Zoogeographical relationships of the littoral ascidiofauna around the Antarctic Peninsula , in the Scotia Arc and in the Magellan region *

The Antarctic Province represents one of the most clearly defined biogeographic divisions in the world. Its northern limit is defined by the Antarctic Convergence (Ekman, 1967, Hedgpeth, 1969; Briggs, 1974; Knox, 1994; Crame, 1999). However, there are several hypotheses about the subdivisions of the Antarctic region, and their relationships with the Scotia Arc and the Magellan region. Ekman (1967) and Briggs (1974) consider two main subdivisions (subregions of Ekman, and provinces sensu Briggs). The latter proposed: i) a high Antarctic subregion or South Polar Province containing the continental coastal area and adjoining islands (including the South Shetland, South Orkney and South Sandwich archipelagos); and ii) a low Antarctic subregion or South Georgia Province containing the SCI. MAR., 69 (Suppl. 2): 215-223 SCIENTIA MARINA 2005


INTRODUCTION
The Antarctic Province represents one of the most clearly defined biogeographic divisions in the world.Its northern limit is defined by the Antarctic Convergence (Ekman, 1967, Hedgpeth, 1969;Briggs, 1974;Knox, 1994;Crame, 1999).However, there are several hypotheses about the subdivisions *Received April 26, 2004. Accepted March 10, 2005. of the Antarctic region, and their relationships with the Scotia Arc and the Magellan region.Ekman (1967) and Briggs (1974) consider two main subdivisions (subregions of Ekman, and provinces sensu Briggs).The latter proposed: i) a high Antarctic subregion or South Polar Province containing the continental coastal area and adjoining islands (including the South Shetland, South Orkney and South Sandwich archipelagos); and ii) a low Antarctic subregion or South Georgia Province containing the South Georgia Archipelago and the Shag Rock Bank.Dell (1972) opposed this hypothesis on the relationships of the Antarctic benthic fauna, emphasising that there is little evidence to warrant biogeographic subdivisions.Recently, Knox (1994), following Hedgpeth's scheme (1969), considered the Subantarctic region to be divided into the Magellan and Kerguelen Provinces or subregions; the Antarctic region in the western Antarctic, with the Antarctic Peninsula and the Scotia Arc (with the South Georgia district); and the Continental Antarctic.Longhurst (1998), basing his conclusions on plankton ecology, proposed that the Scotia Arc and the South Georgia Archipelago be included in the Austral Polar Province (Antarctic continent and adjoining islands).Arntz (1999) stressed the necessity of more effort in taxonomy to improve knowledge on floral and faunal relationships between the Antarctic and Magellan regions.
The Antarctic ascidiofauna has been well studied (see references in Van Name, 1945;Millar, 1960;Kott, 1969;Monniot and Monniot, 1983), and recent studies have focused on the South Shetland Archipelago and Weddell Sea (Rauschert, 1991;Monniot and Monniot, 1994;Lujan and Ramos-Espla, 1996;Tatian et al, 1998;Varela and Ramos-Espla, 2003).Also, the Scotia Arc and the Magellan region have been the subject of recent ascidian studies (Sanamyan and Schories, 2003).(Tatian and Sahade in Arntz and Brey, 2003).This study explores the biogeographical relationships of the Antarctic Peninsula and Scotia Arc with the Antarctic continent, Magellan region and the Subantarctic islands from the point of view of littoral ascidiofauna.

MATERIAL AND METHODS
Ascidians were collected (Table 1) on three Spanish Antarctic cruises: Ant-8611 (Scotia Arc and Elephant Island) with the fisheries vessels 'Pescapuerta IV and 'Nuevo Alcocero'; and Deception and Trinity Island,and Bransfield Strait) with the RV 'Hesperides'.The samples were taken from the continental shelf and from upper slope locations, between 10 and 600 m depth, by SCUBA diving and remote sampling gears (Agassiz trawl, rocky and anchor dredges and Van Veen grab).Macrobenthos (0 > 1mm) was sorted by sieve and the ascidians were separated, anaesthetised in menthol, and fixed/preserved in neutral formalin (10% in sea water with borax).
With regard to the ascidian distribution, we considered the following zones: i) the Antarctic region separating the Antarctic Peninsula (Graham and Palmer Lands, and adjoining islands, including the South Shetland Archipelago) from the Antarctic Continent; ii) the Scotia Arc, including the South Orkney, South Sandwich and South Georgia archipelagos; iii) the Magellan region, south of a line from Chiloe Island (at 42°S) to the mouth of the Rio de la Plata (at 35°S); and iv) the Subantarctic islands (Bouvet, Prince Edwards, Marion, Crozet, Kerguelen, Heard, MacDonald, Macquarie).For the purpose of distributional analysis we considered the shore or littoral ascidiofauna extending from the surface down to 500 m depth.This arbitrary depth was selected because the Antarctic ascidian fauna is rather uniform over this range (Briggs, 1974) and some deep Antarctic ascidian species such as Cibacapsa gulosa reach 500-600 m depth (Monniot andMonniot, 1983, 1994).

Zoogeographical affinities
The MDS analysis between zones (Fig. 2) for the 172 species listed in Table 2 shows the proximity between the Antarctic continent and the Antarctic Peninsula, and a clear separation from the Scotia Arc islands.The South Orkney Archipelago has a similar component to the Antarctic groups, and the ascidian fauna of the South Georgia and South Sandwich archipelagos is intermediate between that of the Antarctic, the Magellan region and the rest of the Subantarctic islands.
The comparison at species level of the main families (with percentages > 5%: Polyclinidae, Styelidae, Molgulidae, Pyuridae, Polycitoridae, Didemnidae, and Agneziidae; Table 3) confirms the differences between the archipelagos of the Scotia Arc.The South Orkney islands show similar percentages of Molgulidae (12%), Styelidae (12%), Polycitoridae (9%) and Didemnidae (6%) to the The biogeographical position of the ascidian fauna in the Scotia Arc is controversial.Reporting on this taxon, Kott (1969Kott ( , 1969a) ) included the Antarctic Peninsula (with the Bellingshausen Sea) and the Scotia Arc in the South Georgia Province (or western Antarctic subregion, cf.Knox, 1994).However, Millar (1971) and Monniot and Monniot (1983) consider the South Georgia Archipelago to occupy an intermediate position between the Antarctic and Magellan regions, and to be a separate province, whereas the South Orkney and South Sandwich islands belong to the Antarctic region, in accordance with Ekman (1967) and Briggs (1974).
The present study supports the inclusion of the Antarctic Peninsula and adjacent islands in the Antarctic Province and the division of the Scotia Arc into two sectors: the South Orkney Archipelago, more related to the Antarctic Province, and the South Georgia Archipelago, a separate area, intermediate between the Antarctic Province and the Magellan region.The position of the South Sandwich islands is doubtful.It is related on the one hand to the South Georgia Archipelago and the Subantarctic islands, and on the other hand to the Antarctic Province.Nevertheless, the Scotia Arc has been poorly sampled, and much more sampling effort (as the LAMPOS cruise) must be carried out to elucidate its biogeographical relationships.participated in these expeditions.We also thank three anonymous referees who improved this paper.These cruises were carried out under the auspices of the Spanish Ministry of Education and Science (DGICYT ANT93-0996 and DGICYT ANT94-1161 projects).
region, the Scotia Arc and/or the Subantarctic islands.