelling, milky secretion that is emitted when animals of this genus are disturbed (Hickman 1998, and pers. observ.). The rectum is generally turgid, filled with a longitudinally ridged faecal string composed of amorphous microscopically particulate faecal material. It extends further forward than is the case in trochoidean taxa, almost reaching the anterior right extremity of the mantle edge. Distally it is free of the mantle cavity wall and tapers to form a nozzle, the tip of which frequently possesses a minute tentacle-like appendage.
The presence of a post-ocular peduncle in chilodontids is a significant feature and one reflective of their seguenzioidean affinity (Kano 2008). This is distinct from, and should not be confused with, the subocular tentacle, a structure found also in many Trochoidea. The function of the post-ocular peduncle is unclear, though a penis-like role as a conduit for sperm transfer during pseudo-copulation has been suggested for a similar structure found in calliotropids (Dall 1889). If such is the case, then its presence in both sexes is puzzling and requires an explanation--something that has not yet been provided. Similar structures are evident in other vetigastropods, some present only in males, but care needs to be taken before inferences relating to the function of these structure and the phylogenetic implications of their presence are drawn, since their structure has not been investigated in detail and their homology has not been demonstrated.
BIOLOGY, ECOLOGY AND ZOOGEOGRAPHY
Chilodontid gastropods inhabit a considerable depth range, from the intertidal zone (e.g. Euchelus, Granata, Herpetopoma and Vaceuchelus) to depths of 2000 m (Danilia) or more (Putzeysia). However, despite the fact that some species are intertidal and may even be common, the biology of the family as a whole is poorly known. The group is thought to have arisen in the Early Mesozoic of Europe, and to have subsequently radiated in the Cretaceous, both in terms of diversity and geographic range (Hickman & McLean 1990), in association with the contemporaneous expansion of the carbonate based habitats on which they lived. Today the greatest diversity, at both genus and species level, remains associated with warm water and hard, carbonate-based substrata, as evidenced by the rich chilodontid fauna of the Philippines (Poppe et al. 2006; Poppe & Tagaro 2008) and the central Indo-West Pacific in general. Mostly this fauna occurs in near-shore and continental shelf ecosystems (<200 m). The greatest depth at which living specimens have been found in the south-western Indian Ocean is 277 m (Clypeostoma salpinx; Table 1). Such is in marked contrast to the Calliotropidae, which are largely bathyal (>200 m), and are associated with unconsolidated substrata and colder water (including the Antarctic).
Consistent with the above, the Chilodontidae is clearly a tropical element within the fauna of the south-western Indian Ocean. Only one species, Vaceuchelus gemmula, ranges beyond the subtropical waters of the east coast of South Africa in to the warm-temperate waters of the southern Cape. This is also the only chilodontid endemic to South Africa. A considerable portion of the regional chilodontid fauna (14 out of 27 recorded species, 51.9 %); however, is endemic to the south-western Indian Ocean (Table 1), indicating a significant focus of endemism in this part of the Indo-West Pacific. This may be due to the prevailing south-westerly current regime, which would tend to restrict nascent taxa to this region. A further five species (18.5 %) are endemic to the western Indian Ocean and eight (29.6 %<>) are more widespread Indo-West Pacific taxa ranging east to south-east Asia. Interestingly, for some of the regionally endemic species there is a closely similar species occurring in the central Indo-West Pacific, suggestive of a sister taxon relationship spanning the Indian Ocean (e.g. Ascetostoma providentiae and A. ringens, and Vaceuchelus cretaceus and V. pagoboorum).
Almost without exception, the chilodontids of the south-western Indian Ocean are associated with hard substrata (Table 1). Commonly these are carbonate-based, but some species occur in habitats lacking and obvious carbonate facies, such as the rocky shore of KwaZulu-Natal and Eastern Cape (Vaceuchelus natalensis and V. gemmula), and the sponge-dominated communities on the edge of the continental shelf in the same region (Clypeostoma salpinx and Danilia textilis). The latter communities, however, often contain a significant octocoral and scleractinian component. Only one species, Granata sulcifera, occurs in habitats dominated by soft substrata, and can be found in sheltered bays with muddy sand, but even here it lives attached to hard objects. This species evinces a somewhat limpet-like mode of life that is taken further by the Australian G. imbricata, which clings tenaciously to the rocks on which it lives and is unable to right itself once dislodged (Hickman 1998). Both species are negatively phototropic and secrete a foul-smelling, milky mucus when disturbed (see external anatomy above), as evidently does G. lyrata (Kano 2008). Unlike G. imbricata, however, G. sulcifera has no difficulty righting itself if upturned, using the highly mobile and extendable anterior portion of the foot.
Almost nothing is known of their diet, though one may speculate, given the robust interlocking of the teeth in the central field of the radula, and the generally strong dentition of the rachidian and lateral tooth cusps, that they graze superficially on the hard substrata on which they occur and on the organisms growing thereon, the long marginal teeth being used to gather up dislodged material. The unusual marginal teeth of Danilia species, however, are suggestive of a more specialised diet, perhaps associated with the deep-water corals with which they commonly co-occur.
As is to be expected in vetigastropods the sexes are separate in chilodontids, but little is known of their reproductive biology. Duch (1969) recorded that in the littoral Euchelus gemmatus (Gould, 1845) (probably a species of Herpetopoma) spawning is associated with the spring equinox and is initiated by pair formation, subsequent to which gelatinous egg masses are deposited on hard surfaces, inside which the embryo grows into a fully developed veliger that hatches as a 'semi-crawling larva' after 1-2 weeks. One may assume, therefore, that, in this species at least, a free-swimming stage, if such exists at all, is at most brief. Such may not be the case in other species, particularly those that possess a prominent beak on the protoconch lip (e.g. Clypeostoma salpinx and Granata sulcifera), suggesting, as mentioned above, a longer, more active swimming phase.
In life, the shell of many species is covered, often entirely so, with a living encrusting sponge, particularly in Euchelus and Herpetopoma (Odhner 1917; Gardner 1975; Walsby & Morton 1982; Hickman & McLean 1990; Wilson 1993; Poppe et al. 2006; Poppe & Tagaro 2008). Others commonly have superficial growths of white or pink encrusting coralline algae.
Family Chilodontidae (1) Wenz, 1938
The use of generic names now considered to belong to the Chilodontidae has been complicated by a lack of precision. Doubtlessly, in many cases this is due to the fact that the type species are poorly known and thus the diagnostic characters of the genera are not clearly identified. Genera such as Turcica (sensu stricto, a calliotropid genus) and Euchelus have been employed as hold-all taxa that in reality represent ill-defined assemblages of dubiously related species. Some clarity is beginning to emerge as more material is studied, but there remain problematic areas where shell characters, when considered across a range of taxa, seem to intergrade between genera, as for example between Euchelus and Herpetopoma. Even within Herpetopoma there is a puzzling diversity of shell form, which suggests that it may be a composite taxon (see below). I propose two new genera for taxa which exhibit distinctive combinations of characters not present in the existing genera. A third new genus is proposed for extant species currently referred to Agathodonta, a fossil genus that I consider is not appropriate for these Recent forms.
The genera are not treated alphabetical order, but are grouped loosely in terms of similarity in shell form. However, this should not be taken to indicate any suggestion of phylogenetic relationship. Such inferences must await analysis of molecular data and a more complete dataset of morphological characters. Species treatments do not contain diagnoses as these tend to be repetitive of information given in the description. Instead, I provide keys to species which summarise the characters important for species discrimination and represent a more useful tool for identification. Further discussion of distinctive features and comparison with similar extralimital species is provided under 'Remarks'. In addition to the Recent species discussed, I draw attention to a fossil taxon Chilodonta (Agathodonta) africana described by Rennie (1930) from Upper Cretaceous deposits on the Pondoland coast, Eastern Cape. The holotype (SAMC 8630) somewhat resembles Ascetosotomaprovidentiae (see below) and it seems correctly referred to the Chilodontidae.
Leptothyra alfredensis Bartsch, 1915, and Cyclostremella alfredensis Bartsch, 1915, both from Port Alfred, Eastern Cape, superficially resemble Vaceuchelus, but show much greater similarity to the skeneid genus Parviturbo Pilsbry & McGinty, 1945 (Hickman & McLean 1990; Waren 1991). Unfortunately, this cannot be confirmed by examination of the radula, as neither species has ever been found alive. Both names may quite possibly apply to the same species. Likewise Vitrinella agulhasensis Thiele, 1925, from the Agulhas Bank, resembles Vaceuchelus, but is also probably a species of Parviturbo.
Stomatella cancellata Krauss, 1848, the type species of Hybochelus Pilsbry, 1890, was erroneously described from Table Bay, Cape Town. The species is not part of the southern African fauna and occurs instead in the central Indo-West Pacific (Wilson 1993; Herbert 1996; Herbert & Waren 1999; Poppe et al. 2006; Poppe & Tagaro 2008).
Material referable to Tibatrochus Nomura, 1940 has been dredged off southern Mozambique and north-eastern South Africa, but is not included in this review as I strongly suspect the genus belongs in the Calliotropidae.
Key to genera of Chilodontidae in the south-western Indian Ocean (applies only to specimens with mature apertural dentition)
1 Columella lacking any form of dentition or with at most a low rounded bulge at its base 2
--Columella with one or more denticles (mature specimens)
2 Shell small (adult diameter <10 mm) and predominantly white, turbiniform; sculpture cancellate or foveolate Vaceuchelus
--Shell larger, depressed turbiniform to ear-shaped, last whorl expanding rapidly and aperture consequently large; interior highly nacreous Granata
3 Aperture lacking a U-shaped notch at junction of basal and columella lips 4
--Aperture with a distinct U-shaped notch at junction of basal and columella lips 6
4 Periphery angled and marked by a well-developed, often keel-like, spiral cord, sometimes stellate; suture strongly channelled; spiral sculpture inside outer lip extending deep within aperture Perrinia
--Periphery more or less rounded 5
5 Adult shell generally exceeding 10 mm in length; operculum oligospiral Euchelus
--Adult shell less than 10 mm in length; operculum Herpetopoma (part)
6 Columella with a single tooth above U-shaped basal notch 7
--Columella with 2 teeth above U-shaped basal notch 8
7 Shell trochiform and with coarsely cancellate sculpture; periphery almost stellate; outer lip with subterminal external varix; parietal callus expanded beyond aperture Pholidotrope
--Shell turbiniform-conical and with finely cancellate or beaded sculpture; periphery more or less evenly rounded; outer lip lacking external varix; parietal callus not expanded Herpetopoma (part)
8 Exterior of outer lip with a strong, rib-like, subterminal varix; apex truncate and protoconch sunken Danilia
--Exterior of outer lip lacking such a varix, at most with a broad, low thickening behind outer lip; apex not truncate and protoconch exsert 9
9 Umbilicus present, lined and apically plugged with callus Ascetostoma
--Umbilicus absent Clypeostoma
Genus Clypeostoma gen. n.
Etymology: From Latin clypeus (a shield) and Greek stoma (a mouth); in reference to the shield-like inductural callus. Gender neuter.
Type species: Turcica salpinx Barnard, 1964.
Diagnosis: Shell moderately elevated; spire whorls flat-sided to weakly convex; suture indented, level with subperipheral cord; sculpture of spiral cords and axial pliculae, cords ornamented with squamose beads where crossed by pliculae; aperture subcircular to obliquely D-shaped; outer lip conspicuously flaring beyond an internal collabral thickening; columella stout and bearing 2 rounded teeth when mature; thickened interior of outer lip with ridge-like denticles, that nearest columella largest and separated from basal columella tooth by deep U-shaped notch; umbilical and parietal region covered by glossy inductural callus shield; shield raised and flaring basally where it joins basal lip; umbilicus absent; circumference of aperture and inductural shield more or less in one plane, obliquely tangential to base of last adult whorl; outer lip not thickened externally; protoconch exsert and with a distinct sinusigera-like projection on terminal lip.
Remarks: Extant species belonging to this genus have previously been referred to Agathodonta Cossmann, 1918 (McLean 1984; Neubert 1998; Vilvens 2001; Vilvens & Heros 2003; Poppe et al. 2006). The type species of Agathodonta, by original designation, is Trochus dentiger d'Orbigny, 1843 (an unjustified emendation of Trochus dentigerus d'Orbigny, 1843) from the Lower Cretaceous (Neocomian) of north-eastern France, which has recently been shown to be a junior synonym of Littorina elegans Deshayes in Leymerie, 1842 (Kollmann 2005). Whilst extant species undoubtedly exhibit some similarity with Agathodonta, examination of the lectotype (designated by Kollmann 2005: 70, pl. 9, fig. 6) and topotypic material of Trochus dentigerus (UCBL, Fig. 7A, B and MNHN, Fig. 7C, respectively) indicates that its shell has more strongly convex whorls, does not develop a flaring aperture or expanded inductural callus shield, has a broad, thickened external varix behind the outer lip, with additional varices on the spire whorls (arrowed in Fig. 7A, C), shows little evidence of any denticles or ridges inside the outer lip, and the columella teeth, instead of being discrete teeth (which develop only at maturity), appear to be the ends of ridges spiralling up the columella. In addition, the sculpture is considerably less coarse, the spiral cords being finely and regularly beaded, and without axial pliculae in their intervals. Since these characters are at variance with those of Recent species, I have chosen not to refer extant material to Agathodonta. The original figure of Trochus dentigerus provided by d'Orbigny (1843) is misleading in not showing the varices (Fig. 7D), but Cossmann's figures clearly do so, though strangely he does not mention them in his description of the genus (Cossmann 1918: pl. vii, fig. 9). Such varices also occur in Calliovarica Vokes, 1939, which is also probably referable to the Chilodontidae (Beu & Raine 2009).
Since no other genus-rank name is available for these extant species, I propose the new genus Clypeostoma, with Turcica salpinx Barnard, 1964, as type species. Other described taxa referable here are Agathodonta elongata Vilvens, 2001, A. meteorae Neubert, 1998, A. nortoni McLean, 1984, Euchelus townsendianus Melvill & Standen, 1903 and perhaps Perrinia cecileae, P. docili and Herpetopoma barbieri all of Poppe, Tagaro & Dekker (2006). Whereas Clypeostoma elongatum, C. reticulatum and C. salpinx are distinctive, the others are confusingly similar and exhibit intergrading variability, suggesting that some at least may prove to be synonyms (see below). When a good series of samples is available for one taxon, e.g. C. salpinx, it is clear that allowance must be made for some individual variation in size at maturity, L/D ratio, spiral cord number and strength of apertural dentition. In the absence of such information for the other taxa, it is not possible to make informed comment on potential synonymies within the genus. In some cases the differences evident between the species are small and perhaps of limited significance. C. meteorae and C. townsendianum were both described from the mid to outer continental shelf of the north-western Indian Ocean and may well be synonyms, even though the holotype of former is smaller than the figured syntype of Euchelus townsendianus (length 7.5 vs 10.4 mm), which I here refigure and designate as lectotype (Fig. 12F, G).
Clypeostoma differs from Danilia in lacking an external varix behind the outer lip and in not having a sunken region median to the thickened edge of the columella pillar. In Danilia the shell apex is also truncate and the protoconch flat or a little sunken. In Ascetostoma gen. n. the apertural dentition is more complex and the umbilicus is retained and lined with callus. Some species referred to Herpetopoma Pilsbry, 1890, also have a strong U-shaped notch at the base of the columella, but such species have only one distinct columella tooth, lack an expanded inductural callus and often retain an open umbilicus.
Key to species of Clypeostoma in the south-western Indian Ocean
1 Columella teeth well developed, of more or less equal size (though not necessarily the same shape) 2
--Columella teeth relatively weak, the lower one usually stronger; basal callus shield extensive, smooth and glossy; South Africa and southern Mozambique salpinx
2 Adult shell relatively large (length > 16 mm); sculpture coarsely reticulate; currently known only from Almirante Leite Knolls reticulatum
--Adult shell smaller (length <15 mm); sculpture relatively fine 3
3 Adult shell <8.0 mm in length; protoconch diameter 220-240 [micro]m; basal callus shield not extensive; dentition of aperture and callus shield well developed meteorae
--Adult shell >9.0 mm in length; protoconch diameter 310-320 [micro]m; basal callus shield extensive, but with weak superficial ridging (except inside aperture) cf. nortoni
Clypeostoma reticulatum sp. n.
Figs 8, 9
Etymology: From Latin reticulatus (netted); in reference to the cancellate, net-like sculpture.
Shell: Large for the genus (length up to 16.7 mm), trochoid-turbiniform, with conical spire and relatively deep, globose last adult whorl (L/D=1.38-1.44, last adult whorl 0.70-0.74 of shell length); teleoconch of up to 6 whorls; first two whorls rounded, whorls 3-4 more or less flat-sided, subsequent whorls convex; apical angle 69-74[degrees]; suture of spire whorls level with and mostly covering subperipheral cord of preceding whorl, demarcated by a well-developed channel below peripheral cord; shell periphery rounded. First teleoconch whorl worn in all available specimens, but traces of curved axial pliculae remaining; 3 spiral cords arise during second whorl, with a fourth appearing below adapical suture toward end of whorl; abapical (peripheral) cord strongest; cords crossed by axial pliculae and beaded at intersections; a fifth cord arising below subsutural cord during third whorl (sometimes a sixth during fourth whorl), penultimate whorl thus with 5-6 spiral cords, an additional cord becoming evident level with abapical suture; sculpture remaining similar but strengthening on fourth and subsequent whorls; beads on subsutural cord developing in to angular nodules (25-30 on last adult whorl); intervals between cords slightly wider than cords themselves (that below peripheral cord widest) and crossed by axial pliculae creating quadrate interstices; alignment of pliculae from interval to interval somewhat variable and irregular due to intercalation of additional pliculae, but general pattern of sculpture clearly reticulate, particularly on spire whorls; beads stronger and axially elongated on latter part of last adult whorl; sutural cord of spire whorls may emerge above suture toward end of last adult whorl. Base rounded with 4-6 spiral cords, the last small and partly concealed by reflected lip of aperture; basal cords somewhat more finely beaded than those above periphery. Peristome markedly oblique, more or less in one tangential plane; aperture subcircular to roundly quadrate; columella lip robust, with 2 relatively massive teeth, the upper one slightly more robust, the lower one somewhat narrower and with a small denticle on its lower margin; parietal and umbilical region covered with well-developed, glossy callus which flares outward over base; callus porcelaneous and mostly smooth, but with some beads of unequal size in region adjacent to columella teeth; one particularly strong bead at base of lower columella tooth; edge of callus raised and flaring where it joins flared margin of outer lip. Outer lip lacking an external varix, but somewhat thickened internally and strongly flaring, with 9 labral denticles; denticle nearest base of columella largest and peg-like, separated from basal columella tooth by a U-shaped notch; remaining labral denticles representing the ends of in-running spiral ridges extending deep into aperture, where there may be additional intermediary ridges; similar ridges present inside parietal portion of aperture; denticle at end of uppermost labral spiral ridge comprises a somewhat disjunct swelling, itself with 2 or 3 smaller granules; ridge below this ending in a single large elongate denticle.
Microsculpture: No fresh material available; all traces of intritacalx lost, but some evidence of scratch-like microsculpture present on later whorls.
Protoconch: Missing in all available specimens.
Colour: Shell mostly uniform cream-white to dirty buff, with some pale orange-brown spotting, particularly on subsutural and basal cords.
Dimensions: Holotype, length 16.4 mm, diameter, 11.4 mm; largest specimen, length 16.7 mm.
Operculum, radula and external anatomy: Unknown.
Holotype (Fig. 9A, B): MOZAMBIQUE: Mozambique Channel, Almirante Leite Knolls (26.200[degrees]S 35.033[degrees]E), -228-230 m, Campagne Mainbaza, st'n DW3167, dredged RV Vizconde de Eza, 16.iv.2009 (MNHN 24646).
Paratypes: Same data as holotype (MNHN 24647, 4 specimens, of which 3 adult, 1 juvenile).
Distribution and habitat (Fig. 8): Known only from Almirante Leite Knolls, approx. 250 km due east of Maputo, Mozambique, -228-230 m; rocky substrata with corals, sponges and gorgonians (Fraussen & Rosado 2011).
Remarks: This species is distinctive amongst south-western Indian Ocean chilodontids on account of its relatively large size, coarse sculpture and well-developed apertural dentition. The most similar species is Clypeostoma elongata (Vilvens, 2001) from Indonesia and the Philippines, but that species is still larger (adult length over 19 mm), has a more acute apical angle (<60[degrees]), a less rounded last adult whorl and only four spiral cords on the spire whorls. In addition, the aperture is transversely elongate and the lower columella tooth is larger than the upper one and bears additional denticles. C. reticulatum shows some variation in the strength of the sculpture (coarseness of reticulation), but this may in part be caused by damage to the outer lip during growth and the subsequent intercalation of additional spiral cords.
That this species has not been found off the Zululand coast of South Africa, a relatively well sampled area not far removed from the type locality, indicates that the isolated Almirante Leite Knolls (a volcanic seamount with numerous craters, rising from -1100 to -80 m) provide habitats of a different nature to those occurring on the continental shelf and upper slope. The station from which this material was obtained also contained an undescribed species of Bolma (Alf et al. 2010) and two new buccinid species (Fraussen & Rosado 2011), and many other new species of benthic invertebrates have been obtained from the locality as a whole (Bouchet pers. comm.), some of which have already been named (Cabezas et al. 2010; Komai & Chan 2010; Richer de Forges 2010).
Clypeostoma salpinx (Barnard, 1964) comb. n. Figs 4D, 6B, 10, 11
Turcica salpinx: Barnard 1964: 19, fig. 3a-d; Kensley 1973: 44, fig. 112. Type loc. (here designated, see below): off Cape Morgan, Eastern Cape, South Africa, 77 fath. [-141 m].
Shell: Trochoid-turbiniform, moderately elevated (L/D=1.25-1.50); teleoconch of 6-7 whorls; apical whorls convex, later ones more flat-sided; apical angle 60-70[degrees]; suture level with subperipheral cord, not itself channelled, but appearing so due to presence of a narrow sulcus below peripheral cord of preceding whorl, particularly pronounced on spire whorls; shell periphery rounded. First teleoconch whorl sculptured by somewhat sinuous axial pliculae; 3 spiral cords arise during second whorl and a fourth, beneath the adapical suture, during whorl three; sculpture of subsequent whorls comprising 4 (rarely five) spiral cords (not counting that level with abapical suture), these crossed by somewhat irregular, prosocline axial pliculae (30-40 on last adult whorl); cords with scale-like nodules where crossed by pliculae, those on uppermost (subsutural) cord frequently larger; intervals between cords slightly wider than cords themselves; cord intervals on last adult whorl occasionally with a weak intermediary cord. Base rounded, sculptured with [+ or -] 6 spiral cords; cords somewhat weaker than those on apical surface, but similarly nodose; umbilicus closed in all but small juveniles (<4 whorls). Peristome markedly oblique, more or less in one tangential plane; aperture subcircular to D-shaped, flattened in the parietal region; columella lip with 2 teeth, the lower one well developed, the upper one less so; teeth bluntly rounded, separated by shallow concavity; parietal and umbilical region covered with well-developed, translucent, smooth, glossy callus; margin of callus raised and flaring basally, running smoothly into basal lip; callus with 1 or 2 small tubercles near base of lower columella tooth. Outer lip lacking external varix, but thickened internally and bearing up to 15 denticles; denticles ridge-like, more or less in a single row and usually alternating in strength, that nearest columella usually strongest and separated from basal columella tooth by a U-shaped notch; edge of outer lip flaring; interior highly nacreous, not spirally lirate (except at apertural thickening), but somewhat angled beneath cords of external surface.
Microsculpture (Fig. 11B, C): Early teleoconch whorls with vermiform spiral threads in intervals between axial pliculae; adult microsculpture of scratch-like marks, filled with intritacalx in fresh specimens, but this seldom well developed and usually obscured by superficial encrustations.
Protoconch (Fig. 11A, C): Translucent white to pale buff; diameter 360-380 um; globose, strongly exsert and somewhat tilted; terminal lip convex with sinusigera-like beak in mid region; surface sculptured with a fine irregular granulation and a faint spiral thread midway between sutures.
Colour: Shell mostly uniform buff to pale orange-brown, occasionally nearly white; cord intervals with slight pink/green iridescence; some specimens with occasional slightly darker spots and blotches beneath suture and at edge of outer lip; live-collected specimens nearly always over-grown with a thin encrusting sponge (Fig. 10E, F).
Dimensions: Largest specimen (NMSA D4231), length 12.6 mm, diameter 9.0 mm; adult length mostly 11.0-12.5 mm, but at the northern limit of the distribution specimens are considerably smaller (adult length 6-7 mm) (Fig. 10G).
Operculum (Fig. 4D): Initially tightly multispiral, but whorls broadening with growth and becoming more openly multispiral.
Radula (Fig. 11D, E): Formula [infinity]+3+1+3+[infinity], with ca 80 transverse rows of teeth; transition from lateral to marginal series not clearly defined. Rachidian shaft with well-developed lateral flanges creating a distinct hood, base of cusp with raised, transverse basal ridge, cusp apex acutely trigonal with a narrow median denticle and 3-5 smaller denticles on each side. Lateral teeth overlapping extensively, their cusps asymmetrically trigonal; central denticle large on first lateral, progressively decreasing in size on second and third; side denticles few on first lateral, but progressively more numerous on laterals 2 and 3. Marginals numerous, the inner ones slightly longer than the laterals, their tips relatively narrow and finely pectinate laterally; marginals progressively longer and more slender in middle of series then shorter with proportionately broader tips at radula margin.
External anatomy (Fig. 6B): Body uniformly yellowish white, no skin pigmentation evident in preserved specimens. Cephalic lappets well developed, almost meeting in mid-line, free margin with fine projections; right post-ocular peduncle present in both sexes, nearly as long or longer than eyestalk, with a faint dorsal groove, peduncle tip often flattened and expanded; no right subocular tentacle evident; snout with extensive ventro-lateral flanges extending well beneath neck lobes, mouth deeply split midventrally; neck lobes broad, anterior two-thirds of right lobe with approx. 5 first-order tentacles with many smaller intermediaries, posterior third minutely fimbriate; anterior three-quarters of left lobe with approx. 10 tentacles in a variety of sizes, posterior quarter more or less smooth; epipodium bearing approx. 7 relatively large, first-order tentacles (not all the same size) with many, very much smaller intermediaries; larger epipodial tentacles usually with an indistinct basal sense organ; no epipodial sense organs evident beneath neck lobes.
Type material: Two syntype lots of Turcica salpinx Barnard, 1964 (SAMC): SOUTH AFRICA: off Cape Morgan, 77 fath. [-141 m], 1 broken adult (A9252); off Hood Point, 49 fath. [-90 m], 2 juvenile, 1 immature, 3 fragments (A9253).
Material examined (all NMSA unless indicated otherwise): MOZAMBIQUE: Inhambane transect, Campagne Mainbaza, dredged RV VizcondedeEza, st'n CP3143 (23.533[degrees]S 35.767[degrees]E), living, -264-277 m, 11.iv.2009 (MNHN); ditto, st'n CP3144 (23.55[degrees]S 35.68[degrees]E), living, -171-180 m, 16.iv.2009 (MNHN); between Inhaca and Inhambane (approx. 25.1[degrees]S 34.8[degrees]E), -225-300 m, lobster traps, J. Rosado, v.1996 (L8399); southern Mozambique [not further localized], living, -250 m, don. F. Amorim, v.1990 (K7361). SOUTH ARICA: KwaZulu-Natal: off Island Rock (27.2800[degrees]S 32.8233[degrees]E), -400 m, sandstone boulders, dredged NMDP, RV MeiringNaude, st'n ZE5, 5.vi.1987 (D6150); NE of Leven Point (27.8983[degrees]S 32.6567[degrees]E), -260 m, sponges, stones, dredged NMDP, RV Meiring Naude, st'n ZL6, 9.vi.1988 (E4447); off Cape Vidal off (28.1383[degrees]S 32.6150[degrees]E), -200 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n ZM9, 11.vi.1988 (E3941); SE of Neill Peak [Cunge Hill] (28.7400[degrees]S 32.5367[degrees]E), -320-340 m, sandy mud, dredged NMDP, RV Meiring Naude, st'n ZP5, 12.vi.1988 (E3979); SE of Port Durnford (29.0150[degrees]S 32.2017[degrees]E), -215 m, glutinous sandy mud, dredged NMDP, RV Meiring Naude, st'n ZQ8, 13.vi.1988 (E3123); ditto (29.0250[degrees]S 32.1967[degrees]E), -310-320 m, glutinous sandy mud, dredged NMDP, RV Meiring Naude, st'n ZQ9, 13.vi.1988 (E3167); off Glenton Reef (29.2450[degrees]S 32.0370[degrees]E), -200-210 m, sandy mud, dredged NMDP, RV Meiring Naude, st'n ZRR9, 18.vi.1989 (S457); off Matigulu River mouth (29.3567[degrees]S 31.9417[degrees]E), -145 m, mud, shell rubble, dredged NMDP, RV Meiring Naude, st'n ZR7, 16.vi.1989 (E8788); ditto (29.3650[degrees]S 31.9367[degrees]E), -200-220 m, mud and coarse sand with Dendrophyllia, dredged NMDP, RV Meiring Naude, st'n ZR9, 16.vi.1989 (E9014); off Umhlanga Rocks (29.7499[degrees]S 31.1685[degrees]E), 59 fath. [-107 m], dredged A.D. Connell, 7.vii.1983 (B6296); off Durban (29.8333[degrees]S 31.2367[degrees]E), living, -130 m, sandstone gravel and some rocks, dredged NMDP, RV Meiring Naude, st'n XX115, 9.vii.1986 (D4231); ditto (29.8400[degrees]S 31.2333[degrees]E), living, -150 m, sandstone gravel and some sponge, dredged NMDP, RV Meiring Naude, st'n XX114, 9.vii.1986 (D4146); off Umlaas Canal (30.0133[degrees]S 31.0600[degrees]E), -150 m, muddy sand and fine pebbles, dredged NMDP, RV Meiring Naude, st'n XX75, 10.vii.1985 (D1155); ditto (30.0183[degrees]S 31.0533[degrees]e), living, -150 m, coarse sand and pebbles with numerous spatangoids, dredged NMDP, RV Meiring Naude, st'n XX70, 9.vii.1985 (E7597, D796); ditto (30.0317[degrees]S 31.0450[degrees]E), -150 m, coarse sand, dredged NMDP, RV Meiring Naude, st'n XX72, 10.vii.1985 (D859); ditto (30.0367[degrees]S 31.0650[degrees]E), living, -250 m, coarse sand, dredged NMDP, RV Meiring Naude, st'n XX67, 9.vii.1985 (D1449); off Amanzimtoti (30.0783[degrees]S 31.0550[degrees]E), -300-305 m, medium sand, dredged NMDP, RV Meiring Naude, st'n XX66, 9.vii.1985 (D1315); ditto (30.0883[degrees]S 31.0417[degrees]E), -260-270 m, medium sand, dredged NMDP, RV Meiring Naude, st'n XX65, 9.vii.1985 (D1180); ditto (30.0967[degrees]S 30.9950[degrees]E), living, -115-125 m, medium sand, dredged NMDP, RV Meiring Naude, st'n XX61, 9.vii.1985 (D1267); ditto (30.1000[degrees]S 31.0267[degrees]E), living, -245-250 m, medium sand, dredged NMDP, RV Meiring Naude, st'n XX64, 9.vii.1985 (D1663); ditto (30.1067[degrees]S 31.0133[degrees]E), -160-170 m, medium sand, dredged NMDP, RV Meiring Naude, st'n XX62, 9.vii.1985 (D1489); off Park Rynie (30.3838[degrees]S 30.8355[degrees]E), living, -140 m, some sand, sponge rubble, dredged NMDP, RV Meiring Naude, st'n X6, 19.viii.1981 (C1590); ditto (30.3838[degrees]S 30.8350[degrees]E), living, -140 m, some sand, sponge rubble, dredged NMDP, RV Meiring Naude, st'n X7, 19.viii.1981 (C1628); Port Edward, slightly north of (31.0967[degrees]S 30.3133[degrees]E), living, -140 m, live sponges, dredged NMDP, RV Meiring Naude, st'n XX57, 8.vii.1985 (D1402); off Port Edward (31.1067[degrees]S 30.3000[degrees]E), living, -125 m, living sponges, dredged NMDP, RV Meiring Naude, st'n XX58, 8.vii.1985 (D906); ditto (31.1133[degrees]S 30.2967[degrees]E), living, -120-125 m, living sponges, dredged NMDP, RV Meiring Naude, st'n XX59, 8.vii.1985 (D1358, E7372); ditto (31.1400[degrees]S 30.2766[degrees]E), living, -160 m, sponge, gorgonians and sand, dredged NMDP, RV Meiring Naude, st'n A18, 7.vii.1986 (C9650, S3190); ditto (31.1650[degrees]S 30.2516[degrees]E), living, -140 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n A6, 18.viii.1981 (E225). Eastern Cape: off Mtentu River (31.2333[degrees]S 30.1833[degrees]E), living, -150-160 m, sponges rocks, dredged NMDP, RV Meiring Naude, st'n XX12, 15.vi.1983 (C5225); off Msikaba River (31.3867[degrees]S 30.0433[degrees]E), living, -100 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n C3, 12.viii.1981 (C1125, E192); ditto (31.4166[degrees]S 29.9833[degrees]E), -150 m, sponge, gorgonians, dredged NMDP, RV Meiring Naude, st'n D23, 2.vii.1986 (C9808); off Port Grosvenor (31.4166[degrees]S 29.9666[degrees]E), living, -95-100 m, coarse sand, very gorgonians, dredged NMDP, RV Meiring Naude, st'n D12, 16.viii.1981 (E209); ditto (31.4166[degrees]S 29.9333[degrees]E), -80-84 m, calcareous nodules, sand, dredged NMDP, RV Meiring Naude, st'n D19, 16.viii.1981 (E196); ditto (31.4317[degrees]S 29.9650[degrees]E), -120-128 m, coarse sand, some mud, solitary coral, shells, dredged NMDP, RV Meiring Naude, st'n D2, 13.viii.1981 (C1151); off Waterfall Bluff (31.5017[degrees]S 29.9200[degrees]E), living, -200 m, some sponges, rocks, dredged NMDP, RV Meiring Naude, st'n E10, 4.vii.1986 (C9837); off Mgazi River (31.7283[degrees]S 29.5317[degrees]E), -140-145 m, glutinous black mud, dredged NMDP, RV Meiring Naude, st'n J13, 4.vii.1985 (C9304, S3192); off Rame Head (31.8450[degrees]S 29.4750[degrees]E), living, -150-160 m, some sponges, dredged NMDP, RV Meiring Naude, st'n K10, 20.vii.1982 (C1896); ditto (31.8583[degrees]S 29.4683[degrees]E), living, -170-200 m, sandstone, yellow hydroids, dredged NMDP, RVMeiring Naude, st'n K11, 20.vii.1982 (C1911); off Ubombo Head (31.9050[degrees]S 29.2967[degrees]E), living, -60-62 m, coarse sand, oyster shell conglomerate, dredged NMDP, RV Meiring Naude, st'n L4, 16.vii.1982 (E248); ditto (31.9700[degrees]S 29.3900[degrees]E), living, -200 m, smooth bedrock, living sponges, dredged NMDP, RV Meiring Naude, st'n L11, 8.vii.1985 (C8970); off Whale Rock (31.9800[degrees]S 29.2800[degrees]E), living, -90 m, sponge rubble, coarse sand, some rocks, dredged NMDP, RV Meiring Naude, st'n M12, 3.vii.1985 (C9502); ditto (32.0166[degrees]S 29.3166[degrees]E), living, -150-200 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n M8, 20.vii.1982 (E289); ditto (32.0283[degrees]S 29.3050[degrees]E), living, -150-165 m, some coarse sand, discoid corals, dredged NMDP, RV Meiring Naude, st'n M9, 20. vii.1982 (C2313); ditto (32.0333[degrees]S 29.3166[degrees]E), living, -200-210 m, sponge rubble, some sandstone rocks, dredged NMDP, RV Meiring Naude, st'n M13, 3.vii.1985 (C9276); off Mncwasa Point (32.1033[degrees]S 29.1083[degrees]E), -68 m, sand, dredged NMDP, RV Meiring Naude, st'n N8, 10.vii.1982 (E337); off Nthlonyane River (32.2183[degrees]S 28.9950[degrees]E), -80 m, sand, broken shell, dredged NMDP, RV Meiring Naude, st'n P6, 17.vii.1982 (C2577); ditto (32.2400[degrees]S 29.0233[degrees]E), living, -95 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n P4, 17.vii.1982 (E329); ditto (32.2567[degrees]S 29.0133[degrees]E), living, -90-95 m, lithothamnion pebbles, dredged NMDP, RV Meiring Naude, st'n P5, 17.vii.1982 (C2565); ditto (32.2750[degrees]S 29.0100[degrees]E), living, -240 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n P1, 17.vii.1982 (C3305); ditto (32.2867[degrees]S 29.0817[degrees]E), living, -220-230 m, branching sponges, gorgonians, dredged NMDP, RV Meiring Naude, st'n P2, 17.vii.1982 (C3300, E264); off Mbashe River (32.3033[degrees]S 29.0683[degrees]E), living, -200-220 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n Q1, 18.vii.1982 (V1944); off Qora River (32.3967[degrees]S 28.8117[degrees]E), living, -196 m, sponge, dredged NMDP, RV Meiring Naude, st'n U9, 14.vi.1983 (C5154); off Nqabara Point off (32.4533[degrees]S 28.9317[degrees]E), living, -250 m, live sponges with some corals, dredged NMDP, RV Meiring Naude, st'n S10, 12.vii.1984 (V4056); ditto (32.4550[degrees]S 28.9267[degrees]E), living, -210 m, live sponges, dredged NMDP, RV Meiring Naude, st'n S9, 11.vii.1984 (W7627); off Shixini Point (32.4900[degrees]S 28.8967[degrees]E), living, -240 m, sponge rubble, some sandstone, dredged NMDP, RV Meiring Naude, st'n T11, 11.vii.1984 (C6301); ditto (32.5233[degrees]S 28.8650[degrees]E), -240 m, sand and old rubble, dredged NMDP, RV Meiring Naude, st'n T13, 11.vii.1984 (C6342); off Qora River (32.5633[degrees]S 28.8067[degrees]E), living, -150-168 m, stones and sponges, dredged NMDP, RV Meiring Naude, st'n U7, 14.vi.1983 (C5095); ditto (32.5700[degrees]S 28.6567[degrees]E), -400 m, sand, dredged NMDP, RV Meiring Naude, st'n U8, 14.vi.1983 (C4877); off Sandy Point (32.6233[degrees]S 28.6150[degrees]E), -90 m, coarse sand, dredged NMDP, RV Meiring Naude, st'n W1, 11.vi.1983 (V2882); off Stony Point (32.6250[degrees]S 28.7333[degrees]E), living, -150-152 m, calcareolite and coral, dredged NMDP, RV Meiring Naude, st'n V9, 14.vi.1983 (C4371); off Qolora River (32.7733[degrees]S 28.5900[degrees]E), living, -174 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n Y2, 12.vi.1983 (C4637); ditto (32.7633[degrees]S 28.6067[degrees]E), living, -240-250 m, live sponges, dredged NMDP, RV Meiring Naude, st'n Y9, 14.vii.1984 (C7131).
Type locality: Barnard's type material originated from two Pieter Faure stations, but he did not select one in particular as the type locality. The given data for one of these stations (off Hood Point (East London), 49 fath. [-90 m]) has since been shown to be erroneous (Herbert 1987: 344). In this instance, the Hood Point locality is not far out of the known range of the species, but for other taxa it is widely so. The dredge haul was more probably taken off Durban. In view of this inaccuracy I here select the provenance of Barnard's other type lot, off Cape Morgan [close to mouth of Great Kei River] 77 fath. [-141 m], as type locality.
Distribution and habitat (Fig. 8): South-eastern Africa; from southern Mozambique (Inhambane) to the Great Kei River, Eastern Cape, South Africa; -60-400 m, living specimens -60-277 m, on hard substrata with abundant marine growths, particularly sponges.
Remarks: When compared to other species here transferred to Clypeostoma, C. salpinx differs from C. meteorae, C. nortoni and C. townsendianum in having slightly flatter whorls, commonly four rather than five or six spiral cords above the suture, fewer tubercles on the callus shield near the basal columella tooth, less elongate, more widely spaced denticles inside the outer lip and a more extensive inductural callus. C. elongata Vilvens, 2001 from Indonesia and C. reticulatum are both considerably larger, have coarser sculpture and more robust columella teeth. Within southern Africa, juvenile C. salpinx may be confused with those of Danilia textilis, but the latter have finer, more regularly cancellate sculpture, a sunken rather than an exsert protoconch and more rounded whorls with a more narrowly channelled suture.
Specimens from the northern limit of the distribution (off Inhambane) are surprisingly small, almost half the size South African specimens (Fig. 10G). However, this is the only significant conchological difference between this material and typical C. salpinx, and I consider it to be insufficient evidence upon which to base the description of a new species. Other species of Clypeostoma also exhibit considerable variation in size at maturity.
Clypeostoma meteorae (Neubert, 1998) comb. n. Figs 8, 12A-E
Agathodontameteorae: Neubert 1998: 465, figs 4-7. Type loc.: 'Bab al-Mandab' [Bab-el-Mandeb] (12[degrees]21.4'N 43[degrees]26.9'E), -45 m, Djibouti, Red Sea-Gulf of Aden.
Material examined: MADAGASCAR: South-west of Cap d'Ambre, Antsiranana (12.133[degrees]S 48.933[degrees]E), -238-249 m, CampagneMiriky, st'n DW3196, dredged, 28.vi.2009 (MNHN); off Mahajamba Bay (14.883[degrees]S 46.933[degrees]E), -90-257 m, Campagne Miriky, st'n DW3245, dredged, 7.vii.2009 (MNHN); East of Faux-Cap (25.633[degrees]S 46.217[degrees]E), -128-133 m, Exped'n Atimo Vatae, st'n CP3561, dredged Nosy Be 11, 6.v.2010 (MNHN); between Lokaro and Ste Luce (24.865[degrees]S 47.467[degrees]E), -80-83 m, Exped'n Atimo Vatae, st'n DW3519, dredged Nosy Be 11, 30.iv.2010 (MNHN).
Remarks: Five shells matching the description of this little-known Red Sea-Gulf of Aden species were obtained during dredging surveys undertaken by the MNHN off Madagascar (Campagne Miriky and Exped'n Atimo Vatae). In this material the aperture is clearly distinct from that of C. salpinx in having stronger columella teeth and a much less extensive inductural callus shield. There are also in-running ridges extending from the labral denticles into the aperture and the callus shield, though less extensive, bears superficial ridges in the parietal region and additional granules, particularly around the columella base. Like the original material of C. meteorae, these specimens have four spiral cords on the spire whorls, but 1-2 intermediary cords appear during the penultimate whorl such that the last adult whorl has 5-6 spiral cords above and including the peripheral one.
In terms of its apertural features, this Malagasy material is also close to C. cf. nortoni (see below) and C. townsendianum, but it is considerably smaller than both. The specimens all have mature apertural dentition and range in length from 5.3-6.9 mm. In addition, the protoconch is also smaller, diameter 220-240 um (compared to 310-320 um for C. cf. nortoni). Specimens at the smaller end of this size range show considerable similarity with Perrinia docili Poppe, Tagaro & Dekker, 2006 from the Philippines, which I strongly suspect may also belong to Clypeostoma.
Clypeostoma cf. nortoni (McLean, 1984) comb. n. Figs 8, 12H-L
Agathodonta nortoni: McLean 1984: 122, figs 1-3; Tsuchida & Kurozumi 1996: 37, fig. 3; Vilvens 2001: figs 11-13; Vilvens & Heros 2003: figs 12, 13; Poppe et al. 2006: 30, pl. 7, figs 3, 4; Poppe & Tagaro 2008: 166, pl. 28, figs 4, 6. Type loc.: off Baltazar Is. (13[degrees]14'N 121[degrees]49'E), Marinduque Province, Philippines.
A number of Clypeostoma shells resembling C. nortoni have been dredged off Madagascar and obtained ex piscibus from the Sofala Bank, off central Mozambique. Similar material evidently also occurs off Reunion and was recorded there by Jay (2009 as Clanculus ceylonicus). Compared with C. meteorae, these specimens are consistently larger (mature length 9.6-14.6 mm) and have a larger protoconch (diameter 310-320 um). In addition, the basal callus shield is more extensive and smoother, bearing (apart from 3-5 strong, in-running parietal ridges overlying the basal sculpture of the previous whorl) only traces of superficial ridging in the parietal region and a small number of granules associated with the medial columella area. C. townsendianum (Fig. 12F, G), from the Persian Gulf, is of a similar size (figured syntype, length 10.4 mm), but has stronger apertural dentition more similar to that of C. meteorae. The form of the apertural dentition in the present material is closest to that of the Philippine C. nortoni, but the shell attains a relatively larger size (length up to 14.6 mm) than does C. nortoni, and the suture is slightly less indented and the whorls thus less convex. With the limited amount of material available and the geographical distances involved, it is difficult to assess the significance of these differences. I cannot thus confidently identify this material as C. nortoni, but rather draw attention to its existence and its similarity to that species.
Material examined: MADAGASCAR: West of Nosy Be (13.400[degrees]S 47.967[degrees]E), -210-310 m, Campagne Miriky, st'n DW3232, dredged, 03.vii.2009 (MNHN); south of Faux-Cap (26.133[degrees]S 45.650[degrees]E), -280-333 m, Exped'n Atimo Vatae, st'n DW3553, dredged Nosy Be 11, 5.v.2010 (MNHN); ditto (26.117[degrees]S 45.650[degrees]E), -264-280 m, Exped'n Atimo Vatae, st'n DW3552, dredged Nosy Be 11, 5.v.2010 (MNHN). MOZAMBIQUE: Between Beira and Bazaruto, Sofala Bank (20.123[degrees]S 35.543[degrees]E), -90-145 m, expisce, J. Rosado 2005-2008 (J. Rosado coll'n).
Genus Danilia Brusina, 1865
Olivia: Cantraine 1835: 387 [non OliviaBertolini, 1810 (?Porifera)]. Type species: Oliviaotaviana-Cantraine, 1835, by monotypy.
Craspedotus: Philippi 1847: 23 [non Craspedotus Schoenherr, 1844 (Coleoptera)]. Type species:Monodonta limbata Philippi, 1844 [= Monodonta tinei Calcara, 1839], by monotypy.
Otavia: Gray 1847: 145 (non Risso, 1826), laps. cal. for Olivia Cantraine, 1835.
?Heliciella: O.G. Costa 1861: 64. Type species: Heliciella costellata O.G. Costa, 1861, by subsequent designation (Dall 1927: 134).
Danilia: Brusina 1865: 25. Type species: Monodonta limbata Philippi, 1844 [= Monodonta tinei Calcara, 1839], by monotypy.
Nomenclatural remarks: Beu and Climo (1974) stated that Brusina (1865) proposed the name Danilia as a replacement name for the homonymous Olivia Cantraine, 1835. However, I can find no evidence in either Brusina (1865) or Brusina (1866) that this was his intention. In fact, he made no mention of Olivia Cantraine, 1835. Had he proposed Danilia expressly as a replacement for Olivia, the type species of Danilia would have been that of Olivia (ICZN 1999: Art. 67.8), namely Olivia otaviana Cantraine, 1835 (by monotypy). Instead, the type species of Danilia (by monotypy) is Monodonta limbata Philippi, 1844 [= Monodonta tinei Calcara, 1839]. The type species of Olivia, O. otaviana, has traditionally also been considered to be synonymous with Calcara's Monodonta tinei and is the earlier name, but Palazzi and Villari (2001) consider O. otaviana to represent a distinct fossil species--an opinion shared by Landau et al. (2003). In such case, the type species of Danilia and Cantraine's Olivia are not the same and they are thus subjective rather than objective synonyms.
Keen (1960) listed Heliciella O.G. Costa, 1861, in the synonymy of Olivia, but Beu and Climo (1974) considered the affinity of this taxon with Olivia and thus with Danilia to be debateable, since its type species, H. costellata O.G. Costa, 1861 (2), was based on a very juvenile shell. (The specimen of H. costellata figured by O.G. Costa (1861) has a diameter of 1.0 mm and comprises only approx. 1.5 teleoconch whorls.) They chose instead to employ the younger name Danilia on account of this uncertainty. However, Monterosato (1884) had earlier stated that H. costellata was a juvenile shell of Danilia tinei and certainly the figures of juvenile D. tinei provided by Scaperrotta et al. (2009) clearly support the view that H. costellata is a juvenile Danilia. However, Palazzi and Villari (2001), recognising H. costellata as a species of Danilia, believed it to be distinct from D. tinei and to represent a second Recent European species of this genus. Thus Heliciella and Danilia are evidently synonymous and since both are valid names, the earlier one, Heliciella, ought to be afforded priority. However, in order to maintain prevailing usage, and in accordance with ICZN, Art. 23.9 (ICZN 1999), the principle of priority can be moderated, provided the conditions of Art. 126.96.36.199 and Art. 188.8.131.52 are met. In this regard, I am not aware that Heliciella has been used as a valid name after 1899 (Art. 184.108.40.206) and Danilia has been used more than 25 times in the last 50 years by at least 10 different authors (Art. 220.127.116.11) (Beu & Climo 1974; Powell 1979; Piani 1980; Aimassi et al. 1983; Guidastri et al. 1984; Spadini 1986; Graham 1988; Vaught 1989; Hickman & McLean 1990; Poppe & Goto 1991; Wilson 1993; Giannuzzi-Savelli et al. 1994; Jansen 1996; Millard 1997; Higo et al. 1999; Sasaki 2000; Palazzi & Villari 2001; Vilvens 2001; Spencer et al. 2002; Landau et al. 2003; Rolan 2005; Vilvens & Heros 2005; Poppe et al. 2006; Crocetta & Spanu 2008; Poppe & Tagaro 2008; Kano 2009; Scaperrotta et al. 2009; Spencer et al. 2009; Bandel 2010; De Simone & Kosuge 2010). Some of these, despite recognising the priority of Heliciella over Danilia, have continued to use the latter as the valid name. Therefore, in accordance with Art. 23.9, Danilia is to be afforded priority over the earlier, but unused Heliciella. Danilia thus becomes a nomen protectum and Heliciella a nomen oblitum.
Remarks: The genus Danilia is known primarily from relatively deep water and has a fossil record extending back to the Lower Cretaceous (Beu & Climo 1974). Nine Recent species are known from the Indo-West Pacific, eight of which were discussed and illustrated (mostly type specimens) in a useful contribution by Vilvens and Heros (2005). The differences between the species are in some cases small, and Beu and Climo (1974) cautioned that without more detailed comparative study, it is impossible to be certain whether each of these nominal taxa represents a genuinely distinct species, or whether there are fewer, more widespread and sculpturally variable ones--an observation with which I concur. The problem is exacerbated by the fact that most descriptions have been based on very few specimens and thus give no indication of intraspecific variability. These difficulties notwithstanding, I describe below two additional species, since neither appears clearly referable to any of the described taxa.
Danilia species generally live on hard substrata. Whilst most species seem to be scarce, a Mediterranean species, tentatively identified as Danilia costellata (O.G. Costa, 1861), may be locally abundant in colonies of the gorgonian Corallium rubrum (Linnaeus, 1758) (Crocetta & Spanu 2008). Schepman (1908), Beu & Climo (1974) and De Simone & Kosuge (2010) also reported species living in deep-water coral communities. D. textilis (below) was found living primarily in sponge dominated communities, but such communities frequently included deep-water corals. Guidastri et al. (1984) and Smriglio et al. (1989) reported a similar association between Putzeysia wiseri (Calcara, 1842) and deep-water Scleractinia in the Mediterranean.
The radula of Danilia tinei (Calcara, 1839) was illustrated by Guidastri et al. (1984 as D. otaviana) and that of D. insperata was figured and discussed in detail by Beu and Climo (1974). That of D. textilis described below is very similar.
Key to species of Danilia in the south-western Indian Ocean
1 Shell relatively small (adult length <6 mm); spiral cords of unequal size, the two largest delimiting a cylindrical peripheral band; axial pliculae fine, close-set and sharp boucheti
--Shell larger (adult length >10 mm); periphery more or less rounded; spiral cords of approximately equal size (peripheral one may be slightly larger), axial pliculae rounded, relatively widely spaced textilis
Danilia boucheti sp. n. Figs 13, 14
Etymology: Named for Prof. Philippe Bouchet (MNHN), well-known malacologist and co-ordinator of numerous important dredging expeditions.
Shell: Turbiniform with moderately elevated spire (L/D=1.09-1.14); teleoconch of 4.25-4.5 whorls; apical angle approx. 75[degrees]; whorls convex and suture indented, level with periphery on spire, but descending below this just prior to outer lip; protoconch sunken and apex thus appearing truncated; outer lip with well-developed subterminal varix. First teleoconch whorl sculptured with approx. 17 orthocline axial pliculae; 4 spiral cords develop during second whorl (P1-P4); additional secondary and tertiary cords arising during subsequent whorls; abapical cord (P4) strongest; axial pliculae narrow, becoming strongly prosocline on later whorls, the crest crisp and leaning forward; spiral cords beaded where crossed by axial pliculae, beads on primary cords scale-like; last adult whorl with P2-P4 (particularly P3 and P4) stronger than other cords; P3 and P4 delimiting an almost cylindrical peripheral band; subsutural cord (P1) also slightly larger and with angular beads that intermesh in a zip-like manner with beads of peripheral cord (P4) of preceding whorl, delimiting a narrow sutural channel; number of intermediary cords variable, generally only 1-2 between P3 and P4; 5-6 between P1 and P3 (including P2); beading of intermediary cords weaker; cords generally somewhat narrower than their intervals and slightly stronger than axial pliculae; interaction of cords and pliculae producing a fine, regular, oblique cancellation with rhomboidal interstices. Base with approx. 10 spiral cords of alternating size; the second one below peripheral cord generally the strongest; axial pliculae continue onto base rendering basal cords finely beaded. Peristome markedly oblique, more or less in one near-tangential plane; aperture roundly D-shaped and flattened in parietal region; columella lip a thickened pillar set with 2 teeth, lower one a well-developed, rounded peg, the upper one smaller; interval between teeth shallowly concave; umbilical region medial to thickened columella pillar sunken, forming an elongate curved pit, broadening basally; umbilical/parietal region covered by a thin, extensive inductural callus; callus translucent, smooth and glossy, raised somewhat basally and confluent with flaring margin of outer lip; interior of outer lip with an in-set varix lying beneath external outer lip varix; internal varix swollen to form prominent bulge near junction of basal and columella lips, creating deep, U-shaped notch below lower columella tooth; apertural varix set with numerous small, ridge-like denticles, these restricted to varix, not running into deeper portions of aperture; interior of aperture with traces of nacre (specimens not fresh), not spirally lirate, though spiral cords of shell exterior visible by translucence; margin of outer lip flaring; external varix ca 0.25 mm wide and set back a similar distance behind lip edge.
Microsculpture: Irregular, vermiform spiral threads present on early whorls; later whorls with evidence of scratch-like sculpture, but sculptural details largely obscured by dirty intritacalx deposit, particularly in interstices.
Protoconch: Present only in holotype; translucent white; diameter 320 um; sunken into first teleoconch whorl and somewhat tilted; terminal lip evidently more or less straight.
Colour: First teleoconch whorl uniform white, later whorls pale buff with faint brown spots on primary spiral cords; flared outer lip and external varix with brown marks in intervals between paler primary spiral cords. One paratype with darker brown subsutural blotches, stronger spots on the primary peripheral cords and irregular brownish mottling on the base.
Dimensions: Holotype (largest specimen), length 5.32 mm, diameter 4.66 mm.
Operculum, radula and external anatomy: Unknown.
Holotype (Fig. 14A-C): MADAGASCAR: Antsiranana, south-west of Cap d'Ambre (12.133[degrees]S 48.933[degrees]E), -238-249 m, Campagne Miriky, st'n DW3196, dredged, 28.vi.2009 (MNHN 24648).
Paratypes: Same data as holotype (MNHN 24649, 2 specimens).
Note: The holotype is the most complete specimen, but its apertural dentition is subadult; one of the paratypes has more mature apertural features (Fig. 14D), but is missing the apical whorls.
Distribution (Fig. 13): Known only from off the coast of the far north of Madagascar, -238-249m.
Remarks: Danilia boucheti is smaller than any of the other described species of Danilia, most of which attain or exceed 10 mm in length (Vilvens & Heros 2005). Perhaps the most similar species is D. angulosa Vilvens & Heros, 2005 from Melanesia and the Philippines, which is also small (length up to 7.9 mm) and has stronger spiral cords associated with the periphery, creating a cylindrical peripheral band. In that species, however, the cords are stronger than in D. boucheti, the whorls more distinctly shouldered, and there is only one intermediary cord between the shoulder cord (upper peripheral cord) and the subsutural cord. In D. boucheti there are 5 or 6 intermediary cords in this region.
Danilia textilis sp. n. Figs 4E, 6C, 13, 15-17
Etymology: From Latin textilis (woven); in reference to the regular, net-like sculpture.
Shell: Trochoid-turbiniform, moderately elevated (L/D=1.10-1.30); teleoconch of 5-6 whorls; whorls rounded, but peripheral spiral cord frequently slightly stronger than others, giving the appearance of a weak peripheral angulation; apical angle approx. 75[degrees]; protoconch sunken and apex thus appearing truncated; suture indented, but somewhat adpressed, narrowly channelled, inserted at level of subperipheral cord, but descending below this just prior to outer lip; outer lip with well-developed subterminal varix. First teleoconch whorl sculptured with 15-18 axial pliculae; 4 spiral cords develop during second whorl, others arising by intercalation with growth; penultimate and last adult whorls with [+ or -] 8 primary spiral cords between suture and periphery (inclusive); cords well defined, narrower than their intervals; an intermediary cord often developing between primary cords in latter half of last adult whorl. Axial sculpture of prosocline pliculae persists throughout growth; pliculae generally slightly weaker than cords, these together producing a regular, oblique cancellation with equilaterally rhomboidal interstices; cords roundly beaded where crossed by pliculae; beads stronger and more angular on subsutural cord and still larger and scale-like or spine-like on peripheral cord. Base with [+ or -] 7 primary spiral cords, their intervals usually with a weaker intermediary in final half whorl; axial pliculae continue onto base but their number appears to double in latter third of last adult whorl, rendering spiral cords more finely and closely beaded; umbilicus lacking in all except small juveniles (<4 whorls), but often obscured by reflected columella even in these. Peristome markedly oblique, more or less in one tangential plane; aperture roundly D-shaped, flattened parietally; columella lip a thickened pillar with 2 well developed teeth separated by a shallow concavity; lower tooth stronger and frequently squarish with a raised ridge along its lower margin, upper tooth weaker and clearly representing the end of a subparietal spiral pleat; pleat separated from paries by a U-shaped notch; umbilical region median to thickened columella pillar sunken, forming an elongate, more or less rectangular pit; pit bordered basally by a medial extension of ridge of lower columella tooth; umbilical/parietal region covered by inductural callus; callus translucent, smooth and glossy, raised somewhat basally and confluent with flaring margin of outer lip; interior of outer lip with a subterminal thickening corresponding in position with the external labral varix; thickening set with numerous elongate denticles and rounded tubercles, these not running into deeper portions of aperture; 2-3 denticles closest to junction of basal and columella lips larger, forming a two- to three-humped bulge, which together with the lower columella tooth delineates a pronounced U-shaped notch; margin of outer lip flaring; interior of aperture nacreous, somewhat angled beneath spiral cords of shell exterior, but not spirally lirate.
Microsculpture (Fig. 16B, C): First teleoconch whorl with vermiform spiral threads, replaced on subsequent whorls by prosocline, scratch-like microsculpture, filled with intritacalx deposit.
Protoconch (Fig. 16A, C): Translucent white; diameter ca 360 um; level -with or slightly sunken below first teleoconch whorl and somewhat down-tilted; degree of tilting variable between individuals; surface mostly worn, but with traces of irregular granulation; terminal lip weakly sinuous.
Colour: Spire apex uniform white, later whorls (3rd onwards) with pale fawn ground patterned with darker, brownish markings; markings initially in the form of axial bands or blotches, but penultimate and last adult whorls rather more randomly mottled; subterminal varix and flaring edge of outer lip with brownish spiral bands in cord intervals; precise shade and density of colour pattern variable between individuals, some specimens very sparsely marked, but outer lip varix evidently patterned in all; cord intervals with pink/green iridescence; living and fresh specimens with a buffish intritacalx deposit. Shell exterior of live-taken specimens often with some encrustation by other marine organisms, but not (in the material available) entirely covered with sponge.
Dimensions: Holotype (largest specimen), length 11.4 mm, diameter 8.8 mm.
Operculum (Fig. 4E): Initially tightly multispiral, but whorls broadening with growth and becoming more openly multispiral.
Radula (Fig. 17): Formula co+4+1+4+co, with ca 60 transverse rows of teeth. Lateral flanges of rachidian well developed creating a distinct hood, base of cusp with well-developed transverse ridge, cutting edge coarsely dentate with elongate-lanceolate central denticle with narrowly acuminate tip, and up to 5 lateral denticles on each side, last of which forms a rounded boss at shoulder of tooth. Laterals with overlapping shafts and trigonal, distinctly asymmetrical cusps, decreasing in size from first to fourth; inner cutting edge curved and finely serrate almost from tip, outer cutting edge nearly straight and set with coarse, close-set, more elongate denticles, starting some distance back from tip; first lateral strongly hunched at outer base of cusp. Marginals very long and slender, the cusp elongate, sides frequently in-rolled, and with a fringe of fine denticles on outer margin and at tip; outer shaft base of inner 1 or 2 marginals expanded and jaggedly serrate; serrations continue sparsely up shaft.
External anatomy (Fig. 6C): Body whitish with some brownish pigmentation on sides of foot, lateral regions of snout and around base of cephalic tentacles. Cephalic lappets distinct, their free margin smooth or at most shallowly lobate (rather than digitate), its edge minutely papillate; lateral expansions of snout broad; right post-ocular peduncle present in both sexes, a longitudinal groove on its upper surface; right subocular tentacle not evident; right neck lobe with approx. 4 moderately large, first-order tentacles anteriorly; left neck lobe with a group of 8-10 slightly smaller first-order tentacles anteriorly, with even smaller intermediaries; left neck lobe with approx. 5 epipodial sense organs on under surface, right lobe with approx. 3; posterior part of both lobes smooth; epipodial fold with 7-9 tentacles of various sizes, but without very small intermediary tentacles; larger tentacles each with a distinct epipodial sense organ at the base.
Holotype (Fig. 15A-C): SOUTH AFRICA: Eastern Cape: off Rame Head (31.8450[degrees]S 29.4750[degrees]E), living, -150-160 m, sponges, dredged NMDP, RV Meiring Naude, st'n K10, 20.vii.1982 (NMSA E7756/T2595).
Paratypes: SOUTH AFRICA: KwaZulu-Natal: SE of Neill Peak [Cunge Hill] (28.7400[degrees]S 32.5367[degrees]E), -320-340 m, sandy mud, dredged NMDP, RV Meiring Naude, st'n ZP5, 12.vi.1988 (MNHN 24817, 1 specimen); SE of Port Durnford (29.0150[degrees]S 32.2017[degrees]E), -215 m, glutinous sandy mud, dredged NMDP, RV Meiring Naude, st'n ZQ8, 13.vi.1988 (NMSA E3119/T2597, 4 specimens); SE of Port Durnford (29.0967[degrees]S 32.1567[degrees]E), -165 m, mud with sand, dredged NMDP, RV Meiring Naude, st'n ZQ15, 17.vi.1989 (NHMUK 20110382, 1 specimen); off Matigulu River mouth (29.3567[degrees]S 31.9417[degrees]E), -145 m, mud, shell-rubble, dredged NMDP, RV Meiring Naude, st'n ZR7, 16.vi.1989 (NMSA E8790/T2599, 1 specimen). Eastern Cape: same data as holotype (NMSA S9912/T2594, 1 specimen); off Qora River (32.3967[degrees]S 28.8117[degrees]E), living, -196 m, sponge, dredged NMDP, RV Meiring Naude, st'n U9, 14.vi.1983 (NMSA C5153/T2593, 1 specimen); off Nqabara Point (32.4533[degrees]S 28.9317[degrees]E), living, -250 m, live sponges, some corals, dredged NMDP, RV Meiring Naude, st'n S10, 12.vii.1984 (NMSA W7501/T2592, 1 specimen); off Nqabara Point (32.4550[degrees]S 28.9267[degrees]E), living, -210 m, live sponges, dredged NMDP, RV Meiring Naude, st'n S9, 11.vii.1984 (NMSA W7502/T2671, 1 specimen); off Qolora River (32.7633[degrees]S 28.6067[degrees]E), -240-250 m, live sponges, dredged NMDP, RV Meiring Naude, st'n Y9, 14.vii.1984 (NMSA C7026/T2598, 1 specimen).
Additional material examined (all NMSA): MOZAMBIQUE: off Beira, -110-145 m, ex pisce, v.2005 (J. Rosado coll'n). SOUTH AFRICA: KwaZulu-Natal: SE of Neill Peak [Cunge Hill] (28.7400[degrees]S 32.5367[degrees]E), -320-340 m, sandy mud, dredged NMDP, RV Meiring Naude, st'n ZP5, 12.vi.1988 (E3977, E3978, E6958) SE of Port Durnford (29.0250[degrees]S 32.1967[degrees]E), -310-320 m, glutinous sandy mud, dredged NMDP, RV Meiring Naude, st'n ZQ9, 13.vi.1988 (E3168); ditto (29.0967[degrees]S 32.1567[degrees]E), -165 m, mud with sand, dredged NMDP RV Meiring Naude, st'n ZQ15, 17.vi.1989 (E8697); off Glenton Reef (29.2450[degrees]S 32.0367[degrees]E), -200-210 m, sandy mud, dredged NMDP, RV Meiring Naude, st'n ZRR9, 18.vi.1989 (S481); off Matigulu River mouth (29.3650[degrees]S 31.9367[degrees]E), -200-220 m, mud and coarse sand with Dendrophyllia, dredged NMDP, RV Meiring Naude, st'n ZR9, 16.vi.1989 (E8989). Eastern Cape: off Rame Head (31.8583[degrees]S 29.4683[degrees]E), -170-200 m, sandstone, yellow hydroids, dredged NMDP, RV Meiring Naude, st'n K11, 20.vii.1982 (S9944); off Mbashe River (32.3033[degrees]S 29.0683[degrees]E), -200-220 m, sponge rubble, dredged NMDP, RV Meiring Naude, st'n Q1, 18.vii.1982 (V495); off Shixini Point (32.5267[degrees]S 28.8833[degrees]E), -500 m, muddy sand, coral rubble, dredged NMDP, RV Meiring Naude, st'n T17, 13.vii.1984 (C7060).
Distribution and habitat (Fig. 13): Known primarily from Zululand to south-western Transkei (Neill Peak [Cunge Hill] to Qolora River), with one additional record from central Mozambique; -110-500 m (living specimens -150-250 m). Evidently a rather rare species, dead shells have been dredged in relatively deep water on rather lifeless muddy substrata (e.g. the Tugela Bank), but the five living specimens available were all collected in sponge communities on hard substrata near the continental shelf break. The absence of records from off central KwaZulu-Natal probably reflects less extensive dredging on the outer continental shelf and upper slope in this area.
Remarks: Danilia textilis closely resembles D. discordata Vilvens & Heros, 2005 from Vanuatu, particularly in terms of the number and relative strength of the spiral cords. In both species there are eight spiral cords above (and including) the periphery (the ninth cord mentioned by Vilvens and Heros (2005) is subperipheral), of which the peripheral cord is somewhat larger and distinctly more spiniform than those above it. Both species also have a strong, squarish lower columella tooth and a colour pattern of brown, frequently axial, markings on a pale fawn ground. However, in D. discordata the six spiral cords between the subsutural and peripheral cords are wider and conspicuously flattened, as are the beads where the cords are crossed by the axial pliculae. In D. textilis the spiral cords are rounded and narrower than their intervals and retain raised rounded beads. The sculpture throughout is of a regular, open, oblique, net-like reticulation. In this regard it resembles D. insperata Beu & Climo, 1974 from New Zealand, but that species has more evenly rounded whorls with fewer spiral cords (six above and including periphery) and a less robust lower columella tooth. D. weberi Schepman, 1908 from Indonesia and the western Pacific has more numerous spiral cords due to intercalation of intermediaries, has beads that are angular rather than rounded and has a broader, more robust basal columella tooth. The recently described D. stratmanni Poppe, Tagaro & Dekker, 2006 from shallower water (-50-150 m) in the Philippines has much finer sculpture, more numerous spiral cords and lacks spiniform beads on the peripheral spiral cord. D. boucheti, the only other species of Danilia recorded from the south-western Indian Ocean, is considerably smaller and has an almost biangular whorl profile.
The precise form of the apertural dentition varies considerably between individuals and is presumably related to maturity. There is evidently further thickening of the columella lip and its teeth, even after the subterminal varix and flared outer lip margin have formed.
Genus Ascetostoma gen. n.
Etymology: From Greek asketos (curiously wrought or ornamented) and stoma (a mouth); in reference to the complex apertural features. Gender neuter.
Type species: Euchelus providentiae Melvill, 1909.
Diagnosis: Whorls rounded, suture indented and somewhat channelled; sculpture of finely beaded spiral cords; umbilicus present, lined and apically plugged with callus; columella pillar with a well developed, squarish, basal tooth and a smaller, more rounded, upper one; parietal region with spreading, glossy, translucent callus deposit bearing short oblique ridges; callus deposit extending around umbilical margin and joining basal lip; umbilical margin with several small denticles and one larger one in parietal region; interior of outer lip subterminally thickened and bearing ridge-like denticles arranged more or less in 2 rows; base of columella with pronounced U-shaped notch between basal columella tooth and first outer lip denticle; exterior of outer lip also with a low, broad, subterminal thickening; suture descending at aperture when mature.
Remarks: Similar to the Herpetopoma group of species, but characterised by the callus lined umbilical depression, spirally ridged parietal callus and denticulate umbilical margin. Danilia also has a callus lined depression in the umbilical region, but in members of that genus it is shallower and trough-like; they also possess a strong, well-defined, rib-like external varix behind the outer lip. Clypeostoma has at most a faint umbilical depression and a much more extensive inductural callus deposit. In Ascetostoma the sculpture is also more finely and closely beaded than in Clypeostoma.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||p. 381-420|
|Date:||Dec 1, 2012|
|Previous Article:||A new stenopterous genus of the tribe Gaetuliini Fennah (Hemiptera, Fulgoroidea, Tropiduchidae) from southern Africa--particular intercontinental...|
|Next Article:||A revision of the chilodontidae (Gastropoda: Vetigastropoda: Seguenzioidea) of Southern Africa and the south-western Indian ocean.|