Bio-Mineralogy As a Structuring Factor for Marine Epibenthic Communities

MARINE ECOLOGY PROGRESS SERIES Vol. 193: 241-249,2000 Published February 28 Mar Ecol Prog Ser

Bio-mineralogy as a structuring factor for marine epibenthic communities

'Istitulo di Scienze del Mare, Universita di Ancona, Via Brecce Bianche, 60131 Ancona, Italy 2~arineEnvironment Research Centre, ENEA Santa Teresa, PO Box 316, 19100 La Spezia, Italy 3~ipartimentoper 10 studio del Territorio e delle sue Risorse, Universita di Genova, Viale Benedetto XV 5, 16132 Genova, Italy

ABSTRACT The meralogical features of the substrate were generally cons~dereda rmnor factor m structunng manne benthic communities The alm of th~swork 1s to venfy whether the presence of quartz nunerals in rock may exphcate dlfferences, usually explamed m terms of substrate roughness or other factors In epibenthic cornrnun~hes Laboratory tests on the hydroid Eudendnum glomeratum showed that ~tsplanulae settle preferentially on carbonat~c,rather than quartzltic substrates To test the Influence of quartz on established communities, we analysed the specles composition and quanti- tat~vestructure of subhttoral sesslle assemblages on Wferent rocks In several locahhes of the L~gurian and Tyrrhenian seas The observed dlfferences appeared to be related to the presence of quartz in the substrate rock The ~nteractionsbetween organisms and minerals (bio-mmeralogy) mlght play a slgnlf- icant role on benthic communities affechng not only the inihal colonlsahon but also later assemblages This potential role has been largely neglected to date and further studies are needed to prove ~tsImpor tance

KEY WORDS: Substrate colonisation . Mineral composition . Marine benthos distnbution . Hard substrates . Bio-mineralogy

INTRODUCTION communities growing on rocks of different nature. A species assemblage, which may be slightly more The spatial distribution and structure of marine ben- attracted to a particular substrate, could affect succes- thic communities are due to numerous abiotic and sion by its subsequent interaction with later assem- biotic factors which, in turn, are influenced by the blages. A similar effect was evidenced in the colonisa- presence of the organisms, in a mutual exchange of tion of artificial substrates, with respect to both species inputs. Among the abiotic factors, the mineralogical composition and abundance (Anderson & Underwood features of the substrate were generally considered of 1994, Holm et al. 1997). Less information is available scarce importance, but recent studies by Cerrano et al. for natural substrates (McGuiness 1989), but it is com- (1998) have shown that the presence of quartz in the mon knowledge that the softness and asperity of a rock sand may affect the initial steps of infauna colonisa- can favour or hamper biotic colonisation through selec- tion. Cerrano et al. (1998) introduced the term bio-min- tive larval settling, retention of water (in the littoral) eralogy to explicate the interrelationships between and organic matter, and provision of refuges from pre- biological systems at different hierarchies (cell, organ- dation or grazing (Den Hartog 1972, Levinton 1982, ism, species, community) and minerals. Walters & Wethey 1996). Bio-mineralogy could influence hard-bottom assem- More is known about the influence of substrate min- blages and explain some 'anomalies' in the structure of eralogy on bioboring, which is prevented by high percentages of quartzitic or pelitic components in the rock. Sublittoral endolitic communities are charac-

0 Inter-Research 2000 Resale of fullarticle not permitted 242 Mar Ecol Prog Ser 193: 241-249. 2000

during November 1998 at 20 to 25 m depth. The fertilised eggs develop in verticils of 5 to 10 fixed to a rudimentary polyp, the blas- tostyle, deprived of mouth and tentacles. Each egg is enveloped by a non-branched spadix. The verticils of eggs were mechani- cally detached from the mother colony and placed in 250 m1 cups filled with filtered natural sea water at a temperature of 15°C. Cerrano et al. (1997) demon- strated that light exposure is nec- essary to trigger egg hatching and therefore we reared the eggs in lit conditions for 3 d. New re- leased planulae were divided into 4 stocks of 60 specimens and each Fig. 1.Sublittoral rock showing differences in the distribution of borers due to the stock was placed for 12 d in an mineral composlt~onof the substrate. The white calcitic vein IS widely bored by the experimental Petri dish in com- sponge Cliona celata which is absent on marl limestone (grey). Herein only the date- plete darkness to avoid problems mussel Lithophaga lithophaga is able to penetrate (arrows) related to their very high pho- totrophy. Petri dishes were pre- pared by sticking a uniform layer terised mainly by clionid sponges and bivalves. Clionid of sand grains with Eukitt; half of the surface was cov- sponges use hatching cell pseudopodes and the pro- ered by quartz sand, and half by sand derived from duction of carbonic acid to penetrate limestone (Riit- Carrara marble. The 2 sand types have the same gran- zler & Rieger 1973, Pomponi 19?9), but Cliona celata ulometric (125 to 250 pm) and morphological features Grant preferentially bores, in the Mediterranean Sea, (roundness, 0.35 to 0.7; projection of sphericity, 0.8 to substrates richer than 60% in carbonates (unpubl. obs.). Bivalves, such as species of Lithophaga, bore with the help of sulphuric acid and/or neutral muco- proteins (Russo & Cicogna 1992), and are also able to penetrate marl limestone, which is impenetrable by clionids (unpubl. obs., Fig. 1). All this has important consequences on the rock texture and roughness and, ultimately, on the community structure. To check the influence of substrate mineralogy on the larval settling, a series of laboratory tests was conducted using planulae of the hydroid Eudendrjum glomeratum (Picard), a very colnmon species of the sublittoral Med- iterranean zoobenthos (Boero et al. 1986). In order to Rome seek a relation between the mineralogy of natural sub- m - strates and the structure of sessile epibenthic communities, the species composition and percent cover of assemblages living on different kinds of rock in the western Mediterranean Sea were compared.

MATERIALAND METHODS

Laboratory. Planulae of the hydroid Eudendrium glomeratum were obtained from female colonies col- Fig. 2. Geographic layout of the fleld-study sites in the Lig- lected on the rockycliff of the Portofino Promontory urian and Tyrrhenian seas (NW Mediterranean) Bavestrello et al.: Structuring of marine epibenthic communities 243

Table 1. Locahties for field studies on subllttoral rocky bottoms

Locality Contrasted mineralogies Depth No. of No. of range (m) stations species

Ligurian Sea Quartzite vs Puddingstone, mar1 5-7 14 60 (Gallinaria Island) limestone and sandstone (Portofino region) Giglio Island Granite (most of vs Limestone (western part 6-10 16 54 the island) of the island) Northeast Sardin~a Granite (most of the vs Limestone-dolom~te 18-34 20 4 4 coast and small islands) (Tavolara Island)

0.9). Both kinds of sand were previously stored at taken into account the epibenthic communities on the 100°C for 24 h, which enabled us to obtain an artificial rocky bottom (Bianchi et al. 1987, Cattaneo-Vietti et al. hard bottom on the entire Petri dish surface, with very 1988). One of the most studied sites is the region similar physical features but with 2 different mineralo- around Portofino (Tortonese 1958, 1962, Morri et al. gies. 1988), but recent research has also been done on the Field. The data matrices were produced using un- Island of Gallinaria (Balduzzi et al. 1994).These stud- published data sets on subtidal epibenthic communi- ies underlined that epibenthc communities on sublit- ties of the Ligurian Sea (Gallinaria Island and Portofino toral rocks at Portofino are dominated by flourishing region) and Tyrrhenian Sea (Giglio Island and north- gorgonian populations, whereas at Gallinaria, gor- east Sardinia) (Fig. 2). The data sets were chosen in gonians are scarce and sponges dominate. From a min- order to be able to contrast communities from an equal eralogical point of view, Gallinaria Island is exclusively number of stations on rocks rich in quartz, such as quartzitic (Orsino 1975), whereas the coast around quartzite and granites, or deprived of that mineral Portofino is characterised by puddingstone, mar1 lime- (Table 1). All data sets consist of percent cover data of stone and sandstone, none of which has a significant conspicuous species (Hiscock 1987), derived from quartz component (Boni et al. 1969). underwater visual inventories by SCUBA divers using The Giglio Island is included in the southern group quadrats (Bianchi et al. 1991). Cover data for a total of of the Tuscan Archipelago, at the border between Lig- 96 species (Table 2) were available, but each data set urian and Tyrrhenian seas. Its rocky coast is mostly was analysed separately according to depth zone and granite, but a small portion on the western side is corn- geographical location, so that communities only posed of limestone (Alvisi et al. 1994). The subtidal showed differences in the mineralogical nature of the epibenthic communities of Giglio Island were first substrate and possibly exposure. To avoid the influ- described by Balduzzi et al. (1996). ence of sedimentation, which may alter the effects The northeast corner of Sardinia is mostly composed of mineralogical composition at the rock surface of granite. A striking exception is the Tavolara Island (Bavestrello et al. 1995a), vertical stations (70 to 95" made by a gigantic limestone-dolomite slab (Lorenzoni slope) were selected. Matrices of species cover data & Chiesura-Lorenzoni 1973). Together with Molara, were compared by correspondence analysis (Legendre Molarotto and other minor islets and rocks (allgranitic), & Legendre 1998), and species numbers and total Tavolara forms a small archipelago, the epibenthic cover by l-way ANOVA. Prior to analysis, percentage communities of which have been studied by Navone et cover values were arcsine transformed to meet the al. (1992). assumption of homogeneity of variances (Underwood 1997). Field-study sites. The Ligurian Sea data derive from RESULTS surveys carried out in shallow water at Gallinaria Island in the summer of 1991 and in the region of Laboratory Portofino Promontory in the summer of 1993. The Tyrrhenian Sea data include information taken at After hatching, the Eudendnum glomeratum planulae infralittoral depths around Giglio Island in September crawled on the bottom of culture vessel for 2 to 3 d. Then, of 1988 to 1991 and at circalittoral depths in north- they fixed to the substrate by the anterior pole and meta- eastern Sardinia in June 1990 (Table 1). morphosed into a planulary polyp. Data showed that The Ligurian Sea has been the object of much crawling planulae had a strong selectivity for the sub- research in marine biology, and many studies have strate, being always about 5 times more abundant on the 244 Mar Ecol Prog Ser 193: 241-249, 2000

Table 2. List of the species in inventories of field studies on sublittoral rocky bottoms. Species codes are those used in Figs. 4. 6 & 7

l Code Phylum Phylum Species/family name Code Species/family name Aab Chlorophyta Ponfera Acetabulana acetabulum (Linnaeus) lor Irc~niaoros (Schrn~dt) P. C. Silva Irclnia sp. Ponfera Aac Porifera Irc Porifera Acanthella acuta Schrnidt Iva Ircin~a variabilis (Schmidt) A cr Arnphiroa cryptarlhrodia Zanardin~ Rhodophyta Janra rubens (Linnaeus) Lamouroux Rhodophyta Ada Ponfera Jru Rhodophyta AxjneUa damrcomis (Esper) Lfr L~thophyllumfrondosum (Dufour)Furnari, Ali Arbacia lixula (Linnaeus) Echinodermata Cormaci et Along] Aor Agelas oroides (Schm~dt] Porifera Lithophyllum rncrustans Philipp~ Rhodophyta Lln A ri Amphiroa rigida Lamouroux Rhodophyta Laurencia obtusa (Hudson) Larnouroux Rhodophyta Lob A ve Axinella verrucosa (Esper) Porifera Leptopsammia pruvoti Lacaze-Duthiers Cnidaria Beu Balanophyllia europaea (Risso) Cnldaria LP~ Leucosolenfa variabiljs Haeckel Ponfera Arthropods L va BP~ Balanus perforatus Bruguiere Mesophyllum lichenoides (Ellis) Lemoine Rhodophyta Mli Cad Codium adhaerens C. Agardh Chlorophyta Mvrjapora truncata (Pallas) Bryozoa Bryozoa Mtr Cbo Caberea boryi (Audouin et Savlgny) M~crocosmusvulgaris HeUer Chordata Chlorophyta Mvu Cbu Codrum bursa (Linnaeus) C. Agardh Parazoanthus axineflae (Schmidt) Cnidaria Pax Cc1 Clathrina clathrus (Schmidt) Porifera Paramuricea clavata (Risso) Cnidaria Pc1 Cc0 Clathnna contorta (Bowerbank) Ponfera Parerythropodium coralloides (Pallas) Cn~dal-ia Pro Ccr Crambe crambe (Schrnidt) Porifera Palmophyllum erassum (Naccari) Chlorophyta Pcr Cde Clavelina dellavalle~(Zirpolo) Chordata Rabenhorst Cel Coraltina elongata Ellis et Solander Rhodophyta Pentapora fascialis (PaUas) Bryozoa Pla Cfi CeUana fistulosa (Linnaeus) Bryozoa Petrosia ficifomis (Poiret) Ponfera r Pfi Cf Codium fragile (Suringar) Hariot Chlorophyta Phorbas ficlitius (Bowerbank) Porifera Pft Cgr CoraUina granifera Ellls et Solander Rhodophyta Pseudochlorodesm~s furcellata Chlorophyta Cin Caq~ophyllralnornata (Duncan) Cnidaria Pfu (Zanard~n~)Borgesen Cla Cladophora sp Chlorophyta Paracentrolus li~lidus(Lamarck) Ech~noderniata Cln PI1 Clionldae spp. Porifera Phyllophora nervosa (De Candolle) Rhodophyta J.Agardh Cmu Cutleria multifida (Smith)Grevill- Phaeophyta Pne Greville er Cn i Padrna pavonica (L~nnaeus)Lamouroux Cliona nigrrcans (Schm~dt) Ponfera Phaeophyta Cnu Peyssonnelia squamaria (Grnelin) Chondrilla nucula Schm~dt Porifera PP~ Rhodophyta Chartella papyrea (Pallas) Bryozoa Decaisne Phorbas tenacior (Topsent) CP~ Cladophora prolrfera (Roth) Kutzlng Chlorophyta Pornaloceros triqueter (Linnaeus) Ponfera Cre Chondrosia reniformis Nardo Ponfera Pte Protula tubularla (Montagu) Annelida C~P Cladostephus spongiosus (Hudson) Phaeophyta Ptr Renrera fulva Topsent Annelida C. Agardh Ptu Rhyncozoon sp. Ponfera Cystoseira zoslerordes (Turner) Phaeophyta Rfu Reptadeonella violacea (Johnston) Bryozoa C. Agardh Rhy Serpulorbis arenaria (Linnaeus) Bryozoa Ddi Dictyota dichotoma (Hudson) Lamouroux Phaeophyta R vi Schizomavella auriculata (Hassalll Mollusca DPO Dictyopteris polypod~ordes(De Candolle) Phaeophyta Sar Bryozoa Larnouroux Sau Smittina cervicorn~s(Pallas) Sphaerococcus coronopifolrus Bryozoa Dve Dasycladus vermicular~s(Scopoli) Krasser Chlorophyta Sce (Goodenough et Woodward) C. Agardh Rhodophyta Dvr Dudresnaia verticillatd (Withering) Rhodophyta Sco Spirastrella cunctatnx Schmidt Le Jol~s Porlfera Eca Eunicella cavolinir (Koch) Cnidaria Scu Salmaana dysteri (Huxley) Annelida Era Eudendrium raremosurn (Gmel~n) Cnidaria S~Y SertularellaMilne-Edwards) ellrwi (Deshayes et Cnidaria Esi Eunicella singularis [Esper) Cnidaria Sel Unidentified serpulids FP~ FlabeUia peliolata rrurra) Nizamudd~n Chlorophyta Savignyella latontii (Audouin et Savigny) Annelida Fm 'Falkenbergia ruloldnosa' (Hantey) Schmilz Rhodophyta ser Schlzoporella longirostns H~ncks Bryozoa Gob Galaxaura oblungata (Ell~set Solander) Rhodophyta Sla Unidentified spirorb~ds Bryozoa Lamouroux Slo Stypocaulon scopan'urn (Linnaeus]Kiitzing Annelida Hro Hemimycale columella (Bowerbank) Ponfera spi Sertella septentrionalis Harmer Phaeophyta H~IHalopteris diaphana (Heller) Ssc Cnidaria Serpula vemiculans Linnaeus Rryozoa jifi Halopteris fihcina (Grateloup) Kutzing Sse Phaeophyta Sargassurn vulgareC. Agardh Annelida Halocynthia paprllosa (Linnaeus) Chordate Sve Vst Vemlliopsis stnatlceps (Gmbe) Phaeophyta I-itu Halirneda tuna (Fllis et Solander) Svu Chlorophyta Vul Valonia utnculansi (Roth) C. Agardh Annelida Lamouroux 4Vpe M'rangclia penjcillala C. Agardh Chlorophyta Unident~fiedhydro~ds Cnldana Rhodophyta carbonatic sediments (Fig.Ja). During the experimental the dishes (Fig. 3b).The ratio between metamorphosed time (1l d), about one-third of the planulae metamor- (settled) and crawling planulae was not sigriificantly phosed and the number of planulary polyps reflected the different in the 2 sections (43.3 * 14.6 and 45.6 * 6% for number of planulae present in the different sections of quartz and carbonate respectively). Bavestrello et al.: Structuring of marine epibenthic communities 245

Aab

Acr Cel hyd R Czo Ffu

lob l va M F~ FN p ser Q Q cmu Htv Q Gob In sseci" tin Q c,,; RM Cla ML Q Aw Hco Cad Pax So0 Sar I W, l '." I L~gurianSea RhY 'Ccocd : spj 1* Ada M

Fig. 4. Correspondence analysis on sublittoral rocky bottom data in the Ligurian Sea (Portofino region and Gallinaria Island). The variance fracbon explained by the first 2 axes is 0 2 4 6 8 10 12 indicated. Station points are identified by bold capital letters DayS according to the nature of the rock. Q: quartzite; S:sandstone; ML: mar1 limestone; P: puddingstone. Species points are iden- Fig. 3. Response of Eudendrium glomeraturn planulae to dif- tified by codes as in Table 2 ferences In the mineral composition of the substrate. (a)Num- ber of crawling planulae on marble (a) and quartz (m)sands. (b) Number of metamorphosed planulae on the same substrates. Data are the means + SE of 4 replicates

Field

The first 2 axes extracted from the correspondence analyses on the 3 data sets were significant in all cases (p < 0.05 according to the tables of Lebart 1975). The fraction of the total variance represented on the plane formed by the first 2 axes was 45.73 % for the Ligurian Sea data set, 43.84 % for Giglio Island, and 32.00% for northeast Sardinia. Correspondence analysis on Ligurian Sea infralittoral epibenthic communities showed, along the 1st axis, a clear-cut separation of the stations of Gallinaria Island from those of the Portofino region, the first site being mainly characterised by quartzitic rock and the second by different rock types (Fig. 4). Along the 2nd axis, the station points of Portofino region tended to spread around according to the nature of their rocky bottoms, mar1 limestone separating at one extreme and sandstone at the other. This last arrangement might also reflect an exposure gradient, sandstone stations being located at more exposed sites that the limestone ones. However, there were station points of equally Ligurian Sea N TyrrhenianSea NE Sardinia exposed sites of Gallinaria, all on quartzitic rock, which remained well separated from the correspond- Fig. 5. Mean (*l standard deviation) number of species (a) and total substratum cover (b) of the epibenthlc assemblages ing ones of Portofino. Mean species richness on the in the 3 study sites, according to rock mineralogy. For each quartzitic rocks of Gallinaria was lower than on the site, quartz-rich rocks (quartzite or granites) (U), and rocks non-quartzitic rocks of the Portofino region (Fig. 5a): poor in or deprived of quartz (m)are represented 246 Mar Ecol Prog Ser 193: 241-249, 2000

Sau Cd

vot Rh,

PflPk L ,j, RV; hfh ba for Ada Pax m Sm Sel Sve Pla

Fig. 6. Correspondence analysis on sublittoral rocky bottom Fig. 7. Correspondence analysis on sublittoral rocky bottom data at Giglio Island (northern Tyrrhenian Sea). The variance data in northeastern Sardinia (Tavolara and Molara Archipel- fraction explained by the first 2 axes is indicated. Station ago). The variance fraction explained by the first 2 axes is points are identified by bold capital letters according to the indicated. Station points are identified by bold capital letters nature of the rock. G: granite; L: lunestone. Species points are according to the nature of the rock. G: granite; LD: limestone- identified by codes as in Table 2 dolomite. Species points are identified by codes as in Table 2

this difference is significant (l-way ANOVA, p = DISCUSSION 0.024). On the contrary, total substratum cover (Fig. 5b) was not significantly different (p = 0.119). The structure of hard bottom communities has been Results from correspondence analysis on infralit- traditionally interpreted as mainly the result of com- toral communities of Giglio Island showed a separa- plex biotic interactions within adult assemblages tion of stationpoints along the 1st axis that was consis- (Connell 1983, Schoener 1983). Only in the last tent with the mineralogical nature of the rock, decade has the importance of external factors, such as opposing limestone to granite (Fig. 6). On the 2nd physical processes and recruitment vagaries, been axis, limestone station points were closely grouped, fully recognised (Barry & Dayton 1991, Bingham 1992, whereas granite station points were more scattered. Bianchi 1997, Guichard & Bourget 1998, Smith & Wit- Looking at the species points (see Table 2 for decod- man 1999). However, our knowledge about the role of ing species names), the 2nd axis opposed encrusting the nature of substratum remains anecdotal. Many invertebrates to algae and erect or massive inverte- general papers on marine ecology report on the influ- brates: the first assemblage was considered the result ence of hard substratum surface texture and physico- of excess sea-urchin grazing by Balduzzi et al. (1996). chemical properties upon community composition and Both the number of species (Fig. 5a) and total cover structure (Den Hartog 1972, Levinton 1982), but stud- (Fig. 5b) were severely reduced on granite as com- ies providing direct evidence are scarce. Comparing pared with limestone (l-way ANOVA, p < 0.001 in the epibiotic communities of different types of sub- both cases). stratum, Connell & Glasby (1999) found differences in Correspondence analysis on circalittoral epibenthic the identity of species and the abundance of individ- communities of northeastern Sardinia (Tavolara and ual taxa. The same authors observed that the charac- Molara Archipelago) showed again the separation teristics of the substratum surface were an important between carbonate (limestone-dolomite) and granite determinant in the structure of assemblages and that station points (Fig. 7). On the contrary, no relationshp certain physical 01- chemical properties (e.g. alkalin- was evident with depth. notwithstandinq the compara- ity) may affect the settlement, growth or survival of tively great range investigated (18 to 34 m). Also in this organisms. On the other hand, Caffey (1982) found no case, the number of species on granite was lower than effect of rock types on the settlement or survival of on limestone-dolomite (Fig. 5a) and, although small, the intertidal barnacle Tesseropora rosea (Krauss). the difference was significant (l-way ANOVA, p = Our experiments with the subtidal hydroid Euden- 0.043). In contrast, no difference was found for total drium glomeratum suggest that the larvae of benthic cover [p = 0.717). organisms could be selected, other environmental con- Bavestrello et al.: Structuring of marine epibenthic cornrnunitles 247

ditions being equal, by the mineral composition of the monly interpreted as a consequence of intense sea- substrate and by the presence of quartz in particular. urchin grazing (Navone et al. 1992, Balduzzi et al. Most hydroids are substrate generalists (Gili & Hughes 1996). However, there is no reason to think that sea 1995) and species of Eudendriurn are probably not urchins graze more heavily on granite that on lime- exceptions (Sommer 1992). Our data for the first time stone; rather, sea-urchin grazing should be considered demonstrate a strong choice of hydroid larvae against as another form of dsturbance that frees patches of the mineral substrates. In this choice no biological (pres- original substrate (Fanelli et al. 1994). In deeper water, ence of a bacterial film) or physical (roughness, grain sea urchins are less abundant and wave effects negli- shape) features may be invoked. The fact that after gible. Disturbance is therefore less frequent and/or 11 d the ratio metamorphosed/crawling planulae intense, and bio-mineralogy looses importance. Per- in each dish section was not significantly different haps this is the reason why, in the whole Mediter- showed that, while the mineral nature of the sub- ranean Sea, the shallowest communities dominated by strate affects the larval behaviour, the metamorphosis gorgonians, which are typical late-successional organ- into planulary polyps was not influenced. It is possible isms, are found at greater depths on granite than on that the production of a perisarcal cover avoids any limestone (unpubl. obs.). Also, our data on circalittoral kind of negative interaction between the polyp and communities of northeast Sardinia showed that the dif- the substrate. ference in the number of species was not as great as for Our field data seem to suggest that the mineralogical the shallower, infralittoral communities. Obviously, properties of the substratum are long lasting, and con- substratum stability is a consequence not only of depth tinue to affect the established epibenthic communities. but also of its intrinsic hardness and consistency, which The contrast between rocks rich in or deprived of are of direct importance with respect to wave action quartz always resuIted evident on the first axis of cor- and bioboring. Hardness and consistency, which are in respondence analysis, indicating that this was the most turn related to the mineralogical composition, should important factor in differentiating the patterns of spe- be higher in granite than in limestone, thus suggesting cies composition and dominance. In all localities, the the reverse of the observed pattern. assemblages on quartz-rich rocks were less diverse In the study of ecological succession, Connell & and showed a simpler physiognomy. This could indi- Slayter (1977) proposed 3 different mechanisms of in- cate the difficulty to reach a 'mature' condition in pres- teraction between organisms colonising hard sub- ence of quartz, which acts as an inhibiting factor. Cer- strates: facilitation, inhibition, and tolerance. We think rano et al. (1998) suggested that crystalline quartz has that similar mechanisms might be recognised in the an evident negative effect on animals that colonise interaction between organisms and the substrate, sands, probably due to both the oxidant properties of which can facilitate, inhibit, or be neutral, not only to the crystal surface, generating silicon-based radicals, borers, but also to epilithic species, therefore affecting and to the formation of .OH radicals in the surround- the development of communities. ing aqueous environment (Marasas & Harington 1960, Animals colonising sands and algae on rocks, as dis- Langer & Nolan 1986, Shi et al. 1988, Vallyathan et al. cussed above, might provide examples of inhibition by 1988). In the case of rocky substrates, it seems reason- the substrate and of neutrality, respectively. A case of able to think that this inhibitory effect acts chiefly on facilitation may occur when an organism, requiring the early stages of colonisation, perhaps via the micro- high concentration of silica, uptakes this mineral di- bial communities of the primary film (Wahl 1989, John- rectly from the substrate, instead of from the water. son et al. 1997), and reduces as succession goes on. Gemmules of Spongilla lacustris L. reared in silica-free Many sessile organisms lay calcareous structures that water may produce a complete spicular complement in turn offer a secondary substrate to later colonists, using different kinds of silicates laid down on the thus annulling the effect of quartz. Most algae, for aquarium bottom as silica source (Jargensen 1944). instance, are not influenced by the mineralogical Volkmer-Ribeiro (pers. comm.) maintains that this phe- nature of the rock (Cabioc'h et al. 1992) and, once nomenon is also common in a natural environment: developed, operate a 'biological conditioning' of the Brazilian fresh water sponges produce normal spicules substrate, which has been considered as the most in temporary water pools filled only by completely sil- important factor structuring fauna1 communities (Ab- ica-free rain. In this habitat the only silica source is the biati et al. 1987. 1991, Simboura et al. 1995). Clearly, quartzitic sand of the bottom of the pools and sedi- disturbance may disrupt biological cover (Sousa 1984), mented diatoms and spicules. To date, the biochemical re-exposing organisms to the direct influence of the mechanism allowing the dissolution and the uptake by original substrate. This may explain why shallow- sponges of particulate silica is not known. Neverthe- water sessile communities on quartz-rich rocks always less the marine sponge Chondrosia reniformis is able appear in an early developmental phase, a fact com- to collect and dissolve quartz grains (Bavestrello et al. 24 8 Mar Ecol Prog Ser 193: 241-249. 2000

1995131, probably due to a considerable production of nuove tecnologie, l'energie e I'ambiente, Collana di studi ascorbic acid (Cerrano et al. 1999) ambientali. Roma The results of this study and the above considera- Bianchi CN, Cocito S, Morri C, Sgorbini S (1991) Rilevamento bionomico subacqueo. In: Abbiati M (ed) Lezioni del corso tions suggest that bio-mineralogy is likely to play a formativo per ricercatore scientific0 subacqueo. Interna- major role on benthic communities, selecting the biota tional School for Scientific Diving, Pisa, p 67-83 and affecting not only the initial colonisation, as seen Bingham %L (1992) Life histories in an epifaunal community: in the experiment of Eudendriumsettlement, but also coupling of adult and larval processes. Ecology 73: 2244-2259 later assemblages, as apparent from our field data. Boero F, Balduzzi A, Bavestrello G, Caffa B, Cattaneo Vietti R This potential role has been largely neglected to date (1986) Population dynamics of Eudendrium glomeratum and further studies are needed to prove its importance. (Cnidaria, Anthomedusae) on the Portofino Promontory (Ligurian Sea). Mar Biol92:81-85 Bon~A. Braga G, Conti S, Gelab R, Marchettl G, Passeri LD Acknowledgements. This research was financially supported (1969) Note illustrative della carta geologica dlItalia. by MURST Italian funds. Field actlvity received support from Foglio 83, Rapallo. Foglio 94, Chiavari. Servizio Geologico ENEA. Many thanks are due to all colleagues that helped d'ltalia, Roma during underwater work. Cabioc'h J, Floc'h JY, Le Toquin A, Boudouresque CF, Meinesz A, Verlaque M (1992) Guide des algues des mers d'Europe. Delachaux et Niestle, Neuchhtel LlTERATURE CITED Caffey HM (1982) No effect of naturally-occurring rock types on settlement or survival in the intertidal barnacle, Abbiati M, Bianchi CN, Castelli A (1987) Polychaete vertical Tesseropora rosea (Krauss). J Exp Mar Biol Ecol 63: zonation along a littoral cliff in the West Mediterranean. 119-132 PSZN I: Mar Ecol 8(1):33-48 Cattaneo-Vietti R, Sirigu AP, Tommei A (1988) Sea of Liguria, Abbiati M. Bianchi CN, Castelli A, Giangrande A, Lardicci C 2nd edn. Research Centre of the Union of Ligurian Cham- (1991) Distribution of polychaetes on hard substrates of ber of Commerce. Genoa the midlittoral-infralittoral transition zone. western Med- Cerrano C, Bavestrello G, Cattaneo-Vietti R (1997) Light iterranean. Ophelia 5(suppl):421-432 influence on planulae emission and settlement in Euden- Alvisi M, Bianchi CN. Colantoni P (1994) Le grotte sommerse driumglomeratum (Cnidaria, Hydrozoa). Biol Mar Medit del10 Scoglio delld Cappa (Isola del Giglio). Mem 1st Ital 4(1):30-33 Speleologia Ser 11 b:25-30 Cerrano C. Arillo A, Bavestrello G, Benatti U, Calcinai B. Cat- Anderson MJ, Underwood AJ (1994) Effects of substratum taneo-Vietti R, Cortesogno L, Gaggero L, G~ovineM, Puce on the recruitment and development of an intertidal S, Sara M (1998) Organism-quartz interactions in structur- estuarine fouling assemblage. J Exp Mar Biol Ecol 184: lng benthic comn~unities.towards a marine bio-mineral- 217-236 ogy? Ecol Lett 2:1-3 Balduzzi A, Bianchi CN, Cattaneo-Vietti K,Cerrano C, Cocito Cerrano C, Bavestrello G, Arillo A, Benatti U, Bonpadre S, S, Cotta S. Degl'Innocenti F, Diviacco G, Morgigni M, Cattaneo-Vietti R, Gaggero L. Giovlne M, Leone L, Luc- Morri C, Pansini M, Salvatori L, Senes L, Sgorbini S, chetti G, Sara M (1999) Calcium oxalate production in the Tunesi L (1994) Primi lineamenti di bionomia bentica marine sponge Cl~ondrosiarenifor~nis. Mar Ecol Prog Ser dell'lsola Gallinaria (Mar Ligure). In: Albertelli G, Catta- 179:297-300 neo-Vletti R, Piccazzo M (eds) Attl del 10" Congresso Connell JH (1983) On the prevalence and relative importance dell'Associazione Italiana di Oceanologia e Limnologia of interspecific competition: evidence from field experi- (AIOL),Genova, p 603-617 ments. Am Nat 122:661-696 Balduzzi A. Bianchi CN, Burlando B, Cattaneo-Vietti R, Man- Connell JH, Slayter R0 (1977) Mechanisms of succession in coni R, Morri C, Pansini M, Pronzato R, Sara M (1996) natural communities and their role in community stability Zoobenthos di substrato duro delle isole d~ Capraia e del and organization. Am Nat 111.1119-1144 Giglio (Arcipelago Toscano). Atti Soc Toscana Sci Nat Connell SD, Glasby TM (1999) Do urban structures influence Mem Ser A 102 (Supp1)(1995).124-135 local abundance and diversity of subtidal epibiota? A case Barry JP, Dayton PK (1991) Physical heterogeneity and the study from Sidney Harbour, Australia. Mar Environ Res organization of marine communities In: Kolasa J, Pickett 47:373-387 STA (eds) Ecological heterogeneity. Springer Verlag, New Den Hartog C (1972) Substratum. In: Klnne 0 (ed) Marme York, p 270-320 Ecology, Vol 1, Part 3. Wiley-Interscience, p 1277-1289 Bavestrello G, Cattaneo-Vietti R, Cerrano C, Danovaro R, Fanelli G, Piraino S, Belmonte G, Geraci S, Boero F (1994) Fabiano M (1995a) Annual sedimentation rates and role of Human predation along Apulian rocky coasts (SE Italy): the resuspension processes along a vertical cliff (Ligurian desertification caused by Lithophaga lithophaga (Mol- Sea, Italy). J Coastal Res 11(3):690-696 lusca) fisheries. Mar Ecol hog Ser 110:l-8 Bavestrello G, Arillo A, Benatti U, Cerrano C, Cattaneo-Vietti Gili JM, Hughes RG (1995) The ecology of marine benthic R. Cortesogno L, Gaggero L, Giovine M, Tonetti M, Sara hydroids. Oceanogr Mar Biol Annu Rev 33:351-426 h'! (19955: Q"=:!z dir~~!??!!"~!he cpnnge Chnndrnqia Guirhard F. Rourget E (19981 Topographic heteroqeneity, reniformis (Podera. Demosponglae). Nature 378:374-376 hydrodynamics, and benthic community structure: a Bianchi CN (1997) Climate change and biological response in scale-dependent cascade. Mar Ecol Prog Ser 171:59-70 the marine benthos. In: Piccazzo M led) Atti del 12" Con- I-Iiscock K (1987) Subtldal rock and shallow sediments using gresso deU'Associazione Italiana di Oceanologia e Lim- diving. In: Baker JM, WaLff WJ [eds) Biological surveys nologia IAIOL), Voi 1. Genova, p 3-20 of estuaries and coasts. Cambridge University Press, Cam- Bianchi CN, Morri C, Peirano A, Romeo G, Tunesi L (1983) bridge, p 198-237 Bibliogralia ecotipolog~ca sul Mar Ligure. Ente per le Holm ER, Cannon G, Roberts D, Schmidt AR, Sutherland JP, Bavestrello et al.: Structuring of marine eplbenthic cornrnunlties 249

Rittschof D (1997) The influence of initial surface chem- excavation of calcium carbonate substrates by boring istry on development of the fouling community at Beau- sponges. J Mar Biol Assoc UK 59:777-784 fort, North Carolina. J Exp Mar Biol Ecol 215:189-203 Russo GF, Cicogna F (1992) Exploitation of the date mussel, Johnson CR, Lewis TE, Nichols DS, Degnan BM (1.997) Bacte- Lithophaga lithophaga: problems and prospects. Boll Mus rial induction of settlement and metamorphosis in marine 1st Biol Univ Genova 56-57 (1990-1991):184-194 invertebrates. In: Lessios HA. Maclntyre IG (eds) Proceed- Riitzler R. Rieger G (1973) Sponge burrowing: fine structure ings of the 8th International Coral Reef Symposium. of Cliona Iampa penetrating calcareous substrata. Mar Smithsonian Tropical Research Institute, Panama, 2: Biol 21:144-162 1219-1224 Schoener TW (1983) Field experiments on interspecific com- Jsrgensen CB (1944) On the spicule-formation of Spongilla petition. Am Nat 122:240-285 lacustris (L.). 1 The dependence of the spicule formation Shi X, Dalal NS, Vallyathan V (1988) ESR evidence for the on the content of dissolved and solid shcic acid of the hydroxyl radical formation in aqueous suspension of milieu. Det Kg1 Dansk Vidensk Selsk Biol Medd 19:2-45 quartz particles and its possible significance to lipid per- Langer AM, Nolan RP (1986) Physicochemical properties of oxidation in silicosis. J Toxicol Environ Health 25:23?-245 quartz controlling biological activity. In: Goldsmith DF, Simboura N, Zenetos A, Thessalou-Legaki M. Pancucci MA, Winn DM, Shy CM (eds) Sihca, shcosis, and cancer: con- Nicolaidou A (1995) Benthic communities of the infra- troversy in occupational medicine. Praeger, New York littoral in the N. Sporades (Aegean Sea): a variety of Lebart L (1975) Validite des resultats en analyse des donnees. biotopes encountered and analysed. PSZN 1: Mar Ecol Centre de recherches et de documentation sur la consom- 16(4):283-306 mation, Paris, LL/cd No 4465 Smith F, Witman JD (1999) Species diversity in subtidal land- Legendre P, Legendre L (1998) Numerical ecology. Second scapes: maintenance by physical processes and larval English edition. Dev Environ Model 20:l-853 recruitment. Ecology 80(1):51-69 Levinton JS (1982) Marine Ecology. Prentice-Hall. Englewood Sornmer C (1992) Larval biology and dispersal of Eudendnurn Cliffs, NJ racemosum (Hydrozoa, Eudendriidae). In: Bouillon J. Lorenzoni GG, Chiesura-Lorenzoni F (1973) Considerazioni Boero F, Cicogna F, G~liJM. Hughes RG (eds) Aspects of conservazionistiche sulle isole di Tavolara, Molara, hydrozoan biology. Sci Mar 56:205-211 Molarotto e relativa costa sarda. In: Scalera-Liaci L (ed) Sousa WP (1984) The role of disturbance in natural communi- Atti del I11 simposio nazionale sulla conservazione della ties. Annu Rev Ecol Syst 15:353-391 natura, Vol2. Cacucci, Bari, p 409-435 Tortonese E (1958) Bionornia marina della regione costiera fra Marasas LW, Harington JS (1960) Some oxydative and Punta della Chiappa e Portofino (Riviera ligure di levante). hydroxylative action of quartz: their possible relationship Archo Oceanogr Limnol 11(2):167-210 to the development of silicosis Nature 188:1173-1174 Tortonese E (1962) Recenti ricerche sul benthos in ambienti McGuiness KA (1989) Effects of some natural and artificial litorali del Mare Ligure. Pubbl Staz Zoo1 Napoli 32 substrata on sessile manne organisms at Galeta Reef, (S~ppl):99-116 Panama. Mar Ecol Prog Ser 52:201-208 Underwood AJ (1997) Experiments in ecology. Cambridge Morri C, Bianchi CN, Damiani V, Peirano A, Romeo G, Tunesi University Press, Cambridge L (1988) L'ambiente marino tra Punta della Chiappa e Ses- Vallyathan V, Shi X, Dalal NS, I W, Castranova V (1988) Gen- tri Levante (Mar Ligure). profilo ecotipologico e proposta eration of free radicals from freshly fractured silica dust: & carta bionomica Boll Mus 1st Biol Univ Genova 52 potential role in acute silica-induced lung injury. Am Rev (Suppl):213-231 Respir Dis 138:1213-1219 Navone A, Bianchi CN, Ornl P, Ulzega A (1992) Saggio di Wahl M (1989) Marine epibiosis. 1. Fouling and antifouling: cartografia geomorfologica e bionomica nel parco marino some basic aspects. Mar Ecol Prog Ser 58:175-189 di Tavolara - Capo Coda Cavallo. Oebaha 17(2)(Suppl): Walters LJ, Wethey DS (1996) Settlement and early post-set- 469-478 tlement survival of sessile marine invertebrates on topo- Orsino F (1975) Flora e vegetazione delle isole Gallinara e graphically complex surfaces: the importance of refuge Bergeggl (Liguria occidentale). Webbia 29595-644 dimension and adult morphology. Mar Ecol Prog Ser 137. Pomponi SA (1979) Ultrastructure of cells associated with 161-171

Editorial responsibiljty: Otto fine (Editor), Submitted: February 26, 1999; Accepted: September 21, 1999 OldendorULuhe, Germany Proofs received from authorls): February 14, 2000