J. Mar. Biol. Ass. U. K. 32001), 81,961^966 Printed in the United Kingdom

Habitat-associated variability in survival and growth of three species of microgastropods

C. Olabarria* and M.G. Chapman Centre for Research on Ecological Impacts of Coastal Cities, Marine Ecology Laboratories A11, University of Sydney, NSW 2006, Australia. *Corresponding author, e-mail: [email protected]

Three species of microgastropods, Eatoniella atropurpurea, Eatonina rubrilabiata and Amphithalamus incidata, are common in various habitats at mid to low levels on intertidal shores in New South Wales, Australia. These habitats include patches of sediment, pebbles and algal turf. These species are very patchy, varying in abundance within and among habitats at scales of centimetres to many metres. This study describes labora- tory experiments which tested hypotheses about di¡erences in mortality and growth rates for each species in three di¡erent habitats: sediment, pebbles and coralline turf. There was greater mortality in coralline turf without sediment for E. rubrilabiata and A. incidata, whereas Eatoniella atropurpurea showed a greater mortality in sediment. Moreover, Eatonina rubrilabiata had a faster rate of growth in sediment, whereas Eatoniella atropurpurea grew more rapidly in coralline turf. The di¡erent rates of mortality and growth for these species in di¡erent habitats provide mechanisms which may partially explain the patterns of abundance in the ¢eld.

INTRODUCTION 1983). Because many small species live in cryptic habitats 3e.g. algal mats), the logistics of doing experiments which The structure of intertidal assemblages results from require growth measures, survival or behaviour of indivi- many interacting processes, including disturbances 3e.g. dual animals, are very di¤cult in the ¢eld. Despite limita- Sousa, 1980; Littler et al., 1983; Chapman & Underwood, tions 3Chapman, 2000), laboratory experiments have 1998), predation or competition 3e.g. Dayton, 1971; therefore often been used to imply processes that deter- Fairweather & Underwood, 1991) and settlement and/or mine patterns of abundance in the ¢eld 3e.g. Fenchel, recruitment 3e.g. Dayton, 1975; Underwood & Fairweather, 1976; Underwood & McFadyen, 1983; Levinton et al., 1989). Understanding the structure and dynamics of these 1985). assemblages is, however, not possible without knowledge Three common microgastropods on intertidal shores of of the ecology of the component species 3Hutchinson, 1961; New South Wales are Eatoniella atropurpurea 3Frauenfeld, Dayton, 1971). Studies of reproduction, recruitment, 1867), Eatonina rubrilabiata Ponder & Yoo, 1980 and growth and mortality have provided great insight into Amphithalamus incidata 3Frauenfeld, 1867). Populations of many interactions that occur within and among the these species are found at mid- to low-shore levels in various trophic levels in assemblages 3Boulding & Van di¡erent habitats, such as sediment, on pebbles or in algal Alstyne, 1993; Otway, 1994). In addition, studies of beha- turf. All three species are particularly abundant in coral- vioural responses of individuals to other individuals or line turf 3i.e. algal beds composed primarily of Corallina habitats provide an understanding of patterns of distribu- o¤cinalis Linnaeus, often containing patches of sediment), tion of organisms and structure of intertidal assemblages although the latter two are also quite abundant in patches 3Creese, 1982; Chapman & Underwood, 1994). of sediment. All three species vary in abundance among Research in New SouthWales, Australia, has focused on patches of habitats at very small spatial scales 3Olabarria various aspects of ecology of large prosobranchs such as & Chapman, 2001). In addition, adults of the three species distribution of species 3e.g. O'Gower & Meyer, 1971), rapidly colonize new boulders in boulder ¢elds 3M.G.C., reproduction 3e.g. Underwood, 1974), growth and mor- unpublished data). tality 3e.g. Fletcher, 1984; Underwood, 1984a) or beha- A number of di¡erent general models can explain these viour 3e.g. Chapman & Underwood, 1994). The same di¡erent patterns of distribution, for example, di¡erent degree of ecological understanding is not available for rates of recruitment or di¡erent survivorship among hab- microgastropods 3i.e. gastropods with adult size approxi- itats. We tested the models that: 3i) all species are more mately 52 mm), even though they are abundant and wide- abundant in algal turf than in sediment because they spread on intertidal shores 3Wigham, 1975; Southgate, survive better in the same habtat; 3ii) E. rubrilabiata and 1982; Borja, 1987; Grahame & Hanna, 1989).This is parti- A. incidata may survive better in sediment than does cularly true for intertidal shores in Australia 3Olabarria & Eatoniella atropurpurea, hence their relative greater abun- Chapman, 2001), where most research has been taxonomic dance in this habitat in comparison to this latter one; and or anatomical 3Beesley et al., 1998). 3iii) all three species may show di¡erent rates of survival in Although ¢eld experiments in intertidal habitats are di¡erent habitats because they grow faster in some habitats increasingly important, most have focused on the largest than in others. This study describes laboratory experi- components of the fauna 3see review Underwood et al., ments to test hypotheses from these models. Speci¢c

Journal of the Marine Biological Association of the United Kingdom 2001) 962 C. Olabarria and M.G. Chapman Survival and growth of microgastropods predictions are made 3see Material and Methods) about preferred to alternatives because it allows accurate survivorship and rate of growth of each species in di¡erent comparison of individuals of slightly di¡erent initial sizes habitats 3i.e. pebbles, sediment and coralline turf ). 3Denley & Underwood, 1979). Individual snails were not

marked and, therefore, all calculations 3values of L0) were based on the mean size of animals for each pot. MATERIALS AND METHODS Data of mortality and rates of growth were analysed Collection of samples using analysis of variance. Samples of Eatoniella atropurpurea, Amphithalamus incidata and Eatonina rubrilabiata were collected from algal turf at Experiment 2 the mid-shore level on one shore 3described in Olabarria Using the data obtained from experiment 1 A. incidata & Chapman, 2001) in the Cape Banks Scienti¢c Marine and Eatonina rubrilabiata survived better in sediment, Research Area, New South Wales in January 2000 and although in the ¢eld they are more abundant in coralline May 2000. turf 3Olabarria & Chapman, 2001) and the observations In the ¢rst experiment, samples of rock chips 31^2cm that 3i) much of the coralline turf in the ¢eld contains deep, 5 cm diameter) with turf were collected by chiselling sediment 3personal observation); and 3ii) Eatoniella from a patch of mid-shore coralline turf. Sediment and atropurpurea was mainly found in coralline turf, a second pebbles of similar size to the rock chips with a ¢ne cover experiment was done. This tested the predictions that: 3i) of ¢lamentous algae were collected from nearby sites at the A. incidata and Eatonina rubrilabiata would have greater same level on the shore. All samples were taken to the survival and a faster rate of growth in sediment, whether laboratory, examined under a microscope and all visible or not this occurred with coralline algae; and 3ii) Eatoniella organisms removed. Each sample, a chip of rock with atropurpurea would have greater survival and rates of growth coralline algae, a pebble or approximately 2 cm of sedi- in coralline algae, whether sediment was present in the ment, was then placed in a transparent pot 36.5 cm algae or not. diameter, 8 cm high). These were then covered with a This experiment used three di¡erent habitats: 3i) sedi- 200-mm mesh, and maintained with a continuous £ow of ment that had been removed from coralline turf; 3ii) seawater. coralline turf with associated sediment; and 3iii) coralline In the second experiment, the coralline turf was turf from which the sediment had been removed. The turf collected using a 6-cm diameter corer. For treatments in was collected using cores 3see earlier), taken to the labora- which turf without sediment was needed, the sediment tory and then washed with seawater under high pressure to was washed out of the turf using running seawater. Both eliminate most of the organisms. This also removed the the algae and the sediment were defaunated, removing all sediment from the cores. The sediment was sieved visible organisms under a microscope, before being placed through 60-mm mesh and examined under a microscope into the transparent pots. to remove all visible organisms. Subsequently, the cores were treated in one of three di¡erent ways, depending on Experiment 1 the type of substratum needed: 3i) the sediment was defau- nated, the algae scraped from the surface of the core and This experiment evaluated whether these species would the sediment was placed on the core; 3ii) the algae and survive and grow under laboratory conditions and tested sediment were defaunated and the sediment was put back the hypothesis that each species would show greater into the coralline algae on the core; 3iii) the sediment was survival and faster growth on the chips of rock covered eliminated and coralline algae was defaunated. with coralline algae than on the other two habitats. The samples were thenput into separate pots 3Nˆ10) and Eight animals 3four adults and four juveniles where maintained with a continuous £ow of seawater. In this possible; in most of the pots) of each species were placed experiment, the animals 3eight per replicate) were individu- in each pot; Nˆ10 pots for each species on each ally marked using permanent markers. Once the shell was substratum. The pots were placed in ¢ve tanks, with two dry, a colour code of up to four dots of permanent marker replicates of each treatment for each species in each tank. were carefully applied to each shell under a dissecting The experiment was done under 12 h L:12 h D. The dura- microscope. Six colours were used to mark the snails: tion of the experiment was chosen to be 28 days based on blue, red, yellow, green, pink and orange. Snails were previous studies of growth rates of small gastropods immersed in seawater as soon as the paint was dry and carried out in the laboratory, which showed measurable any that did not emerge within two min were discarded. growth after a few weeks 3Underwood & McFadyen, Because of the large mortality in experiment 1, the pots 1983; Boulding & Van Alstyne, 1993). were assigned at random in the experimental area and At the end of the experiment, the length of each shell re-randomized into di¡erent positions once a week to was measured from the apex to the lower lip on the oper- minimize any potential e¡ects of position on survival. cular side, using a binocular microscope ¢tted with an Rates of growth and mortality after 35 days 3see above, eyepiece micrometer 3measurement error 0.001mm). The experiment 1) were analysed using analysis of variance. rate of growth of each mollusc was calculated as: RESULTS R ˆ ln 3Lt/L0)/t 31) Experiment 1 where R is the instantaneous rate of growth per unit length, Lt and L0 are shell lengths at times t and 0, respec- There were signi¢cant di¡erences in mortality after 28 tively. This model assumes exponential growth and is days among the di¡erent habitats 3Table 1). Each species

Journal of the Marine Biological Association of the United Kingdom 2001) Survival and growth of microgastropods C. Olabarria and M.G. Chapman 963

Table 1. Analysis of mortality of Amphithalamus incidata Table 2. Analysis of rates of growth of Amphithalamus and Eatonina rubrilabiata in three di¡erent habitats incidata and Eatonina rubrilabiata in three di¡erent Nˆ10). Species two levels) and habitat three levels) are habitats rate of growth is in mm 103). Because there was large ¢xed factors; tank ¢ve levels) is a random factor. Variances mortality in many pots, rates of growth were from those pots with were homogeneous Cochran's test, P40.05). Student^ 450% survival Nˆ2). Tank two levels) is a random factor Newman^Keuls tests were used to identify signi¢cant di¡erences nested in species two levels, ¢xed) and habitat three levels, among means see text). ¢xed). Variances were homogeneous Cochran's test, P40.05). Student^Newman^Keuls tests were used to identify signi¢cant Source df MS F-ratio P di¡erences among means see text).

Sp 1 0.15 0.02 0.907 Source df MS F-ratio P T 4 2.98 0.76 0.562 Hb 2 40.72 10.32 0.006** Sp 1 24.00 8.47 0.021* SpÂT 4 9.82 2.49 0.064 Hb 2 1.04 0.37 0.082 SpÂHb 2 1.25 0.14 0.870 T3SpÂHb) 6 2.83 0.35 0.546 TÂHb 8 3.95 1.00 0.457 SpÂHb 2 3.88 1.37 0.075 SpÂTÂHb 8 8.85 2.24 0.062 Residual 12 8.08 Residual 30 3.95 Hb, habitat; Sp, species; T, tank; *, P50.05, **, P50.01, Hb, habitat; Sp, species; T, tank; *, P50.05; **, P50.01; ***, P5 0.001. ***, P50.001.

in sediment than in coralline turf or on pebbles although showed smaller mortality in sediment than in coralline the di¡erences were not signi¢cant. turf or on pebbles 3Student^Newman^Keuls 3SNK) tests, P50.05; Figure 1A). Each species also showed slightly smaller average mortality on pebbles although the di¡er- Experiment 2 ences were not signi¢cant 3Figure 1A). Eatonina rubrilabiata and A. incidata had signi¢cantly There were signi¢cant di¡erences in rates of growth smaller mortality in sediment and coralline algae plus among species 3Table 2), but not among the di¡erent habi- sediment, than in coralline algae by itself 3SNK tests, tats. Eatonina rubrilabiata grew faster than Amphithalamus P50.05; Figure 2A; see interaction in Table 3). Eatoniella incidata 3Figure 1B). The two species grew slightly faster atropurpurea, in contrast, showed signi¢cantly smaller

Figure 2. Results from experiment 2. 3A) Mean mortality Figure 1. Results from experiment 1. 3A) Mean mortality of of Amphithalamus incidata, Eatonina rubrilabiata and Eatoniella Eatonina rubrilabiata and Amphithalamus incidata in three di¡erent atropurpurea in three di¡erent habitats 3Nˆ10); 3B) mean rate habitats 3Nˆ10); 3B) mean rate of growth of E. rubrilabiata and of growth of A. incidata, Eatonina rubrilabiata and Eatoniella A. incidata in 3Nˆ6). atropurpurea in three di¡erent habitats 3Nˆ18).

Journal of the Marine Biological Association of the United Kingdom 2001) 964 C. Olabarria and M.G. Chapman Survival and growth of microgastropods

Table 3. Analysis of mortality of Amphithalamus DISCUSSION incidata, Eatonina rubrilabiata and Eatoniella atropurpurea in three di¡erent habitats Nˆ10). Species These experiments clearly showed di¡erences in three levels) and habitat three levels) are ¢xed factors. mortality and rates of growth among the di¡erent types Variances were homogeneous Cochran's test, P40.05). of habitat for the three species. Eatonina rubrilabiata and Student^Newman^Keuls tests were used to identify signi¢cant Amphithalamus incidata survived better in sediment than in di¡erences among means see text). algal turf without sediment or on pebbles coated with a ¢ne cover of ¢lamentous algae, whether the sediment was Source df MS F-ratio P with or without algae. Eatoniella atropurpurea, in contrast, showed greater mortality in sediment, surviving better in Sp 2 8.63 4.70 0.012* coralline turf, whether sediment was present in the turf Hb 2 8.93 4.87 0.011* or not. Furthermore, Eatonina rubrilabiata grew faster in SpÂHb 4 14.67 7.99 0.000*** sediment, whereas Eatoniella atropurpurea grew faster in Residual 81 1.84 coralline algae. Amphithalamus incidata, in contrast, grew Hb, habitat; Sp, species; *, P50.05, **, P50.01, ***, P50.001. at similar rates on all three habitats. In general, all species showed greater rates of survival in those habitats where they showed greater rates of growth. These results contrast with their distribution among habitats in the mortality in coralline algae, with or without sediment, ¢eld, where all three species are more common in algal than in sediment alone 3SNK tests, P50.05; Figure 2B). turf than in sediment 3Olabarria & Chapman, 2001). In addition, di¡erences in rates of mortality among the The availability of food is one of the most important three species depended on the habitat 3Figure 2A). For factors a¡ecting the rates of growth and survival of example, A. incidata and Eatonina rubrilabiata had the di¡erent species of gastropods 3Sutherland, 1970; Creese, greatest mortality in coralline algae by itself, whereas 1981; Underwood, 1984a). Therefore, the patterns of Eatoniella atropurpurea survived best in coralline algae. survival and growth of Eatonina rubrilabiata, A. incidata The initial size of these animals did not vary signi¢- and Eatoniella atropurpurea in di¡erent habitats could be cantly among treatments for each species, i.e. there was explained in terms of availability of food. Although there no interaction 3F ˆ0.43, P40.05). Because the rates of 4, 81 are no quantitative data on feeding in these species, all growth of individuals did not di¡er signi¢cantly among three species are assumed to be microphagous feeders, pots in each treatment 3F ˆ1.05, P40.05) and to 81, 630 feeding on diatoms, micro-algae and detritus 3Beesley et balance the data 3the minimal number of individuals al., 1998). Whether they do eat the same range of food is which survived in one of the habitats was 18), 18 not, however, known. randomly-selected individuals of each species from each Although the three species have taeniglossan radulae, substratum were used to test the hypothesis of di¡erences there are small di¡erences in the structure of these which in rates of growth among habitats for each species. could be related to di¡erent types of feeding. Thus, the Rates of growth varied interactively between species rachidian tooth of E. atropurpurea is large and square, with and habitat 3Table 4). Amphithalamus incidata grew at three cusps with strong basal processes and lateral and similar rates among the di¡erent habitats 3SNK tests, marginal teeth with few cusps. In addition, this species P50.05; Figure 2B). Eatonina rubrilabiata, in contrast, has jaws with chitinous rods 3Beesley et al., 1998). Eatonina grew signi¢cantly faster in sediment, with or without rubrilabiata and A. incidata have radulae with smaller coralline algae, than in algae alone 3SNK tests, P50.05) central teeth with small obsolete cusps and lateral and and Eatoniella atropurpurea grew faster in coralline algae, marginal teeth with more cusps. Snails with pluricuspid with or without sediment than in sediment alone 3SNK teeth such as these are more likely to be grazers of micro- tests, P50.05). algae and ¢lamentous algae, whereas herbivores with fewer cusps and robust teeth are more likely to consume tougher algae, including articulated corallines 3Steneck & Table 4. Analysis of rates of growth of Amphithalamus Watling, 1982). Thus, Eatoniella atropurpurea could use its incidata, Eatonina rubrilabiata and Eatoniella radula to scrape microalgae and feed on the branches of the atropurpurea in three di¡erent habitats Nˆ18) rate of coralline turf, thus explaining its greater rates of survival 3 growth is in mm 10 ). Species three levels) and habitat three and growth when maintained in algae. Amphithalamus levels) are ¢xed factors; pot was not a signi¢cant factor see incidata and Eatonina rubrilabitata could use their radulae text) and was consequently eliminated from this analysis. to feed on detritus and diatoms in the sediment, thereby Variances were homogeneous Cochran's test, P40.05). explaining their greater rates of survival and growth in Student^Newman^Keuls tests were used to identify signi¢cant this habitat. di¡erences among means see text). Coralline turf and other complex biogenic structures can be a trap for sediment and food 3Taylor & Littler, Source df MS F-ratio P 1982; Grahame & Hanna, 1989; Akioka et al., 1999). Therefore, microgastropods feeding on micro-algae Sp 2 127.94 3.38 0.036* Hb 2 151.39 4.00 0.020* associated with sediments may obtain adequate food from SpÂHb 4 166.52 4.40 0.002** the sediment in coralline turf. On the relatively exposed Residual 153 37.82 shore used in this study, the sediment is rather coarse- grained and loosely packed. This may make it susceptible Hb, habitat; Sp, species; *, P50.05, **, P50.01, ***, P5 0.001. to disturbances from movement of water, which can limit

Journal of the Marine Biological Association of the United Kingdom 2001) Survival and growth of microgastropods C. Olabarria and M.G. Chapman 965 production of micro-algae 3Riznyk et al., 1978; Varela & This work was supported by funds from the Australian Penas, 1985). Grain-size can be a limiting factor for primary Research Council, the Institute of Marine Ecology and the production because of the strong attachment of the micro- Centre for Research on Ecological Impacts of Coastal Cities. We phytobenthos to coarse sediment 3Barranguet et al., 1998). are grateful to Sophie Diller and Stefanie Arndt for help with sampling, and particularly Chris McKindsey who worked hard Factors such as these may prevent micro-algal grazers from in the ¢eld and provided many hepful comments. We also thank occupying patches of sediment in the ¢eld. In these lab- A.J. Underwood and B. Kelaher for commenting on an earlier oratory experiments, the sediment was not disturbed and draft of this paper and two anonymous referees for comments there was continuous gentle £ow of water providing nutri- that improved the manuscript. ents and adequate light, any of which may have increased productivity, or provided greater opportunity to feed. REFERENCES Coralline turf can also provide di¡erent microhabitats because it is an heterogeneous substratum, which may Akioka, H., Baba, M., Masaki,T. & Johansen, H.W.,1999. Rocky o¡er refuge from predation 3Hassell & May, 1973; shore turfs dominated by Corallina 3Corallinales, Rhodophyta) Murdoch, 1977; Edgar et al., 1994). Heterogeneous hab- in northernJapan. Phycological Research, 47,199^206. itats can, in contrast, provide shelter for predators, thus Barranget, C., Kromkamp, J. & Peene, J., 1998. Factors control- increasing rates of predation around the source of hetero- ling primary production and photosynthetic characteristics of intertidal microphytobenthos. Marine Ecology Progress Series, 173, geneity 3Fairweather, 1988; Underwood, 1999). Coull & 117 ^126. Wells 31983), in ¢eld and laboratory experiments, showed Beesley, P.L., Ross, G.J.B. & Wells, A., 1998. Mollusca: the southern complex habitats o¡ered a signi¢cant refuge for meiofauna synthesis. Fauna of Australia, vol. 5. Melbourne: CSIROPublishing. from predation by ¢sh. Therefore, algal turf may be an Borja, A., 1987. Biology and ecology of three species of intertidal important refuge from predation for the microgastropods gastropod molluscs Rissoa parva, Barleeia unifasciata and Bittium in this study, but, as yet, any predators of these species reticulatum.II.Growth.Cahiers de Biologie Marine, 28,351^360. have not been identi¢ed. Boulding, E.G. & Van Alstyne, K.L., 1993. Mechanisms of Processes that can in£uence the distribution of inverte- di¡erential survival and growth of two species of Littorina on brates, such as predation 3Fairweather & Underwood, 1991; wave-exposed and on protected shores. Journal of Experimental Osman & Whitlatch, 1996), competition 3Underwood, Marine Biology and Ecology, 169,139^166. Chapman, M.G., 2000. Poor design of behavioural experiments 1984b; Fletcher & Underwood, 1987), morphology, life gets poor results: examples from intertidal habitats. Journal of history 3Boulding & Van Alstyne, 1993), behaviour Experimental Marine Biology and Ecology, 250,77^95. 3Chapman & Underwood, 1994) are very complex and Chapman, M.G. & Underwood, A.J., 1994. Dispersal of the interactive 3e.g. Menge, 1992, 1997). Most advances in intertidal snail, Nodilittorina pyramidalis, in response to topo- understanding of these complex interactions have been graphic complexity of the substratum. Journal of Experimental obtained from experimental studies of relatively large Marine Biology and Ecology, 179,145^169. animals 3but see Creese, 1981; Fairweather et al., 1984; Chapman, M.G. & Underwood, A.J., 1998. Inconsistency and Boulding & Van Alstyne, 1993). It is reasonable to assume variation in the development of rocky intertidal algal assem- that similar processes operate to determine the distribu- blages. Journal of Experimental Marine Biology and Ecology, 224, tion of small species, although the scales over which they 265^289. operate are likely to be smaller 3Lawton, 1990; Cotgreave, Cotgreave, P., 1993. The relationship between body size and population abundance in animals. Trends in Ecology and 1993). Therefore, more understanding of the ecology of Evolution, 8,244^248. these species will also depend on our ability to use Coull, B.C. & Wells, J.B.J., 1983. Refuges from ¢sh predation: manipulative experiments to test speci¢c hypotheses experiments with phytal meiofauna from the New Zealand 3Underwood, 1990). Although ¢eld experiments provide rocky intertidal. Ecology, 64, 1599^1609. more reliable information about ecological processes Creese, R.G., 1981. Patterns of growth, longevity and recruit- 3Crowe & Underwood, 1998), they may not be feasible ment of intertidal limpets in New South Wales. Journal of for tests of many hypotheses when the animals are small Experimental Marine Biology and Ecology, 51,145^171. and live in cryptic habitats. In this case, laboratory experi- Creese, R.G., 1982. The distribution and abundance of the limpet ments may be the only practical approach, although a Patelloida latistrigata, and its interaction with barnacles. combination of laboratory and ¢eld experiments would Oecologia, 52,85^96. Crowe, T. & Underwood, A.J., 1998. Testing behavioural `prefer- be more desirable 3Chapman, 2000). ence' for suitable microhabitat. Journal of Experimental Marine The present study has provided data about mortality Biology and Ecology, 225,1^11. and growth of three common microgastropods, which can Dayton, P.K.,1971. Competition, disturbance and community orga- be used with previous data on patterns of distribution in nisation: the provision and subsequent utilization of space in a the ¢eld 3Olabarria & Chapman, 2001) to develop rocky intertidal community. Ecological Monographs, 41,351^389. further models on the processes that most a¡ect these Dayton, P.K., 1975. Experimental studies of algal canopy interac- species. It has also been demonstrated that these species tions in a sea otter dominated kelp community at Amchitka survive and grow in laboratory conditions and that indivi- Island, Alaska. Fisheries Bulletin. USA, 73,230^237. duals can be marked and tracked through time in di¡erent Denley, E.J. & Underwood, A.J., 1979. Experiments on factors types of habitat. Nevertheless, in the ¢eld these parameters in£uencing settlement, survival and growth of two species of are likely in£uenced by many interacting variables. barnacles in New South Wales. Journal of Experimental Marine Biology and Ecology, 36,269^293. Further tests of natural spatio^temporal change in their Edgar, G.J., Shaw, C., Watson, G.F. & Hammond, L.S., 1994. populations, together with ongoing laboratory tests of Comparisons of species richness, size-structure and production hypotheses of habitat-choice and competition, will contri- of benthos in vegetated and unvegetated habitats in Western bute important information to the ecology of these Port, Victoria. Journal of Experimental Marine Biology and common, but poorly understood, gastropods. Ecology, 176,201^226.

Journal of the Marine Biological Association of the United Kingdom 2001) 966 C. Olabarria and M.G. Chapman Survival and growth of microgastropods

Fairweather, P.G., 1988. Movements of intertidal whelks 3Morula Riznyk, R.Z., Edens, J.I. & Libby, R.C., 1978. Production of marginalba and Thais orbita) in relation to availability of prey epibenthic diatoms in a southern California impounded and shelter. Marine Biology, 100,63^68. estuary. Journal of Phycology, 14,273^279. Fairweather, P.G. & Underwood, A.J., 1991. Experimental Sousa, W.P., 1980. The responses of community to disturbance: removals of a rocky intertidal predator: variations within two the importance of successional age and species life histories. habitats in the e¡ects on prey. Journal of Experimental Marine Oecologia, 45,72^81. Biology and Ecology, 154,29^75. Southgate, T., 1982. The biology of Barleeia unifasciata Fairweather, P.G., Underwood, A.J. & Moran, M.J., 1984. 3Gastropoda: Prosobranchia) in red algal turfs in south-west Preliminary investigations of predation by the whelk Morula Ireland. Journal of the Marine Biological Association of the United marginalba. Marine Ecology Progress Series, 17,143^156. Kingdom, 62,461^468. Fenchel, T., 1976. Evidence for exploitative interspeci¢c competi- Steneck, R.S. & Watling, L., 1982. Feeding capabilities and limita- tion in mud snails 3Hydrobiidae). Oikos, 27,367^376. tion of herbivorous molluscs: a functional group approach. Fletcher, W.J., 1984. Intraspeci¢c variation in the population Marine Biology, 66,299^319. dynamics and growth of the limpet, Cellana tramoserica. Sutherland, J.P., 1970. Dynamics of high and low populations of Oecologia, 63,110^121. the limpet Acmaea scabra 3Gould). Ecological Monographs, 40, Fletcher, W.J. & Underwood, A.J., 1987. Interspeci¢c competi- 169^188. tion among subtidal limpets: e¡ect of substratum Taylor, P.R. & Littler, M.M., 1982. The roles of compensatory heterogeneity. Ecology, 68,387^400. mortality, physical disturbance and substrate retention in Grahame, J. & Hanna, F.S., 1989. Factors a¡ecting the distribu- community development and organization of a sand-in£u- tion of the epiphytic fauna of Corallina o¤cinalis 3L.) on an enced rocky intertidal community. Ecology, 63,135^146. exposed rocky shore. Ophelia, 30,113^129. Underwood, A.J., 1974. The reproductive cycles and geographic Hassell, M.P. & May, R.M., 1973. Stability in insect host^ distribution of some common eastern Australian prosobranchs parasite models. Journal of Animal Ecology, 42,693^726. 3Mollusca: Gastropoda). Australian Journal of Marine and Hutchinson, G.E., 1961. The paradox of the plankton. American Freshwater Research, 25,63^88. Naturalist, 95,137^145. Underwood, A.J., 1984a. Microalgal food and the growth of the Lawton, J.H., 1990. Species richness and population dynamics of intertidal gastropods Nerita atramentosa Reeve and Bembicium animal assemblages. Patterns in body size: abundance space. nanum 3Lamarck) at four heights on a shore. Journal of PhilosophicalTransactions of the Royal Society B, 330, 283^291. Experimental Marine Biology and Ecology, 79,277^291. Levinton, J.S., Stewart, S. & Dewitt, T.H., 1985. Field and Underwood, A.J., 1984b. Vertical and seasonal patterns in laboratory experiments on interference between Hydrobia competition for microalgae between intertidal gastropods. totteni and Ilyanassa obsoleta 3Gastropoda) and its possible rela- Oecologia, 64,211^222. tion to seasonal shifts in vertical mud£at zonation. Marine Underwood, A.J., 1990. Experiments in ecology and manage- Ecology Progress Series, 22,53^58. ment: their logics, functions and interpretations. Australian Littler, M.M., Martz, D.R. & Littler, D.S., 1983. E¡ects of recur- Journal of Ecology, 15,365^389. rent sand deposition on rocky intertidal organisms: Underwood, A.J., 1999. History and recruitment in structure of importance of substrate heterogeneity in a £uctuating environ- intertidal assemblages on rocky shores: an introduction to ment. Marine Ecology Progress Series, 11,129^140. problems for interpretation of natural change. In Aquatic life Menge, B.A., 1992. Community regulation: under what condi- cycle strategies: survival in a variable environment 3ed. M. Whit¢eld tions are bottom-up factors important on rocky shores? et al.), pp. 79^96. Cambridge: Cambridge University Press. Ecology, 73,755^765. Underwood, A.J., Denley, E.J. & Moran, M.J., 1983. Menge, B.A., 1997. Detection of direct versus indirect e¡ects: Experimental analyses of the structure and dynamics of mid- were experiments long enough? American Naturalist,149,801^823. shore rocky intertidal communities in New South Wales. Murdoch,W.W., 1977. Stabilizing e¡ects of spatial heterogeneity in Oecologia, 56,202^219. predator prey systems.Theory of Population Biology, 11,252^273. Underwood, A.J. & Fairweather, P.G., 1989. Supply-side ecology O'Gower, A.K. & Meyer, G.R., 1971. The ecology of six species and benthic marine assemblages.Trends in Ecology and Evolution, of littoral gastropods. III. Diurnal and seasonal variations in 4,16^19. densities and patterns of distribution in the environment. Underwood, A.J. & McFadyen, K.E., 1983. Ecology of the inter- AustralianJournal of Marine and Freshwater Resources, 22,35^40. tidal snail Littorina acutispira Smith. Journal of Experimental Olabarria, C. & Chapman, M.G., 2001. Comparison of patterns Marine Biology and Ecology, 66,169^197. of spatial variation of species of microgastropods between two Varela, M. & Penas, E., 1985. Primary production of benthic contrasting intertidal habitats. Marine Ecology Progress Series, microalgae in an intertidal sand £at of the Ria de Arosa, NW 220,201^211. Spain. Marine Ecology Progress Series, 25, 111 ^119. Osman, R.W. & Whitlatch, R.B., 1996. Processes a¡ecting Wigham, G.D., 1975. The biology and ecology of Rissoa parva 3da newly-settled juveniles and the consequences to subsequent Costa) 3Gastropoda: Prosobranchia). Journal of the Marine community development. Invertebrate Reproduction and Biological Association of the United Kingdom, 55,45^67. Development, 30,217^225. Otway, N.M., 1994. Population ecology of the low-shore chitons Onithochiton quercinus and Plaxiphora albida. Marine Biology, 121, 105^116. Submitted 30 November 2000. Accepted 12 October 2001.

Journal of the Marine Biological Association of the United Kingdom 2001)