Page 2 - Pardi&al_2006

Posidonia oceanica epiphytes of Sicily Pardi, Piazzi, Balata, Papi, Cinelli & Benedetti-Cecchi

et al. 2005). Although ESL have been investigated repeat- scales (e.g. Terlizzi et al. 2005), although this is not a
edly in space and time (Casola et al. 1987; Mazzella et al. general phenomenon (Chapman et al. 1995).
1989; Kendrick & Burt 1997; Reyes et al. 1998; Trautman
& Borowitzka 1999; Vanderklift & Lavery 2000; Lavery & This paper examines the general hypothesis that variab-
Vanderklift 2002; Esposito et al. 2003; Saunders et al. ility in ESL occurs at different spatial scales. This hypo-
2003), a detailed quantification of natural variability thesis was tested with a multifactorial sampling design
over a broad range of spatial scales, including differences that generated quantitative data on variability among: (i)
between the small islands and continental coasts, has geographical areas 100s of kilometres apart, (ii) habitats
never been undertaken. (islands and mainland), (iii) meadows 10s of kilometres
apart, (iv) sites 100s of metres apart, (v) plots 2–5 m
Numerous ecological processes can be responsible for apart and (vi) shoots 10s of centimetres apart. Sicily is
geographical patterns of variability in P. oceanica ESL. particularly suited for this purpose because it is surroun-
Differences in hydrodynamic conditions, exposure to ded by small islands, it separates different Mediterranean
Atlantic currents and gradients in water temperature, may basins and it is large enough to be considered a continen-
generate latitudinal gradients and differences among the tal habitat.
basins in the Mediterranean Sea (Margalef 1985). Posi-
donia oceanica shows genetic disjunctions among the Material and Methods
meadows of the different Mediterranean regions (Procac- This study was carried out between July and September
cini & Mazzella 1998), confirming that different popula- 2000 in three geographical areas around Sicily, the nor-
tions thrive in different geographical areas. Differences thern coast (Tyrrhenian Sea), the southern coast (Sicily
could be expected on this basis for epiphytic assemblages Strait) and the western coast (at the boundary between
also. A previous study showed differences in the structure the two basins) (Fig. 1). Two habitats were sampled in
of epiphytic assemblages of rhizomes among the Mediter- each geographical area, represented by islands and by
ranean regions (Piazzi et al. 2002), while no comparable locations on the mainland, with three replicate meadows
information is available to date on ESL. in each combination of geographical area · habitat. Five
sites were randomly selected in each meadow at 10 m
Assemblages of islands can differ from those inhabiting depth and five plots were randomly selected in each site.
locations on the mainland for a variety of reasons, inclu- Five vertical (orthotropous) shoots 10s of centimetres
ding isolation from important sources of colonists, type apart were collected in each plot by scuba diving and
of habitats available for colonization and exposure to fixed in 4% formalin seawater. Sampling was completed
human disturbance. These issues have been investigated within 1 month.
mostly for terrestrial biota (Whittaker 1998), while a few
studies have been carried out on marine assemblages (e.g. Epiphytes were examined on the more external leaf of
Piazzi et al. 2004b; Rindi & Guiry 2004). In the Mediter- each shoot (the oldest one), to ensure that the analysis
ranean Sea, Benedetti-Cecchi et al. (2003) have highligh- focused on a mature assemblage (Reyes et al. 1998;
ted differences in shallow rocky shore assemblages
between two islands of the Tuscan Archipelago and two Fig. 1. Map of Sicily. Symbols indicate locations of the studied Posi-
locations on the mainland. On the contrary, no signifi- donia oceanica meadows. Grey pins: islands; black pins: mainland
cant differences have been detected between deep rocky coasts.
assemblages of islands and continental coasts in the same
geographic area (Piazzi et al. 2004c).

In addition to large-scale patterns of variability,
small-scale spatial heterogeneity is also a prominent fea-
ture of benthic assemblages (Fraschetti et al. 2005). Sev-
eral processes can account for this type of variability,
including vagaries in the dispersal, supply and settle-
ment of propagules, post-recruitment biological proces-
ses (Coleman 2002 and references therein) and
disturbances (Airoldi & Cinelli 1997). In addition to
providing a better understanding of relevant ecological
processes, proper quantification of small-scale spatial
variability in assemblages is also important for the
detection of human impacts. Some studies have shown
that certain types of anthropogenic disturbances may
alter natural patterns of variability over small spatial

398 Marine Ecology 27 (2006) 397–403 ª 2006 The Authors. Journal compilation ª 2006 Blackwell Publishing Ltd