Page 5 - Russo_Terlizzi_1998

Structural patterns in the Posidonia oceanica beds rnolluscs

sample-points were more closely arranged than the inter-site ical aspect is of particular interest and may be a promising fleld
ones. As a conseguence, the hypothesis that the 'inter-site' vari- of investigation. In fact, up to now most malacological research
ability of the malacological associations is the effect of a bias activity has focused on investigating the historical aspects of
introduced by the sampling method should be excluded. The single species, such as their evolution and biogeography, with
large-scale variability of the malacofauna of the seagrass beds little or no attention paid to the study of the history of assem-
seems to be so intrinsic to the associations, and the structural blages they form.
'signal' so strong, that they are recognisable independently of Last, but not least, the large variability in the composition of
the sampling method utilised. Consequently, the strong struc- the assemblages described in the present paper implies a great
tural differences between the Mediterranean sites seem to be difflculty in recognising a unique and well-defined malacologi-
related mainly to the edaphic factors characterizing each sea- cal taxocoene for the Posidonia beds. Species co-occurring in the
grass bed. The density of the prairie seems to have an important different beds consti tute a very low percentage of the total.
role in determining the strong srructural differentiation of the Therefore, the present analysis of one of the main faunistic
malacological assemblage of the Medas, with respect to all other components of the vagile fauna does not agree with the model
sites. In this site, in fact, the air-lift was far more efficient than of PÉRÈS & PICARD (1964), w h ere the Medi terranean Posidonic1
the hand-net, as indicated by the high presence of molluscs beds are considered as a single biocoenotic and bionomic unit
more strictly related to the substrate, such as bivalves and, (HP). Present a-ecologica! data agree better with the 'poly-bio-
among gastropods, the families of Alvaniidae and Marg inelli- coeni tic' hypothesis (BIANCHI et al., 1989), although b-ecology
dae. Apart from the vegetal cover of the substrate (e.g. density needs to be focused in the future in order to gain a better under-
of plant shoots), the guality of the substrate itself, its exposure standing of the ecologica! processes underlying the structural
to water movements and the related sedimentation rates may be patterns described.
among the main edaphic factors affecting the quali-quantitative
composition of the malacological associations and determining CONCLUSIONS
the variety of functional adaptation. In fact, the dominance of - The suction-sampler and the hand-towed net discriminate the
herbivore-deposit feeders (e.g. Bittùtm spp.) is typical of sea- intra-site variability, as they sample different parts of the habi-
grasses under conditions of low environmental energy, while, on tat produced by the seagrass. This gives rise to 'method-depen-
the contrary, brusher-herbivores (e.g.} u;ubin/IS spp. and Tricolia dent' patterns which correspond to a real 'topological' stratiflca-
spp.), are abundant in conditions of high water movement. tion of the malacological association within the seagrasses.
In addition, the information collected suggests the presence Therefore, the study of this type of variability needs sampling
of a latitudinal coenotic gradient. This is fairly evident in the by both the above complementary technigues.
ordination model, where the station points show saturation - When inter-site comparisons are performed, a new and stronger
along Fl consistent with their geographical (i.e. longitudinal) structural variability is added, against which the discriminating
positions. There are no species with narrow biogeographical dis- power of the two methods becomes ineffective, as they produce
tributions affecting the composition of the assemblages. There- similar patterns. This new variability may be considered as
fore, this latitudinal pattern is likely to be of the structural 'edaphic', being related to local factors affecting the seagrasses.
type, a product of the way widely distributed species are sorted Consequently, for the study of this second type of variability, one
together, rather than the result of the presence of endemie of the two techniques may be adopted indifferently.
species in a narrow region of the Mediterranean. This geograph- - A longitudinal trend was observed even if no species with nar-
row biogeographical distributions were recorded. Therefore this
air-lift pattern is more likely to be related to the 'history' of the assem-
blage (e.g. successional stages) rather than to the history of some
of the species belonging to them (e.g. evolutionary ecology).
- The high variability in the composition of the malacological
assemblages does not fit with the model by Pérès and Picard
Mcda.c; (57 _(,Ci%)
which considers the Mediterranean systems of Posidonia oceanica
beds as belonging to a sing le bionomic unit; on the contrary,
our findings represent additional evidence in support of the
hypothesis considering the system as an 'ecologica! cross-roads'.

Meda> (40 15%)
REFERENCES

BIANCHI C. N., BEDULLI D, MORRI C., 0CCHPINTI AMBROGI A., 1989
Marctt1mo (l 7.42(1/u)
- L'herbier de posidonies: ecosysteme ou carrefour ecologiqueì
International \Vorkshop on Posidonia Beds, Boudoresque C.F.,
Figure 5. Samples by air-l1ft: (a) quaiiranve Meinesz A. , Fresi E. & Gravez V. edit., GIS Posidonie Pubi. ,
and (b) quanritative dominances among rhe sires. Fr., 2: 257-272.

> 93<