Hostname: page-component-74656f9884-npj9w Total loading time: 0 Render date: 2026-11-07T22:05:46.061Z Has data issue: false hasContentIssue false
  • English
  • Français

Differences among fish assemblages associated with a nearshore vermetid reef and two other rocky habitats along the shores of Cape Milazzo (northern Sicily, central Mediterranean Sea)

Published online by Cambridge University Press:  25 March 2008

P. Consoli ,
T. Romeo ,
U. Giongrandi  and
F. Andaloro
P. Consoli*
Affiliation:
ICRAM, Central Institute for Marine Research, Via dei Mille 44, 98057 Milazzo (Me), Italy
T. Romeo
Affiliation:
ICRAM, Central Institute for Marine Research, Via dei Mille 44, 98057 Milazzo (Me), Italy
U. Giongrandi
Affiliation:
Datalog Technology Ltd, Oil and Gas Services' Company, Unit 1, Cadleigh Close, Lee Mill Industrial Estate, Plymouth, PL21 9GB, UK
F. Andaloro
Affiliation:
ICRAM, Central Institute for Marine Research, Via dei Mille 44, 98057 Milazzo (Me), Italy
*
Correspondence should be addressed to: P. Consoli ICRAM, Central Institute for Marine Research Via dei Mille 44, 98057 Milazzo (Me) Italy email: pierpaoloconsoli@hotmail.com
Rights & Permissions [Opens in a new window]

Abstract

Fish assemblages associated with shallow rocky bottoms along the coast of Cape Milazzo were investigated within three selected habitats, each represented by a different degree of complexity and namely vermetid reef (VR), rocky-algal reef (RR) and boulder and pebbles (BP). The area, located along the northern coast of Sicily (southern Tyrrhenian Sea) was studied by using a non-destructive diver visual census methodology evaluating species composition and abundance. A total of 39 fish taxa belonging to 15 families were recorded over the three habitats investigated. Multivariate and univariate approaches detected significant differences among the three habitats. These differences were either in fish assemblage structure, in the assemblage parameters or in the abundances of some common species. A multidimensional scaling performed on the entire fish density data set showed distinct groupings for the three habitats. The highest values of mean species richness (S), diversity (H′), evenness (J) and density (N) occurred over the more complex habitat, represented by VR, than over rocky-algal reef habitat or BP. Different habitat preferences were evidenced for several fish species. Diplodus sargus, Serranus scriba, Tripterygion tripteronotus, T. delaisi and T. melanurus were significantly more abundant over VR whereas Thalassoma pavo and Symphodus tinca were more abundant both over VR and BP habitats. Statistical comparison did not detect habitat preferences in Symphodus roissali, Oblada melanura, Diplodus vulgaris and Sarpa salpa. As far as the size is concerned, small, medium and large-sized specimens of D. sargus and only large-sized individuals of D. vulgaris were significantly more abundant over the VR habitat side than over the other two. These results suggest that differences in assemblage parameters are related to habitat structure. Greater habitat complexity means greater surface available, thus providing additional resources for adults and juveniles of many species: food, cavities, caves and then new niches, refuges from predation, from artisan fishery nets, as well as resting or mating sites, boosting the total number of individuals and species. The results of this study highlight the importance of VR in sustaining the biodiversity of fish assemblages and emphasize the need for their careful management in an area already proposed as a Site of Community Interest.

Keywords

fish assemblagesvermetid reefPendropoma petracumCape Milazzovisual census

Information

Type
Research Article
Information
Journal of the Marine Biological Association of the United Kingdom , Volume 88 , Issue 2 , March 2008 , pp. 401 - 410
Copyright
Copyright © Marine Biological Association of the United Kingdom 2008

INTRODUCTION

Trottoir à vermets (Pérès & Picard, Reference Pérès and Picard1952) are biogenic constructions (sensu Laborel, Reference Laborel1987) located in warm-temperate areas such as the Atlantic Ocean and the Mediterranean Sea and formed by the gastropod species Dendropoma petraeum (Monterosato, 1884).

They grow perpendicular to the coastline, from land to sea, forming flat horizontal reefs lining most rocky calcareous shores in which a well developed community is found (Orlando, Reference Orlando1978; Chemello, Reference Chemello1989; Safriel & Ben-Eliahu, Reference Safriel, Ben-Eliahu, Bell, McCoy and Mushinsky1991; Mannino, Reference Mannino1992; Pandolfo et al., Reference Pandolfo, Chemello and Riggio1992a, Reference Pandolfo, Chemello and Riggiob).

These highly indented structures are documented to inhabit the rocky coasts of temperate and tropical seas (Keen, Reference Keen1961), with the isolated exceptions of the Caribbean Sea (Jones & Hunter, Reference Jones and Hunter1995) and the Island of Fernando de Noroña in Brazil (Kempf & Laborel, Reference Kempf and Laborel1968); the most important constructions are found in the Mediterranean Sea. In this basin vermetid platforms are always located along the shores below the 14°C winter isotherm (Chemello et al., Reference Chemello, Dieli and Antonioli2000) and were first reported from along the coast of Isola delle Femnine, near Palermo (northern Sicily, central Mediterranean) by the French naturalist De Quatrefages (Reference De Quatrefages1854). A century later, Molinier & Picard (Reference Molinier and Picard.1953) gave a detailed description of the same locality and suggested a correlation with similar bio-constructions located along the Algerian coast (Perés & Picard, Reference Pérès and Picard1952). In the eastern Mediterranean, Safriel (Reference Safriel1966, Reference Safriel1974, Reference Safriel1975) described intertidal reef-like structures from along the coasts of Israel; vermetid platforms are present along the coast of Lebanon (Dalongeville, Reference Dalongeville1977; Bitar & Bitar-Kouli, Reference Bitar and Bitar-Kouli1995a, Reference Bitar and Bitar-Koulib), southern Turkey (Laborel, Reference Laborel1987), Crete (Kelletat, Reference Kelletat1979; Laborel, Reference Laborel1987), Malta (Azzopardi, Reference Azzopardi1992), Algeria (Molinier & Picard, Reference Molinier and Picard.1953), Corsica (Molinier, Reference Molinier1955a, Reference Molinierb) and Spain (Templado et al., Reference Templado, Templado, Calvo, Giusti and Manganelli1992).

In Sicily trottoir à vermets are located along Tyrrhenian coasts, between the Cefalu promontory and Trapani coast, and inside the Egadi Islands. Some of the thick formations are placed to the south-east Ionian coast, near Siracusa; some other small constructions are situated along the southern shores of the Lampedusa and Ustica Islands.

The vermetid formations located at the north side of the Capo Milazzo promontory have never been studied. This area, a large plateaux (45.18 ha) located on the northern coast of Sicily (southern Tyrrhenian Sea), particularly interesting for marine and terrestrial fauna and flora, was proposed as a Site of Community Interest (SIC; Council Directive 92/43/EEC of 21 May 1992). The Directive is intended to ensure biodiversity by conserving natural habitats and wild fauna and flora in areas such as this.

In the Mediterranean, vermetid reefs have been studied concerning malacofauna (Orlando, Reference Orlando1978; Chemello, Reference Chemello1989; Pandolfo et al., Reference Pandolfo, Chemello and Riggio1992a, Reference Pandolfo, Chemello and Riggiob; Reference Pandolfo, Chemello, Ciuna, Lo Valvo and Riggio1996), dimensional analyses of the trottoir (Chemello, Reference Chemello1989; Badalamenti et al., Reference Badalamenti, Chemello, Gristina, Riggio and Toccaceli1992a, Reference Badalamenti, Chemello, Gristina, Riggio and Toccacelib;), polychaete communities (Ben-Eliahu, Reference Ben-Eliahu1975, Reference Ben-Eliahu1976, Reference Ben-Eliahu1977a, Reference Ben-Eliahub; Safriel & Ben-Eliahu, Reference Safriel, Ben-Eliahu, Bell, McCoy and Mushinsky1991; Badalamenti et al., Reference Badalamenti, Chemello, D'Anna and Riggio1998) and algae (Mannino, Reference Mannino1992).

Even though in the Mediterranean Sea, several authors have investigated fish fauna from different rocky bottoms (Bell, Reference Bell1983; Harmelin, Reference Harmelin1987; Garcia-Rubies & Zabala, Reference Garcia-Rubies and Zabala1990; Francour, Reference Francour1994; MacPherson, Reference MacPherson1994; Dufour et al., Reference Dufour, Jouvenel and Galzin1995; Fasola et al., Reference Fasola, Canova, Foschi, Novelli and Bressan1997; Sala, Reference Sala1997; Sala & Boudouresque, Reference Sala and Boudouresque1997), studies on fish assemblages associated with vermetid reef habitats are scanty (Goren & Galil, Reference Goren and Galil2001) and comparisons among fish assemblages associated with vermetid reefs and those from other nearshore rocky bottoms are not available.

Therefore, the main objectives of this study were to: (i) characterize the fish assemblages inhabiting the shallow rocky bottoms of Cape Milazzo; (ii) assess differences in the assemblage parameters and total density among a habitat characterized by vermetid formations and two other rocky bottom habitats; and (3) determine if there were habitat preferences of some common inshore fish species.

MATERIALS AND METHODS

Study area

Fish assemblages associated with shallow (i.e. 0–3 m depth) rocky bottoms along the coast of Cape Milazzo (Figure 1) were investigated at three selected habitats, each represented by a different degree of complexity and namely vermetid reef (VR), rocky-algal reef (RR) and boulder and pebbles (BP). All three habitat types were sampled in the same area located along the northern coast of Cape Milazzo.

Fig. 1.

Sampling area: Cape Milazzo with the three investigated habitats: vermetid reef (VR), rocky-algal reef (RR) and boulder and pebbles (BP).

The first habitat (VR) was represented by a vermetid reefs formation rich in crevices and other shelters. The trottoir of Milazzo Cape, as well those in north-western Sicily, are complex systems which share the major features of mid and infralittoral levels and include many microhabitats. In fact, according to Chemello (Reference Chemello1989), a typical vermetid platform can be divided into a number of different zones: (i) Lythophillum lichenoides cushions, the biological marker of the inner margin of the vermetid platforms; (ii) one or more ‘cuvettes’, little pools which remain covered by seawater and separated from each other by small vermetid crests; (iii) an elevated outer margin; and (iv) the upper infralittoral fringe characterized by a ‘facies’ of the Phaeophycaea Cystoseira amentacea var. amentacea. At the beginning, the bottom goes down suddenly to 1–2 m and then gently, and it is composed primarily of rocky boulders of diverse size. Moreover, vermetid ledges growing also on the surface abrasion blocks scattered at the base of the cliff, yield mushroom structures.

The RR was represented by a heterogeneous hard substrate. The third habitat (BP) was characterized by a gently slopping bottom with pebbles and boulders of different size.

Cover types at the sampling habitats were quite similar; algal cover was mainly composed of a small turf of brown algae, such as Dictyota dichotoma (Hudson) Lamoroux and Cystoseira spp.; small patches of Acetabularia acetabulum (L.) Silva and Padina pavonica (L.) Thivy were also present on flat rocks and boulders.

Sampling techniques

Sampling was carried out in June 2005, within the space of a short time lapse (three weeks) to prevent temporal bias and only during excellent weather conditions (i.e. low winds and hydrodynamics and high water transparency). The survey was conducted at three selected habitats, during day time (10 a.m. and 2 a.m.), using a strip transects underwater visual census technique (Harmelin-Vivien et al., Reference Harmelin-Vivien1985). Given the shallow depth (0–3 m), assessment was carried out by snorkelling to evaluate species' composition and number of fish along 20 m long and 2 m wide transects (total surface area 40 m2) parallel to the coastline. At the VR habitat, transects were performed immediately under the upper infralittoral fringe of the platforms.

The ‘transect’ technique was chosen from among the available visual approaches as it is the most suitable for both smaller specimens and fast-swimming species (Harmelin-Vivien & Francour, Reference Harmelin-Vivien and Francour1992; Francour, Reference Francour1997).

Twelve replicates were performed at each habitat by a single diver, in order to minimize the bias associated to the subjectivity of different observers.

Census accuracy is influenced by the observer's swimming speed (Lincoln Smith, Reference Lincoln Smith1988, Reference Lincoln Smith1989; De Girolamo & Mazzoldi, Reference Di Girolamo and Mazzoldi2001), which must be adjusted in relation to the target species' behaviour and substratum heterogeneity. It is clear that evaluating all the species occurring along a transect using one census only, although less time consuming, can give very biased estimates. To avoid such a bias two censuses were carried out along the same transect: one at higher speed (6 m/minute), to census nectobenthic fish and one at low speed (3.5 m/minute), to census benthic and cryptobenthic fish (De Girolamo & Mazzoldi, Reference Di Girolamo and Mazzoldi2001). Each fish recorded was assigned to one of three size-classes, i.e. small, medium, and large, corresponding to the lower, intermediate and upper 33% range of sizes reported in the literature for each species (Fischer et al., Reference Fischer, Bauchot and Schneider1987).

Juvenile stages (i.e. settlers and recruits) were excluded from the study, since their numerical contribution may greatly influence mean values.

Abundance of fish was estimated using a scale based on a geometric progression factor approximately of 2 (1, 2–5, 6–10, 11–30, 31–50, 51–100, 101–200, 201–500, >500 individuals), and fish density (number of individuals 40 m−2) was estimated by considering the mid-point of each abundance class (Harmelin-Vivien et al., Reference Harmelin-Vivien1985; Francour, Reference Francour1997).

As a descriptor of habitat complexity we measured rugosity which was estimated as the ratio L/l between real length (L) measured with a chain contoured over the bottom and extended transversely to the transect centre line, and linear distance between the chain's endpoints (l = 2 m) (Luckhurst & Luckhurst, Reference Luckhurst and Luckhurst1978). This measure was repeated 5 times at regular intervals along all transects. All these measurements were averaged to obtain the final complexity value for each habitat type.

Statistical analysis

The study used an experimental design with a single factor ‘habitat’ and 3 levels: RR, VR and BP.

The Kruskal–Wallis test was applied to assess differences between habitats complexity.

Permutational multivariate analysis of variance (PERMANOVA) (Anderson, Reference Anderson2001a; McArdle & Anderson, Reference McArdle and Anderson2001) was used to analyse the multivariate data set in response to the complete experimental design. This method allows multivariate data to be analysed on the basis of any distance or dissimilarity measure of choice, in response to any complex experimental design, with P values obtained using permutations.

There were 39 species variables, each transformed to log (x + 1) so that each species contributed fairly evenly to each analysis (Clarke, Reference Clarke1993). Bray–Curtis dissimilarities were calculated between all pairs of observations for ensuing analyses (Bray & Curtis, Reference Bray and Curtis1957). For PERMANOVA we used 4999 unrestricted random permutations of the raw data (Anderson, Reference Anderson2001b).

Non-metric multidimensional scaling (MDS; Kruskal & Wish, Reference Kruskal and Wish1978) was used to visualize multivariate patterns among plots on the basis of the Bray–Curtis dissimilarities. MDS is an unconstrained ordination method and thus allows patterns in overall dispersion and potential differences in relative within-group dispersion to be visualized (Willis & Anderson, Reference Willis and Anderson2003).

The assemblage structure for each transect was developed by calculating species richness (S = number of species), diversity (by the Shannon–Wiener index, H′ sp = −∑pi log2 pi, where pi is the proportion of the total sample belonging to the ith species), evenness (J = Hs′ p log S−1) (Pielou, Reference Pielou1966), total density and species density (number of individuals 40 m−2). At the three habitats, mean density and frequency of occurrence were evaluated for all the investigated species.

To test for differences over habitats in the assemblage variables (S, H′, and J) and total density, a one-way analysis of variance (ANOVA) was used. Homogeneity of variances was tested by Levene's test; whenever necessary, data were transformed (generally to log x + 1) and tested again (Underwood, Reference Underwood1997). When transformation did not produce homogeneous variances, α = 0.01 was used to make up for the increased likelihood of Type 1 error (Underwood, Reference Underwood1981). Post hoc multiple comparisons after ANOVA were made by the Student–Newman–Keuls test (Underwood, Reference Underwood1997). ANOVA was also used to test the null hypothesis that there were no differences in the abundances of 12 of the most abundant species among habitats. As regards Diplodus vulgaris and D. sargus ANOVA was also used to test that there were no differences in the abundance of each size-class among habitats. ANOVAs were performed using the Statistica software package.

RESULTS

The mean complexity value, standard deviation and range for each habitat are represented in Figure 2. The Kruskal–Wallis test detected significant differences between habitats complexities (Kruskal–Wallis-H 2,180 = 116.408, P = 0.000) and, as expected, the habitat with the highest complexity was represented by the VR formations (mean value = 2.27; SD = ±0.38).

Fig. 2.

Box and whisker plot showing mean values of complexity for each habitat (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

A total of 39 fish taxa belonging to 15 families was recorded (Table 1) over the three habitats investigated. Atherinidae and most of Mugilidae species were not identified at species level due to the difficulty of the specific determination by direct visual observations. A comparable total number of species (total species richness) was recorded at each habitat (28 at BP, 29 at VR and 31 at RR). Twenty-two species were common among the three habitats. The family with a higher number of species were wrasses (Labridae with ten species), sea breams (Sparidae with six species) and combtooth blennies (Blenniidae with 5 species). Quantitative data on mean species density and frequency of occurrence are reported in Table 1. The most abundant species were Chromis chromis, Symphodus tinca, Thalassoma pavo, Symphodus roissali and Coris julis with a total density (all the habitats pooled) ranging between 18.4 and 4.4 individuals per 40 m2. Of the 39 taxa recorded, S. tinca was ubiquitous (f = 100%) being observed within each habitat and site. Salaria pavo, Paralipophrys trigloides, Scartella cristata, Gobius cruentatus, Dicentrarchus labrax, Mugilidae, Muraena helena, Sparisoma cretense, Serranus cabrilla and Diplodus puntazzo occurred only on one habitat with a low density.

Fig. 3.

Non-metric multidimensional scaling plot of fish assemblages at the three selected habitats of Cape Milazzo. Stress value = 0.08 (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

Table 1.

Mean density (number of individuals per 40 m2, ± SE) and frequency of occurrence of the fish species observed over the three habitats (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

SE, standard error.

The PERMANOVA test (Table 2) detected significantly different fish assemblages between the three habitats. Pairwise comparisons (Table 3) showed significant differences in assemblages of fish among the three habitats selected. The results of statistical tests were confirmed by the (unconstrained) metric MDS plot of the Bray–Curtis distances (Figure 3).

Table 2.

Non-parametric multivariate analysis of variance (permutational multivariate analysis of variance) on the basis of Bray–Curtis dissimilarities for fish assemblages after log (x + 1) transformation.

*P < 0.05; **P < 0.01; ***P < 0.001.

Table 3.

P-values for non-parametric multivariate analysis of variance pair-wise comparisons among habitats, using 4999 permutations.

Significant levels: P < 0.05 significant; P < 0.01 very significant; P < 0.001 highly significant (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

The ANOVA results (Table 4) indicated significant differences among habitats in most of the species indicated in Figure 4. Diplodus sargus, S. scriba and the three Tripterygiidae species were more abundant at VR. Thalassoma pavo and S. tinca were more abundant over the VR and BP habitats side of Cape Milazzo than over the RR. In particular, the peacock wrasse reached the highest density on the BP habitat, where, on average, 22 specimens per 40 m2 were evaluated. Statistical comparison did not detect habitat preferences in S. roissali, O. melanura, D. vulgaris and S. salpa. As far as the size is concerned, small, medium and even more, large-sized specimens of D. sargus were significantly more abundant over the VR habitat side than the other two (Table 5). Diplodus vulgaris showed the same patterns of spatial distribution related to size as observed for D. sargus. Large-size individuals of D. vulgaris, in fact, were significantly more abundant over the VR habitat, while small and medium D. vulgaris did not display any differences among habitats.

Fig. 4.

Mean density (± SE; number of individuals 40 m−2) of the most abundant fish species observed over the three habitats along the island coasts. For Diplodus sargus and Diplodus vulgaris average densities are given also in relation to size (S, small; M, medium; L, large; BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

Table 4.

Results of analysis of variance used to test for differences in the density of dominant species over the investigated habitats.

*P < 0.05; **P < 0.01; ***P < 0.001; NS, not significant; BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef; SNK, Student–Newman–Keuls.

Table 5.

Results of analysis of variance comparing density of the two important commercial species Diplodus sargus and Diplodus vulgaris in relation to the size-class in each of the three habitats (BP, boulder and pebbles; VR, vermetid reef; RR, rocky-algal reef).

NS, not significant.

Average values of density (N), species richness (S), diversity (H′) and evenness (J) of the whole fish assemblage by habitats are given in Figure 5. Significant differences in total density were recognized by ANOVA in relation to habitats (Table 6). Mean fish density was significantly higher both over VR and BP than over RR. Species richness and diversity (H′) were usually higher over the VR habitat than at the other two: these differences between habitats were highly significant. The value of J varied significantly among habitats: the highest value was recorded over VR whereas no differences among site occurred for J. In summary, the highest assemblage variable values were found over the VR habitat.

Fig. 5.

Mean ± SE density (N), species richness (S), diversity (H′) and evenness (J) for the three habitats (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

Table 6.

Results of analysis of variance and multiple comparisons used to test for differences in the assemblages variables and fish density over the investigated habitats.

*p < 0.05; **p < 0.01; ***p < 0.001; NS, not significant; BP, boulder and pebbles; RR, rocky-algal reef; VR, vermetid reef.

DISCUSSION

The present study detected significant differences among the three identified habitats. These differences, observed both by multivariate and univariate approaches, were either in fish assemblages' structure, in the assemblage parameters or in the abundances of some common species.

The results indicate firstly that shallow rocky shores around Cape Milazzo were inhabited by a diversified benthic fish assemblage made up of 39 fish taxa. The value of species richness (29) recorded at the vermetid habitat was, anyway, lower than that recorded by Goren & Galil (Reference Goren and Galil2001) along a VR off the coast of Israel (36 fish taxa).

In our study the highest values of mean species richness (S), diversity (H′), evenness (J) and density (N) occurred over a VR habitat than over an RR habitat or BP. According to Safriel & Ben-Eliahu (Reference Safriel, Ben-Eliahu, Bell, McCoy and Mushinsky1991), in fact, vermetid reefs exercise considerable influences on superior mediolittoral and infralittoral communities, resulting in an increase of associated population diversity (Pandolfo et al., Reference Pandolfo, Chemello, Ciuna, Lo Valvo and Riggio1996; Badalamenti et al., Reference Badalamenti, Chemello, D'Anna and Riggio1998).

Many environmental factors could explain the observed differences between fish assemblages associated with the three selected habitat. For reef-associated fish, indeed, there is abundant evidence that the structure of the habitat has important effects on spatial distributions of populations, both in tropical coral reefs (Roberts & Ormond, Reference Roberts and Ormond1987; Tolimieri, Reference Tolimieri1995; Caley & St John, Reference Caley and St John1996; Friedlander & Parrish, Reference Friedlander and Parrish1998; Tolimieri, Reference Tolimieri1998; Holbrook et al., Reference Holbrook, Forrester and Schmitt2000; McClanahan & Arthur, Reference McClanahan and Arthur2001) and in temperate rocky reef systems (Choat & Ayling, Reference Choat and Ayling1987; Jones, Reference Jones1988; Carr, Reference Carr1989; Holbrook et al., Reference Holbrook, Schmitt and Ambrose1990; Connell & Jones, Reference Connell and Jones1991; Levin & Hay, Reference Levin and Hay1996; Tupper & Boutilier, Reference Tupper and Boutilier1997; García-Charton & Pérez-Ruzafa, Reference García-Charton and Pérez-Ruzafa2001).

Moreover, in different fish assemblages, many researches have pointed out that both S and H′ generally increased with habitat complexity (Luckurst & Luckurst, Reference Luckhurst and Luckhurst1978; MacPherson, Reference MacPherson1994; García Charton & Pérez Ruzafa, Reference García-Charton and Pérez-Ruzafa1998, Reference García-Charton and Pérez-Ruzafa2001; La Mesa & Vacchi, Reference La Mesa and Vacchi2005) and our results are in accordance with those obtained by the above mentioned authors.

Indeed, as shown by our results, there are natural differences in habitat complexity (or bottom relief) among VR, RR and BP habitats. In fact VR provides a more complex three-dimensional biogenic structure (Huston, Reference Huston1994). By building a framework, bioconstructors generate physical structures thus modifying the habitat and its structural complexity. This implies direct provision of living space by the physical structure and quantitative changes in the amount of living space as the framework grows (Cocito, Reference Cocito2004). In addition, vermetid ledges growing on the abrasion blocks scattered at the base of the cliff, yield mushroom structures (Antonioli et al., Reference Antonioli, Chemello, Improta and Riggio1999). These formations, observed along the coasts of Cape Milazzo, determine a further increasing of vermetid habitat complexity which was observed neither on RR habitat nor on BP habitat.

The main feature of the rock-like structure contributed to the reefs by the vermetid Dendropoma is the combination of hardness and porosity. The Dendropoma worm-like empty tubes are inhabited by animals of several groups, of which worms, especially polychaetes, are dominant in numbers and biomass. Moreover, the high wave energy to which the reefs are exposed causes sediment, detritus particles and live plankton to be continually pumped into, and trapped within the reefs' pores and crevices (Safriel & Ben-Eliahu, Reference Safriel, Ben-Eliahu, Bell, McCoy and Mushinsky1991).

Greater habitat complexity means greater surface available, thus providing additional resources for adults and juveniles of many species: food, cavities, caves and then new niches, refuges from predation, from artisan fishery nets, as well as resting or mating sites etc., and boosting the total number of individuals and the number of species.

The greater abundance of the whole fish assemblage over VR is probably due to the high presence of crevices in this habitat. Some authors have suggested that habitat complexity can positively affect fish density through diverse mechanisms, for instance by increasing shelter opportunities (MacPherson, Reference MacPherson1994). Offering protection from predation, wave scour and high light irradiance, crevices represent a better habitat to many organisms, which are far more abundant in crypts than on open surfaces. Habitat complexity can also modify the effects of predation on fish by providing refuges (e.g. Connell & Jones, Reference Connell and Jones1991; Hixon & Beets, Reference Hixon and Beets1993; Caley & St John, Reference Caley and St John1996; Beukers & Jones, Reference Beukers and Jones1997; Tupper & Boutilier, Reference Tupper and Boutilier1997; Steele, Reference Steele1999). It is possible to presuppose that these refuges could also protect fish from set nets; as regards, D. sargus and D. vulgaris, two species of great commercial value targeted by many kinds of set nets along the coasts of Milazzo, the large size-classes were more abundant over vermetid habitat.

Although the data presented here are indicative of habitat effects, we cannot eliminate the alternative possibility that the composition and structure of fish assemblages are also influenced by other environmental factors such us depth, bottom type, algal cover and hydrodynamism (Luckurst & Luckurst, Reference Luckhurst and Luckhurst1978; Bell, Reference Bell1983; Roberts & Ormond, Reference Roberts and Ormond1987; Connell & Jones, Reference Connell and Jones1991; MacPherson, Reference MacPherson1994; García Charton & Pérez Ruzafa, Reference García-Charton and Pérez-Ruzafa1998, Reference García-Charton and Pérez-Ruzafa2001; La Mesa & Vacchi, Reference La Mesa and Vacchi2005). This means that the potential positive influence of high complexity on assemblage variables might be overcome by other environmental features, such as wave exposure. Among the physical factors, it has been demonstrated that wave exposure affects shallow water habitats and associated sessile and vagile faunas in the reef habitats (Williams, Reference Williams and Sale1991; Denny, Reference Denny, Wainwright and Reilly1994). With respect to the fish fauna, it is well known that the effects of wave exposure are one of the main factors which influence the distribution patterns of fish associated with coral reef habitats (Bellwood et al., Reference Bellwood, Wainwright, Fulton and Hoey2002). According to Safriel & Ben-Eliahu (Reference Safriel, Ben-Eliahu, Bell, McCoy and Mushinsky1991) a factor affecting the distribution and dimension of the trottoir à vermets at small scale, in the Mediterranean Sea, is the superficial hydrodynamism: D. petraeum is a sedentary, aggregative plankton-feeding snail restricted to sites with the most extreme wave action. That is why in Sicily, vermetid platforms, are found along all north-west coasts (Chemello et al., Reference Chemello, Dieli and Antonioli2000). Even if all the three habitats were placed on the northern side of the promontory, the vermetid habitat is located exactly on the north-western side probably where the hydrodynamism is maximum (as reported by the winds chart, in the area of Cape Milazzo the 60% of winds blow from the west: north-westerly, westerly and south-westerly winds).

It is possible to suppose that the main hydrodynamism on the north-western slope promontory creates very good platform development conditions, the complexity of which influences fish assemblages, associated with this habitat.

Complexity as well as hydrodynamism may just be some components of differences among the habitats studied. Only specific studies could allow the separation of effects caused by the various components potentially involved. Anyway, it is important not to infer too much concerning processes that may be underlying observed patterns in the results reported here.

To date, no study has undertaken an extensive survey of the shallow water fish assemblage along the Milazzo promontory in reference to habitat. The data collected showed a high fish biodiversity associated with the trottoir à vermets. This habitat, characterized by a great structural complexity, appears fragile and particularly sensitive to human pressures such as pollution, unsustainable tourism and illegal fishery. Considering the environmental importance of these habitats and their role for the Mediterranean biodiversity, specific conservation strategies are desirable. As regards Capo Milazzo trottoir, located close to the urban development of this industrial area, we suggest that this area should become, as soon as possible, a marine protected area.

References

REFERENCES

Anderson, M.J. (2001a) A new method for non-parametric multivariate analysis of variance. Australian Ecology 26, 32–46.Google Scholar
Anderson, M.J. (2001b) Permutation tests for univariate or multivariate analysis of variance and regression. Canadial Journal of Fisheries and Aquatic Sciences 58, 626–639.CrossRefGoogle Scholar
Antonioli, F., Chemello, R., Improta, S. and Riggio, S. (1999) Dendropoma lower intertidal reef formations and their palaeoclimatological significance, NW Sicily. Marine Geology 161, 155–170.CrossRefGoogle Scholar
Azzopardi, L. (1992) Aspects of the ecology of vermetid gastropods on Maltese rocky shores. Dissertation, Department of Biology, University of Malta, 163 pp.Google Scholar
Badalamenti, F., Chemello, R., Gristina, M., Riggio, S. and Toccaceli, M. (1992a) Caratterizzazione delle piattaforme a Molluschi Vermetidi nella costa della Riserva Naturale dello Zingaro (TP). Oebalia 17, 543–545.Google Scholar
Badalamenti, F., Chemello, R., Gristina, M., Riggio, S. and Toccaceli, M. (1992b) Caratterizzazione delle Piattaforme a Molluschi Vermetidi nella costa tra Capo Gallo ed Isola delle Femmine (PA): area proposta come riserva naturale marina. Oebalia 17, 547–549.Google Scholar
Badalamenti, F., Chemello, R., D'Anna, G. and Riggio, S. (1998) Diversity of the polychaete assemblage in the hard bottom mediolittoral along the north-western Sicilian coast: the role played by the vermetid bioconstruction. In Abstract of the 1th CONISMA symposium ‘Diversità e Cambiamento’, Ischia, 11 November 1998.Google Scholar
Bell, J.D. (1983) Effects of depth and marine reserve fishing restriction on the structure of a rocky reef fish assemblage in the north-western Mediterranean Sea. Journal of Applied Ecology 20, 357–369.CrossRefGoogle Scholar
Bellwood, D.R., Wainwright, P.C., Fulton, C.J. and Hoey, A. (2002) Assembly rules and functional groups at global biogeographical scales. Functional Ecology 16, 557–562.CrossRefGoogle Scholar
Ben-Eliahu, M.N. (1975) Polychaete cryptofauna from rims of similar intertidal vermetid reefs on the Mediterranean coast of Israel and in the Gulf of Elat: Nereide (Polychaeta Errantia). Israel Journal of Zoology 24, 177–191.Google Scholar
Ben-Eliahu, M.N. (1976) Errant polychaete cryptofauna (excluding Syllidae and Nereidae) from rims of similar intertidal vermetid reefs on the Mediterranean coast of Israel and in the Gulf of Elat. Israel Journal of Zoology 25, 156–177.Google Scholar
Ben-Eliahu, M.N. (1977a) Polychaete cryptofauna from rims of similar intertidal vermetid reefs on the Mediterranean coast of Israel and in the Gulf of Elat: Syllinae and Eusyllinae (Polychaeta Errantia: Syllidae). Israel Journal of Zoology 26, 1–58.Google Scholar
Ben-Eliahu, M.N. (1977b) Polychaete cryptofauna from rims of similar intertidal vermetid reefs on the Mediterranean coast of Israel and in the Gulf of Elat: Exogoninae and Autolytinae (Polychaeta Errantia: Syllidae). Israel Journal of Zoology 26, 59–99.Google Scholar
Beukers, J.S. and Jones, G.P. (1997) Habitat complexity modifies the impact of piscivores on a coral reef fish population. Oecologia 114, 50–59.CrossRefGoogle Scholar
Bitar, G. and Bitar-Kouli, S. (1995a) Aperçu de bionomie bentique et répartition des différents faciès de la roche littorale à Hannouch (Liban, Méditerranée orientale). Rapport Commission Internationale pour la Mer Méditerranée 34, 19 pp.Google Scholar
Bitar, G. and Bitar-Kouli, S. (1995b) Impact de la pollution sur la répartition des peuplemets de substrat dur à Beyrouth (Liban, Méditerranée orientale). Rapport Commission Internationale pour la Mer Méditerranée 34, 19 pp.Google Scholar
Bray, J.R. and Curtis, J.T. (1957) An ordination of upland forest communities of southern Wisconsin. Ecological Monographs 27, 325–349.CrossRefGoogle Scholar
Caley, M.J. and St John, J. (1996) Refuge availability structures assemblages of tropical reef fishes. Journal of Animal Ecology 65, 414–428.CrossRefGoogle Scholar
Carr, M.H. (1989) Effects of macroalgal assemblages on the recruitment of temperate zone reef fishes. Journal of Experimental Marine Biology and Ecology 126, 59–76.CrossRefGoogle Scholar
Chemello, R. (1989) La bionomia bentonica ed i Molluschi. 5. Il piano Infralitorale: il marciapiede a vermeti. Notiziario SIM 7, 167–170.Google Scholar
Chemello, R., Dieli, T. and Antonioli, F. (2000) Il ruolo dei ‘reef’ a molluschi vermetidi nella valutazione della biodiversità. In ICRAM (ed.) Mare e cambiamenti globali. Rome, pp. 105–118.Google Scholar
Choat, J.H. and Ayling, A.M. (1987) The relationship between habitat structure and fish faunas on New Zealand reefs. Journal of Experimental Marine Biology and Ecology 110, 257–284.CrossRefGoogle Scholar
Clarke, K.R. (1993) Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18, 117–143.CrossRefGoogle Scholar
Cocito, S. (2004) Bioconstruction and biodiversity: their mutual influence. Scientia Marina 68, 137–144.CrossRefGoogle Scholar
Connell, S. and Jones, G.P. (1991) The influence of habitat complexity on postrecruitment processes in a temperate reef fish population. Journal of Experimental Marine Biology and Ecology 151, 271–294.CrossRefGoogle Scholar
Dalongeville, R. (1977) Formes littorales de corrosion dans le roches carbonatées du Liban: étude morphologique. Méditerranée 3, 21–33.CrossRefGoogle Scholar
Di Girolamo, M. and Mazzoldi, C. (2001) The application of visual census on Mediterranean rocky habitats. Marine Environmental Research 51, 1–6.CrossRefGoogle Scholar
Denny, M.W. (1994) Roles of hydrodynamics in the study of life on wave–swept shores. In Wainwright, P.C. and Reilly, S.M. (eds), Ecological morphology: integrative organismal biology. Chicago: University of Chicago Press, pp. 169–204.Google Scholar
De Quatrefages, A. (1854) Souvenirs d'un naturaliste. 1. Charpentier: Paris.Google Scholar
Dufour, V., Jouvenel, J.Y. and Galzin, R. (1995) Study of a Mediterranean reef fish assemblage. Comparisons of population distributions between depths in protected and unprotected areas over one decade. Aquatic Living Resources 8, 17–25.CrossRefGoogle Scholar
Fasola, M., Canova, L., Foschi, F., Novelli, O. and Bressan, M. (1997) Resource use by a Mediterranean rocky slope fish assemblage. Marine Ecology —PS ZNI 18, 51–66.CrossRefGoogle Scholar
Fischer, W., Bauchot, M.I. and Schneider, M. (1987) Fiches FAO d'identification des espèces pour les besoins de la pêche (Révision 1). Méditerranée et mer Noire. Zone de pêche 37. 1. Vegetaux et invertebres. Publication préparée par la FAO (Project GCP/INT/422/EEC). Rome, FAO, 760 pp.Google Scholar
Francour, P. (1994) Pluriannual analysis of the reserve effect on ichthyofauna in the Standola natural reserve (Corsica, northwestern Mediterranean). Oceanologica Acta 17, 309–317.Google Scholar
Francour, P. (1997) Fish assemblages of Posidonia oceanica beds at Port-Cros (France, NW Mediterranean): assessment of composition and long-term fluctuations by visual census. Marine Ecology—PS ZNI 18, 157–173.CrossRefGoogle Scholar
Friedlander, A.M. and Parrish, J.D. (1998) Habitat characteristics affecting fish assemblages on a Hawaiian coral reef. Journal of Experimental Marine Biology and Ecology 224, 1–30.CrossRefGoogle Scholar
García-Charton, J.A. and Pérez-Ruzafa, A. (1998) Correlation between habitat structure and a rocky reef fish assemblage in the southwest Mediterranean. Marine Ecology—PS ZNI 19, 111–128.CrossRefGoogle Scholar
García-Charton, J.A. and Pérez-Ruzafa, A. (2001) Spatial pattern and the habitat structure of a Mediterranean reef fish local assemblage. Marine Biology 138, 917–934.CrossRefGoogle Scholar
Garcia-Rubies, A. and Zabala, M. (1990) Effects of total fishing prohibition on the rocky fish assemblages of Medes Islands marine reserve (NW Mediterranean). Scientia Marina 54, 317–328.Google Scholar
Goren, M. and Galil, B.S. (2001) Fish biodiversity in the vermetid reef of Shiqmona (Israel). Marine Ecology— PS ZNI 22, 369–378.CrossRefGoogle Scholar
Harmelin, J.G. (1987) Structure at variabilité de l'ichtyofaune d'une zone rocheuse protégée en Méditerranée (Parc national de Port-Cros, France). Marine Ecology—PS ZNI 8, 263–284.CrossRefGoogle Scholar
Harmelin-Vivien, M.L. and Francour, P. (1992) Trawling or visual censuses? Methodological bias in the assessment of fish populations in sea grass beds. Marine Ecology—PS ZNI 13, 41–51.CrossRefGoogle Scholar
Harmelin-Vivien, M.L. et al. (1985) Evaluation visuelle des peuplements et populations de poissons: méthodes et problèmes. Revue d'Ecologie (Terre Vie) 40, 467–539.CrossRefGoogle Scholar
Hixon, M.A. and Beets, J.P. (1993) Predation prey refuges and the structure of coral-reef fish assemblages. Ecological Monographs 63, 77–101.CrossRefGoogle Scholar
Holbrook, S.J., Schmitt, R.J. and Ambrose, R.F. (1990) Biogenic habitat structure and characteristics of temperate reef fish assemblages. Australian Journal of Ecology 15, 489–503.CrossRefGoogle Scholar
Holbrook, S.J., Forrester, G.E. and Schmitt, R.J. (2000) Spatial patterns in abundance of a damselfish reflect availability of suitable habitat. Oecologia 122, 109–120.CrossRefGoogle ScholarPubMed
Huston, M.A. (1994) Biological diversity. The coexistence of species in changing landscapes. Cambridge: Cambridge University Press.Google Scholar
Jones, B. and Hunter, I.G. (1995) Vermetid buildups from Grand Cayman, British West Indies. Journal of Coastal Research 11, 973–983.Google Scholar
Jones, G.P. (1988) Ecology of rocky reef fish of north-eastern New Zealand: a review. New Zealand Journal of Marine and Freshwater Research 22, 445–462.CrossRefGoogle Scholar
Keen, A.M. (1961) A proposed reclassification of the gastropod family Vermetidae. Bulletin of the British Museum (Natural History) 7, 181–213.Google Scholar
Kelletat, D. (1979) Geomorphologishe Studien an den Küsten Kretas. Abhandlungen der Gesellschaft der Wissenschaften in Göttingen, Mathematisch-Physikalische Klasse. 3° Folge, 32 pp.Google Scholar
Kempf, M. and Laborel, J. (1968) Formations de vermets et d'algues calcaires sur les côtes du Bresil. Recueil des Travaux de la Station Marine d'Endoume 43, 9–23.Google Scholar
Kruskal, J.B. and Wish, M. (1978) Multidimensional scaling. Beverly Hills, CA: Sage Publications.CrossRefGoogle Scholar
Laborel, J. (1987) Marine biogenic constructions in the Mediterranean. A review. Scientific Reports of the Port-Cros National Park, France 13, 97–126.Google Scholar
La Mesa, G. and Vacchi, M. (2005) Analysis of the blennoid assemblages associated with different rocky shore habitats in the Ligurian Sea. Journal of Fish Biology 66, 1300–1327.CrossRefGoogle Scholar
Levin, P.S. and Hay, M.E. (1996) Responses of temperate reef fishes to alterations in algal structure and species composition. Marine Ecology Progress Series 134, 37–47.CrossRefGoogle Scholar
Lincoln Smith, M.P. (1988) Effects of observer swimming speed on sample counts of temperate rocky reef fish assemblages. Marine Ecology Progress Series 43, 223–231.CrossRefGoogle Scholar
Lincoln Smith, M.P. (1989) Improving multispecies rocky reef fish censuses by counting different groups of species using different procedures. Environmental Biology of Fishes 26, 20–37.CrossRefGoogle Scholar
Luckhurst, B.E. and Luckhurst, K. (1978) Analysis of the influence of substrate variables on coral reef fish communities. Marine Biology 49, 317–323.CrossRefGoogle Scholar
MacPherson, E. (1994) Substrate utilization in a Mediterranean littoral fish community. Marine Ecology Progress Series 114, 211–218.CrossRefGoogle Scholar
Mannino, A.M. (1992) Studio fitosociologico della vegetazione mesolitorale a Lithophyllum lichenoides PHILIPPI (Rhodophyceae, Corallinales). Naturalista Siciliano, Palermo, s. IV, 16, 3–25.Google Scholar
McArdle, B.H. and Anderson, M.J. (2001) Fitting multivariate models to community data: a comment on distance-based redundancy analysis. Ecology 82, 290–297.CrossRefGoogle Scholar
McClanahan, T.R. and Arthur, R. (2001) The effect of marine reserves and habitat on populations of east African coral reef fishes. Ecological Applications 11, 559–569.CrossRefGoogle Scholar
Molinier, R. (1955a) Les platformes et corniches recifales de Vermets (Vermetus cristatus Biondi) en Mediterranée occidentale. Les Comptes Rendus de l'Académie des Sciences 240, 361–363.Google Scholar
Molinier, R. (1955b) Deux nouvelles formations organogenes biologiques construites en Mediterranée occidentale. Les Comptes Rendus de l'Académie des Sciences 240, 2166–2168.Google Scholar
Molinier, R. and Picard., J. (1953) Notes biologiques a propos d'un voyage d'étude sur les cotes de Sicile. Annales de l'Institut Océanographique 28, 163–188.Google Scholar
Orlando, V.E. (1978) Malacofauna del ‘Trottoir’ a vermetidi nella Sicilia occidentale. Naturalista Siciliano 2, 87–96.Google Scholar
Pandolfo, A., Chemello, R. and Riggio, S. (1992a) Prime note sui popolamenti associati ai ‘trottoir’a vermetidi delle coste siciliane: i Molluschi. Oebalia 17, 379–382.Google Scholar
Pandolfo, A., Chemello, R. and Riggio, S. (1992b) Notes sur la signification écologique de la malacofaune d'un ‘Trottoir à Vermets’ le long de la côte de Palerme (Sicile). Rapport Commission Internationale pour la Mer Méditerranée 33, 47 pp.Google Scholar
Pandolfo, A., Chemello, R., Ciuna, I., Lo Valvo, M. and Riggio, S. (1996) Analisi della distribuzione dei molluschi nella zona di transizione tra mesolitorale ed infralitorale superiore lungo le coste della Sicilia. Biologia Marina Mediterranea 3, 78–87.Google Scholar
Pérès, J.M. and Picard, J. (1952) Les corniches calcaires d'origine biologique en Méditerranée occidentale. Recueil des Travaux de la Station Marine d'Endoume 4, 2–34.Google Scholar
Pielou, E.C. (1966) The measurement of diversity in different types of biological collections. Journal of Theoretical Biology 13, 131–144.CrossRefGoogle Scholar
Roberts, C.M. and Ormond, R.F.G. (1987) Habitat complexity and coral reef fish diversity and abundance on Red Sea fringing reefs. Marine Ecology Progress Series 41, 1–8.CrossRefGoogle Scholar
Safriel, U. (1966) Recent vermetid formation on the Mediterranean shore of Israel. Proceedings of the Malacological Society of London 37, 27–34.Google Scholar
Safriel, U. (1974) Vermetid gastropods and intertidal reefs in Israel and Bermuda. Science 186, 1113–1115.CrossRefGoogle ScholarPubMed
Safriel, U. (1975) The role of vermetid gastropods in the formations of Mediterranean and Atlantic reefs. Oecologia 20, 85–101.CrossRefGoogle Scholar
Safriel, U.N. and Ben-Eliahu, M.N. (1991) The influnce of habitat structure and environmental stability on the species diversity of polychaetes in vermetid reefs. In Bell, S.S. McCoy, E.D. and Mushinsky, H.R. (eds.), Habitat structure: the physical arrangements of objects in space. London: Chapman and Hall, Ltd, pp. 349–370.CrossRefGoogle Scholar
Sala, E. (1997) The role of fish in the organization of a Mediterranean sublittoral community. I: epifaunal communities. Journal of Experimental Marine Biology and Ecology 212, 45–60.CrossRefGoogle Scholar
Sala, E. and Boudouresque, C. (1997) The role of fish in the organization of a Mediterranean sublittoral community. I: algal communities. Journal of Experimental Marine Biology and Ecology 212, 25–44.CrossRefGoogle Scholar
Steele, M.A. (1999) Effects of shelter and predators on reef fishes. Journal of Experimental Marine Biology and Ecology 233, 65–79.CrossRefGoogle Scholar
Templado, J., Templado, D. and Calvo, M. (1992) The formations of vermetid gastropod Dendropoma petraeum (Monterosato, 1884) on the coasts of the Iberian Peninsula (Western Mediterranean). In Giusti, F. and Manganelli, G. (eds), Abstracts of the 11th Internetional Malacological Congress, Siena, pp. 514–515.Google Scholar
Tolimieri, N. (1995) Effects of microhabitat characteristics on the settlement and recruitment of a coral reef fish at two spatial scales. Oecologia 102, 52–63.CrossRefGoogle ScholarPubMed
Tolimieri, N. (1998) Contrasting effects of microhabitat use on large-scale adult abundance in two families of Caribbean reef fishes. Marine Ecology Progress Series 167, 227–239.CrossRefGoogle Scholar
Tupper, M. and Boutilier, R.G. (1997) Effects of habitat on settlement, growth, predation risk and survival of a temperate reef fish. Marine Ecology Progress Series 151, 225–236.CrossRefGoogle Scholar
Underwood, A.J. (1981) Techniques of analysis of variance in experimental marine biology and ecology. Oceanography and Marine Biology: an Annual Review 19, 513–605.Google Scholar
Underwood, A.J. (1997) Experiments in ecology: their logical design and interpretation using analysis of variance. Cambridge: Cambridge University Press.Google Scholar
Williams, D.M. (1991) Patterns and processes in the distribution of coral reef fishes. In Sale, P.F. (ed.), The ecology of fishes on coral reefs. San Diego: Academic Press, pp. 437–474.CrossRefGoogle Scholar
Willis, T.J. and Anderson, M.J. (2003) Structure of cryptic reef fish assemblages: relationships with habitat characteristics and predator density. Marine Ecology Progress Series 257, 209–221.CrossRefGoogle Scholar
Figure 0

Fig. 1. Sampling area: Cape Milazzo with the three investigated habitats: vermetid reef (VR), rocky-algal reef (RR) and boulder and pebbles (BP).

Figure 1

Fig. 2. Box and whisker plot showing mean values of complexity for each habitat (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

Figure 2

Fig. 3. Non-metric multidimensional scaling plot of fish assemblages at the three selected habitats of Cape Milazzo. Stress value = 0.08 (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

Figure 3

Table 1. Mean density (number of individuals per 40 m2, ± SE) and frequency of occurrence of the fish species observed over the three habitats (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

Figure 4

Table 2. Non-parametric multivariate analysis of variance (permutational multivariate analysis of variance) on the basis of Bray–Curtis dissimilarities for fish assemblages after log (x + 1) transformation.

Figure 5

Table 3. P-values for non-parametric multivariate analysis of variance pair-wise comparisons among habitats, using 4999 permutations.

Figure 6

Fig. 4. Mean density (± SE; number of individuals 40 m−2) of the most abundant fish species observed over the three habitats along the island coasts. For Diplodus sargus and Diplodus vulgaris average densities are given also in relation to size (S, small; M, medium; L, large; BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

Figure 7

Table 4. Results of analysis of variance used to test for differences in the density of dominant species over the investigated habitats.

Figure 8

Table 5. Results of analysis of variance comparing density of the two important commercial species Diplodus sargus and Diplodus vulgaris in relation to the size-class in each of the three habitats (BP, boulder and pebbles; VR, vermetid reef; RR, rocky-algal reef).

Figure 9

Fig. 5. Mean ± SE density (N), species richness (S), diversity (H′) and evenness (J) for the three habitats (BP, boulder and pebbles; VR, vermetid reef; RR, rocky algal reef).

Figure 10

Table 6. Results of analysis of variance and multiple comparisons used to test for differences in the assemblages variables and fish density over the investigated habitats.