Biodiversity & Conservation

The rockpool environment varies depending on factors such as pool depth, surface area, volume, orientation to sunlight, shading, internal topography, sediment content and type, together with wave exposure, shore height, and hence flushing rate, and the presence of absence of freshwater runoff. As a result, no two rockpools exhibit exactly the same physio-chemical conditions and exhibit large spatial variation in community structure, even between adjacent pools at the same shore height (Ganning, 1971; Metaxas & Scheibling, 1993; Metaxas et al., 1994). Any given rockpool is effectively unique (Metaxas & Scheibling, 1993). Therefore, while the organisms detailed below have been recorded in this rockpool biotope, not all, or in some case few, of the organisms, and hence the relationships between them, may occur in any one rockpool.

As communities in rockpools remain constantly submerged and the danger of desiccation is absent, it might be expected that rockpools provide an easier environment in which to live for marine life than drying rock surfaces, and that species from regions lower on the shore would be able to extend much further up the shore. However, the rockpool environment differs from that of the surrounding emergent rock surfaces, so that not all species that thrive on the surrounding rock occur abundantly in rockpools and much of the lower shore open rock fauna is absent from rockpools (Lewis, 1964). Rockpools constitute a distinct environment for which physiological adaptations by the flora and fauna may be required (Lewis, 1964; Metaxas & Scheibling, 1993). The following description is based on reviews by Lewis (1964), Ganning (1971) and Metaxas & Scheibling (1993), the species listed in the MNCR database (JNCC, 1999) and additional references as cited. Macroalgae such as kelps, fucoids, red and green algae, erect and encrusting corallines provide primary productivity either directly to grazing invertebrates and fish or indirectly, to detritivores and decomposers, in the form of detritus and drift algae or as dissolved organic material and other exudates. Benthic microalgae and phytoplankton (e.g. diatoms) also add to primary productivity.

Where present, large macroalgae such as Halidrys siliquosa and laminarians (e.g. Laminaria digitata, Saccharina latissima and Saccorhiza polyschides) and fucoids (e.g. Fucus serratus, Fucus vesiculosus) shade the substratum (depending on density) so that understorey plants tend to be shade tolerant red algae. Understorey algae, by effectively restricting access to the substratum, may also inhibit or restrict recruitment of other species of macroalgae (Hawkins & Harkin, 1985; Hawkins et al., 1992).

Macroalgae compete for space with sessile invertebrates such as sponges, hydroids, ascidians and bryozoans.

Macroalgae provide substrata and refuges for a variety of invertebrates and epiphytic algae. The stipes and lamina of Laminaria spp. may support bryozoans (e.g. Membranipora membranacea or Electra pilosa) and grazing blue-rayed limpets (Helcion pellucida), while their holdfasts provide additional refuges for meiofauna and small invertebrates. If present, the stipes of Laminaria hyperborea may support numerous epiphytes such as Palmaria palmata, Phycodrys rubens and Cladophora rupestris (Goss-Custard et al., 1979). Where present, Halidrys siliquosa provide substratum for epiphytes, depending on location, including microflora (e.g. bacteria, blue green algae, diatoms and juvenile larger algae), Ulothrix and Ceramium sp., hydroids (e.g. Obelia spp.), bryozoans (e.g. Scrupocellaria spp.), and ascidians (e.g. Apilidium spp., Botryllus schlosseri, and Botrylloides leachi) (Lewis, 1964; Moss, 1982; Connor et al., 1997a).

The macroalgae provide refuges for small invertebrates, such as isopods, amphipods, ostracods and copepods. Corallina officinalis provides a substratum for small spirorbids e.g. Spirorbis corallinae, which is only found on Corallina officinalis. Increasing density of Spirorbis corallinae was shown to increase the species richness of the epiphytic fauna (Crisp & Mwaiseje, 1989). The invertebrate fauna of Corallina officinalis is detailed in £ELR.Coff£.

Amphipods, isopods (e.g. Idotea granulosa) and other mesoherbivores graze the epiphytic flora and senescent macroalgal tissue, which may benefit the macroalgal host, and may facilitate dispersal of the propagules of some macroalgal species (Brawley, 1992b; Williams & Seed, 1992). Mesoherbivores also graze the macroalgae but do not normally adversely affect the canopy (Brawley, 1992b).

Grazers of periphyton (bacteria, blue-green algae and diatoms) or epiphytic algae include harpacticoid copepods, the limpets Patella vulgata and Patella ulyssiponensis, the blue-rayed limpet Helcion pellucidum, and gastropods such as Gibbula cineraria, Gibbula umbilicalis, Littorina saxatilis, Littorina littorea, Littorina obtusata and Rissoa spp. Limpets and littorinids also graze macroalgal sporelings and green algae especially.

Coralline algae are probably relatively grazing resistant (Littler & Kauker, 1984) and few species graze the corallines directly except perhaps chitons and limpets of the genus Tectura.

Grazing by littorinids and gammarid amphipods has been shown to significantly affect macroalgal abundance and diversity. For example in cage experiments in littoral fringe pools, Parker et al. (1993) found that gammarid amphipods significantly reduced the erect macroalgal canopy, while littorinids grazed microalgae and macroalgal sporelings, and prevented the establishment of erect and encrusting algal canopies. Both groups reduced the species richness of the algal canopy (Parker et al., 1993). In tidepools in Nova Scotia, Chapman (1990) and Chapman & Johnson (1990) reported that grazers (especially littorinids) reduced the abundance of Fucus spp. sporelings and juveniles but increased the abundance of ephemeral algae, while having no effect on the encrusting red alga Hildenbrandia rubra (Metaxas & Scheibling, 1993). Conversely, Lubchenco (1978) noted that the addition of littorinids to mid-shore pools in Massachusetts decreased the abundance of dominant Ulva spp. (as Enteromorpha spp.) in favour of Chondrus crispus (Metaxas & Scheibling, 1993) a less palatable red alga. Wolfe & Harlin (1988a) noted that Rhode Island tidepools with the highest littorinid densities had the lowest abundance of macroalgae. Similarly, removal of the limpet Patella vulgata from high tidal pools at Lough Ine resulted in an increased abundance of Ulva (as Enteromorpha) sp. (Goss-Custard et al., 1979). Where present, suspension feeders include barnacles (e.g. Semibalanus balanoides), the mussel Mytilus edulis, hydroids, tubeworms ( e.g. Spirorbis spp. and Pomatoceros spp.), ascidians, bryozoans and sponges. However, the abundance of barnacles and mussels in rockpools is usually low (Lewis, 1964), presumably due to heavy predation on juveniles by the dog whelk Nucella lapillus and crabs (e.g. Carcinus maenas and Cancer pagurus).

The sediment provides habitat for deposit feeding annelids e.g. Arenicola marina and terebellids.

Scavengers include shrimp (e.g. Palaemon serratus), brittlestars (e.g. Amphipholis squamata), cushion stars (Asterina gibbosa) and hermit crabs (e.g. Pagurus bernhardus).

Invertebrate predators include turbellarians and nemerteans feeding on small invertebrates such as copepods and small gastropods. Lower to mid shore rockpools provide refugia for dog whelks Nucella lapillus feeding on barnacles and small mussel within the rockpool and /or leaving the rockpool to forage at high tide. Similarly, crabs such as Carcinus maenas and Cancer pagurus are generalist predators of gastropods (e.g. littorinids) and bivalves as well as scavengers. Passive carnivores include sea anemones such as Anemonia viridis, Actinia equina and Urticina felina. In addition, intertidal fish such as the shanny Lipophrys pholis and gobies prey on small invertebrates such as copepods, amphipods and isopods.

As with grazing, predation pressure is potentially higher in mid to low shore rockpools, since predators can continue feeding irrespective of the state of the tide. For example, in New South Wales, whelks were shown to reduce the abundance of barnacles, tubeworms and limpets (Fairweather, 1987; Metaxas & Scheibling, 1993). Dethier (1984) concluded that harpacticoid copepod abundance in mid to low shore pools was low due to the presence of predators such as fish and to a lesser extent anemones. The reduced abundance of barnacles and mussels observed in rockpools (Lewis, 1964) is probably partly due to increased predation pressure.

Conditions within rockpools are the consequence of prolonged separation from the main body of the sea, and physico-chemical parameters within them fluctuate dramatically (Pyefinch, 1943; Ganning, 1971; Daniel & Boyden, 1975; Goss-Custard et al., 1979; Morris & Taylor, 1983; Huggett & Griffiths, 1986; Metaxas & Scheibling, 1993; Metaxas et al., 1994). In general, larger and deep rockpools low on the shore tend to correspond to the sublittoral habitat with a more stable temperature and salinity regime. In contrast, small and shallow pools higher on the shore are especially influenced by insolation, air temperature and rainfall, the effects of which become more significant towards the high shore, where pools may be isolated from the sea for a number of days or weeks (Lewis, 1964).

Weather conditions exert a considerable influence on temperature and salinity. Water temperature in pools follows the temperature of the air more closely than that of the sea.

Temperature
The temperature of pools fluctuates with air temperature and sunlight, and tend to warm throughout the day, especially if in direct sunlight (Daniel & Boyden, 1975; Goss-Custard et al., 1979). Shaded pools exhibit less dramatic changes in temperature (Daniel & Boyden, 1975). For example, the temperature of an high shore pool exposed to direct sunlight rose quickly in the morning to a maximum of 25 °C, while a shaded high shore pool only rose by 1 °C, even though air temperatures reached 20 °C (Daniel & Boyden, 1975). In addition, deeper pools may become stratified, with warmer water near the surface and cooler near the bottom (Daniel & Boyden, 1975), primarily due to sunlight. They noted that pool temperatures remained almost constant at night and suggested that pool temperatures would fluctuate slowly during the day under overcast conditions. In deeper pools, the vertical temperature gradation present in summer, may reverse during winter owing to density stratification, so that ice may form (Naylor & Slinn, 1958). Morris & Taylor (1983) reported warmer water at the bottom of the pools and cooler at the surface, which they attributed to cooling of the surface water by wind. Examples of temperature ranges reported for mid to low shore pools include annual maxima and minima of 1-25 °C and 2-22 °C (Morris & Taylor, 1983), a diurnal range of 24 °C (day) and 13 °C (night) for a mid shore pool (Daniel & Boyden, 1975), and surface water temperature ranges of 14-19.25 °C and 15.5-20.75 °C in mid shore pools (Pyefinch, 1943). Smaller (or shallow pools) are more affected by climatic change in temperature than larger and/or deeper ones (Ganning, 1971). Morris & Taylor (1983) noted that temperature showed the greatest seasonal variation of all the physical parameters examined. In summer, the minimum recorded temperatures were greater than the maximum temperatures recorded in winter, and the daily temperature ranges were greater in summer than in winter, in both high and low shore pools (Morris & Taylor, 1983).

Salinity
High air temperatures cause surface evaporation of water from pools, so that salinity steadily increases, especially in pools not flooded by the tide for several days. Alternatively, high rainfall will reduce pool salinity or create a surface layer of brackish/nearly fresh water for a period. In addition, freezing of surface water increases the salinity of the underlying water (Ganning, 1971; Metaxas & Scheibling, 1993). Daniel & Boyden (1975) and Morris & Taylor (1983) reported little variability in salinity over one tidal cycle, and Ganning (1971) suggested that changes in salinity were of limited importance. However, heavy rain resulted in a layer of low salinity water on the surface of pools. Morris & Taylor (1983) stated that a low salinity layer of 2-10 mm was normal but after one storm the low salinity layer increased in depth, eventually resulting in a homogeneous pool of brackish water. Morris & Taylor (1983) reported an annual salinity range in mid to low shore pools of 26-36.5 ppt.

The extent of temperature and salinity change is affected by the frequency and time of day at which tidal inundation occurs. If high tide occurs in early morning and evening the diurnal temperature follows that of the air, whilst high water at midday suddenly returns the temperature to that of the sea (Pyefinch, 1943). Rockpools in the supralittoral, littoral fringe and upper eulittoral are liable to gradually changing salinities followed by days of fully marine or fluctuating salinity at times of spring tide (Lewis, 1964).

Oxygenation
The biological community directly affects oxygen concentration, carbon dioxide concentration and pH, and are themselves affected by changes in the chemical parameters. Throughout the day, algae photosynthesize and produce oxygen, the concentration of which may rise to three times its saturation value, so that bubbles are released (Pyefinch, 1943; Ganning, 1971; Daniel & Boyden, 1975; Goss-Custard et al., 1979; Morris & Taylor, 1983; Metaxas & Scheibling, 1993). In summer, rockpools are likely to be supersaturated with oxygen during the day (Pyefinch, 1943). For example, the greatest range of oxygen saturation of 101.7% occurred in a seaweed dominated, sediment floored pool, which reached over 190%saturation on some days (Pyefinch, 1943). Daniel & Boyden (1975) noted that a mid shore, seaweed dominated pool reached 194% saturation (ca 15 mg O₂/l) but that oxygenation was also marked in shaded pools. A pool with dense fauna exhibited a maximum saturation of 210% (Pyefinch, 1943). During photosynthesis algae absorb carbon dioxide and as concentrations fall, the pH rises. Morris & Taylor (1983) recorded pH values >9 in rockpools on the Isle of Cumbrae.

At night changes occur in the opposite direction as respiration utilizes much of the available oxygen and pH decreases. Daniel & Boyden (1975) reported oxygen depletion at night, with mid to low shore pools reduced to 8-44% saturation. They noted that the crab Carcinus maenas leaves the pools at night, and that other species with the ability to air-breathe could also do so, e.g. limpets, littorinids, and the shanny Lipophrys pholis. They also observed that shrimps gathered at the edge of high shore pools at night, presumably to take advantage of the better oxygenated surface layer (Daniel & Boyden, 1975). Daily fluctuation in oxygen concentration and pH also vary seasonally, and with the height of the pool on the shore or pool depth (Daniel & Boyden, 1975; Morris & Taylor, 1983; Metaxas & Scheibling, 1993). Morris & Taylor (1983) noted variation in oxygen concentration with pools with pool depth and proximity to pool algae. Again, inundation by the rising tide causes sudden changes in oxygenation, either sudden drops during the day or increases at night.

Ganning (1971) noted that the temperature of sediment at the bottom of pools showed little variation in temperature, and noted that diurnal fluctuations of greater than 1 °C were rarely observed in bottom sediments of pools deeper than 0.5 m, although a temperature of fall 0.5-1 °C was observed at the water/sediment interface. The presence of sediment in pools increases the risk of scour, which is likely to vary seasonally, increasing in winter storms. In addition, anoxic conditions within the sediment could potentially cause significant and sudden decreases in oxygen levels if the sediment was stirred up by wave action, although no evidence of this effect was found.

Seasonal change in communities
Tidepool community structure has been show to vary markedly over time, particularly with season (Metaxas & Scheibling, 1993). However, most studies have examined tidepools overseas, or different tidepools communities to those that occur in this biotope (Dethier, 1984; Wolf & Harlin, 1988a,b; Kooistra et al., 1989; Metaxas et al., 1994) so that the species concerned are very different. Seasonal changes in macroalgal cover and diversity were associated with changes in water temperature, light intensity and day length, and reduced grazing pressure from littorinids in winter, although the community types were relatively stable over time in Rhode Island pools (Wolfe & Harlin, 1988a,b). Metaxas et al. (1994) noted that sheet forming algae (e.g. Ulva lactuca) were found throughout the year, filamentous forms (e.g. Cladophora and Ceramium) were present in late spring and summer (although other studies found them to be present from late spring to late autumn) while thick leathery and encrusting forms did not vary seasonally. However, macroalgal diversity was lowest in summer and autumn, especially in mid shore pools, probably due to grazing. Metaxas et al. (1994) also noted that the abundance of mussels, littorinids and whelks in Rhode Island peaked in summer due to recruitment but varied significantly between pools.

Red algae exhibit seasonal variation in growth and reproduction and red algal turf declines in abundance during the winter months, partly due to die back and abrasion during winter storms. Although protected from wave action in deep pools, macroalgae will be particularly susceptible to damage and abrasion by wave action and winter storms in shallow sediment filled pools. For example, maximum growth of Furcellaria lumbricalis occurs in March/April (Austin, 1960b) and release of carpospores and tetraspores occurs in December/January (Bird et al., 1991). Reproductive bodies are present on the gametophytes of Ahnfeltia plicata between July and January and mature carposporophytes occur between October and July (Maggs & Pueschel, 1989). However, in the Bristol Channel, Bamber & Irving (1993) noted that the biomass of Corallina officinalis increased steadily through spring and summer and began to decline after July. Mastocarpus stellatus (as Gigartina stellata) was reported have a perennial holdfast, losing many erect fronds in winter, which grow back in spring (Dixon & Irvine, 1977). Osmundea pinnatifida also shows seasonal variation in growth, expanding its perennial holdfast in June to September, and producing erect fronds from October onwards reaching a maximum in February to May (Maggs & Hommersand, 1993). Corallina officinalis may be overgrown by epiphytes, especially during summer. This overgrowth regularly leads to high mortality of fronds due to light reduction (Wiedemann, pers. comm.). The ephemeral green seaweeds Ulva intestinalis and Ulva lactuca are likely to be more abundant in summer depending on grazing pressure. In summer, erect and encrusting corallines may be bleached (especially in shallow pools) and loose their pink pigment but in some species, e.g. Phymatolithon, this does not necessarily result in death of the plant and pigment may be re-synthesized (Little & Kitching, 1996).

As already noted, no two rockpools exhibit exactly the same physio-chemical conditions and exhibit large spatial variation in community structure, even between adjacent pools at the same shore height (Ganning, 1971; Metaxas & Scheibling, 1993; Metaxas et al., 1994). Habitat complexity will vary with macroalgal cover, the dominant macroalgal species, shore height, pool depth and topography. Deeper pools are likely to be more complex than shallow pools. In this biotope, the influence of sediment and scour on the community is likely to increase with the sediment depth (thickness) and/or with the shallowness of the pool, i.e. shallow pools will exhibit simple communities dominated by sediment resistant red algae. However, a variety of intermediate communities could also be represented by this biotope.

In deep pools the underlying rock is likely to be covered by encrusting corallines. Large macroalgal species (e.g. kelps and fucoids) may dominate the surface of the pool. Their depth within the pool (vertical zonation) is limited by self-shading so that only corallines and red algae occur beneath them. The interface between the bottom sediment and the rock surface is likely to support only sand resistant red algae and fauna, e.g. sand-tolerant algae such as Furcellaria lumbricalis, Polyides rotundus, Ahnfeltia plicata, Rhodothamniella floridula, and the anemone Urticina felina. The sediment may support infauna such as lugworm (e.g. Arenicola marina), the sand mason worm Lanice conchilega, terebellids and meiofauna. The upper limit of some species of algae within the pool may be limited by the summer surface water temperatures, and or desiccation after evaporation (e.g. corallines). Grazing intensity due to littorinids may also affect the abundance of fleshy macroalgae, so that the pools may be dominated by less palatable red algae (e.g. Chondrus crispus and Mastocarpus stellatus). Vertical surfaces within deep pools, and crevices or overhangs present, are likely to be dominated by encrusting fauna e.g. the sponges Halichondria panicea and Hymeniacidon perleve, tubeworms and anemones. The surface of larger stones and pebbles may support tubeworms and the holdfasts of kelps or Chorda filum. The holdfasts of kelps and fucoids, and fronds of filamentous species and erect corallines provide refuges for small invertebrates (e.g. amphipods, isopods and small gastropods) or meiofauna (e.g. copepods) (see ecological relationships above). In addition, the shade of macroalgae provide refuges for shrimps (e.g. Palaemon spp.) and intertidal fish (e.g. blennies and gobies), while crevices and underboulder habitats provide additional refuges for crabs.

Rockpool species also display zonation patterns, similar to the emergent species. For example, brown algae and corallines are usually dominant in mid to low shore pools, while green algae tend to dominant high shore pools (Metaxas & Scheibling, 1993). Kooistra et al. (1989) noted vertical zonation within pools and found that macroalgal communities could be allocated to different depths within pools in the lower or higher parts of the shore. However, the communities studied in Brittany differed markedly from those found in this biotope. Similarly, zonation patterns have also been reported in flatworms, rotifers, oligochaetes, cladoceans, copepods, ostracods, barnacles, amphipods, isopods, chironomid larvae and fish (see Metaxas & Scheibling, 1993). Littorina littorea, mussels, whelks, limpets and sea urchins tend to dominate in lower shore pools, while other littorinids dominate higher on the shore (Metaxas & Scheibling, 1993). Nevertheless, Metaxas et al. (1994) noted that horizontal spatial variability between pools within the same shore height appeared to be as great as variability along the intertidal gradient, and suggested that the physical setting of the pool may be of primary importance in determining the macroalgal abundance. Dethier (1984) examined the effect of natural disturbance rockpool communities in the coast of Washington State. She noted that disturbance such as heat stress in summer and wave action in winter occurred regularly (ca 1.6 times per year per pool). The observed disturbances affected dominant species, so that no one dominant species could occupy all the pools within the tidal range at any one time. None of the tidal pool assemblages observed were stable over many generations and disturbances resulted in a mosaic of species assemblages within pools in any one region (Dethier, 1984).

Although little information on rockpool productivity was found, rocky shore communities are highly productive and are an important source of food and nutrients for members of neighbouring terrestrial and marine ecosystems (Hill et al., 1998; Raffaelli & Hawkins, 1999). Macroalgae exude considerable amounts of dissolved organic carbon which are absorbed readily by bacteria and may even be taken-up directly by some larger invertebrates. Only about 10% of the primary production is directly cropped by herbivores (Raffaelli & Hawkins, 1999). Dissolved organic carbon, algal fragments and microbial film organisms are continually removed by the sea. This may enter the food chain of local, subtidal ecosystems, or be exported further offshore. Rocky shores make a contribution to the food of many marine species through the production of planktonic larvae and propagules which contribute to pelagic food chains. Accumulations of algal debris are also likely in rockpools and such detrital material contributes to overall productivity.

Information specific to the community was not found but Workman (1983) gave an estimate of primary production by microalgal films on the lower shore in the British Isles to be in the region of 100 g C/m²/yr, much of which will be utilized directly by grazers, while primary productivity for fucoids on sheltered shores was estimated to be 1250 g C/m²/yr (Hawkins et al., 1992) and for encrusting corallines to be 1000 g C/m²/yr (Dawes et al., 1991; Raffaelli & Hawkins, 1999). Ganning & Wulff (1970) reported primary productivity values in terms of gross photosynthesis of between 2 and 3.5 O₂/m³/ hr in brackish water rock pools dominated by green algae. Overall, deep rockpools with abundant macroalgae are likely to be highly productive mesocosms on the shore. However, shallower pools, with only sparse macroalgal cover due to sediment scour are likely to be far less productive.

As with the emergent rock surfaces, variation and unpredictability in recruitment of species contributes to the variation in dominant species within rockpool communities and hence spatial variation between rockpools (Dethier, 1984). Recruitment processes of characteristic or dominant species are described here.

Flora

• Vadas et al. (1992) reviewed recruitment and mortality of early post settlement stages of benthic algae. They concluded that grazing, canopy and turf effects were the most important but that desiccation and water movement may be as important for the early stages. The review indicated that recruitment is highly variable and episodic and that mortality of algae at this period is high. Chance events during the early post settlement stages are therefore likely to play a large part in survival.
• Laminaria hyperborea and Laminaria digitata produce vast numbers of spores, however they need to settle and form gametophytes within about 1 mm of each other to ensure fertilization and therefore may suffer from dilution effects over distance. However, Chapman (1981) demonstrated that substantial recruitment of Laminaria digitata plants to areas barren of kelp plants was possible up to 600 m away from reproductive plants. Saccharina latissima (studied as Laminaria saccharina), is short-lived (2-4 years), reaches maturity quickly and colonizes bare substratum rapidly (ca 6 months) (Kain, 1975; Kain, 1979; Birkett et al., 1998; see MarLIN review) and may be regarded as relatively opportunistic.
• Settlement of algal spores is partly dependant on their motility (if any) and adhesive properties together with preferences for topography (surface roughness), the chemical nature of the substratum and water movement (Norton 1992; Fletcher & Callow 1992). As with all red algae, the spores of Ahnfeltia plicata, Chondrus crispus, Furcellaria lumbricalis and Polyides rotundus are non-flagellate and therefore dispersal is a wholly passive process (Fletcher & Callow, 1992). In general, due to the difficulties of re-entering the benthic boundary layer, it is likely that successful colonization is achieved under conditions of limited dispersal and/or minimum water current activity. Norton (1992) reported that although spores may travel long distances (e.g. Ulva sp. 35 km, Phycodrys rubens 5 km), the reach of the furthest propagule does not equal useful dispersal range, and most successful recruitment occurs within 10 m of the parent plants. It is expected, therefore, that recruitment of Ahnfeltia plicata, Chondrus crispus, Furcellaria lumbricalis, Polyides rotundus and the majority of other macroalgae in the biotope would occur from local populations on the surrounding emergent rock but that establishment and recovery of isolated populations would be patchy and sporadic. Scrosati et al. (1994) commented that viability of spores of Chondrus crispus was low (<30%) and suggested that reproduction by spores probably does not contribute much to maintenance of the intertidal population in Nova Scotia, compared to vegetative growth of gametophytes. However, macroalgae are highly fecund and widespread in the coastal zone so that recruitment may be still be rapid, especially in the rapid growing ephemeral species such as Ulva spp., which reproduce throughout the year with a peak in summer. Similarly, Ceramium species produce reproductive propagules throughout the year, while Mastocarpus stellatus produce propagules from February to December, and exhibit distinct reproductive papillae in summer (Dixon & Irvine, 1977; Burrows, 1991; Maggs & Hommersand, 1993).
• Lithophyllum incrustans, representing the encrusting corallines in the biotope, it has been calculated that 1 mm x 1mm of reproductive thallus produces 17.5 million bispores per year with average settlement of only 55 sporelings/year (Edyvean & Ford, 1984). Dispersal is likely to be in excess of 5 km and spores will settle and new colonies will arise rapidly on bare substratum, although the growth rate is slow (2-7 mm per annum - see Irvine & Chamberlain 1994). The erect coralline Corallina officinalis has isomorphic sexual (gametophyte) and asexual (sporophyte) stages (see MarLIN review). Settlement and development of fronds is optimal on rough surfaces but settlement can occur on smooth surfaces (Harlin & Lindbergh 1977; Wiedemann pers comm.). Corallina officinalis settled on artificial substrata within 1 week of their placement in the intertidal in New England summer suggesting that recruitment is high (Harlin & Lindbergh 1977).

Fauna

• Gastropods exhibit a variety of reproductive life cycles. The common limpets Patella vulgata and Patella ulyssiponensis, the top shell Gibbula umbilicalis, and Littorina littorea have pelagic larvae with a high dispersal potential, although recruitment and settlement is probably variable. However, Littorina obtusata lays its eggs on the fronds of fucoids from which hatch crawl-away miniature adults. Similarly, the dog whelk Nucella lapillus lays egg capsules on hard substrata in damp places on the shore, from which crawl-always emerge. Therefore, their dispersal potential is limited but probably designed to colonize an abundant food source. In addition, most gastropods are relatively mobile, so that a large proportion of recruitment of available niches would involve migration. Nucella lapillus is an exception, as they generally do not move far, averaging 100 mm /tidal cycle, or between 30 cm or 10 m per year when in the vicinity of an abundant food source (see MarLIN reviews for details; Fish & Fish, 1996).
• Arenicola marina lays its eggs in its burrows, while adjacent males release sperm at the sediment surface, the resultant juveniles developing within the burrows before migrating upshore. Long distance recruitment of lugworm from sedimentary areas is probably sporadic, and reproduction within a rockpool is only likely if both sexes are present within the same pool. However, Lanice conchilega produce dispersive planktonic larvae.
• Many species of mobile epifauna have long lived pelagic larvae and/or are highly motile as adults. Gammarid amphipods brood their embryos and offspring but are highly mobile as adults and probably capable of colonizing new habitats from the surrounding area (e.g. see Hyale prevostii review for example). Similarly, isopods such as Idotea species brood their young. Idotea species are mobile and active swimmers and probably capable to recruiting to new habitats from the surrounding area by adult migration. Hicks (1985) noted that epiphytic harpacticoid copepods lack planktonic dispersive larval stages but are active swimmers, which is therefore the primary mechanism for dispersal and colonization of available habitats. Some species of harpacticoids are capable to moving between low and mid-water levels on the shore with the tide, while in other colonization rates decrease with increasing distance form resident population. Overall immigration and in situ reproduction were thought to maintain equilibrium populations exposed to local extinction, although there may be local spatial variation in abundance (see Hicks, 1985).
• Emson & Whitfield (1989) noted that a population of Amphipholis squamata from a mid shore rock pool (in south Devon) displayed more seasonal reproduction and short lifespans than other populations studied, which they suggested may be due to the more stressful environmental conditions provided by the rockpool. Amphipholis squamata is a simultaneous hermaphrodite, that broods its young in genital bursae. Emson & Whitfield (1989) noted that recruitment occurred in mid-summer, grow in summer and autumn, mature ( at a 1.6-1.7 mm, ca 8-12 months old) and reproduce in the following spring, release juveniles in summer and then die in autumn and winter, a life span of ca 13 -17 months. Individuals brood up to an average of 10 embryos per individual depending on size, larger individuals brooding more embryos. Regardless of the short life span and brooding habit, the population in the rockpool was highly abundant and had maintained its abundance between 1975 and 1987 (Emson & Whitfield, 1989). Although, a brooding species it is widespread with a cosmopolitan distribution suggesting that dispersal occurs by migration of adults.
• Epiphytic and sessile fauna, such as sponges, hydroids, bryozoans and ascidians, have pelagic but short lived larvae with relatively short effective dispersal ranges, depending on the local hydrography. However, most epiphytic species are widespread and ubiquitous and would probably recruit rapidly from adjacent or nearby populations.

Development of the community in this biotope, from bare or denuded rock, is likely to follow a similar successional pattern to emergent sheltered rocky shore communities. The loss of grazing species results in an initial proliferation of ephemeral green then fucoid algae, which then attracts mobile grazers, and encourages settlement of other grazers. Limpet grazing reduces the abundance of fucoids allowing barnacles to colonize the shore. After the Torrey Canyon oil spill in March 1967, recovery rates were dependant on local variation in recruitment and mortality so that sites varied in recovery rates, for example maximum cover of fucoids occurred within 1-3 years, barnacle abundance increased in 1-7 years, limpet number were still reduced after 6-8 years and species richness was regained in 2 to >10 years (depending on the shore and the degree of disturbance) (Southward & Southward, 1978; Hawkins & Southward, 1992; Raffaelli & Hawkins, 1999). Where only the rockpool is affected, recruitment is potentially good, especially in motile species. However, variability in rockpool communities and their response to disturbance (recovery) has been attributed to variation and seasonality in recruitment (Metaxas & Scheibling, 1993).

Dethier (1984) noted that few rockpool populations, even of dominant species, remained static over time, based on long-term observations over several years. On the coast of Washington State, partial disturbance (a reduction in abundance or cover) resulted in relatively rapid recovery of the community for example; encrusting corallines recovered in over 2 months and erect corallines attained 87% of their original cover in 2 years. The red algae Rhodomela sp. exhibited 39% recovery from total loss after 2 years and Cladophora sp. exhibited 77% recovery after total loss. Dethier (1984) concluded that disturbance was a factor that resulted in a mosaic of different communities in rockpools within an area and that, at any point in time, separate rockpool communities were probably in different stages of recovery.

None