Platyla jordai, Altaba, Cristian R., 2013

Altaba, Cristian R., 2013, A new species of Platyla (Mollusca: Gastropoda: Aciculidae) fills a biogeographic gap in the Mediterranean, Zootaxa 3683 (1), pp. 87-91 : 87-91

Platyla jordai sp. nov.

Figures 1 , 2.

Holotype. MZB 2013 -0005 (fig. 1), fresh, 3.2 mm shell height (SH), collected among water-borne debris close to the seashore mouth of the temporary stream Torrent de s’Aladernar, on the NW side of La Victòria peninsula on the northern coast of Mallorca (39º 52’ 12.93” N, 3º 9’ 25.67” E; UTM 31T EE 3845 3637), 14 February 2013.

Paratypes. Three older shells collected with holotype, 3.2 to 3.3 mm SH (Paratype 1 MZB 2013-1926; paratypes 2 and 3 CRA-13803). One shell, also rather old, 3.5 mm SH, found under palmetto ( Chamaerops humilis ) shrubs at 280 m altitude, in the steep canyon walls at the uppermost reach of the dry seasonal mountain brook tributary to the left of Torrent de s’Aladernar, on the N side of Talaia d’Alcúdia, the area's highest peak (39º 52’ 2.39” N, 3º 10’ 9.87” E; UTM 31T EE 4642 1310), 29 May 2009 (paratype 4, CRA-13711, fig. 2)

Diagnosis. A slender, pale yellowish brown except for the nearly ivory white aperture, medium-sized (for the genus) Platyla species with about six progressively flattened whorls, a deep suture underlined by a narrow but conspicuously raised and delimited keel, strong parietal callus with inconspicuous angular denticle, and a strongly raised, posteriorly concave cervical crest taller than the reflected, ortho- to opisthocline peristome.

Description. Shell narrow conical with wide blunt apex, pale yellowish brown, semitranslucent, 3.2–3.5 mm high, 1.2–1.3 mm wide. Spire with 5½– 6 1/4 whorls, initially well inflated but progressively flattened, especially in middle part. Last whorl slightly less than half shell height (46–48%) and noticeably flattened laterally, giving appearance of a blunt angle as it joins more strongly curved area around closed umbilicus. Shell surface smooth, with general waxy shine due to almost imperceptible growth lines, rather regularly spaced (about every 0.01 mm in penultimate whorl). Texture only interrupted by very few (six) radial growth-rest lines (about one per whorl). Suture deep, indented, underlined by a narrow, low, but conspicuous subsutural keel, starting as a weak structure at the onset of teleoconch and gradually increasing in height until penultimate whorl; in the last whorl shallower. Subsutural keel clearly delimited below by thin incised spiral furrow, briefly overhanging each whorl; thus suture has stepped shape and whorls appear slightly imbricated. Protoconch with 1 1/4 whorl, initially small but greatly increasing, smooth with silky sheen. Aperture (8.7–1.0 mm high, 0.7–0.8 mm wide) wide oval, pointed adapically where it forms a small narrow embayment. In lateral view, aperture rises slightly over penultimate whorl, without forming any sinulus. Peristome little thickened and conspicuously reflected, in lateral view slightly sinuous with upper half ortho- to opisthocline. Parietal callus moderately strong, extending over the columella, joining smoothly onto the peristome at the top of its columellar edge, enlarged adapically up to an inconspicuous angular denticle. Umbilicus completely covered by thin, concave callus, adapically narrowing over the parietal callus until reaching the middle of it, and laterally joining peristome through a smooth, well curved bend throughout its columellar margin. Behind aperture and surrounding it on the outside is conspicuous cervical crest, a rather wide (1/5 of shell diameter) semitorus belt, its shape following the curvature of the peristome. Peristome slightly thickened, preceded by narrow but smoothly curved annular depression, rising suddenly but smoothly into very strong cervical crest. Highest part of cervical crest near its posterior third, bulging as rounded raised annular keel. Rear side of cervical crest turns back abruptly, thus making a concave posterior ring. In frontal view, cervical crest protrudes (a bit more so abapically) around peristome, which thus appears as doubled. At suture, cervical crest terminates in nearly semispherical bulge, but reaching columella it submerges under umbilical callus (which is continuous with peristome in this area). No noticeable reflection of cervical crest structure inside the aperture. Peristome and cervical crest, and less so umbilical and parietal callus, stand out through their glossy enamel white color.

Habitat. The narrow, steep canyon inhabited by P. j o rd a i is excavated in karstified Upper Jurassic marls. The Aleppo pine ( Pinus halepensis ) forest cover is much degraded through the combined effect of repeated wildfires, uncontrolled feral goats and illegal activities aimed at favoring hunted wildlife. In the peak area it is completely substituted by a mosaic of Mediterranean garrigue (Cneoro tricocci-Ceratonietum siliquae community) dominated by the shrubs Pistacia lentiscus , Ampelodesmos mauretanicus and Chamaerops humilis , and rich in Balearic endemics such as Hypericum balearicum . Inside this northward canyon, rupicolous plants such as Micromeria filiformis constituting the Saturejo filiformis-Asplenietum petrarchae community attest to a more humid microclimate. It is likely that P. j o rd a i lives in crevices or even underground in this small refuge. In spite of the fact that this area is included in the Natura 2000 network of European areas for biodiversity conservation ( Evans 2012), threats to the native fauna have not been controlled. The rarity, extremely narrow range (likely less than 1 Km2) and seriously degraded habitat of P. j o rd a i suggest it may be critically endangered.

Comparisons. Platyla jordai differs from all other congeners by its combination of 1) flattened, gently increasing last whorls; 2) enlarged and posteriorly concave cervical crest; 3) prominent subsutural keel; 4) sinuous, opisthocline peristome; and 5) very light brown color. Out of the 29 Recent Platyla species, uneven anatomical data are available for only three of them, thus precluding a meaningful phylogenetic analysis. However, since its earliest appearance in the Eocene this genus has experienced a remarkable stasis. Thus, structures in the shell may prove useful to tentatively infer relationships within this obscure group.

P. j o rd a i is most similar to P. subdiaphana (Bivona, 1839) , a very rare species restricted to a limestone massif in NW Sicily. However, in this Tyrrhenian species the suture is shallower, the peristome is prosocline and regularly arched in lateral view, the aperture is bluntly trapezoidal, the umbilical callus is thinner, and the cervical crest is much lower and not concave behind. Another similar species is P. gracilis (Clessin, 1877) , ranging throughout the mountain ranges surrounding the Adriatic Sea in Italy, Austria, Slovenia, Croatia and Greece. This continental species can be easily distinguished by its decidedly prosocline, evenly curved peristome, its more cylindrical shape (most clearly in the last three whorls), and reddish to light brown color; it also differs in its small but conspicuous angular denticle, thinner umbilical callus, peristome running more smoothly into the columellar callus, and cervical crest not (or barely) wider than the thickened peristome. Both P. subdiaphana and P. gracilis share with P. j o rd a i a slender conical shape, flattened last whorls, and deep suture underlined by a narrow but prominent, well-defined subsutural keel. These traits are probably synapomorphies, although it is unclear whether they might be developmentally linked. Also similar is P. callostoma (Clessin, 1911) from the eastern Pyrenees and central Cantabrian Mountains, but it differs in lacking a subsutural keel or line, and having a clearly recognizable angular denticle, a small sinulus, and a wider, not so prominent, more rounded cervical crest.

Three other extant species from southern Europe recall P. jordai by their enlarged cervical crest that frames the peristome in frontal view. P. pezzolii Boeters, Gittenberger & Subai, 1989 , apparently endemic to a small area in NE Italy, has rounder, fewer whorls, a wider last whorl, an almost orthocline peristome, a conspicuous angular denticle, a wider, stronger posterior part of the cervical crest, and a feeble subsutural keel. P. elisabethae (Pintér & Szigethy, 1973) from southern Dalmatia is larger, more slender and light brown, has no trace of subsutural keel or furrow, has a narrower, orthocline aperture with a small adapical sinulus, and its cervical crest is less prominent, narrower, keeled and wider at its posterior end. P. talentii Bodon & Cianfanelli, 2008 from the southernmost Italian peninsula is quite similar to P. elisabethae , differing in its even less prominent, slightly broader cervical crest, followed by a smaller groove. In addition, P. fa l kn er i Boeters, Gittenberger & Subai, 1989 from the Miocene of Poland is larger and clearly more tumid overall, has less rounded upper whorls, its peristome is arched in lateral view and runs more smoothly into the columellar callus, and its cervical crest is flaring on top of the lower end of the columella. The disparity of these four species and P. jordai suggests that an enlarged, bulging cervical crest is a homoplasious trait with no phylogenetic value, probably related to adaptation to habitats with seasonal drought.

Etymology. Dedicated to the late Cristòfol Jordà Barreras, in recognition for his lifelong dedication to nature study and conservation activities in the Catalan Countries. The species name is a masculine genitive adjective.

Remarks. It may appear weak science to describe a new species based on very few empty shells. There are, however, two reasons to do so. The first one is theoretical —although sample size is certainly important, its relevance does depend on the range of interspecific variation. In other words, the comparative or explicit classification criteria of taxonomists, intended to predict an underlying phylogeny that is a highly structured object, may be seen as estimators of the cost of incorrect classification, and thus are likely a case of Cover-Hart loss function. In this large class of functions, a small sample turns out to be empirically close to indefinitely larger samples; even a sample size of just one contains at least half the information in an infinite sample ( Gneiting 2012). In taxonomy, the empirical difference between small and large samples depends on the information content; i.e., the number, discreteness and independence of diagnostic traits allowing correct identification. This does not affect the preceding inequality; instead it means that distinctive taxa may be correctly identified and described on the basis of very few, perhaps only one specimen. Essentially, what matters is the strength of inferences made from the available materials.

The second reason is empirical—the Aciculidae are extremely rare but quite distinctive. Most species are known from a handful of specimens, and the vast majority of them have never been collected alive. However, in contrast to other land snail shells where subtle differences in shape require statistics (e.g., Altaba 2007) or may prevent the description of new species (e.g., Paul & Altaba 1992), aciculid shells do possess rather constant diagnostic features. Actually, about 10 % of aciculids have been described on the basis of single shells (three in the monographic revision by Boeters et al. 1989), because they are distinctive enough, lying way out of any others' variation.

The probable closest relatives of P. j o rd a i live in the central Mediterranean region. Their dispersal abilities are exceedingly limited, so it is likely that the current distributions of these species are largely the result of vicariant events preceding the latest Miocene. This appears to be the case in non-marine taxa with adequate fossil records and that are unlikely to have been translocated by human activities, including caenogastropods such as freshwater melanopsids ( Altaba 1998) and most of the terrestrial pomatiasids of the genus Tudorella Fischer, 1885 , endemic to the Western Mediterranean ( Paul & Altaba 1992; Pfenninger et al. 2010), as well as other taxa such as the anuran Discoglossus ( Pabijan et al. 2012) . Although there are no living melanopsids in Sardinia, fossils show that they occurred there until the Pleistocene. Likewise, although the Tudorella living locally in Sardinia belong to a species having experienced much anthropogenic dispersal ( Jesse et al. 2011), there are Plio-Pleistocene fossils closely similar to the Balearic endemic T. ferruginea (Lamarck, 1810) . This faunistic similarity is a reflection of geological history, because Sardinia was the last terrane to remain united with the Balearics until ca. 30 Ma (except for a brief connection to southeastern Iberia around 14 Ma; Altaba 1997, submitted). Aciculids are indeed represented in Sardinia ( Cianfanelli et al. 2000), although by two species apparently not closely related to P. j o rd a i. The new species of Platyla from the largest of the Balearic Islands fills a biogeographic gap and supports the role of Neogene tectonics in the diversification of the Mediterranean fauna.

Figures

FIGURE 1. Platyla jordai, sp. nov. Holotype (3.2 mm).