Echinolittorina

Reid, David G., 2009, The genus Echinolittorina Habe, 1956 (Gastropoda: Littorinidae) in the western Atlantic Ocean 2184, Zootaxa 2184 (1), pp. 1-103 : 33-35

Subgenus Amerolittorina new subgenus

Type species: Trochus ziczac Gmelin, 1791 .

Etymology: America and Littorina , in reference to the New World occurrence of the majority of the assigned species.

Diagnosis: Shell with spiral ribs and grooves, sometimes a pseudumbilicus, never granulose or nodulose; colour white with spots, oblique axial or spiral lines of brown to black; aperture brown with white band at base and usually also at shoulder. Penis usually bifurcate with mamilliform gland and penial glandular disc, but either or both may be absent. Pallial oviduct: copulatory bursa opens at posterior or anterior end of straight section. Atlantic and eastern Pacific.

Remarks: There are no diagnostic morphological characters for this subgenus ( Reid 2002a). Most have a strong black-and-white shell colour pattern. There is usually not only a basal white band within the aperture, but also another white band at the shoulder, and this is rare elsewhere in the genus (only in some Indo-West Pacific species of the E. (Granulilittorina) leucosticta group; Reid 2007); however, this second band is not always visible in the small members of the E. (Amerolittorina) porcata group ( Reid 2002b).

In the first comprehensive molecular phylogeny of Echinolittorina , the species now classified in Amerolittorina fell into three clades of unresolved relationship: E. peruviana ; the E. porcata group; and the black-and-white shelled species including E. ziczac , E. aspera ( Philippi, 1846a) and E. punctata ( Gmelin, 1791) ( Williams & Reid 2004) . In the most recent analysis ( Williams & Duda 2008) these three clades form a monophyletic group with significant support ( Fig. 1 ). An argument could be made for recognizing the six members of the well-defined E. porcata clade as a separate subgenus, but the monophyly of the remainder has low support ( Fig. 1 ). Therefore all 24 members are retained in the single subgenus Amerolittorina .

In the western Atlantic there are five members of Amerolittorina . Together with the Brazilian E. (Lineolittorina) lineolata these all share superficially similar shells with moderately tall spires, weak to strong spiral grooves, and black-and-white shell coloration that often forms striking axial stripes. Consequently there has been a long history of taxonomic confusion among these six species, informally known as the ‘ Littorina ziczac species complex’ (T.V. Borkowski & M.R. Borkowski 1969). The first to be named was Trochus ziczac by Gmelin (1791), but even then there was ambiguity, because one of the figures referred to by Gmelin represents another species, probably E. interrupta . Emphasizing shell coloration, Lamarck (1822) included shells now classified in two genera, Littoraria tessellata and E. interrupta , in his Phasianella lineata , but although this mistake was soon pointed out ( Deshayes 1843) the error persisted (Philippi 1847). Paying closer attention to shell shape and sculpture, d’Orbigny (1841) distinguished three species in the ziczac complex— Littorina zigzag , L. carinata and L. lineata (respectively E. ziczac , E. angustior and a mixture of E. angustior and E. jamaicensis ). He also inadvertently named the Brazilian species Littorina lineolata d’Orbigny, 1840 (an error for L. lineata ). The number of named species within this complex proliferated in subsequent conchological monographs; Philippi (1846 b, 1847) accepted six, Mörch (1876) nine ( ziczac , riisei , jamaicensis , carinata , angustior , floccosa , lineolata , pusilla and an unnamed species), and Weinkauff (1878, 1882) seven. Curiously, Reeve (1857) included only two (both now recognized as E. ziczac ) in his monograph. The trend towards subdivision was comprehensively reversed by Tryon (1887), who included all four of the Caribbean members of the ziczac complex (plus E. lineolata and a Littoraria species from the Indian Ocean) under the name L. ziczac .

A broad species concept persisted well into the twentieth century. Martens (1900) recognized only two species, L. ziczac and L. lineata (the latter a combination of three species here recognized), and these were listed as subspecies by Johnson (1934). Both Bequaert (1943) and Abbott (1954a) recognized only a single, variable, black-and-white species in the western Atlantic from Florida and Texas to Brazil, L. ziczac . Bequaert (1943) did, however, make the suggestion that some of the perceived variability could be explained by sexual dimorphism, because male shells were said to be more strongly sculptured, with sharper peripheral keel and lower spire. This idea was maintained, despite the observation of distinct zonation patterns in the two forms in Venezuela, and of three ‘morphological forms’ in Florida ( Rodriguez 1959). It was only when dissection revealed that both sexes were present in each of Bequaert’s shell types that two species were demonstrated ( Abbott 1964; Fraenkel 1968), a conclusion supported by contrasting egg capsules ( Lewis 1960) and opercula ( Fraenkel 1968). Abbott (1964) retained the name L. ziczac for the taller form (a concept that has persisted ever since), whereas the smaller, keeled, sculptured form was identified as L. lineolata (a combination of E. angustior , E. jamaicensis and E. placida new species). Based on shells, he later distinguished a third species as L. floccosa (now E. jamaicensis ) ( Abbott 1968). Shortly afterwards, detailed studies of shell shape, sculpture and egg capsules revealed three species in Florida— L. ziczac , L. lineolata and L. lineata (corresponding to E. ziczac , E. jamaicensis plus E. lineolata , and E. angustior , respectively) (T.V. Borkowski & M.R. Borkowski 1969; T.V. Borkowski 1971). Working in Colombia, Curaçao and the Bahamas, and describing characters of the shells, egg capsules and radulae, Bandel (1974a) accepted these same three species (but renamed L. lineata as L. jamaicensis ), and also added a fourth unnamed species ( E. interrupta ). Meanwhile the Brazilian species was named L. ziczac brasiliensis by Vermeij & Porter (1971). However, T.V. Borkowski (1975) considered both Bandel’s (1974a) unnamed species and the Brazilian subspecies ( E. lineolata ) to fall within the range of variation in shell, radula and egg capsule shown by his L. lineata (i.e. E. angustior ).

The next development was the publication of a detailed monographic revision of all the western Atlantic species then assigned to Nodilittorina (i.e. the ziczac complex, the nodulose species of Echinolittorina , and Tectarius antonii ) by Bandel & Kadolsky (1982). This influential work revised nomenclature, reviewed the available published information on egg capsules and penes, and illustrated shells and radulae from throughout the region. The correct names and distributions of N. ziczac , N. angustior , N. interrupta and the Brazilian N. lineolata were all established. However, such was the importance ascribed to radular tooth form that three species were discriminated ( N. riisei , N. glaucocincta , N. mordax ) within the species here classified as E. jamaicensis . Meanwhile, other authors were emphasizing the penis as a specific mate recognition character in littorinids. This character led Reid (1989) to reunite these three under the name N. riisei , leaving a total of five species within the ziczac complex— N. ziczac , N. angustior , N. riisei , N. interrupta and N. lineolata . This conclusion was restated by Reid (2002a), in the light of new information about radular plasticity in littorinids ( Padilla 1998; Reid & Mak 1999), while pointing out that genetic data were required to resolve the issue conclusively. The first large-scale molecular phylogenetic study of Echinolittorina supported these five taxa ( Williams & Reid 2004). Recently, genetic data have shown that the form of E. interrupta from the Gulf of Mexico is genetically distinct (Reid, in Williams & Duda 2008), and this is here described as E. placida new species. Only now are genetic data available from animals of known radular type, showing that E. jamaicensis is indeed a single taxon with pronounced radular variation.

Figures

FIGURE 1. Molecular phylogeny of all 60 known species of worldwide Echinolittorina (after Williams & Duda 2008), with geographical distributions and subgeneric groups. This is a chronogram, produced by BEAST analysis of sequence data from 28S, 12S and COI genes, which employs Bayesian inference and an uncorrelated, relaxed, log-normal clock; error bars are 95% highest posterior density intervals, and numbers are Bayesian posterior probabilities (see Williams & Duda 2008 for methods). The time calibration of the phylogeny may be underestimated; see Remarks on subgenus Echinolittorina.