A SYSTEMATIC REVISION OF THE VITRINID SEMISLUGS OF THE AZORES (GASTROP0DA:PULMONATA) PETER B. MORDAN' AND ANTONIO M.F. MART INS^ '~e~artmentof Zoology, Cromwell Road, London S W7 5BD, UK E-mail: [email protected]. uk. 2~epartmentof Biology, University of the Azores, Ponta Delgada, S6o Miguel, Azores, Portugal. E-mail: [email protected] (Received 19 November 2000; accepted 9 February 2001) ABSTRACT A total of six species, all endemic, are recognised in a comprehensive taxonomic revision of the vitrinid semi-slugs of the Azores. All are referred to the genus Plutonia on the basis of the disposition of the penial retractor muscle. One species, P. brevispira, is highly unusual in lacking a vaginal stimulator and having a large, elaborate penis which would appear to function as an intromittant organ during copulation. INTRODUCTION 1987; Morales, Ibaiiez & Alonso, 1988; Valido, Alonso & Ibaiiez, 1990; Valido, Groh, Ibaiiez & Alonso, 1993) The principal island groups of the eastern north- and Madeira (Groh & Hemmen, 1986), there is no Atlantic, the Azores, the Canaries and Madeira, pro- comprehensive modern account of the Vitrinidae of vide an outstanding opportunity to study a rich variety the Azores. The present paper addresses this gap, and of evolutionary radiations in areas relatively close to presents a revision of the vitrinid semi-slugs of the continental Europe. Together with the Cape Verdes Azores based on extensive material collected over the and the Selvagens they form a region called Macaro- last thirty years from throughout the archipelago. nesia, though the biogeographic utility of such a grouping has been questioned (Beyhl, Mies & Ohm, 1995). These archipelagos all exhibit very high levels of THE HISTORY OF THE AZOREAN endemism, and their biotas have a relictual appear- VITRINIDAE ance, with those of Madeira and the Azores showing strong affinities with the Tertiary of western Europe The first description of vitrinids from the Azores and the Mediterranean (WaldCn, 1963; 1983). The was published by Morelet (1860) in his monumental Canaries, on the other hand, show an obvious relation- Histoire Naturelle des A~ores,following a trip in 1857 ship with north-west Africa, and the Cape Verdes with in which all the islands of the archipelago except tropical Africa (WaldCn, 1983). Surprisingly, their SZo Jorge were visited. He described seven species in affinities with North America are insignificant (WaldCn, Vitrina Draparnaud, 1801, all new, from the islands of 1963; 1983). Santa Maria, SZo Miguel, Terceira and Flores: pelagica The oldest of the nine islands which comprise the Morelet, 1860, brumalis Morelet, 1860, laxata Morelet, Azores, Santa Maria, is thought to be aged around 1860, brevispira Morelet, 1860, mollis Morelet, 1860, 8 million years (Serralheiro & Madeira, 1993). As such, jinitima Morelet, 1860 and angulosa Morelet, 1860. the Azores archipelago is younger than either the Morelet's system for the Azorean Vitrina remained Canaries or Madeira, and this, together with its greater unchallenged until 1889 when Simroth (1 889:5) grouped distance from the continent, have been suggested as all seven Azorean species, together with a further three reasons why species diversity there is relatively lower from Madeira (ruivensis Gould, 1848, marcida Gould, (WaldCn, 1984). Nevertheless, there are over one hun- 1848, and nitida Gould, 1848) and five from the dred species of land mollusc recorded from the Azores, Canaries (lamarcki FCrussac, 1821, canariensis Mous- approximately half of which are endemic. Major son, 1872, reticulata Mousson, 1872, latebasis Mous- endemic evolutionary radiations have taken place in son, 1872 and blauneri Shuttleworth, 1852), under the the pulmonate families Helicidae, Vitrinidae, Bulimin- name Vitrinapelagica. He thus considered there to be a idae, Zonitidae and Vitrinidae. single species of vitrinid semislug throughout the Whilst there have been recent systematic revisions of Atlantic islands. Hesse (1923) included all the Atlantic the Vitrinidae of both the Canary Islands (Ibaiiez, vitrinid semislugs (together with the Arabian ones) in Morales & Alonso, 1987; Alonso, Ibaiiez & Morales, his new genus Insulivitrina, but recognised distinct J. Moll. Stud. (2001), 67,343-368 O The Malacological Society of London 2001 PETER B. MORDAN & ANTONIO M.F. MARTINS species for each of the three main archipelagos; the of Phenacolimax (e.g. Zilch, 1959; Backhuys, 1975). Azorean species were all synonymised underpelagica. Indeed, excepting Plutonia and Guerrina Odhner, 1954, Six years later, Hoffmann (1929) revised the all the vitrinid semislugs of the northern Atlantic Atlantic-island vitrinids. His solution was to reduce islands were originally united by Hesse (1923) in his Morelet's seven original Azorean species of Vitrina to new genus Insulivitrina, with lamarcki FCrussac from three: pelagica (including laxata, as possibly a local the Canary Islands as the type species. Hesse also form); brumalis (including brevispira andjinitima, both originally included the Arabian species arabica Thiele, thought to be juvenile forms); and mollis (including 1910, but this was subsequently removed from Insu- angulosa, which he also considered to be juvenile). livitrina to a new genus, Arabivitrina, by Thiele (193 1). Hoffmann's arrangement was subsequently taken up However, there have been problems with the precise by Fischer-Piette (1946). diagnosis of Insulivitrina, and defininition of its taxo- The most recent taxonomic treatment of the vitrinids nomic limits. Recently two cladistic studies have clari- of the Azores is that of Backhuys (1975). He reduced fied this situation considerably (Alonso, Valido, Groh Morelet's seven species of Vitrina to two, one of which & Ibanez, 2000; Hausdorf, in press). he considered to be a species complex, both placed Alonso et al. (2000) addressed the problem of rela- within Insulivitrina Hesse, 1923, which he treated as a tionships within the vitrinid subfamily Phenacolimaci- subgenus of Phenacolimax Stabile, 1859. His pelagica nae of Schileyko (1986). They united the Macaronesian was restricted to Santa Maria and Sgo Miguel, and vitrinids in a single genus Plutonia, comprising five sub- included laxata. The brumalis species complex incor- genera: Plutonia sensu stricto, Guerrina, Insulivitrina, porated the remaining taxa, namely mollis, jinitima, Madeirovitrina Groh and Hemmen, 1986, and a new brevispira, and angulosa. He considered pigmentation subgenus, Canarivitrina. Plutonia sensu lato was to be a valuable diagnostic feature in separating these defined by the course of penial retractor muscle which, two taxa. Backhuys (1975:143), however emphasised when present, 'loops round the optic nerve, passing that his conclusions were provisional, and stated that below the right ommatophore retractor, then running 'only future research can bring the problem to a defin- backwards (left of the same right ommatophore retrac- ite solution'. tor)' before inserting on the diaphragm. This is the con- In addition to the seven Vitrina species, Morelet dition in all those Azorean vitrinids having a penial (1 860) described a new species of vitrinid slug under the retractor. In the analysis of Alonso et al. (2000) Pluto- name Viquesnelia atlantica Morelet, 1960. Viquesnelia nia s.1. appeared in an unresolved trichotomy with Deshayes, 1857 was based on a Romanian fossil shell. Arabivitrina and Oligolimax Fischer, 1878 (syn:Gallan- This attribution was shown to be erroneous by Stabile dia Bourguignat, 1880) which, together with Phena- (1864) who erected a new monotypic genus, Plutonia, colimax, comprised their subfamily Plutoniainae (syn: for atlantica. Zilch (1959) treated it as a subgenus of Phenacolimacinae). Phenacolimax. Shileyko (1986) included it as a full Hausdorf (in press) undertook a phylogenetic analy- genus in his subfamily Phenacolimacinae, whereas sis of the entire family Vitrinidae. He followed Alonso Vaught (1989) placed it in its own subfamily Pluto- et al. (2000) in uniting all Macaronesian vitrinids under niinae, both within the Vitrinidae. Plutonia. Plutonia was found to be the sister group of Plutonia atlantica has recently been comprehensively Oligolimax in his analysis, based on the course of the redescribed by Wiktor & Backeljau (1995) who consid- right ommatophoral retractor muscle, which 'does not ered it to be sufficiently distinct from all other vitrinids pass between the penis and the female genitalia'. Haus- to be placed in its own monotypic family Plutoniidae dorf recommended against the use of subfamily groups (formerly proposed as a subfamily of the Limacidae by in the Vitrinidae. Cockerell, 1893). Although its genital anatomy is Thus it would appear that the Macaronesian vitrin- remarkably similar to that of Insulivitrina, Wiktor and ids represent a monophyletic assemblage defined by Backeljau believed the the lack of a penial retractor the course of the penial retractor muscle, and should muscle, first noted by Simroth (1891:228), was suffi- correctly be grouped together in the genus Plutonia. cient to justify the new family. We will not attempt any Alonso et al. (2000) included all the Azorean semislugs, further description of P. atlantica, which we believe to and some from the Canaries, in the subgenus Insulivit- be a vitrinid. rina. However, unlike Madeirovitrina and Canari- vitrina, Insulivitrina does not appear to be supported by The validity of Insulivitrina any autapomorphic characters in their analysis. Haus- Since Hesse (1923), the vitrinid semi-slugs