SPATIAL PATTERNS OF SHELL SHAPE OF THREE SPECIES OF CO-EXISTING LITTORINID SNAILS IN NEW SOUTH WALES, AUSTRALIA

The shape and relative weight of the shell have been shown tovary intraspecifically and interspecifically in a number ofspecies of gastropods, including many different littorinids.These differences give rise to different shell forms in differenthabitats. In those species which have non-planktotrophic development,differences in shell form among shores have been usually explainedin terms of natural selection because exposure to waves supposedlyfavours light shells with large apertures, while predation bycrabs on sheltered shores favours elongated, thick shells withsmaller apertures. Differences in shell shape among speciesfound at different heights on the shore have been explainedin terms of resistance to desiccation and temperature. Suchvariables would tend to act on a relatively broad-scale, i.e.causing differences among heights on a shore or among shores.Rates of growth, which might vary at much smaller scales withina shore, have also been shown to affect the shapes of many shells. In this study, the shape and relative weight of shells of threespecies of co-existing littorinids (Littorina unifasciata, Bembiciumnanum and Nodilittorina pyramidalis) were measured. These speciesall haveplanktotrophic development and they are found on manyshores where there is no evidence that they are preyed uponby crabs. Before explanations of shell shape are proposed, itis necessary that patterns of variation, within different partsof ashore and among different shores are clearly documented.These patterns were measured at a number of different spatialscales within and among replicate shores with different amountsof wave exposure. Large and small specimens were included toallow intraspecific comparisons among snails of different sizesfound at different heights on the shore. The results showedsignificant differences among shores in shape and relative weightof shells, but these differences could not be explained by exposureto waves. In addition, snails of different sizes and differentspecies did not show the same patterns although they were collectedfrom the same sites. Importantly, the shell shape of Liuorinaunifasciata varied significantly among sites at approximatelythe same height within a shore. These differences could notbe clearly correlated with density, mean size nor exposure towaves. The only consistent pattern was a decrease in relativeaperture size in specimens living higher on the shore. Modelsthat have commonly been proposed to explain shape and relativeweight of shells in other species of gastropods are not adequateto explain the small- and large-scale variation of the measurementsdescribed here. It is proposed that any selective advantageof shell morphology and the effects of any variables on thedevelopment of shell morphology in these species can only beidentified after appropriately designed and replicated fieldexperiments. (Received 4 March 1994; accepted 13 September 1994)     

Phenotypic homogeneity of two intertidal snails across a wave exposure gradient in South Australia

Dislodgement by the large drag forces imparted by breaking waves is an important cause of mortality for intertidal snails. The risk of drag-induced dislodgement can be reduced with: (1) a smaller shell of lower maximum projected surface area (MPSA); (2) a streamlined shell shape characterized by a squatter shell; and/or (3) greater adhesive strength attained through a larger foot area or increased foot tenacity. Snails on exposed coasts tend to express traits that increase dislodgement resistance. Such habitat-specific differences could result from direct selection against poorly adapted phenotypes on exposed shores but may reflect gastropod adaptation to high wave action achieved through phenotypic plasticity or genetic polymorphism. With this in mind, we examined the size, shape and adhesive strength of populations of two gastropod species, Austrocochlea constricta (Lamarck) and Nerita atramentosa (Reeve), from two adjacent shores representing extremes in wave exposure. Over a 5 day period, maximum wave forces were more than 10 times greater on the exposed than sheltered shore. Size-frequency distributions indicate that a predator consuming snails within the 1.3-1.8 cm length range regulates sheltered shore populations of both snail species. Although morphological scaling considerations suggest that drag forces should not place physical limits on the size of these gastropods, exposed shore populations of both snails were small relative to the maximum size documented for these species. Therefore, selective forces at the exposed site might favour smaller individuals with increased access to microhabitat refuges. Unexpectedly, however, neither snail species exhibited between-shore differences in shape, foot area or foot tenacity, which are likely to have adaptive explanations. Hence, it is possible that these snails are incapable of adaptive developmental responses to high wave action. Instead, the homogeneous and wave-exposed nature of Australia's southern coastline may have favoured the evolution of generalist strategies in these species.     

MORPHOLOGICAL VARIATION IN THE PAINTED TOPSHELL, CALL1OSTOMA ZIZYPHINUM (L.) (PROSOBRANCHIA: TROCHIDAE), FROM INTER-TIDAL RAPIDS ON THE IRISH COAST

Shell shape variation in Calliostoma zizyphinum (L.) is examinedwith different techniques of analysis (profile, discriminant,canonical variate and ratio analysis). Shells are collectedfrom five tidal rapids systems, all in very sheltered inlets,where C. zizyphinum is to be found at shore levels higher thanthe normal shore distribution of LWNT to the sublittoral. Shellmorphology varies consistently with the local habitat conditions,such as water flow rate, relative exposure and shore level.populations can be distinguished using various shell parametersand shell profiles. Shells from two different levels in thesame rapids show the effects of their subtly different habitats. (Received 5 February 1987;     

SEMI-LUNAR AND LUNAR SPAWNING PERIODICITY IN SOME TROPICAL LITTORINID GASTROPODS

Littoraria strigata, Nodilittorina millegrana and N. pyramidalisall spawned regularly in a rock-flanked Malaysian bay in March-June1985, whereas none had spawned there in July-October 1981. N.millegrana showed the least precise timing with some irregularpreponderance during the days between 2d before and 7d afternew or full moon and their associated spring tides. Spawningof N. pyramidalis was more strictly confined to spring tideswith 90% of eggs appearing between 2d before and 5d after newor full moon. There was no apparent difference between alternatespawnings on new and on full moons. L. strigata also spawnedregularly on the spring tides with 89% of eggs appearing between1 and 5 days after new and full moon. In this case, however,full moon spawnings (when spring tides rose highest) yieldedsome 7 times more eggs on average than the weaker spring tidesat new moons. In laboratory aquaria, the three species all released eggs duringboth night and daytime and through periods of both spring andneap tides with no evidence that periodicity persisted in constantconditions. The different degrees of 2-weekly spawning periodicity, togetherwith littorinids which spawn only monthly, may represent a rangeof adaptation to life at increasing shore heights. (Received 26 November 1985;     

Ecology of the intertidal snail Littorina acutispira Smith

Littorina acutispira Smith, a minute gastropod of < 3 mm shell height, lives at great densities in pools and on rock-surfaces at the highest levels of sea-shores in New South Wales. Populations from pools and dry areas were sampled on two shores for 18 months to investigate seasonal changes in density, size-structure, rates of growth and reproductive biology of the snails. Densities of snails increased between February and May, due to an influx of juveniles, and then declined until the following February, when they increased again. The decrease in density was due to the death of the largest snails at the end of summer, and the mortality of medium-sized snails between June and January. Longevity was estimated as 1–2 yr, but most individuals died by ≈ 16 months from settlement on the shore. Newly-settled snails grew to merge in size with those of the previous year's population by winter. During the summer months, the rate of growth of snails from a sheltered shore was greater than that of snails on a shore exposed to wave-action. Laboratory experiments revealed that this could be attributed to the presence of better quality food, or food in greater abundance on the sheltered shore, compared with the exposed shore.During winter months, but not during the summer, snails from dry areas grew more slowly than those from pools. An experiment demonstrated that some snails from dry areas might be able to compensate for reduced periods of feeding by being able to feed faster when submersed. This could not explain the differences in natural rates of growth.L. acutispira bred from October–November to March–April. Spawning in the laboratory was greatest during late summer (January to March). The percentage of mature oocytes in the gonads was small in winter and increased in early summer. Among the largest-sized snails, females outnumbered males. Two experiments, on unsexed and pre-sexed snails, demonstrated that the biased sex-ratio of the largest snails was due to faster growth by females.There was a greater density of snails on the exposed shore, which was correlated with the presence of barnacles. When barnacles were removed from experimental areas, the density of the snails declined within 24 h. This suggested that barnacles provided a refuge from wave-shock, rather than shelter from desiccation or high temperature. In laboratory experiments, snails were exposed to higher temperatures and less humidity than they would normally encounter on the shore. There was negligible mortality of small or large snails after 24 h of these conditions.This minute species grows quickly, recruits annually and has a short life-span. This type of life-history is discussed in comparison with similar small species from other habitats.     

SIZE-ASSORTATIVE MATING IN A NATURAL POPULATION OF VIVIPARUS ATER (GASTROPODA: PROSOBRANCHIA) IN LAKE ZURICH, SWITZERLAND

The number of mating pairs, the size of the mating partners,and the distribution of individuals of Vivi-parus ater on agrid in Lake Zürich were recorded during one breeding seasonin 1990. There was positive assortative mating with respectto shell size. The proportion of copulating individuals rangedfrom 1% to 6% (average 3%) of the active population at any onetime. Individual snails copulated 60 times on average from Apriluntil November. Snails were abundant and copulated in shallowwater close to the shore in Spring. They moved towards deeperareas in Autumn. V. ater copulated on all substrates at anydepth (1–9 m)of the grid. The spatial distribution ofcopulations throughout the summer reflected the pattern of snailabundance. (Received 29 January 1993; accepted 14 November 1994)     

Effects of habitat on growth and shape of contrasting phenotypes of Bembicium vittatum Philippi in the Houtman Abrolhos Islands,Western Australia

Translocation experiments were used to test the effect of habitat on growth and shape of three contrasting phenotypes of Bembicium vittatum: dwarf, highly domed snails from an usually dry tidal pond; large, moderately domed snails from a sheltered, regularly inundated pond; and relatively flat snails from a vertical, exposed shore. Snails from both ponds grew nearly twice as fast in the wet pond as in the dry pond, indicating a high degree of plasticity of growth. Associated with these changes in growth rates was convergence of shape. Under conditions of rapid growth, the dwarf snails became relatively flatter, and hence more similar to the native snails at that site. These results indicate that the dwarf phenotype is largely a plastic stunting in response to conditions of little submersion time. The snails from the exposed shore also grew faster in the sheltered, wet pond than at their native site. However, they not only retained their flat shape, but actually became flatter (and hence divergent from the pond snails) when grown in the pond. Thus, variation in shell shape was due to interactions between source population and a common plastic association of flatter growth profile with more rapid growth. Previous experiments had demonstrated high heritability of the flat phenotype, while the present results show that the expression of the genetically different types is affected substantially by the conditions of growth, and that phenotypic differences among populations may either overestimate or underestimate the underlying genetic differences. This unpredictability of the relationship between variation in shell form and its underlying genetic basis complicates interpretations of geographical variation or palaeontological sequences based on shell form.     

Population dynamics,growth and reproductive rates of Littorina nigrolineata Gray from a moderately sheltered locality in North Wales

The population dynamics of Littorina nigrolineata Gray on a moderately sheltered boulder shore was characterized by an almost constant mortality rate of post-juvenile snails both within and between years. Cohorts of post-juvenile snails had a “half life” of about 5 months. Juvenile snails had higher mortality rates than post-juveniles. Mortality rate was independent of size in post-juveniles. The population of post-juvenile snails fluctuated only about two-fold during the 3 yr of study. Maxima of population size occurred each September as a result of an autumnal pulse of recruitment into the post-juvenile size classes. These recruits originated from eggs laid in the summer of the previous year. The size-frequency structure of the population lacked well defined modes except during the spring and summer when the pulse of recruits formed a distinct mode, which later disappeared due to growth and mortality of the recruits. Oviposition continued throughout the year with a pulse in summer. Newly laid eggs took 4–7 wk to hatch. The hatchlings had a mean shell height of 0.5 mm and were estimated to take 6–9 months to grow to a shell height of about 4 mm. Growth increments over periods of 3 months. estimated from tagged cohorts of snails ranging from 3 to 16 mm, released on a stone jetty, revealed seasonal and size-specific changes in growth rate. Growth rate was significantly lower between December and March than at other times of the year. Seasonal differences in growth rate did not, however, significantly affect the yearly size increments of snails hatching in summer or winter. Growth increments did not decline linearly with increasing body size, as is assumed by the von Bertalanffy growth model, but increased to a maximum at about 6–7 mm shell height. Beyond the maximum, the growth increments decreased linearly with increasing shell height. This non-linearity caused an inflexion in the growth curve, so that the shell height. This non-linearity caused an inflexion in the and 12.4 mm by the von Bertalanffy growth curve. The deviation from the von Bertalanffy growth model was possibly due to departures from the assumptions that the assimilation rate is proportional to the square of a linear size dimension and that the metabolic rate is proportional to the cube of the same dimension. A von Bertalanffy growth curve, fitted to growth increments measured from “growth checks” on tagged snails on the boulder shore, predicted slightly faster growth rates than the von Bertalanffy growth curve fitted to increments of mean cohort size on the stone jetty. The closeness of the two curves, however, suggested that both methods of measuring growth increments were tolerably accurate.     

COMPARISONS OF THE BIOLOGY OF THE INTERTIDAL SUBANTARCTIC LIMPETS NACELLA CONCINNA AND KERGUELENELLA LATERALIS

Two limpet species occur intertidally on subantarctic SouthGeorgia, the patellid Nacella concinna and the siphonarlid Kerguelenellalateralis. N. concinna is confined to the lower shore closeto LWS; K. lateralis occurs in middle shore pools, so theirdistributions do not overlap. N. concinna has a much narrowerthermal niche (–12.9°C to +15.6°C) than K. lateralis(–17.8°C to +31.8°C). Environmental data are presentedto show that the upper lethal temperature of N. concinna islow enough to prevent the limpet living higher on the shore.Both limpet species are slow-moving, but K. lateralis showsincreasing speed with rising temperature, peaking at 15–20°C.In contrast, N. concinna moves actively down to –1.9°C(when sea water freezes), but there is a steady decrease inspeed of locomotion above +2°C. Locomotion ceases at 14°Cin N. concinna (c.f. 30°C in K. lateralis). Both speciesexhibit very low tenacities, but in N. concinna tenacity decreaseswith increasing shell length. In K. lateralis there is no effectof temperature on tenacity. Both species show a positive allometricrelationship between foot area and shell length. N. concinnafeeds upon microbial films and microepiflora, but K. lateraliseats colonial diatoms and Enteromorpha bulbosa. Observationson shell middens of the kelp gull Larus dominicanus showed thatthe gulls did not eat K. lateralis, though they ate great quantitiesof the less accessible N. concinna. Gulls ate N. concinna assmall as 11 mm shell length (within the size range of K. lateralis).Experiments on gulls demonstrated an unwillingness to eat K.lateralis, probably because the siphonariid extrudes a viscidwhite mucus when the foot is touched. (Received 9 May 1996; accepted 8 July 1996)     

PHENOTYPIC CHANGES IN THE PROGENY OF NUCELLA LAPILLUS (GASTROPODA) TRANSPLANTED FROM AN EXPOSED SHORE TO SHELTERED INLETS

Nucella lapillus adults were transplanted from the exposed northCornish coast (Bude) to two sheltered inlets in south Devon(Dart Estuary and Plymouth Sound), sites where the native populationshad ceased breeding as a result of sterilisation by tributyltin(TBT) pollution. The transplants produced progeny, many of whichsurvived to maturity and beyond (6 + years). The shell formof Bude individuals is typical of animals exposed to severewave-action, i.e. of light construction with a large apertureto accommodate a broad foot allowing for strong adhesion. However,in contrast, the shell form of the progeny differed markedlyfrom that of their parents; they had the characters associatedwith development under sheltered conditions and subject to severecrab predation, i.e. a robust structure and small aperture.Statistical analysis of the shell characters of the two generationsindicated highly significant differences. It is concluded theseobservations provide evidence of the remarkable plasticity inthe shell form of N lapillus in that both ‘exposed morph’and ‘sheltered morph’ can be produced by the samegene pool when subject to the actions of the appropriate selectionpressures. (Received 25 September 1992; accepted 14 October 1992)     

LIFE CYCLE AND HABITAT SHIFT OF THE TROCHID SNAIL DILOMA SUAVIS (PHILIPPI) WITHIN INTERTIDAL MUSSEL ZONES

The life cycle of the trochid snail Diloma suavis (Philippi, 1849),was studied on an intertidal rocky shore at Shirahama, Wakayama Prefecture,where two mytilid bivalves, Septifer virgatus (Wiegmann) andHormomya mutabilis (Gould), formed vertically contiguous musselbeds in the upper-middle and lower zones, respectively. At lowtide in April, the snail density increased with decreasing shoreheight and was greatest at the middle level of the H. mutabilisbed. Then, the density decreased towards the lower littoralfringe. Newly settled juveniles smaller than 2 mm in shell heightappeared abundantly in late summer and autumn within algal turfon the lower shore. As snails grew larger than 2 mm, they appearedwithin the gaps of the H. mutabilis bed and the S. virgatusbed. They increased in size monotonically towards the next summer,but rate of growth in shell height tended to be great in autumnand small in winter. Seasonal change in the density of snailsfound within the gaps of the mussel beds was remarkable during athree year period, increasing from autumn to winter and then decreasingtowards next summer. Reproduction occurred in summer, and adultsnails disappeared by September. It is thus suggested that this specieshas a one year lifespan and shows a habitat shift from algal turfto the gaps of the mussel beds with growth. (Received 12 October 1998; accepted 2 March 1999)     

Habitat-specific size structure variations in periwinkle populations (<Emphasis Type="Italic">Littorina littorea)</Emphasis> caused by biotic factors

Shell size distribution patterns of marine gastropod populations may vary considerably across different environments. We investigated the size and density structure of genetically continuous periwinkle populations (Littorina littorea) on an exposed rocky and a sheltered sedimentary environment on two nearby islands in the south-eastern North Sea (German Bight). On the sedimentary shore, periwinkle density (917 ± 722 individuals m⁻²) was about three times higher than on the rocky shore (296 ± 168 individuals m⁻²). Mean (9.8 ± 3.9 mm) and maximum (22 mm) shell size of L. littorea on the sedimentary shore were smaller than on the rocky shore (21.5 ± 4.2 and 32 mm, respectively), where only few small snails were found. Additionally, periwinkle shells were thicker and stronger on the rocky than on the sedimentary shore. To ascertain mechanisms responsible for differences in population structures, we examined periwinkles in both environments for growth rate, predation pressure, infection with a shell boring polychaete (Polydora ciliata) and parasitic infestation by trematodes. A crosswise transplantation experiment revealed better growth conditions on the sedimentary than on the rocky shore. However, crab abundance and prevalence of parasites and P. ciliata in adult snails were higher on the sedimentary shore. Previous investigations showed that crabs prefer large periwinkles infested with P. ciliata. Thus, we suggest that parasites and shell boring P. ciliata in conjunction with an increased crab predation pressure are responsible for low abundances of large periwinkles on the sedimentary shore while high wave exposure may explain low densities of juvenile L. littorea on the rocky shore. We conclude that biotic factors may strongly contribute to observed differences in size structure of the L. littorea populations studied on rocky and sedimentary shores.     

The effects of exposure and predation on the shell of two British winkles

Changes in shell size and shell shape of the two British winkles Littorina nigrolineata and L. rudis were studied in relation to exposure and to crab-size. In both species, shells from exposed shores are smaller and more globose than those from sheltered shores. Also, in rudis of exposed shores the mouth is relatively wider. In shores of equally sheltered conditions, shells are bigger at those localities where crabs are large than at those localities where they are small. The largest shells are found in those localities where it is extremely sheltered, and the crabs are very large.
It is argued that on exposed shores, small shells are favoured because they have more possibilities than large ones to shelter in crevices and in barnacle interspaces, from the impact of winds and waves. A globose shell could accommodate more foot muscle and thus enable a stronger adherence to the rock; and an increased mouth diameter would increase the area of foot adherence to the rock. On sheltered shores, on the other hand, large, narrow-mouthed shells are favoured because they discourage crab predation, large crabs being abundant mainly on sheltered shores.
The possible significance of shell size and shape in relation to zonation is discussed, in view of the different predatory and physical conditions which prevail in different zones of the shore, and the different shell specializations which these conditions would require.     

Distribution in relation to life history in the directdeveloping gastropod Batillaria cumingi (Batillariidae) on two shores of contrasting substrata

Population structure and spatial distribution with growth ofthe direct-developing gastropod, Batillaria cumingi, were investigatedon two shores of differing substrata. Sand-mud shore and rockyshore populations differed in size structure; first-year snailswere ca. 7 mm in shell length (SL) in both populations, whereassecond-year snails, merging with older cohorts, measured 15–25mm SL in the sand-mud shore population and ca. 15 mm SL in therocky shore population. Egg distribution matched adult distributionin the sand-mud shore population, but was more restricted thanthat of adults in the rocky shore population. The distributionof newly-hatched juveniles (0–1 mm SL) was restricted inboth populations, but the growth stage at which snails extendedtheir distribution differed between the two populations; 1–2.5mm SL on the sand-mud shore and 5 mm SL on the rocky shore. Floatingachieved by early juveniles (ca. 2 mm SL), was commonly observedin the sand-mud population, but rarely in the rocky shore population.The sudden expansion in distribution of the 1–2.5 mm SLgrowth stage in the sand-mud shore population is consideredto have been caused by floating, while expansion of the distributionof older growth stages (>5 mm SL) in the rocky shore populationprobably occurs by crawling. (Received 13 May 1998; accepted 25 August 1998)     

Influence of shell morphometry,microstructure, and thermal conductivity on thermoregulation in two tropical intertidal snails

Tropical intertidal organisms tolerate large fluctuations in temperature and high desiccation rates when exposed during low tide. In order to withstand the short‐term heat stress, intertidal organisms adopt behavioral responses to maximize their survival. Our previous research showed that tropical littorinids found at the upper and lower intertidal shores in Singapore exhibited different behavioral adaptations during low tide. Most of the upper‐shore Echinolittorina malaccana kept a flat orientation, with the aperture against the substrate and the long axis of the shell towards the sun, whereas a majority of the lower‐shore individuals of Echinolittorina vidua stood with the edge of the aperture perpendicular to the substrate on the rocky shore during low tide. This prompted analyses of the shells of these two species to determine whether the differences in the shell morphometry, microstructure, and thermal conductivity of shells of E. malaccana and E. vidua were associated with their respective behavioral responses to thermal stress. Analyses of shell morphometry and thermal conductivity showed that shells of E. malaccana were more likely to minimize heat gain, despite having a higher thermal conductivity on the outer surface, due to their light‐gray, elongated shell. By contrast, the dark‐colored, globose shells of E. vidua probably gain heat more readily through solar radiation. Scanning electron microscopy images of the shells of both littorinid species further revealed that they have cross‐lamellar structure; however, only individuals of E. vidua showed the presence of disjointed rod layers and a pigmented inner shell surface. Individuals of E. malaccana had a rough outer shell surface with holes that inter‐connect to form water‐trapping channels that probably aid cooling. Individuals of E. vidua, however, had a smooth outer surface with rows of kidney‐shaped depressions as microsculptures which probably help to stabilize shell shape. In both Echinolittorina species, behavioral responses were used to overcome thermal stress during low tide that was associated with shell morphometry and shell thermal conductivity. Such combined adaptations increase survivability of the littorinids at their respective tidal levels.     

SEXUAL DIMORPHISM OF NUCELLA LAPILLUS (GASTROPODA: MURICIDAE) IN NORTH WALES, UK

Multivariate statistical methods were employed to examine sexualdimorphism in size and shape of Nucella lapillus collected from 16sheltered sites along coasts of Anglesey and the Lleyn Peninsula, NorthWales, UK. Females were significantly larger than males in overallsize; among 12 relative measures of shell shape, two ratios (shellwidth/shell length and aperture length/shell length) were significantlydifferent between males and females, but these differences usuallydecreased with increasing age (shell length). The observed hypoallometricdimorphism could be a result of selection on increased femalefecundity, which may be positively correlated with shell sizein N. lapillus as in other gastropod species. (Received 22 November 1999; accepted 10 April 2000)     

The role of growth,predation, and habitat selection in the population distribution of the crown conch,Melongena corona Gmelin

The intertidal gastropod Melongena corona Gmelin exhibits a size gradient along the shore as a function of habitat. Small, juvenile snails were found on the sand beach and larger adults were found on hard substratum habitats (shell rubble beach and oyster bar). A transplant experiment was performed to test three proximate explanations for this pattern:

• 1.(1) differential growth rate of conchs in different habitats,
• 2.(2) differences in predation intensity between habitats and
• 3.(3) active habitat selection by snails.

Shell scars on the last shell whorl were used as an index of predation pressure. Growth and predation were not significantly different for snails of similar size in different habitats, but snails were found to return to their original habitat when displaced. The ability to home did not differ between sexes or juveniles and adults. Active habitat selection appears to be a significant proximate factor maintaining the population distribution. A number of potential ultimate causes of the size class segregation are suggested.     

Variability of foraging in highshore habitats: dealing with unpredictability

Intertidal areas are habitats at the border of two very different environments: the marine environment and the terrestrial environment. In contrast to many habitats at borders, intertidal areas are very variable in space and time. They have upshore and alongshore gradients of environmental conditions, which change through time in predictable (due to changing tides) and unpredictable (due to changing weather) ways. Because most animals and plants on rocky shores are marine in origin, extreme highshore levels are generally considered more harsh and unpredictable environments than are mid- or lowshore levels. In this study, the linear distances and directions dispersed by the littorinid Littorina unifasciata while foraging were compared across randomly-chosen mid- and highshore replicated sites to test the hypothesis that movement during foraging was more variable from place to place in complex midshore habitats. Experiments were repeated on different days to test the hypothesis that temporal variability in movement was greater at high- than at midshore levels because environmental conditions necessary for foraging were more variable high on the shore. Finally, the data were used to test models about differences in variability among individuals in the same patch of habitat according to their recent history of submersion/emersion. In contrast to expectations, dispersal was more variable in space and time within and among highshore sites. The implications of such variable behaviour are discussed with respect to the generalization of patterns of behaviour from sparse data and the levels of replication needed in the design of experiments to investigate behaviour of intertidal animals.     

Maintenance of zonation patterns in two species of flat periwinkle,Littorina obtusata and L. mariae

The zonation patterns of Littorina obtusata (L.) and Littorina mariae Sacchi et Rastelli were shown to be quite distinct on a sheltered rocky shore. L. obtusata was found at all the heights sampled; it reached peak numbers at mid shore on the alga Ascophyllum nodosum L. (Le Jol). There was no difference in the tidal height occupied by adults or juveniles; or in the mean size of L. obtusata along the vertical gradient of the shore. In contrast L. mariae occurred exclusively low on the shore, on Fucus serratus L. Translocation of the two species within their respective levels resulted in random movement after 4 days, although initial movements after 1 and 2 days were sometimes directional. Animals transplanted to the normal level of the other species showed directional movement towards their home zone; this was most pronounced after 4 days. There was no difference in the distance moved by the two species, although the distance moved did vary with tidal height, both species moving further at mid shore than low shore. Distances moved by littorinids at replicate areas in the low shore were similar but those at mid shore did vary. There was an interaction between the species and the different tidal heights which revealed that transplanted species moved further than translocated species at the same tidal level. However, this was only significant in the case of L. mariae. It is suggested that the close relationship between the winkles and their host algae may direct the homing behaviour of displaced individuals.     

Variation in adult shell morphology and life-history traits in the land snail Oreohelix cooperi in relation to biotic and abiotic factors

In this study, we explore factors influencing variation in adultshell morphology and life-history characteristics (offspringsize and number) in the ovoviviparous land snail, Oreohelixcooperi, from the Black Hills, South Dakota and Wyoming, USA.To date, no study of shell morphology and life-history characteristicsin ovoviviparous land snails has included a combination of datasetsincluding genetic data, life-history traits, shell morphologyand multiple environmental factors. We report that differencesin shell size among populations are strongly related to meanannual temperature (and the highly correlated variable elevation)and population density (measured as shell density). In addition,shell size varies among populations, despite an apparent lackof population genetic differentiation. Common factors thoughtto influence adult shell size, like precipitation and calciumlevels, do not have a significant effect in this study. Adultsize strongly influences per-clutch reproductive output, withlarger snails having larger and more offspring. As mean annualtemperature and shell density affect adult shell size, theyalso indirectly affect per-clutch output. The results suggestthat a large portion of the life-history variation in O. cooperiis environmentally induced, as has been found in oviparous landsnails and brooding freshwater bivalves. (Received 20 July 2006; accepted 10 January 2007)