Effects Of Disturbance, Life Histories, And Overgrowth On Coexistence Of Algal Crusts And Turfs

Effects Of Disturbance, Life Histories, And Overgrowth On Coexistence Of Algal Crusts And Turfs

Laura Airoldi

LAURA AIROLDI [1]

Abstract. Coexistence of species is generally attributed to the interacting roles of competition, predation, and disturbance. Overgrowth is considered to be an important mechanism of competition for space, and species are often ranked in hierarchies based on their abilities to overgrow. In some marine habitats, however, encrusting algae dominate primary substrata despite a dense permanent cover of epiphytes, suggesting that factors other than competition could be important in influencing their distributions.

The spatial relationships and competitive interactions between encrusting algae and overgrowing filamentous, turf-forming algae were investigated on a subtidal rocky reef (Mediterranean Sea, Italy). Quantitative observations and field experiments were done from 1992 to 1998: (1) to investigate the relative patterns of distribution and abundance of crusts and turf and how they differed across space and time, (2) to test whether spatial relationships between crusts and turf were influenced by various characteristics of the substratum and by disturbance from wave action, (3) to investigate spatial and temporal patterns of recruitment and the mechanisms by which crusts colonize space, and (4) to test whether crusts and turf compete for space.

Crusts were always among the first colonizers of available bare rock and were subsequently overgrown by turf. Despite variations in spatial and temporal recruitment of crusts, this pattern was never reversed. Covers of crusts and turf were not significantly affected by disturbance. There was some evidence for competition, but this did not result in the local exclusion of either crusts or turf. Despite extensive and persistent cover of turf, encrusting algae were abundant and were able to live, grow, and reproduce beneath the turf over long periods with little adverse effect. Similarly, crusts did not limit the distribution of turf, although they slowed its rate of growth.

The capabilities of turf and crusts to overgrow and tolerate overgrowth, respectively, were identified as the major determinants of the structure of this two-layered assemblage and the probable basis of the success of these species in coexisting as the dominant algal forms. Overgrowth was not equivalent to competitive subordination and displacement, emphasizing the need for caution when interpreting competitive abilities from observed patterns of overgrowth. Overall, results suggested that, in some habitats, tolerance may be more important than competition in maintaining coexistence of species and influencing community structure.

Key words: algae, encrusting vs. turf-forming, coexistence of species; colonization; competition; disturbance effects; field experiment; life histories; overgrowth; Polysiphonia setacea; recruitment; reef, subtidal; rocky reef; spatial coexistence; tolerance.

INTRODUCTION

Space on the primary substratum is often a limiting resource in many marine and terrestrial habitats (Connell 1961, 1976, Dayton 1971, Yodzis 1978, Jackson 1979, Sebens 1986). In a variety of assemblages, manipulative experiments have shown that competition for space, or space-associated resources, may lead to the progressive elimination of all but a few competitively dominant species (Connell 1961, Paine 1966, 1984, Dayton 1971, Lubchenco 1978, Sousa 1979, Bertness and Ellison 1987). The existence of transitive hierarchical ranking of competitive abilities is an important component of models predicting the greatest number of species at intermediate levels of disturbance (Quinn 1982, Pickett and White 1985, Petraitis et al. 1989, Tilman 1990). According to these models, coexistence in space-limited habitats occurs as a result of selective damage to competitively superior species and/or of increased environmental heterogeneity through physical and biological disturbance (Connell 1961, 1978, Paine 1966, Menge 1976, Lubchenco 1978, Sousa 1979, 1984, Tilman 1982, Bertness and Ellison 1987). Alternative models have also been proposed to explain coexistence of species when disturbance is small (Jackson and Buss 1975, Buss and Jackson 1979, Vance 1984, Warner and Chesson 1985). According to these models, coexistence in space-limited habitats may occur because often no clear competitive dominants exist (Sutherland 1974, Jackson 1979, Buss 1980, Tanner et al. 1994). This view is supported by the observational and/or experimental evidence that competitive relationships may be non-hierarchical, and reversals in the outcomes of competition or competitive equivalence among species are frequent (Buss and Jackson 1979, Jackson 1979, Liddell and Brett 1982, Russ 1982, Goldberg and Werner 1983, Chornesky 1989, Benedetti-Cecchi and Cinelli 1996, Harris 1996, but see Paine 1984).

Overgrowth of one organism by another is considered one of the most frequent mechanisms of interference competition between space-limited organisms (Stebbing 1973, Jackson 1979, Schoener 1983, Denley and Dayton 1985, Sebens 1986, Olson and Lubchenco 1990). Overgrowth is assumed to be a disadvantage for the underlying organism. The negative effects of overgrowth may include reduced growth and fecundity, structural damage or death, and local extinction of the overgrown species (Seed and O’Connor 1981, Sebens 1986, Stevens 1987, Dittman and Robles 1991, Harris 1996). For these reasons, in some epibiotic communities, even marginal overgrowth of the surface of an organism is considered as evidence of its loss in an encounter with other organisms and of its competitive subordination (Wethey and Walters 1986), and competitive abilities of species are often ranked based on observed patterns of overgrowth (Stebbing 1973, Jackson 1979, Liddell and Brett 1982, Russ 1982, Breitburg 1984, Sebens 1986, Steneck et al. 1991 , Harris 1996). Some organisms, including sponges, barnacles, and encrusting coralline algae, can, however, survive overgrowth, without apparent damage, for indeterminate periods of time (Rutzler 1970, Sara 1970, Jernakoff 1985, Sebens 1986, Miles. and Meslow 1990, Morcom et al. 1997) or may even benefit from being overgrown (Vance 1978, Osman and Haugsness 1981, Bertness et al. 1983). Some species of sponges have been defined as specialized “supporters” because they have never been observed other than completely overgrown Rutzler 1970). In some habitats, apparently epilithic organisms are instead epiphytes of encrusting coralline algae that cover and replace bare rock, to the point that coralline crusts have been considered equivalent to primary substrata with respect to settlement of organisms (Dayton 1971, 1975, Menge 1976, Benedetti-Cecchi and Cinelli 1996). Other than a few proposals that epibiosis could be a solution to competition in some assemblages (Rutzler 1970, Lee and Ambrose 1989, Lohse 1993), th e possibility that overgrowth represents a mechanism of coexistence in space-limited habitats has been largely overlooked.

Understanding the mechanisms of competition and coexistence used by organisms is important for developing realistic models of community structure and interpreting how diversity is maintained in natural systems (Petraitis et al. 1989). Encrusting algae are good subjects to address questions about competition for space. Their essentially two-dimensional mode of growth often brings them into contact with other benthic organisms that tend to overgrow them. For this reason, encrusting algae are widely considered as subordinate to algae of other morphologies in the hierarchy of ability to compete for space (Littler and Littler 1980, Breitburg 1984, Dethier 1994, Steneck and Dethier 1994, but see Underwood 1980 and Padilla 1984). This supposed competitive inferiority contrasts with the evidence that encrusting algae are among the major holders of primary hard substrata in marine habitats within the euphotic zone (Littler 1972, Adey and Macintyre 1973, Steneck 1986, Dethier 1994, Kaehler and Williams 1997) and that they often coexist with their epiphytes (Miles and Meslow 1990, Parker and McLachlan 1990, Figueiredo 1997). This would suggest that the common interpretation of competitive ranking based on the ability to overgrow may not be adequate to explain patterns of distribution of encrusting algae and their interactions with other species.

Several mechanisms have been proposed to account for spatial dominance of encrusting algae despite their tendency to be overgrown by other species. These include dependence on disturbance or stressful conditions to remove or suppress epiphytes (Paine 1980, Steneck 1982, 1983, 1986, Hawkins and Harkin 1985, Kendrick 1991, Dethier 1994), morphological, anatomical, or chemical properties that slow or hinder overgrowth (Breitburg 1984, Johnson and Mann 1986, Figueiredo et al. 1996, Keats et al. 1997), and rapid recruitment and opportunistic life histories (Dethier 1981, 1994, Figueiredo 1997). All these models assume that competition has important effects on the relative distribution of encrusting algae and their epiphytes, but no consistent experimental evidence supports the generality of this assumption. Competition could be less important, or absent, in assemblages where encrusting algae tolerate overgrowth over long periods (Sebens 1986, Miles and Meslow 1990).

In this study, I investigated the spatial relationships and competitive interactions between encrusting and overgrowing filamentous, turf-forming algae on a subtidal rocky reef (Ligurian Sea, Italy). Encrusting algae and filamentous turfs often occur together in different habitats (Parker and McLachlan 1990, Williams and Carpenter 1990, Kendrick 1991, Airoldi et al. 1995, Figueiredo 1997), but the mechanisms that allow the persistent coexistence of these two growth forms have received little attention. I use Welden and Slauson’s 1986:25) definition of competition, i.e., “the induction of strain in one organism as a result of the use, defense, or sequestering of resource items by another organism,” where strain is a “deleterious, suboptimal physiological state” that is sometimes translated “into measurably reduced growth rate, size, or reproduction, or into increased mortality or emigration and thus into reduced fitness.” I report here the results of quantitative observations and field experiments done over 6 yr, from 1992 to 1998 (Table 1). The specific aims of these studies were: (1) to investigate the relative patterns of distribution and abundance of encrusting and turf-forming algae and how they change across space and over time; (2) to test whether spatial relationships between encrusting and turf-forming algae are influenced by certain characteristics of the substratum and by disturbance induced by storms; (3) to investigate spatial and temporal patterns of recruitment and the mechanisms by which encrusting algae colonize space in patches of bare rock; and (4) to test whether competition is important in influencing the distribution of encrusting and turf-forming algae in the study area.

STUDY AREA AND SPECIES

The research was done by scuba diving on a wave-exposed rocky cliff south of Livorno, Italy (43[degrees]30′ N, 10[degrees]20′ E, locality Calafuria, Fig. 1A and B), at depths of 10-18 m. This area is characterized by moderately large sediment loads, compared to other shallow coastal regions (Airoldi et al. 1996). The bottom consists of a sandstone platform that extends [tilde]300 m seaward to a depth of about 40 m. From [tilde]5 to 20 m, the platform is gently sloping ([tilde]10[degrees]-30[degrees]), and has eroded to form fields of dense outcrops. All outcrops have similar heights (around 50-70 cm). Based on the dimensions of their upper horizontal surfaces, outcrops can be separated into three size categories: small (upper surface [less than]1000 [cm.sup.2]), medium (upper surface 1000-2000 [cm.sup.2]) and large (upper surface [greater than]2000 [cm.sup.2]). From a random sample of 70 outcrops, the relative frequencies of these three size categories were estimated as 37.3%, 41.8%, and 20.9%, respectively. A total of 11 fields of outcrops, each [tilde]500 [m.sup.2] were used in the different studies (Fig. 1B). Hereafter, these will be indicated as “stations,” while the term “site” will refer to any smaller areas (6-25 [m.sup.2] located at random within each station (Fig. 1C).

On their upper surfaces, the outcrops support a rich assemblage of encrusting calcareous algae (Airoldi et al. 1995). These consist of non-geniculate coralline algae, and some species of Peyssonnelia, mostly P. harveyana P.L. et H.M. Crouan ex J. Agardth and P. squamaria (Gmelin) Decaisne. Hereafter, the non-specific term “crust” will be used to designate collectively this group of encrusting algae. Crusts are hardly visible because they are covered by a dense filamentous turf, consisting mostly of the species Polysiphonia setacea Hollenberg (Airoldi et al. 1995). This species is widely distributed on subtidal reefs in various areas of the Mediterranean basin. It occurs on rocks and as an epiphyte on crusts and develops dense mats of upright filaments up to 1 cm high from prostrate branches firmly attached to the substratum. A few erect algae, mostly Halimeda tuna (Ellis et Solander) Lamouroux, Dictyota dichotoma (Hudson) Lamouroux, and Padina pavonica (Linnaeus) Lamouroux are also present, but they do not f orm a canopy (Airoldi et al. 1995). Specific experiments on these components of the assemblage and on their interaction with the turf have been reported in previous papers (Airoldi and Cinelli 1997, Airoldi 1998).

At the depths where the study was done, macro-herbivores are scarce during most of the year, and these are mainly represented by herbivorous fishes. Sea urchins, chitons, limpets, and snails are frequent at shallower depths (Benedetti-Cecchi et al. 1998, L. Airoldi, personal observation), but they become virtually absent below 5 m in depth (personal observation). During the six years of this study, I have seen very few specimens of Sphaerechinus granularis (Lamark) on the bottom between the outcrops or on their vertical sides, but I never observed any urchins or other macro-herbivores on the upper surfaces of the outcrops. I have occasionally seen specimens of the herbivorous fish Sarpa salpa (Linnaeus) grazing on this assemblage, mainly during the spring, but this fish is otherwise rare during the year. So far, the causes of the notable lack of macro-grazers in the area are not known, but similar scarcity of herbivores has been observed also in other reefs impacted by sediments (D’Antonio 1986, McGuinness 1 987). It has been hypothesised that grazing by urchins or other herbivores might be deterred by the large amounts of sediment entrapped in the filamentous turf (Airoldi 1998, Airoldi and Virgilio 1998).

METHODS

Distribution and dynamics

Cover by turf made it difficult to quantify crusts. The distribution and dynamics of crusts and turf were therefore studied by combining nondestructive and destructive methods: the former allowed estimates of the abundance of turf and exposed crusts (i.e., not overgrown by turf), while the latter allowed estimates of the overall biomass and cover of crusts. Some data concerning turf have been reported in previous papers (Airoldi et al. 1995, Airoldi and Cinelli 1997, Airoldi 1998). Methods and results relative to those studies will be summarized here.

Covers of turf, exposed crusts, and bare rock were each estimated during three different periods (Table 1). Initially, observations were done only at one to two stations, while a larger range of spatial scales was considered in the third period. In the first period (July 1992 to June 1993), sampling was done monthly using photographs of areas of 13 X 17 cm (Airoldi and Cinelli 1997). Six plots were permanently located at one station (Fig. 1B, station 1). Percentage covers were estimated using the point-intercept technique. In the second period (June 1994 to October 1995), areas were 10 X 10 cm (Airoldi 1998). Three plots were permanently located at each of two stations (Fig. 1B, stations 2), about 100 m apart. Percentage covers were estimated in the field every 1–2 mo by means of the visual-estimation method (Dethier et al. 1993). In the third period (November 1995 to July 1997), sampling was done every 1-2 mo by photographs of areas of 10 X 10 cm. Nine plots were permanently located at each of three statio ns (Fig. 1B, stations 3), each hundreds of meters apart. The nine plots within each station were in three sites (tens of meters apart) with three plots per site (Fig. 1C and D). Percentage covers were estimated using the point-intercept technique.

The biomasses of crusts and turf were estimated at station 1 (Fig. 1B) by scraping three random areas of 13 X 17 cm in each of June, October, and December 1992 and March and June 1993 (Airoldi et al. 1995). The biomass of algae was measured as dry mass. Calcareous crusts were decalcified using HC1 to obtain a biomass comparable to that of turf.

Finally, the cover of crusts was estimated by a semidestructive method in July 1995. Sampling was done in areas of 5 X 5 cm after removing manually all the turf by using forceps. Percentage cover of crusts was then estimated in the field by the visual-estimation method. A plastic grid with 25 small quadrats (each corresponding to 4% of the total cover) was used. A score from 0 to 4% was given to crusts in each small quadrat, and these were summed to obtain total cover. Cover of crusts was quantified at different stations and in relation to the size of the rocky outcrops and the position on the surface of the outcrop, in order to test whether the spatial distribution of crusts varied in relation to the characteristics of the substratum. Sampling was done at three stations of 50 X 10 m (Fig. 1B, stations 4), randomly placed in the study area [tilde]50–100 m apart. At each station, sampling was done on the horizontal upper surfaces of six randomly chosen outcrops of each size category (small, medium, and large ). Three outcrops of each size were sampled in their centers, the remaining three along their edges. Two replicate areas were sampled on each outcrop. Data were analyzed using a four-way, mixed-model analysis of variance (ANOVA), with station (three levels) and outcrop (three levels) as random factors and size (small, medium, and large) and position (center and edge) as fixed factors. Station, size, and position were orthogonal to each other, while outcrop was nested in their interaction. Student-Newman-Keuls (SNK) tests were used for a posteriori multiple comparisons of means. In this and each of the following analyses, the assumption of homogeneity of variances was examined using Cochran’s C test.

Effects of disturbance

It has been suggested that the distribution of encrusting algae may be influenced by the regime of physical and/or biological disturbance that periodically removes their epiphytes (Paine 1980, Steneck 1986, Kendrick 1991, Dethier 1994). In the study area, herbivores were scarce (see Study area and species, above), and experiments have shown that disturbance by sediments affects the biomass of turf but not its cover (Airoldi 1998, Airoldi and Virgilio 1998). On this heavily wave-exposed reef, storms can remove patches of turf up to [tilde]500 [cm.sup.2] (personal observation), but the potential importance of these effects was not quantified in previous studies. Effects of storms vary over time and space and may also depend on the topography of the substratum (Denny 1995). The following studies were done to test whether disturbance by storms and some characteristics of the substratum (i.e., the size of the rocky outcrops and the position on the surfaces of the outcrops) influenced the abundance of turf and exp osed crusts (Table 1).

Cover of turf was measured from October 1995 to June 1997 at different sites and in relation to the size of the rocky outcrops, the position on the surface of the outcrop, and the potential disturbance by storms. Sampling was done at station 5 (Fig. 1B). This was divided into a grid of 20 sites, each 5 X 5 m. Sites were marked and numbered to guarantee that they were sampled only once during the research. Effects of storms on the abundance of turf were tested by sampling its cover immediately after a storm or after the sea had remained calm (relative to effects at the study depth) for at least one up to three months. Sampling was done at three randomly chosen times for each condition of sea. At each time, two sites were randomly sampled, up to a total of 12 sites sampled out of those available. In each site, sampling was done on the horizontal upper surfaces of eight randomly chosen outcrops of each size category (small, medium, and large). Four outcrops of each size were sampled in their centers, the remain ing four along their edges. One area of 5 x 5 cm was sampled on each outcrop. Percentage cover of turf in each area was estimated using the visual-estimation method. Data were analyzed using a five-way, mixed-model ANOVA, with disturbance (storm and calm), size (small, medium, and large) and position (center and edge) as fixed factors and time (three levels) and site (two levels) as random factors. Disturbance, size, and position were orthogonal to each other, while time and site were nested in disturbance and time, respectively.

Cover of exposed crusts was measured in the same station and sites used to quantify the distribution of turf, following the same sampling design. Sampling was done in randomly chosen patches naturally free of turf, In these patches, the percentage covers of live and dead crusts were quantified in areas of 5 X 5 cm using the visual-estimation method. Data were analyzed using the same ANOVA model as that used for cover of turf.

Reproduction and recruitment

Opportunistic life histories may account for persistence of competitively subordinate species (Connell and Slatyer 1977), but little is known about reproduction and recruitment of encrusting algae (Steneck 1986). The following studies were done to investigate spatial and temporal patterns of recruitment of crusts and to evaluate the importance of life history in influencing spatial dominance by crusts.

Spatial and temporal patterns of recruitment of crusts were investigated during three experiments in which patches of bare rock were produced by the use of hammer and chisel at different times and different locations (Table 1). Methods of the first two studies have been described previously (Airoldi and Cinelli 1997, Airoldi 1998) and are summarized here. The third experiment is part of a larger study about life histories and recruitment of algae in this assemblage. Major results are reported here, while spatial and temporal variability over different scales is discussed in Airoldi (in press). In all the experiments, new recruits of crusts were overgrown by turf. Therefore, I could not follow their growth after 2-3 mo. Destructive methods were used to investigate if crusts had been growing under the turf over longer periods.

In the first experiment (Airoldi and Cinelli 1997), six patches of 16 X 20 cm were scraped on two occasions (June and December 1992) at one station (Fig. 1B, station 1). Recruitment after 2 mo was sampled by photographs. Percentage covers were estimated using the point-intercept technique. Patches scraped in June 1992 were destructively sampled after 1 year (June 1993), and biomass of crusts was measured as decalcified dry mass. Patches scraped in December 1992 were sampled after 5 yr (January 1998). At that time, percentage covers of crusts were quantified by the visual-estimation method, after removal of overlying turf.

In the second experiment, patches of 20 x 20 cm were scraped in June and December 1994 (Airoldi 1998). Each time, three patches were scraped at each of two stations (Fig. 1B. stations 2) [similar to]100 m apart. Recruitment after 2 mo was sampled visually in the field. Percentage covers were quantified using the visual-estimation method. Patches were sampled again in January 1998 by the visual-estimation method, after removal of turf.

In the third experiment, patches of 15 X 15 cm were scraped at eight different times from October 1995 to September 1996. Each time, nine patches were scraped at each of three stations (Fig. 1B. stations 3) hundreds of meters apart. The nine patches within each station were in three sites (tens of meters apart) with three plots per site (Fig. 1C and D). Recruitment after 2-3 mo was sampled by photographs. Percentage cover of crusts was estimated by the point-intercept technique. Data were analyzed using a three-way ANOVA, with time (eight levels), station (three levels), and site (three levels) as random factors. Time and station were orthogonal to each other, while site was nested in each combination of times and stations.

Reproduction of crusts was measured in parallel with the third experiment. Crusts removed by scraping were preserved and examined for presence of reproductive structures under a stereomicroscope. Crusts were decalcified, and fertile and nonfertile portions of the thallus were separated. The wet biomass of each component was weighed. Fertility was measured as ratio of the biomass of fertile portions over the total biomass of crusts (fertile + nonfertile), expressed as a percentage. Data were analyzed using the same ANOVA model as that used for recruitment of crusts in the third experiment.

Competitive interactions

Overgrowing species capture space by smothering organisms that already occupy it, and overgrown species defend space by inhibiting other organisms (Denley and Dayton 1985, Olson and Lubchenco 1990). In both cases, organisms should be negatively affected by competition. Interactions between crusts and turf were investigated by two field experiments, in which these two components of the assemblage were manipulated independently (Table 1). I predicted that, if competition were occurring, removal of competing species should have positive effects (e.g., on development, abundance, reproduction) on the species not manipulated (Underwood 1986).

The first experiment lasted from August 1995 to July 1996 and tested the hypothesis that crusts actively inhibit the development of turf. The experiment was done using specimens of the most abundant coralline crust in the study area (named melobesloid sp. 1, identification in progress). This species was chosen because it forms large, thick crusts that are easier to manipulate than other crusts. Many non-geniculate coralline algae have been suggested to actively inhibit colonization of epiphytes by shedding epithallial cells (Millson and Moss 1985, Johnson and Mann 1986, Keats et al. 1994, Figueiredo et al. 1997). Shedding of melobesioid sp. 1 has been observed in the study area during summer months (personal observation) and was occurring at the beginning of this experiment for some of the crusts used.

The experiment was done across an area spanning stations 1, 2, and 4 (Fig. 1B). Twenty-four living crusts of [tilde]10 X 10 cm, naturally free of turf, were permanently marked and were assigned at random, six to each of the following treatments:

1) crusts removed using a hammer and chisel;

2) crusts dead, in order to separate the effects of crusts as living biological surfaces from effects possibly related to the morphology or other properties of inert crusts. Crusts were detached from the substratum, taken to the laboratory, dried at 60[degrees]C for 48 h until bleached and dead, then transplanted back to their original position using epoxy putty;

3) crusts live but manipulated, in order to detect any artefacts due to epoxy putty. Crusts were detached from the substratum as in the previous treatment but were immediately transplanted back to their position using epoxy putty; and

4) unmanipulated crusts, to measure the development of turf on natural living crusts.

The development of turf in the experimental plots was examined every 2-3 mo until July 1996. Photographs were taken of smaller central surfaces of 6 X 6 cm in order to avoid edge effects. Percentage cover of turf was recorded in the laboratory using the visualestimation method, by projecting the slides onto a grid of 25 small quadrats.

Effects of treatments on the rate of growth of the turf were analyzed using one-way ANOVA on the slopes of the regression lines of percentage covers over time. To test if initial effects resulted in significant differences at the end of the experiment, ANOVA was also done on percentage covers measured after 1 yr in July 1996.

The second experiment tested the hypotheses that overgrowth affects crusts negatively, and that their abundance, vitality, or fertility are, therefore, influenced by the regime of disturbance that removes overlying turf. The experiment was done from September 1995 to October 1996 at station 6 (Fig. lB), in 16 nearby sites of [tilde]6 [m.sup.2] In each site, three quadrats of 10 X 10 cm, naturally colonized by dense turf, were permanently marked with epoxy putty. Four sites were allocated at random to each of four treatments. This procedure should have minimized possible initial differences among treatments in the cover of crusts. Treatments were:

1) initial removal of turf. In these treatments, at the beginning of the experiment, turf was manually removed from the experimental quadrats using forceps. This was done very carefully to avoid damage to underlying crusts;

2) repeated removal of turf. Careful removal of turf as above was repeated once each month in order to keep cover by turf [less than]40% over the whole experiment;

3) abrasion. In these treatments, turf was thinned by monthly rubbing with a cotton cloth for 1 mm. This treatment affected cover of turf only marginally, mimicking effects of disturbance by sediments (Airoldi and Virgilio 1998); and

4) unmanipulated control. Turf was not manipulated, and crusts remained overgrown by turf during the entire experiment. In these sites, a greater number of quadrats were initially marked. At the end of the experiment, sampling was done in three quadrats selected at random from those with [greater than]95% cover of turf during the entire year.

The experimental plots were sampled one year after the beginning of the experiment, in October 1996. Before sampling the crusts, turf was removed from quadrats where it was present. Abundance of crusts was quantified in terms of both percentage cover and biomass. Percentage covers of living and dead crusts were measured in the field in smaller central areas of 5 X 5 cm using the visual-estimation method. Mortality was estimated as the ratio of cover of dead crusts over the total cover of crusts (dead + alive), expressed as a percentage. Three samples were then collected at random within each quadrat by driving a metal corer of 19-mm diameter through the crusts and into the substratum. Samples were analyzed in the laboratory for biomass and fertility using the same methods described previously.

Percentage cover and mortality were analyzed with a two-way ANOVA, with removal of turf (four levels of removal) as a fixed factor and site (four sites) as a random factor nested in turf. Biomass and fertility were analyzed with a three-way, mixed-model ANOVA, with removal of turf (four levels of removal) as a fixed factor, site (four sites) as a random factor nested in turf, and quadrat (three quadrats) as a random factor nested in site.

RESULTS

Distribution and dynamics

Turf grew on rocks and on crusts, and its average cover (including primary and secondary cover) was generally [greater than]80% (Fig. 2). Percentage covers of bare rock and exposed crusts were, therefore, very small. The amount of bare rock varied from virtually 0 to [less than]10%, with an exceptional peak of 20% in September 1994 (Fig. 2). Percentage cover of exposed crusts varied on average from virtually 0 to 24% (Fig. 2). Bare rock and exposed crusts were generally more abundant during autumn and winter (Fig. 2), when turf is thinner (Airoldi and Virgilio 1998). Dead crusts were also observed, but their cover was generally small (Fig. 2).

Despite the fact that crusts were largely and persistently overgrown by turf, they were abundant on the rocky substrata sometimes with biomass comparable to that of overlying filamentous algae (Fig. 3). Beneath the turf, crusts covered, on average, 28-73% of the rocky substratum (Fig. 4). Their distribution was heterogeneous, varying from station to station as a function of the size of the rocky outcrops and of the lateral or central position on their upper surfaces (Table 2, significant Station x Size x Position interaction). SNK tests, however, did not define specific patterns among means.

Effects of disturbance

Wave-caused disturbance during storms had no significant effects on turf cover (Fig. 5, Table 3). Effects of storms were generally small and localized. Cover of turf varied from time to time and as a function of the size of the rocky outcrops and of the position on their surfaces (Table 3, significant Size X Position X Time interaction). SNK tests indicated that differences between outcrops were more pronounced during periods of calm sea, but no consistent patterns were detected. Conversely, no significant differences were observed at larger spatial scales among sites (Table 3), in agreement with observations from previous studies (Airoldi 1998, Airoldi and Virgilio 1998).

Percentage covers of exposed crusts were not influenced by disturbance from storms and were rather constant over time (Table 4). Patches of substratum not covered by turf were colonized, on average, by 56% of live crusts (Fig. 6). Some dead crusts were also present, but their abundance changed significantly from time to time (Table 4), probably reflecting a major presence of dead crusts during the summer. The abundance of exposed crusts varied significantly from site to site as a function of the size of the outcrops and the position on their surfaces (Table 4, significant Site X Size X Position interaction). SNK tests indicated only few significant differences, generally when the abundance of crusts was smaller along the edges of outcrops.

Reproduction and recruitment

Crusts were always among the first colonizers of bare rock, reaching up to 57% cover in 2 mo (Fig. 7). After that time, crusts became largely covered by turf, which always regained space very quickly by lateral vegetative encroachment of prostrate axes (Airoldi and Cinelli 1997, Airoldi 1998; Airoldi, in press). Recruitment of crusts onto filamentous turf was never observed. Crusts occupied space mostly by recruitment of propagules. Slow lateral invasion by surrounding specimens was also occasionally observed, but, owing to cover of turf, it was not possible to estimate the contribution of this process to the long-term recovery of crusts.

After 1 yr the average biomass of crusts that had recruited in patches cleared in June 1992 was 1.1 [plus or minus] 0.68 g/[m.sup.2] (means [plus or minus] 1 SE, n = 6). This value was relatively small compared to the biomass of adult crusts measured in the surrounding assemblage in June 1993 (Fig. 3). Long-term observation of plots cleared in December 1992, June 1994, and December 1994, however, indicated that initially settled crusts continued to grow beneath the turf over long time. In January 1998, after 3-5 yr from the beginning of colonization, percentage cover of crusts had reached values of 30 [plus or minus] 5.1% (n = 6), 41 [plus or minus] 4.3% (n = 6), and 39 [plus or minus] 5.2% (n = 9) in plots cleared in December 1992, June 1994, and December 1994, respectively. These values represented a large increase with respect to initial covers measured after 2 mo of colonization in the same plots (Fig. 7A and B).

Early patterns of recruitment of crusts were largely variable in time and space. Average percentage covers of crusts after 2-3 mo from the beginning of colonization ranged from 1.5% in patches cleared in January 1995 to 57% in patches cleared in September 1996 (Fig. 7C, Table 5). Recruitment was generally less in patches cleared in the winter (Fig. 7C), but temporal patterns differed across stations (Table 5, significant Time X Station interaction). Rates of recruitment also differed among sites within stations (Table 5). This variability was not related to differences in the fertility of crusts (correlation analysis; [r.sup.2] = 0.01, P [greater than] 0.05). Although fertility varied through time (Fig. 8), differences among times were not significant (Table 6). Fertility was similar across different sites, but varied significantly among stations (Table 6).

Competitive interactions

The rate of growth of the turf was influenced by the presence of melobesioid sp. 1 (ANOVA on slopes of the regression lines: [F.sub.3,20] = 9.25, P [less than] 0.001). Turf colonized faster in plots where melobesioid sp. 1 had been removed (SNK test: Removed [greater than] Dead = Live = Unmanipulated), reaching percentage covers of 100% after only 9 mo (Fig. 9). Cover increased over time in the other treatments, reaching mean percentage covers of 50-72% after 1 yr (Fig. 9), but cover was still significantly higher in removal treatments (ANOVA on data from July 1996: [F.sub.3,20] = 5.54, P [less than] 0.01). The effects of melobesioid sp. 1 were apparently not related to its properties as a living biological surface. No differences were observed between treatments with live as opposed to dead specimens (Fig. 9), suggesting that reduced growth of turf on this encrusting alga might be influenced by physical and/or structural characteristics of the surface. No differences were observed among live and unmanipulat ed crusts (Fig. 9) indicating that the use of epoxy putty in transplanting did not produce any important artefact.

Complete overgrowth by turf over 1 yr did not affect cover, mortality, or fertility of underlying crusts (Fig. 10). No significant differences for any of these variables were observed among unmanipulated control plots and experimental treatments where turf had been removed or abraded (two-way ANOVAs, Cover: [F.sub.3,12] = 2.08, P [greater than] 0.05, Mortality: [F.sub.3,12] = 0.29, P [greater than] 0.05; three-way ANOVA, Fertility: [F.sub.3,12] = 1.4, P [greater than] 0.05). In contrast, the biomass of crusts was significantly greater in initial and repeated removal and abrasion treatments than in the unmanipulated control plots (Fig. 10; three-way ANOVA, [F.sub.3,12] = 3.85, P [less than] 0.05; SNK test: Initial = Repeated Abrasion [grater than] Control). Since covers of crusts were similar among treatments, increased biomass in removal and abrasion treatments probably reflected an increase in the thickness of the thallus.

DISCUSSION

Primary space was a limited resource in the study area. The amount of bare rock was always small, varying on average from virtually 0 to [less than]10%. Under space-limited conditions, coexistence of species is generally interpreted as a result of several mechanisms leading to the spatial and/or temporal segregation of potential competitors, including disturbances or consumers that create open space, trade-offs in the competitive abilities of species to gain and hold available space, environmental heterogeneity of resources, and fluctuating rates of recruitment (Dayton 1971, Menge 1976, Connell 1978, Lubchenco 1978, Sousa 1979, Tilman 1982, 1990, Pickett and White 1985, Warner and Chesson 1985, Bertness and Ellison 1987, Petraitis et al. 1989, Lavorel and Chesson 1995). In this algal assemblage, variability in rates of recruitment and in the characteristics of the substratum probably influenced the heterogeneous distribution of crusts (see also Kaehler and Williams 1997). Covers of crusts and turf, however, w ere not significantly affected by the regime of disturbance, and competition for space did not result in the local exclusion of either of these growth forms, at least within the range of temporal scales considered. Crusts and turf persisted in the study area as the dominant algal forms, and were able to coexist on the same space. The capability of crusts and turf to tolerate each other appeared to be the major determinant of the structure of this two-layered assemblage.

Crusts were very efficient in procuring and holding primary substratum. Despite spatial and temporal variability in fertility and rates of recruitment, their propagules were always among the first colonizers of patches of bare rock, as observed also in other studies (Adey and Vassar 1975, Dethier 1981, Kendrick 1991, Figueiredo 1997). This life-history trait allowed crusts to occupy space before relatively slower encroachment of the turf by lateral vegetative propagation (Airoldi 1998; Airoldi, in press). Large reproductive output and rapid recruitment are generally interpreted as opportunistic strategies adopted by competitively inferior species to avoid extinction in marine and terrestrial systems (Connell and Slatyers 1977, Grime 1977, Bazzaz 1979, Sousa 1979, 1984, Sebens 1986; but see Davis and Wilce 1987, Underwood and Anderson 1994). It is generally believed that, once encrusting algae become overgrown, they decay unless disturbance removes their epiphytes or creates new open space for recruitment (Steneck 1986, Kendrick 1991, Dethier 1994). Growth of encrusting algae, however, had never been followed over long periods of time. The results of semidestructive samples after 3-5 yr from the beginning of colonization indicated that crusts not only were able to hold initially procured primary substratum over long periods, but were even able to grow and colonize additional space despite being covered by turf. Although extensively and persistently overgrown, crusts occupied on average [greater than]50% of the rocky substratum, and thus constituted the major occupants of primary space. These results clearly indicate that being overgrown is not always equivalent to competitive subordination and displacement, and emphasize the need for caution when interpreting competitive abilities from observed patterns of overgrowth (Buss and Jackson 1979, Jackson 1979, Liddell and Brett 1982, Quinn 1982, Russ 1982, Breitburg 1984, Wethey and Walters 1986). Crusts were much more efficient in exploiting space than generally predic ted on the basis of hierarchies of abilities to overgrow.

Encrusting algae, like other organisms that tend to be overgrown, have been suggested to compete for space through preemption of primary substratum and subsequent inhibition of recruitment of other species (Breitburg 1984, Dethier 1994, Figueiredo et al. 1996). One mechanism often proposed to explain how encrusting algae can inhibit their potential competitors is thallus shedding, which is well documented among nongeniculate coralline algae (Johnson and Mann 1986, Keats et al. 1994, 1997, Figueiredo et al. 1997). In this study, the rate of growth of turf was significantly greater on bare rock than on melobesioid sp. 1. This effect however, was not related to some property of this coralline crust as a living biological surface. This suggests that observed shedding of the thallus was not responsible for slowing down the growth of turf, and that effects of melobesioid sp. 1 were probably related to its smooth morphology or to other characteristics of its surface (see also Figueiredo et al. 1997). The ecological significance of thallus shedding was not specifically investigated in this study. Nevertheless, I observed that shedding generally occurred at the beginning of the summer, which might suggest a role in the seasonal regulation of thickness of crusts, as proposed by Keats et al. (1994).

Although crusts slowed down the growth of turf, this interaction had weaker effects on the distribution of turf than suggested by other researchers (Breitburg 1984, Figueiredo et al. 1996), and did not result in a situation of “standoff” or “delay/tie” (sensu Connell 1976 and Russ 1982, respectively). Turf covered on average [greater than]80% of space, despite the presence of underlying crusts. Cover of turf was not even significantly influenced by intense disturbance from storms, and previous experiments indicated that turf was resistant to disturbance by sediments (Airoldi and Cinelli 1997, Airoldi and Virgilio 1998). Life-history traits associated with the capability of turf to dominate space and resist disturbance have been discussed in a previous paper (Airoldi 1998). Of particular importance was the ability of the dominant filamentous species Polysiphonia setacea to propagate vegetatively by physically resistant, prostrate axes. This ability might also have reduced the importance of interactions with c rusts. By colonizing space through lateral encroachment, P. setacea was relieved from attachment and settlement of propagules, which in other species of algae are much affected by the characteristics of substratum (Fletcher and Callow 1992). It has also been suggested that the duration of studies can influence the perception of the importance and nature of interactions among species (Brooker and Callaghan 1998). Previous experiments showing strong negative effects of encrusting algae on their epiphytes were generally very short (Breitburg 1984, Figueiredo et al. 1996, 1997). In my present study, effects of crusts on rate of growth of turf were intense at the beginning of the experiment, but decreased during the following months. A similar temporal trend was observed also by Underwood (1980).

Overgrowth could potentially be a disadvantage for the underlying organisms, especially if they are photoautotrophic. In some marine and terrestrial systems it has been shown that epibionts may reduce the inputs of light, oxygen, and nutrients to the host plant, resulting in its reduced growth, reduced fecundity, structural damage, or even death (Seed and O’Connor 1981, D’Antonio 1985, Stevens 1987, Dittman and Robles 1991, Harris 1996). In my present study, however, crusts survived, grew, and reproduced beneath the turf with little adverse effect. Overgrowth by turf limited the biomass of crusts, but did not affect their cover or fertility and did not result in increased mortality. While tolerance of encrusting algae to overgrowth has also been observed by other authors (Underwood 1980, Scanes 1986, Sebens 1986, Miles and Meslow 1990, Kendrick 1991), surprisingly I have not found any study quantifying negative effects of overgrowth on encrusting algae. Perhaps negative effects are rarely investigated becaus e it is commonly assumed that there will be negative effects of overgrowth.

Mechanisms that allow crusts to survive, grow, and reproduce under a dense cover of epiphytes are not known. Steneck (1986) and Miles and Meslow (1990) have discussed the role that secondary pit connections and cell fusions present in thalli of crustose corallines could have on the lateral translocation of photosynthates from parts of the thallus exposed to light to parts overgrown, but the importance of this process has not been quantified. Effects of turf could also be influenced by its thickness. Williams and Carpenter (1990) showed that the amount of light penetrating filamentous turfs to the underlying encrusting algae varied largely, depending on intensity of grazing by herbivores that reduce the thickness of the turf. In the current study, presence of herbivores was generally very low (see Study area and species, above), but thickness of turf varied over time and space probably in response to variable disturbance by sediments (Airoldi and Virgilio 1998). Tolerance of epiphytes by encrusting algae migh t probably also be favored by their adaptation to shaded environments and, in particular, by their high photosynthetis efficiency at low light intensities (Littler et al. 1986). This characteristic allows encrusting algae to typically dominate areas that are unproductive due to low light availability (reviewed in Vadas and Steneck [1988] and Steneck and Dethier [19943), including deep habitats, cryptic environments, and canopy understoreys.

Negative effects of secondary cover could also be masked by possible direct or indirect benefits. For example, turf could protect underlying crusts from abrasion by sediments, which is intense in the area (Airoldi et al. 1996) and seems to negatively affect recruitment and growth of crusts (L. Airoldi, unpublished data). Another possibility is that overgrowth by turf might give indirect advantages to crusts in relation to competition with other algae with an erect morphology. Results from other experiments in the study area suggest that erect algae need bare rock to settle; thus, like crusts, they exploit primary substratum (Airoldi 1998; in press). Erect algae, however, are quickly outcompeted by turf (Airoldi 1998), while crusts are able to persist over long periods without important damage by turf. Thus, turf might relieve crusts from competition for primary space by outcompeting erect algae. So far, fast propagation of turf has made my manipulations of its abundance inadequate to test this hypothesis ove r long periods. Nevertheless, the hypothesis that overgrowth might confer indirect advantages during exploitation for space has also been suggested for other assemblages (Osman and Haugsness 1981, Lee and Ambrose 1989, Miles and Meslow 1990). This suggests that understanding the ecological role of overgrowth may require consideration not only of direct effects of contact between encountering species but also of indirect interactions with other species.

In this algal assemblage the success of crusts and turf in coexisting on the same space over several years as the dominant algal forms and in the absence of significant effects of disturbance appeared to depend on their life histories. Of particular importance were the respective abilities of turf and crusts to overgrow and to tolerate overgrowth. These abilities minimized the importance of competition, and allowed crusts and turf to occupy collectively more space than they could have done by using only primary substratum. This conclusion has important implications for the interpretation of mechanisms of spatial coexistence. Regardless of different roles attributed to competition and disturbance, most of the classical models assume that primary space is a potentially limiting resource and explain coexistence as a consequence of processes allowing its partitioning among species (Dayton 1971, Yodzis 1978, Buss and Jackson 1979, Sousa 1979, Paine 1984, Pickett and White 1985). The results of the present study a nd other studies, however, indicate that overgrowth is not always equivalent to competitive displacement, and that tolerance may allow long-term spatial overlapping among species (Rutzler 1970, Sara 1970, Osman and Haugsness 1981, Miles and Meslow 1990, Parker and McLachlan 1990). Thus, use of secondary substrata may be a determinant in maintaining spatial coexistence between species (Lee and Ambrose 1989, Lohse 1993). If secondary substrata are taken into account, then more space is available, and competition becomes less important than generally assumed.

This study emphasizes that an understanding of the mechanisms of coexistence between species requires attention to life histories and to how available resources are used. The strength and nature of interactions between dominant species may in fact change, depending on how they influence each others’ abilities to acquire resources (Vance 1984, Holmgren et al. 1997, Brooker and Callaghan 1998). While current models of community organization emphasise competitive and consumer-prey interactions between organisms (but see Bertness [1998] and references therein), this study suggests that, in some assemblages, tolerance might be more important than competition in maintaining co-existence of species and influencing community structure. I suggest that the potential importance of life-history traits that enhance reciprocal tolerance of species should be further explored.

ACKNOWLEDGMENTS

I am grateful to M. Abbiati and L. Benedetti-Cecchi for stimulating discussions and ideas. I wish to thank M. Abbiati, M. J. Anderson, M. Beck, L. Benedetti-Cecchi, T. Glasby, B. Menge, A. J. Southward, A. J. Underwood, and two anonymous reviewers for their helpful comments and careful reviews of the manuscript, and D. W. Keats for his advice on melobesioid sp. 1. M. Abbiati, F Lenzi, A. Vannucci, and M. Virgilio helped with the field work. F. Lenzi and A. Vannucci also helped with analysis of data and preparation of graphs. F Cinelli and A. J. Underwood provided space and facilities at the Dipartimento di Scienze dell’Uomo e dell, Ambiente (Universita di Pisa) and at the Centre for Research on Ecological impacts of Coastal Cities (University of Sydney), respectively. The work was supported by a Postdoctoral Fellowship of the Universita di Pisa and by a Postdoctoral Fellowship of the University of Sydney (the latter while writing the paper). Some of the data were part of my Dissertation at the University of Genova.

(1.) Dipartimento di Scienze dell’ Uomo e dell’ Ambiente, Universita di Pisa, Pisa, Italy and Centre for Research on Ecological Impacts of Coastal Cities, University of Sydney, Sydney, New South Wales, Australia

(1.) Present address: Scienze Ambientali, Universita di Bologna, Via Tombesi dall’Ova 55, I-48100 Ravenna, Italy. E-mail: lairoldi@ambra.unibo.it

LITERATURE CITED

Adey, W. H., and I. G. Macintyre. 1973. Crustose coralline algae: a re-evaluation in the geological sciences. Geological Society of America Bulletin 84:883-904.

Adey, W. H., and J. M. Vassar. 1975. Colonization, succession and growth rates of tropical crustose coralline algae (Rhodophyta, Cryptonemiales). Phycologia 14:55-69.

Airoldi, L. 1998. Roles of disturbance, sediment stress, and substratum retention on spatial dominance in algal turf. Ecology 79:2759-2770.

Airoldi, L. In press. Responses of algae with different life-histories to temporal and spatial variability of disturbance in subtidal reefs. Marine Ecology Progress Series.

Airoldi, L., and F. Cinelli. 1997. Effect of sedimentation on subtidal macroalgal assemblages: an experimental study from a Mediterranean rocky shore. Journal of Experimental Marine Biology and Ecology 215:271-290.

Airoldi, L., M. Fabiano, and E Cinelli. 1996. Sediment deposition and movement over a turf assemblage in a shallow rocky coastal area of the Ligurian Sea. Marine Ecology Progress Series 133:241-251.

Airoldi, L., F. Rindi, and F. Cinelli. 1995. Structure, seasonal dynamics and reproductive phenology of a filamentous turf assemblage on a sediment influenced, rocky subtidal shore. Botanica Marina 38:227-237.

Airoldi, L., and M. Virgilio. 1998. Responses of turf-forming algae to spatial variations in the deposition of sediments. Marine Ecology Progress Series 165:271-282.

Bazzaz, F. A. 1979. The physiological ecology of plant succession. Annual Review of Ecology and Systematics 10:351-371.

Benedetti-Cecchi, L., F. Bulleri, and F. Cinelli. 1998. Density dependent foraging of sea urchins in shallow subtidal reefs on the west coast of Italy (western Mediterranean). Marine Ecology Progress Series 163:203-211.

Benedetti-Cecchi, L., and F. Cinelli. 1996. Patterns of disturbance and recovery in littoral rock pools: nonhierarchical competition and spatial variability in secondary succession. Marine Ecology Progress Series 135:145-161.

Bertness, M. D. 1998. Searching for the role of positive interactions in plant communities. Trends in Ecology and Evolution 13:133-134.

Bertness, M. D., and A. M. Ellison. 1987. Determinants of patterns in a New England salt marsh plant community. Ecological Monographs 57:129-147.

Bertness, M. D., P. O. Yund, and A. F. Brown. 1983. Snail grazing and the abundance of algal crusts on a sheltered New England rocky beach. Journal of Experimental Marine Biology and Ecology 71:147-164.

Breitburg, D. L. 1984. Residual effects of grazing: inhibition of competitor recruitment by encrusting coralline algae. Ecology 65:1136-1143.

Brooker, R. W., and T. V. Callaghan. 1998. The balance between positive and negative plant interactions and its relationships to environmental gradients: a model. Oikos 81:196-207.

Buss, L. W. 1980. Competitive intransitivity and size frequency distributions of interacting populations. Proceedings of the Natural Academy of Sciences (USA) 77:5255-5259.

Buss, L. W., and J. B. C. Jackson. 1979. Competitive networks: nontransitive competitive relationships in criptic coral reef environments. American Naturalist 113:223-234.

Chornesky, E. A. 1989. Repeated reversals during spatial competition between corals. Ecology 70:843-855.

Connell, J. H. 1961. The influence of interspecific competition and other factors on the distribution of the barnacle Chthamalus stellatus. Ecology 42:710-723.

Connell, J. H. 1976. Competitive interactions and the species diversity of corals. Pages 51-58 in G. O. Mackie, editor. Coelenterate ecology and behavior. Plenum, New York, New York, USA.

Connell, J. H. 1978. Diversity in tropical rain forest and coral reefs. Science 199:1302-1310.

Connell, J. H., and R. O. Slatyer. 1977. Mechanisms of succession in natural communities and their role in community stability and organization. American Naturalist 111:1119-1144.

D’Antonio, C. M. 1985. Epiphytes on the rocky intertidal red alga Rhodomela larix (Rhodophyta, Ceramiales): negative effects for the host and food for herbivores. Journal of Experimental Marine Biology and Ecology 26:197-218.

D’Antonio, C. M. 1986. Role of sand in the domination of bard substrata by the intertidal alga Rhodomela larix. Marine Ecology Progress Series 27:263-275.

Davis, A. N., and R. T. Wilce. 1987. Floristics, phenology, and ecology of the sublittoral marine algae in an unstable cobble habitat (Plum Cove, Cape Ann, Massachusetts, USA). Phycologia 26:23-34.

Dayton, P. K. 1971. Competition, disturbance, and community organization: the provision and subsequent utilization of space in a rocky intertidal community. Ecological Monographs 45:137-159.

Dayton, P. K. 1975. Experimental evaluation of ecological dominance in a rocky intertidal algal community. Ecological Monographs 45:137-159.

Denley, E. J., and P. K. Dayton. 1985. Competition among macroalgae. Pages 511-530 in M. M. Littler and D. S. Littler, editors. Handbook of phycological methods. Ecological field methods: macroalgae. Cambridge University Press, Cambridge, UK.

Denny, M. 1995. Predicting physical disturbance: mechanistic approaches to the study of survivorship on wave-swept shores. Ecological Monographs 65:371–418.

Dethier, M. N. 1981. Heteromorphic algal life histories: the seasonal pattern and response to herbivory of the brown crust, Ralfsia californica. Oecologia 49:333–339.

Dethier, M. N. 1994. The ecology of intertidal algal crusts: variation within a functional group. Journal of Experimental Marine Biology and Ecology 177:37–71.

Dethier, M. N., E. S. Graham, S. Cohen, and L. M. Tear. 1993. Visual versus random-point percent cover estimations: “objective” is not always better. Marine Ecology Progress Series 110:9–18.

Dittman, D., and C. Robles. 1991. Effects of algal epiphytes on the mussel Mytilus californianus. Ecology 72:286-296.

Figueiredo, M. A. de O., 1997. Colonization and growth of crustose coralline algae in Abrolhos, Brazil. Proceedings of the Eighth International Coral Reef Symposium 1:689–694.

Figueiredo, M. A. de O., J. M. Kain (Jones), and T. A. Norton. 1996. Biotic interactions in the colonization of crustose coralline algae by epiphytes. Journal of Experimental Marine Biology and Ecology 199:303–318.

Figuciredo, M. A. de O., T. A. Norton, and J. M. Kain (Jones). 1997. Settlement and survival of epiphytes on two intertidal crustose coralline alga. Journal of Experimental Marine Biology and Ecology 213:247–260.

Fletcher, R. L., and M. E. Callow. 1992. The settlement, attachment and establishment of marine algal spores. British Phycological Journal 27:303–329.

Goldberg, D. E., and P. A. Werner. 1983. Equivalence of competitor plant communities: a null hypothesis and a field experimental approach. American Journal of Botany 70: 1098–1104.

Grime, J. P. 1977. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 111:1169–1194.

Harris, P. M. 1996. Competitive equivalence in a community of lichens on rock. Oecologia 108:663–668.

Hawkins, S. J., and E. Harkin. 1985. Preliminary canopy removal experiments in algal dominated communities low on the shore and in the shallow subtidal on the Isle of Man. Botanica Marina 28:223–230.

Holmgren, M., M. Scheffer, and M. A. Huston. 1997. The interplay of facilitation and competition in plant communities. Ecology 78:1966–1975.

Jackson, J. B. C. 1979. Overgrowth competition between encrusting cheilostome ectoprots in a Jamaican cryptic reef environment. Journal of Animal Ecology 48:805–823.

Jackson, J. B. C., and L. W. Buss. 1975. Allelopathy and spatial competition among coral reef invertebrates. Proceedings of the Natural Academy of Sciences (USA) 72: 5160–5163.

Jernakoff, P. 1985. The effects of overgrowth by algae on the survival of the intertidal barnacle Tesseropora rosea Krauss. Journal of Experimental Marine Biology and Ecology 94:89–97.

Johnson, C. R., and K. H. Mann 1986. The crustose coralline alga, Phymatolithon Foslie, inhibits the overgrowth of seaweeds without relying on herbivores. Journal of Experimental Marine Biology and Ecology 96:127-146.

Kaehler, S., and G. A. Williams. 1997. Do factors influencing recruitment ultimately determine the distribution and abundance of encrusting algae on seasonal tropical shores? Marine Ecology Progress Series 156:87–96.

Keats, D. W., M. A. Knight, and C. M. Pueschel. 1997. Antifouling effects of epithallial shedding in three crustose coralline algae (Rhodophyta, Corallinales) on a coral reef. Journal of Experimental Marine Biology and Ecology 213: 281–293.

Keats, D. W., P. Wilton, and G. Meneveldt. 1994. Ecological significance of deep-layer sloughing in the eulittoral zone coralline alga, Spoagites yendoi (Foslie) Chamberlain (Corallinacee, Rhodophyta) in South Africa. Journal of Experimental Marine Biology and Ecology 175: 145–154.

Kendrick, G. A. 1991. Recruitment of coralline crusts and filamentous turf algae in the Galapagos archipelago: effect of simulated scour, erosion and accretion. Journal of Experimental Marine Biology and Ecology 147:47–63.

Lavorel, S., and P. Chesson. 1995. How species with different regeneration niches coexist in patchy habitats with local disturbances. Oikos 74:103–114.

Lee, H., and W. G. Ambrose. 1989. Life after competitive exclusion: an alternative strategy for a competitive inferior. Oikos 56:424–427.

Liddell, W. D., and C. E. Brett. 1982. Skeletal overgrowths among epizoans from the Silurian (Wenlockian) Waldron Shale. Paleobiology 8:67–78.

Littler, M. M. 1972. The crustose corallinacee. Oceanography and Marine Biology Annual Review 10:311-347.

Littler, M. M., and D. S. Littler. 1980. The evolution of thallus form and survival strategies in benthic marine macroalage: field and laboratory tests of a functional form model. American Naturalist 116:25–44.

Littler, M. M., D. S. Littler, S. M. Blair, and J. N. Norris. 1986. Deep-water plant communities from an uncharted seamount off San Salvador Islands, Bahamas: distribution, abundance, and primary productivity. Deep-Sea Research 33:881–892.

Lohse, D. P. 1993. The importance of secondary substratum in a rocky intertidal community. Journal of Experimental Marine Biology and Ecology 166:1–17.

Lubchenco, J. 1978. Plant species diversity in a marine intertidal community: importance of herbivore food preference and algal competitive abilities. American Naturalist 112:23–39.

McGuinness, K. A. 1987. Disturbance and organisms on boulders. I. Patterns in the environment and the community. Occologia 71:409–419.

Menge, B. A. 1976. Organization of the New England rocky intertidal community: role of predation, competition, and environmental heterogeneity. Ecological Monographs 46: 355-393.

Miles, A. K., and E. C. Meslow. 1990. Effects of experimental overgrowth on survival and change in the turf assemblage of a giant kelp forest. Journal of Experimental Marine Biology and Ecology 135:229-242.

Millson, C., and B. L. Moss. 1985. Ultrastructure of the vegetative thallus of Phyinatolithon lenormandii (Aresh. in J. Ag.) Adey. Botanica Marina 28:123-132.

Morcom, N. F, S. A. Ward, and W. J. Woelkerling. 1997. Competition of epiphytic nongeniculate corallines (Curallinales, Rhodophyta): overgrowth is not victory. Phycologia 36:468–471.

Olson, A. M., and J. Lubehenco. 1990. Competition in seaweeds: linking plant traits to competitive outcomes. Journal of Phycology 26:1–6.

Osman, R. W., and J. A. Hougsness. 1981. Mutualism among sessile invertebrates: a mediator of competition and predation. Science 211:846–848.

Padilla, D. K. 1984. The importance of form: differences in competitive ability, resistance to consumers and environmental stress in an assemblage of coralline algae. Journal of Experimental Marine Biology and Ecology 79:105–127.

Paine, R. T. 1966. Food web complexity and species diversity. American Naturalist 100:65–75.

Paine, R. T. 1980. Food webs: linkage, interaction strength and community infrastructure. Journal of Animal Ecology 49:667-685.

Paine, R. T. 1984. Ecological determinism in the competition for space. Ecology 65:1339-1348.

Parker, T., and J. McLachlan. 1990. Relationship between cover of Chondrus crispus (Gigartinales, Rhodophyta) and Phymatolithon (Corallinales, Rhodophyta) on friable rock substrata. Hydrobiologia 204-205:247-251.

Petraitis, P. S., R. E. Latham, and R. A. Niesenbaum. 1989. The maintenance of species diversity by disturbance. Quarterly Review of Biology 64:393-418.

Pickett, S. T. A., and P. S. White, editors. 1985. The ecology of natural disturbance and patch dynamics. Academic Press, London, UK.

Quinn, J. F. 1982. Competitive hierarchies in marine benthic communities. Oecologia 54:129-135.

Russ, G. R. 1982. Overgrowth in a marine epifaunal community: competitive hierarchies and competitive networks. Oecologia 53:12-19.

Rutzler, K. 1970. Spatial competition among porifera: solution by epizoism. Oecologia 5:85-95.

Sara, M. 1970. Competition and co-operation in sponge populations. Pages 273-284 in W. G. Fry, editor. The biology of porifera. Symposium of the Zoological Society of London Number 25.

Scanes, P. R. 1986. Interactions of crustose coralline algae with grazers. Thesis. University of Sydney, Sydney, New South Wales, Australia.

Schoener, T. W. 1983. Field experiments on interspecific competition. American Naturalist 122:240-285.

Sebens, K. P. 1986. Spatial relationships among encrusting marine organisms in the New England subtidal zone. Ecological Monographs 56:73-96.

Seed, R., and R. J. O’Connor. 1981. Community organization in marine algal epifaunas. Annual Review of Ecology and Systematics 12:49-74.

Sousa, W. P. 1979. Disturbance in marine intertidal boulder fields: the nonequilibrium maintenance of species diversity. Ecology 60:1225-1239.

Sousa, W. P. 1984. Intertidal mosaics: patch size, propagule availability, and spatially variable patterns of succession. Ecology 65:1818-1935.

Stebbing, A. R. D. 1973. Observations on colony overgrowth and spatial competition. Pages 173-183 in G. P. Larwood, editor, Living and fossil bryozoa. Recent advances in research. Academic Press, New York, New York, USA.

Steneck, R. S. 1982. A limpet–coralline alga association: adaptations and defenses between a selective herbivore and its prey. Ecology 63:507-522.

Steneck, R. S. 1983. Escalating herbivory and resulting adaptive trends in calcareous algal crusts. Paleobiology 9:44-61.

Steneck, R. S. 1986. The ecology of coralline algal crusts: convergent patterns and adaptive strategies. Annual Review of Ecology and Systematics 17:273-303.

Steneck, R. S., and M. N. Dethier. 1994. A functional group approach to the structure of algal-dominated communities. Oikos 69:476-498.

Steneck, R. S., S. D. Hacker, and M. N. Dethier. 1991. Mechanisms of competitive dominance between crustose coralline algae: an herbivore-mediated competitive reversal. Ecology 72:938-950.

Stevens, G. C. 1987. Lianas as structural parasites: the Barsera simarubra example. Ecology 68:77-81.

Sutherland, J. P. 1974. Multiple stable points in natural communities. American Naturalist 108:859-873.

Tanner, J. E., T. P. Hughes, and J. H. Connell. 1994. Species coexistence, keystone species, and succession: a sensitivity analysis. Ecology 75:2204-2219.

Tilman, D. 1982. Resource competition and community structure. Princeton University Press, Princeton, New Jersey, USA.

Tilman, D. 1990. Constraints and trade-offs: toward a predictive theory of competition and succession. Oikos 58:3-15.

Underwood, A. J. 1980. The effects of grazing by gastropods and physical factors on the upper limits of distribution of intertidal macroalgae. Oecologia 46:201-213.

Underwood, A. J. 1986. The analysis of competition by field experiments. Pages 240-268 in J Kikkawa and D. J. Anderson, editors. Community ecology: pattern and process. Blackwells, Melbourne, Australia.

Underwood, A. J., and M. J. Anderson. 1994. Seasonal and temporal aspects of recruitment and succession in an intertidal estuarine fouling assemblage. Journal of the Marine Biological Association of the United Kingdom 74:563-584.

Vadas, R. L., and R. S. Steneck. 1988. Zonation of deep water benthic algae in the Gulf of Maine. Journal of Phycology 24:338-346.

Vance, R. R. 1978. A mutualistic interaction between a sessile marine clam and its epibionts. Ecology 59:679-685.

Vance, R. R. 1984. Interference competition and the coexistence of two competitors on a single limiting resource. Ecology 65:1349-1357.

Warner, R. R., and P. L. Chesson. 1985. Coexistence mediated by recruitment fluctuations: a field guide to the storage effect. American Naturalist 125:79-787.

Welden, C. W., and W. L. Slauson. 1986. The intensity of competition versus its importance: an overlooked distinction and some implications. Quarterly Review of Biology 61:23-44.

Wethey, D. S., and L. J. Walters. 1986. Quantifying spatial patterns of overgrowth in epibenthic communities. Marine Ecology Progress Series 29:271-278.

Williams, S. L., and R. C. Carpenter. 1990. Photosynthesis/photon flux density relationships among components of coral reef algal turfs. Journal of Phycology 26:36-40.

Yodzis, P. 1978. Competition for space and the structure of ecological communities. Lecture notes in Biomathematics. Volume 25. Springer-Verlag, New York, New York, USA.

Results of ANOVA on percent cover of crusts at

different stations and in relation to the size of

the rocky outcrops and the position on the surface

of the outcrops.

Source of variation [+] df MS F Denominator MS for F ratio

Station, St 2 3027 4.82 [*] Outcrop(St X S X P)

Size, S 2 861 0.51 St X S

Position, P 1 1076 5.83 St X P

St X S 4 1684 2.68 [*] Outcrop(St X S X P)

St X P 2 184 0.29 Outcrop(St X S X P)

S X P 2 21 0.01 St X S X P

St X S X P 4 1467 2.40 [*] Pooled term

Outcrop(St X S X P) 36 629 1.05 Residual

Residual 54 599

Pooled term [++] 90 611

Notes: Cochran’s C test: C = 0.101, P [greater than] 0.05. There was no need to transform the data for analysis.

(*.)P [less than] 0.05.

(+.)Factors are: Station (three stations; random), Size (small, medium, and large; fixed), Position (edge and center; fixed), and Outcrop (three outcrops, nested in the interaction Station X Size X Position; random).

(++.)Pooled term = Outcrop(St X S X P) + Residual.

Results of ANOVA on percent cover of

turf at different sites, at different

times, and in relation to the size of

the rocky outcrops, the position on

the surface of the outcrops, and the

potential disturbance by storms.

Denominator MS for

Source of variation [+] df MS F F ratio

Disturbance, D 1 4942 1.42 T(D)

Size, S 2 797 2.13 T(D) X S

Position, P 1 340 6.41 T(D) X P

D X S 2 427 1.14 T(D)X S

D X P 1 158 2.97 T(D) X P

S X P 2 22 0.05 T(D) X S X P

D X S X P 2 131 0.26 T(D) X S X P

Time(D), T(D) 4 3483 14.45 [**] A(T(D))

Site(T(D)), A(T(D)) 6 241 1.07 Residual

T(D) X S 8 374 1.31 A(T(D)) X S

A(T(D)) X S 12 286 1.27 Residual

T(D) X P 4 53 0.61 A(T(D)) X P

A(T(D)) X P 6 87 0.39 Residual

T(D) X S X P 8 497 4.11 [*] A(T(D)) X S X P

A(T(D)) X S X P 12 121 0.54 Residual

Residual 216 225

Notes: Cochran’s C test: C = 0.0822, P [greater than] 0.05. There was no need to transform the data for analysis.

(*.)P [less than] 0.05,

(**.)P [less than] 0.01.

(+.)Factors are: Disturbance (storm and calm; fixed), Size (small, medium, and large; fixed), Position (edge and center; fixed), Time (three times; random, nested in Disturbance), and Site (two sites; random, nested in Time).

Results of ANOVA on percent cover of crusts live

and dead in patches naturally free of turf.

Source of Alive Dead

variation df MS F MS F

Disturbance, D 1 3998 1.36 913 0.10

Size, S 2 2570 1.44 1682 1.81

Position, P 1 2734 1.69 282 0.30

D X S 2 1927 1.08 157 0.17

D X P 1 659 0.41 40 0.04

S X P 2 2939 3.19 1986 1.02

D X S X P 2 486 0.53 243 0.12

Time(D), T(D) 4 2943 2.04 9030 8.45 [*]

Site(T(D)), A(T(D)) 6 1440 1.63 1069 2.37 [*]

T(D) X S 8 1790 1.25 931 1.31

A(T(D)) X S 12 1433 1.62 711 1.58

T(D) X P 4 1622 1.60 931 0.64

A(T(D)) X P 6 1012 1.14 1459 3.24

T(D) X S X P 8 922 0.47 1947 3.24 [**]

A(T(D)) X S X P 12 1966 2.22 [*] 1192 1.63

Residual 216 885 450 2.65 [**]

Notes: Sampling was done at different sites, at different times, and in relation to the size of the rocky outcrops, the position on the surface of the outcrops, and the potential disturbance by storms. Cochran’s C test: for alive, C = 0.04, P [greater than] 0.05; for Dead, C = 0.08, P [greater than] 0.05. There was no need to transform the data for analysis. Denominator MS values for F ratios are as in Table 3.

(*.)P [less than] 0.05, (**.)p [less than] 0.01.

(+.)Factors are: Disturbance (storm and calm; fixed), size (small, medium, and large; fixed), Position (edge and center; fixed), Time (three times; random, nested in Disturbance), and Site (two sites; random, nested in Time).

Results of ANOVA on ANOVA on percent cover of crusts

recruited after 2-3 mo in patches of bare rock cleared at

different times and different locations.

Denominator

Source of MS for

variation [+] df MS F F ratio [++]

Time, T 7 1.927 19.96 [***] A(T X St)

Station, St 2 0.001 0.01 A(T X St)

T X St 14 0.097 3.87 [***] A(T X St)

Site(T X St),

A(T X St) 48 0.025 2.58 [***] Residual

Residual 144 0.009

Notes: Cochran’s C test: C = 0.107, P [greater than] 0.05. For purposes of analysis an angular transformation was applied to the data.

(***.)P [less than] 0.001.

(+.)Factors are: Time (eight times; random), Station (three stations; random), and Site (three sites; random, nested in Time X Station).

Results of ANOVA of fertility of crusts at

different times and different locations.

Source of

variation [+] df MS F

Time, T 7 3419 1.40

Station, St 2 16280 6.69 [**]

T X St 14 2434 1.38

Site (T X St) 48 1767 1.27

Residual 144 1387

Notes: Cochran’s C test: C = 0.03, P [greater than] 0.05. There was no need to transform the data for analysis.

(**.)P [less than] 0.01.

(+.)Factors are: Time (8 times; random), Station (3 stations; random), and Site (3 sites; random, nested in Time X Station). Denominator MS values for F ratio are as in Table 5.

COPYRIGHT 2000 Ecological Society of America

COPYRIGHT 2000 Gale Group