Locomotory behavior and habitat selection in littoral gastropods on Caribbean limestone shores.

ABSTRACT Microhabitat selection and diurnal and nocturnal movement patterns in two tropical littoral gastropods (Nerita versicolor and Tectarius antonii) were compared in relation to shore position allowing for an assessment of the influence of physical stressors (heat and desiccation) on their behaviors and vertical distributions. Monitoring of individually marked specimens indicated that movement, and presumably feeding activities of tropical littoral gastropods, is initiated during periods of decreased desiccation and thermal stress. Near continuous movement was observed in snails inhabiting moist eulittoral (midshore) habitats. In contrast, snails inhabiting the upper eulittoral fringe (high-shore) exhibited movement only during periods of rain or when shores were moistened by increased wave action. Twenty-four hour surveys indicated a preference for nocturnal activity in mid- and high-shore habitats. The propensity for increased activity during conditions of reduced desiccation stress was also supported by an experiment in which wetting of the substrate (simulating inundation) encouraged activity in mid- and high-shore species. Mid- and high-shore species exhibited a preference for sheltered microhabitat and avoided exposed surfaces.

KEY WORDS: habitat selection, Littorinidae, locomotory behavior, Neritidae, rocky shores 

INTRODUCTION

Rocky shore intertidal habitats are subject to varying intensities of physical stress, particularly with respect to temperature extremes and desiccation, which increase with increasing shore height (reviewed by Lewis 1964, Newell 1979, McMahon 1988, McMahon 1990). This gradient of physical stress is partially responsible for the zonation of species in rocky shore habitats because each vertical zone is set apart by a unique suite of selective pressures (McMahon 1990). Littoral gastropods have accumulated a number of specialized adaptations in response to periodic tidal emergence, many of which correlate with vertical shore position (reviewed by Underwood 1979, McMahon 1988, McMahon 1990). Indeed, many morphologic, physiologic, and/or behavioral adaptations have been interpreted as responses to desiccation and heat stresses encountered during tidal emergence.

Shell morphologies, which have been suggested to mitigate effects of thermal and desiccation stress include shape, pigmentation and/or ornamentation, size, and presence of opercula (Vermeij 1973, Heath 1975, Britton 1995). Such morphologies are generally assumed to ameliorate the effects of heat stress while eliminating the need for evaporative cooling and allowing for extended periods of emergence while conserving water (McMahon 1990, Britton 1995, Lang et al. 1998). Physiologic adaptations among gastropods in response to emersion-induced thermal and desiccation stress include metabolic thermal regulation (McMahon & Russell-Hunter 1977) and increased thermal (McMahon 1991, McMahon 2001) and desiccation (Britton 1992) tolerance. Behavioral adaptations to reduce the effects of thermal and desiccation stress in tidally emersed gastropods include selection of favorable microhabitats (Kensler 1967, Raffaelli & Hughes 1978), spatial distribution (Chapman & Underwood 1996), individual posture (Britton 1995), evaporative cooling (McMahon 1988), production of mucus holdfasts (Bingham, 1972), and restricting movement to periods of reduced stress (Peckol et al. 1989, Lang et al. 1998).

While behavioral adaptations in littoral gastropods have been explored, very few studies have examined locomotory rates for individual species, particularly with respect to shore position. Only by simultaneously comparing species from varying shore heights can the adaptive significance of behavioral responses in relation to shore position be assessed in a biologically meaningful manner. In this study, locomotory behavior and habitat selection were used tO examine the role of physical stressors in structuring behavioral adaptations in gastropods from eulittoral and eulittoral-fringe areas.

METHODS

Study Area

Comparisons of locomotory behavior and habitat selection in eulittoral and upper eulittoral-fringe species (referred to hereafter as mid and high-shore species, respectively) were examined on natural limestone hard shores of the eastern Yucatan Peninsula, Caribbean Sea.

Most Caribbean rocky shores are remnants of ancient coral reef communities consolidated over time by intense geologic pressure (Kaplan 1988). The resulting limestone rock is continuously weathered by wave action creating pits, cracks, and tide pools that provide a range of habitats for many littoral organisms (Kaplan 1988). The zonation scheme used to describe tropical hard-shore communities is based on the distinct gradations in rock color from weathered white rock surfaces in the highest zone, to distinctly colored growths on rock surfaces in low shore zones (Stephenson & Stephenson 1950, Kaplan 1988). The white zone is the uppermost region above the upper eulittoral-fringe and in many localities is occupied by upland and/or maritime vegetation. The gray and black zones of tropical hard shores correspond to the upper eulittoral-fringe of more traditional classification schemes. The gray zone is the highest and often widest portion of the upper eulittoral-fringe and because of its height on shore, remains dry for most of the year, except when inundated by rainfall and/or tropical storm surges. The black zone occupies the lower portion of the upper eulittoral-fringe and remains mostly dry due to a lack of wave spray,and daily tidal inundation. However, during spring high tides and storm surges, the black zone can remain wet for extended periods. Below the upper eulittoral-fringe (gray and black zones) is the yellow zone. This is the true intertidal zone (=eulittoral) and exhibits the greatest species diversity. Kaplan (1988) describes a pink zone below the eulittoral corresponding to the lower eulittoral fringe of more traditional classification schemes (Stephenson & Stephenson 1950).

Two rocky points, Punta Yu Yum and Punta Xamach, hereafter referred to as study sites 1 and 2, respectively, were sampled in the Sian Ka'an Biosphere Reserve on the eastern side of the Yucatan Peninsula (Fig. 1). Site 1 (19[degrees]55'N, 87[degrees]28'W) is a large, gently-sloped rocky shore that widens to form two large rocky outcrops. All five intertidal color zones are represented and visibly distinct. Site 2 (19[degrees]55 'N, 87[degrees]26'W), situated to the south of Site 1 (Fig. 1), is similar with respect to shore steepness and zone width except that a sandy beach transcends from the black zone directly into upland vegetation effectively replacing the gray zone. Both rocky shores are bordered on the north and south sides by large lagoon-type bays with sandy bottoms and seagrasses. Rocky shores in this region of Mexico exhibit gradual vertical relief, slowly increasing with distance from the water (i.e., platform-like).

[FIGURE 1 OMITTED]

The prosobranch gastropods, Nerita versicolor Gmelin, 1791 (family Neritidae) and Tectarius antonii Philippi, 1846 (family Littorinidae), dominated the mid and high-shore areas of the two study sites, respectively. Although the distribution of these two species is overlapping for a portion of their vertical range, Tectarius antonii is distributed higher on the shore than Nerita versicolor (Peckol et al. 1989, McMahon 2001). It has been reported that snails in the genus Nerita occupy high-shore positions (Garrity 1984), but at the sites reported herein and shores in Jamaica (McMahon 2001) and the Bahamas (Peckol et al. 1989), species of Nerita occupy lower shore positions within areas effected by daily tidal inundation.

Site and Shore Level Characterization

Rugosity and zone width measurements were used to characterize the substrata of mid and high-shore zones. Rugosity is defined as the length of surface contour divided by the total linear length measured (Rooker et al. 1997). Nine measurements of rugosity in the mid and high-shore zones of each site (n = 36) were estimated by haphazardly placing a meter-long stick with an attached piece of twine directly onto the shore. The twine was used to measure the surface contour of the rock immediately below and between the ends of the stick. Zone width (m) was estimated from nine measurements taken within the mid and high-shore areas of each site (n = 36). The eulittoral zone width was measured as the linear distance from the uppermost boundary of the pink zone to the uppermost boundary of the yellow zone. The upper eulittoral fringe zone width was measured as the distance from the uppermost boundary of the yellow zone to the uppermost boundary of the black zone.

The temperatures of rock surfaces were measured in each zone to test the assumption that substratum temperature increased with shore height. Each day, the bulb of a mercury thermometer was placed against the rock surface at three locations in each zone for 5 min and the temperatures were recorded.

The assumption that desiccation stress increases with increasing shore height was tested using a sponge-survey method (Minton & Gochfeld 2001) and by capturing wave splash and spray with rain gauge-type collection devices. For the sponge-survey method, synthetic, kitchen-type sponges were alphabetically marked, dried, and placed in individual gallon size zip-loc bags (with a corresponding alphabetic mark). Each bag was weighed on an electronic balance. Three sponges, each removed from its bag and wetted with 75 mL of seawater, were placed on the substrate in each zone (n = 6). After 1 h, the sponges were collected and reweighed. All sponges were kept sealed in plastic bags except for the duration of the hour-long survey to eliminate the further loss or gain of water between weight measurements. The amount of water lost or gained was calculated by subtracting the combined weight of the sponge and 75 mL of seawater (76.74 g at 35[per thousand]) from the final weight.

Seawater spray resulting from crashing waves was collected using rain gauge-type devices constructed from PVC pipe and secured into cement bases that were arbitrarily placed above and below the mean high water line. At each site, the devices were deployed for a period of 1 h on three different days to capture the day-to-day variation in wave strength over the course of the 12-day experiment (3 devices x 3 different days x 2 zones = 18 total measurements at each site). The volume (mL) of spray collected above and below the mean high water line was used as an additional indicator to define varying degrees of moisture conditions occurring among mid and high-shore zones. One-way analyses of variance (ANOVA) were used to compare mid and high-shore zones with respect to rugosity, zone width, rock surface temperature, and desiccation potential (sponge-survey and seawater spray collection).

Movement and Microhabitat Selection

Field observations were made from May 12-25, 2002. Three subsites from mid and high-shore positions were sampled at each study site (3 midshore x 3 high-shore x 2 study sites = 18 total subsites). Within each subsite, five adult individuals of each species were arbitrarily selected and marked with a uniquely colored enamel paint to ensure individual distinction (methodology after Britton & McMahon 1992, Gochfeld & Minton 2001). Marked specimens of N. versicolor and T. antonii had similar size distributions as individuals in the natural population (16.6 [+ or -] 1.8 mm & 13.9 [+ or -] 1.6 mm shell length, respectively). Initial location of each snail was marked by drilling a hole approximately 5 cm south of each individual and installing a colored nail corresponding to the color of the painted snail (home mark). Three initial observations were recorded for each marked specimen: (1) distance to the water (m); (2) compass heading to the water (degrees); and (3) a description of the microhabitat in which it was found. All distances were estimated using a fiberglass tape measure. Microhabitat was classified based upon surface topography and substratum moisture (after Minton & Gochfeld 2001). Surface topography was classified as exposed surface (flat rock), crevice (depression in the rock surface large enough for two or more individuals to occupy), or pit (depression in rock large enough for a single individual to occupy). Microhabitats were further classified as wet (wet but no standing water), at the waterline, underwater, or dry.

Each subsequent day, as many individual snails as possible were recovered. A second nail, of the same color, was used to mark the new location of each individual (day mark). Distance and direction traveled from the previous day's mark (colored nail) as well as the microhabitat occupied were recorded dally for each individual located. Distance and direction from the home mark was also recorded to determine a "home area" measurement. For those individuals that moved, the day mark was repositioned at the snail's new location.

Daily movement rates were calculated by dividing the distance traveled by the time elapsed since the previous day's record. Average daily movement rates (cm x [h.sup.-1]) were estimated from a linear mixed model that accounted for the serial correlation of errors due to the repeated sampling of individuals in time. The model used a reference cell parameterization and indicator variables to analyze the effects of shore position (mid vs. high) on rate of movement. The model was estimated using maximum likelihood via the MIXED procedure in SAS (Cary, North Carolina). Directional preference was examined using a circular analysis (Oriana for Windows, Wales, United Kingdom). All headings were standardized by adjusting the heading so that the direction to the waterline was zero degrees. For example, if circular analysis indicated the 95% confidence interval of the mean heading was 102[degrees] to 200[degrees], it was concluded that there was a movement away (180[degrees]) from shore. Directional measurements from each site were pooled and analyzed with respect to shore position.

The observed versus expected proportions of microhabitat use were compared using a method based on Bonferroni's inequality after Haney and Solow (1992). Expected proportions of available microhabitat were estimated using a 25 x 25 cm PVC quadrat divided into 25 equal 5 x 5 cm squares by a plastic cord. Sixteen points were created where each of the plastic cords intersected one another. The quadrat was haphazardly placed on the substrate and the microhabitat and degree of substratum moisture (see above) present beneath each of the 16 points were recorded. A total of 192 points (12 quadrats) were sampled in each zone at each site.

The distances between daily locations and home marks were analyzed for the presence of homing behavior by calculating the percentages of home area measurements falling into each of 4 categories (<1, 1-3, 3-5, and >5 m). Simple comparisons of these percentages were made between mid and high-shore zones at each site.

Twenty-four Hour Surveys

Two 24-h diurnal surveys of locomotory behavior were conducted to test the hypothesis that movement increases when desiccation and heat stress are reduced. Site 1 was chosen for these diurnal surveys. Surveys were conducted from May 16-17 and 21-22, 2002. The movements of 15 mid-shore N. veriscolor and 15 high-shore T. antonii were monitored at 4 h intervals for a 24 h period using the same methodology described early for daily locomotory rates. The linear mixed model described earlier for daily rate measurements was similarly used to compare movement rates among mid and high-shore species and time of day (day vs. night).

Density and Aggregation

Spatial patterns of distribution in mid- and high-shore species were analyzed using Goodall (1974) variation of Pielou (1964) two-phase patchworks method. The modification uses presence/ absence data in random pairs of quadrats (without replacement) over increasing quadrat distances to determine patch size. In this study, a 25 x 25-cm PVC quadrat, separated into 25 equal 5 x 5-cm cells, was haphazardly placed on the substrate and the presence or absence of each snail species was noted in each of the 25 squares. Random pairing of cells was analyzed over distances of 5-10, 10-15, 15-20, and 20-25 cm. The observed proportions of specimens occupying cell pairs significantly exceeding expected proportions were taken as evidence of a continuous patch. Density was estimated by counting the number of individuals of N. versicolor and T. antonii occurring within 12 quadrats (25 x 25 cm) in the mid and high-shore zones, respectively at each study site.

Wetting Experiment

A wetting experiment was conducted to estimate the probability of snails moving under different environmental conditions (dry-substratum, freshwater moistened substratum, and seawater moistened substratum) (modified from Gochfeld & Minton 2001). Forty-five individuals of each species were collected from their respective mid- and high-shore zones at Site 1. Five snails were randomly assigned to each of the three treatment groups (15 total) and placed on an outdoor testing surface (concrete platform). Ten minutes prior to testing, each group of snails was placed onto one of three 50 x 50 cm treatment squares that were drawn on the testing surface. Five minutes prior to testing, the snails were moved to the middle of their respective squares and left untouched for the duration of the test. One of three treatments was randomly applied to each group. Treatments tested were dry-substratum (control), 250 mL of fresh water, and 250 mL of saltwater. Freshwater and saltwater treatments were designed to represent 1 mm of rainfall and saltwater inundation and/or spray, respectively and were applied evenly to the entire treatment box using a perforated-capped water bottle. Specimens were observed for periods of 10-min during which the total number of individuals exhibiting movement was recorded. The experiment was replicated three times for each species (15 snails x 3 replicates = 45 individuals of each species). A weighted-least-squares analysis was used to model the probability of moving under the various treatments via the CATMOD procedure in SAS (Cary, North Carolina).

RESULTS

Site and Shore Level Characterization

The rocky shore habitats of mid and high-shore species were similar with respect to substrate morphology (measured by rugosity) and area (zone width; Table 1). The substratum surface in both zones was rough, weathered rock with many small pits and sharp edges. In contrast, mid- and high-shore zones differed significantly with respect to desiccation and heat stress. Rock temperature increased significantly with increasing shore height as did desiccation potential, measured as the amount of seawater lost by evaporation (or gained from sea spray) using the sponge-survey method (Table 1). On average, sponges in midshore zones absorbed an additional 0.27 g of seawater, whereas sponges in the high-shore zone lost 6.97 g of seawater (Table 1). No wave splash and spray were collected in high-shore zones during the 3-day survey period, whereas a total of 337 mL were collected in low-shore zones. The lack of wave splash and spray in high-shore areas precluded comparing zones by ANOVA, yet demonstrated the dry conditions prevailing in high-shore habitats.

Daily Movement and Direction

All 60 snails marked at Sites 1 and 2 (15 mid and high-shore specimens at each site) were subsequently recovered at least once over the course of the 12-day observation period. A total of 593 (86%) of the 690 possible observations (30 individuals x 11 days at Site 1 + 30 individuals x 12 days at Site 2) were recorded over both sites. Similar numbers of mid and high-shore individuals were subsequently recovered during surveys (e.g., 84% of N. versicolor and 89% of T. antonii).

Average daily movement rates were similar among snails within mid and high-shore habitats at both sites (t = 0.61, DF = 559, P = 0.544), thus these data were pooled to analyze the effects of shore position. N. versicolor and T. antonii moved during 94% and 53% of the daily intervals examined, respectively. The mean daily movement rate of the midshore species (3.12 cm x [h.sup.-1]) was significantly higher than the rate of the high-shore species (0.27 cm x [h.sup-1]; t = 10.79, DF = 58, P < 0.0001).

Interestingly, a storm developed 6 days into the observation period (May 19, 2002) beginning with 2 days of rainfall followed by 4 days of increased sea state resulting increased wave heights and subsequent wetting of high shore zones. This provided a unique opportunity to examine the effects of increased wave action on the behavior of midshore species and an increase in substratum moisture on behavior of the high-shore species. Using the same analysis, but introducing a new variable that divided the daily observations into normal and increased wave heights, the effect of shore position on the rate of movement was re-evaluated with respect to increased substratum moisture. The increased wave heights affected the movement rates of mid and high-shore species differently (Fig. 2). Under heavy sea conditions, the rate of movement in the midshore species was similar to that recorded during normal sea state, averaging 2.79 cm x [h.sup.-1] (t = -0.77, DF = 502, P = 0.444). In contrast, the high-shore species exhibited an increase in mobility to 1.63 cm x [h.sup.-1]. Although the locomotory rates of the high-shore species continued to exceed those of the midshore species (t = 2.83, DF = 502, P = 0.005), the mobility of T. antonii increased 6-fold over that observed during normal sea conditions (t = 6.83, DF = 502, P < 0.0001).

[FIGURE 2 OMITTED]

Daily directional movement was analyzed as a 2-factor experiment with respect to shore position and substratum moisture. Daily calculations having <10 observations were removed from the analysis as suggested by Underwood and Chapman (1985). Midshore N. versicolor exhibited directional movement to a higher shore position during the period of increased wave action, otherwise no specific patterns of movement were observed (Fig. 3). Prior to the wetting of the high shore, few movements (favoring no particular direction) were observed in high-shore T. antonii. As wave action increased, some directional movements were evident, hut followed no discernible pattern (Fig. 3).

[FIGURE 3 OMITTED]

Percentages of home area measurements falling into each of 4 categories (< 1, 1-3, 3-5, or >5 m) indicated that 29% of midshore individuals remained within 1 m of their initial mark, 55% ranged from 1-3 m from their home mark, and 16% ventured farther than 3 m. A few N. versicolor (8%) were observed to travel more than 5 m from their home mark with a maximum distance of 17 m recorded. In contrast, 83% of high-shore individuals were found within 1 m of their home mark and the remaining 17% never moved farther than 3 m.

Microhabitat Selection

Availability versus use of the three microhabitat categories (exposed rock, crevice, and pit) were compared among mid- and high-shore species using quadrat data pooled over both sites. Exposed rock surfaces comprised the majority of available microhabitat in mid- and high-shore zones (54% and 66%, respectively), followed by crevice (43% and 30%), and pit (3% and 4%; Table 2). Prior to the increase in wave heights, the midshore species showed a preference for pit (26%) and avoided exposed surfaces (20%; Fig. 4). During the storm-induced increase in wave heights and subsequent increase in water level on the shore, N. versicolor no longer exhibited preference or avoidance for any particular microhabitat, the majority of individuals were found on exposed surfaces (44%) and within crevices (43%), with fewer using pit (12%) microhabitats. The increase in substratum moisture had little effect on microhabitat use by high-shore individuals. Specimens of T. antonii avoided exposed surface (27% and 29%) and preferred crevice habitat (63 and 56%) during times of normal and increased wave action, respectively (Fig. 5).

[FIGURES 4-5 OMITTED]

Proportions of individuals occupying each substratum moisture category (wet, waterline, underwater, and dry) were calculated only for those observations taken prior to the period of increased wave action. During this period, wet substrata comprised 95% of available microhabitats in mid-shore zones (moist and immersed rocks) and only 24% of available microhabitats in high-shore zones. No preference was detected for any of the substratum moisture categories in mid- or high-shore species (Table 3).

Twenty-four Hour Surveys

Surveys were conducted at 4-h intervals over two 24-h periods at Site 1; 3 intervals during the day (1100, 1500, and 1900) and 3 during the night (2300, 0300, and 0700). Wave heights were typical during the first 24-h survey but markedly higher throughout the second. Thus, data from the first and second 24-h observation periods were analyzed separately to allow the effects of the increased wave heights and subsequent increases in water level on the shore to be evaluated (Fig. 6).

[FIGURE 6 OMITTED]

Under the more typical shore conditions of the first 24-h survey, the midshore species increased the rate of movement nearly 3-fold between day (3.96 cm x [h.sup.-1]) and nighttime (11.63 cm x [h.sup.-1]) observations (t = 3.62, DF = 133, P = 0.0004). In contrast, movement rate of the high-shore species was similar among day (0.13 cm x [h.sup.-1]) and night (0.01 cm x [h.sup.-1]) observations (t = -1.04, DF = 133, P = 0.3025). A higher rate of movement was observed for the midshore species when compared with the high-shore species during both daytime (t = 4.79, DF = 27, P < 0.0001) and nocturnal observations (t = 6.23, DF = 133, P < 0.0001).

The increase in wave action occurring during the second 24 h survey resulted in marked responses among mid- and high-shore species. Midshore N. versicolor again exhibited a significant preference for nocturnal movement (t = 3.35, DF = 121, P = 0.0011), increasing from 0.93 cm x [h.sup.-1] in daylight hours to 4.46 cm x [h.sup.-1] during nocturnal observations. High shore T. antonii also exhibited a preference for nocturnal movement (t = 2.20, DF = 121, P = 0.0296), increasing from 2.74 cm x [h.sup.-1] in the daytime to 5.77 cm x [h.sup.-1] at night. In contrast to the first 24 h observation, mid- and high-shore locomotory rates did not differ from one another during daylight or nocturnal surveys (Day: t = -1.86, DF = 25, P = 0.0751; Night: t = -0.65, DF = 121, P = 0.5141).

The increased wave action had a marked effect on the locomotory behavior of midshore N. versicolor with diurnal rates decreasing 4-fold and nocturnal rates decreasing 3-fold when compared with rates recorded under normal wave conditions. In contrast, increased substratum moisture in high-shore habitats resulting from the increased wave heights resulted in greater movement. Locomotory rates of high shore T. antonii increased by 21 times over diurnal rates and 824 times over nocturnal rates recorded under dry-shore conditions.

Density and Aggregation

Density of N. versicolor (50 x [m.sup.-2]) was significantly less than that of T. antonii (313 x [m.sup.-2]; F = 12.94, DF = 47, P = 0.001). There was no statistical evidence for aggregation by N. versicolor during typical sea conditions (P > 0.05; Table 4), however, when water level and wave action increased, N. versicolor was observed to aggregate on the high points of the karst rocks. High-shore T. antonii were observed to aggregate in 5-10 cm diameter groups (DF = 84, P = 0.017) mainly in areas adjacent to tide pools and deep crevasses where water frequently collected (Table 5).

Wetting Experiment

The behavioral responses of snails varied significantly ([chi square] = 1331.68, DF = 2, P < 0.0001) across the 3 experimental treatments (saltwater, freshwater, and dry-air control). However, responses to treatments did not vary across species ([chi square] = 0.04, DF = 1, P = 0.844). Only those treatments simulating seawater moistened substratum elicited clear movement responses in both mid- and high-shore species when compared with freshwater treatments ([chi square] = 1306.31, DF = 1, P < 0.0001) and dry-substratum treatments ([chi square] = 255.8, DF = 1, P < 0.0001). The behavioral responses of snails exposed to freshwater and dry-substratum treatments were similar with few, if any, individuals moving ([chi square] = 2.32, DF = 1, P = 0.127). The model-estimated movement probabilities of N. versicolor (0.980) and T. antonii (0.983) following application of saltwater were higher than after exposure to freshwater (0.004 and 0.009) and dry-air (0.087 and 0.092) treatments.

DISCUSSION

Rocky intertidal habitats have been the focus for numerous studies over the past century. The intense gradient of environmental factors characteristic of intertidal habitats have allowed researchers to empirically evaluate numerous ecologic theories related to abiotic and biotic interactions among species, including gastropods (Underwood 2000). Research related to adaptation of rocky shore organisms is important because these habitats are transitional areas between aquatic and terrestrial ecosystems. Thus, the adaptations developed therein can be observed as they transcend this evolutionary gradient. Before considering locomotory behavior as a measure of adaptation in response to environmental stress, it is important to recognize the circumstances that motivate snails to move. Snails move for a variety of reasons including feeding, activities dealing with reproduction, and responses to stress (physical and biologic). It has been estimated that the energy expended to support movement in actively foraging gastropods is 15 times greater than resting values (Newell 1970). Therefore, to maximize energy reserves, littoral gastropods must use adaptations that conserve stored energy between feeding migrations. It is generally assumed that when snails are moving they are feeding (Forrest et al. 2001).

Physiologic and behavioral responses to desiccation stress are important factors in structuring distribution of littoral gastropods on rocky shores (e.g., Garrity 1984, Menge et al. 1986). This study and others (Kensler 1967, Garrity 1984, Garrity & Levins 1984, Lang et al. 1998) clearly indicate that the potential for desiccation stress increases with increasing shore height, however, comparative studies among littoral gastropods have failed to demonstrate a vertical continuum of increasing capacity adaptation to desiccation stress (McMahon 1990, McMahon & Britton 1991, McMahon 2001). The data gathered in this study indicate that desiccation stress, regulated by the presence of substratum moisture, is an important factor influencing locomotory behavior and microhabitat selection of mid- and high-shore gastropods.

Activity in intertidal gastropods has been shown to increase during periods of reduced desiccation stress such as occurs during nocturnal hours (Ruwa & Jaccarini 1988) and periods of rainfall and increased humidity (Voltolina & Sacchi 1990, Britton & McMahon 1992, Emson et al. 2002). The hypothesis that snails move in response to decreased temperature and desiccation stress is supported by our diurnal and 24-h surveys. The first 5 days of the survey were conducted during a dry period with normal wave activity, which effectively moistened midshore zones but left high-shore habitats dry. Under these conditions, N. versicolor was locomoting continuously whereas little if any movement was observed in the high-shore species, T. antonii. The remainder of the study was conducted during a period of heavy wave action that carried moisture into high-shore zones. The resulting increase in substratum moisture stimulated movement of the high-shore species. In fact, the level of activity in specimens of T. antonii was comparable to that of N. versicolor during this period (Fig. 6).

Mid-shore N. versicolor were rarely found on dry substrate, and as a result, were moving, and presumably feeding, continuously. In contrast, specimens of T. antonii exhibited a lack of movement while occupying dry substrata. Previous research has shown that feeding productivity of high-shore species is low during prolonged periods of tidal emersion (Ruwa & Jaccarini 1988, Norton et al. 1990, Britton & McMahon 1992), suggesting that energy intake is also minimized. Because high-shore gastropods are emersed more than 86% of the time (McMahon 1988), they must be able to rapidly respond to favorable conditions to maximize energy intake. Indeed, high-shore T. antonii responded immediately to increases in substratum moisture in field and laboratory experiments. The locomotory rate of T. antonii specimens increased 6-fold immediately following the increase in substratum moisture with rates equal to or exceeding that of N. versicolor. The rapid response of high-shore snails to increased substratum moisture is believed to be an adaptation that allows them to take advantage of the highly ephemeral feeding periods characteristic of their high-shore habitats (Underwood & McFadyen 1983, McMahon 1990). Our 24 h surveys also supported the hypothesis that snails move in response to reduced environmental stress. N. versicolor and T. antonii exhibited preference for nocturnal activity under conditions of increased substratum moisture, suggesting that activity is further stimulated by the combination of reduced temperature and desiccation stress.

Results from wetting experiments further support the hypothesis that movement occurs when desiccation stress is reduced. In this study, mid and high-shore gastropods were observed to initiate movement immediately following application of saltwater treatments, but no movement was recorded after freshwater applications. Similarly, individuals of the high-shore littorine, Nodilittorina exigua, retracted into their shell and sealed off the aperture when freshwater was applied directly in rainfall simulated laboratory experiments (Ohgaki 1988). These freshwater treatments likely fail to simulate actual conditions during rainfall events due to the lack of crystallized salt that would be present on the shore. Although this study presents no direct evidence that salts are readily accumulated on the shores, this suggestion can be derived from the results of our field experiments and those of related studies. For example, in-situ applications of fresh and saltwater treatments were observed to initiate movement in the high-shore gastropod, Cenchritis muricatus (Gochfeld & Minton 2001). Thus, salt was most likely present on the rock surface and when combined with the freshwater treatment, actually bathed the snails in diluted "salted" water. Further, during a heavy rain on Jamaican shores McMahon (2003) observed the high-shore gastropods, C. muricatus and T. antonii, actively foraging in runoff salinities of 0.5-1.5 psu and 3-7 psu, respectively. The coincidence of rain and snail movement has also been observed for other high-shore gastropods (Britton & McMahon 1992, Chapman 1994, Gochfeld & Minton 2001, Emson et al. 2002).

Synchronous directional movements were only observed among individuals of N. versicolor and may be related to the increased wave action associated with the onset of the offshore storm. While the majority of N. versicolor were found on wet surfaces (82%), relatively few were found immersed (14%), indicating a clear preference for emersed habitats. Because increased wave heights inundated midshore zones, the movement of N. versicolor away from the shore was most likely a behavioral response to remain emersed. Indeed, N. versicolor were observed to move to and aggregate at higher levels on shore as if to escape the flooding water levels. Lang et al. (1998) also reported that mid-shore species sought shelter at higher shore positions away from increased wave action.

The majority of lateral and vertical movements recorded for both species within their vertical zones appeared to be random. Underwood (1977) also reported random directional movements in three species of intertidal gastropods when monitored over a 1- to 3-day period. The random directional movements made by N. versicolor and T. antonii could have been a result of topography in and around each zone. Rugosity measurements of the substrate were characterized as having high relief in mid- and high-shore zones from tide pools and/or cavernous extrusions. Thus, the surface topography would allow for individual snails to move to places of higher or lower shore position without actually moving towards or away from the shore.

Many studies have documented the use of sheltered microhabitats by littoral gastropods as protection from physical (desiccation, solar radiation, wave dislodgement) and biologic stressors (predation from crabs, birds, and fish; Raffaelli & Hughes 1978, Garrity 1984, Peckol et al. 1989, Rochette & Dill 2000, Minton & Gochfeld 2001). The behaviors of mid-shore N. versicolor and high-shore T. antonii in this study are consistent with those observed in previous studies, but the mechanism influencing microhabitat choice may vary between different proposed hypotheses. Our data support the hypothesis maintaining that crevice and pit microhabitat selection on tropical rocky shores is a behavioral adaptation to reduce desiccation and heat stress (Garrity 1984, Peckol et al. 1989, Gochfeld & Minton 2001). Movement in high-shore snails was initiated by the onset of wet-shore conditions. As high-shore snails went from a state of repose (e.g., 0.27 cm x [h.sup.-1]) to actively moving (e.g., 1.63 cm x [h.sup.-1]) over the 4 days of increased wave activity, they continued to exhibit a preference for crevice microhabitats and generally avoided exposed rock surfaces similar to the behavior of N. versicolor prior to inundation of midshore zones. Thus, it is likely that the continued preference for crevice microhabitat observed in T. antonii is related to physical attributes in forming reservoirs of water (Britton et al. 1991) and rehydrating their food supply (Gochfeld & Minton 2001) thereby optimizing feeding rather than serving as refugia from predators or wave action. For example, during periods of normal wave activity and subsequent dry-shore conditions, both mid and high-shore species exhibited preferences for pit and crevice microhabitats and avoided exposed portions of the shore (Fig. 4, Fig. 5; Table 2). However, during periods of increased wave exposure and subsequent inundation of midshore zones, a greater proportion of N. versicolor occurred on exposed rock surfaces with fewer individuals utilizing pit microhabitats. It is reasonable to assume that if the observed preference for pit microhabitat prior to mid-shore inundation was related to predator avoidance, then preference for pit microhabitat would have persisted during periods of increased inundation when snails would have been more vulnerable to attack by fish, crabs, and shell boring gastropod predators. Further, if preference for pit microhabitat in the mid-shore species was a behavioral adaptation to reduce dislodgement then it would have been evident during the period of increased wave activity. The sudden appearance of snails on exposed rock surfaces in mid-shore zones during the onset of increased wave activity (Fig. 4) was most likely related to the general movement of individuals up the shore and away from the rising water (days 8-12; Fig. 3).

Reduced mortality from desiccation and heat stress has been previously documented as a result of intertidal snails seeking out crevice and pit microhabitats between feeding periods (Kensler 1967). Indeed, surface temperatures of carbonate rocks are cooler in depressions and warmer on promontories (Britton et al. 1991). Norton et al. (1990) reported that gastropods occupying eulittoral areas locate an adjacent food supply (e.g., algae mat) and feed on it during acceptable conditions (e.g., immersion, rainfall, etc.) and sometimes return to their previous location. Thus, it is believed that T. antonii remained within 1 m of their initial mark as a defense against environmental stress by residing in their microhabitat of choice during hot days and making proximal feeding excursions when desiccation stress was reduced. Other evidence reported herein also supports the previous statement. If the high-shore specimens move at an average rate of 1.6 cm x [h.sup.-1], after 24 h they could potentially move 0.38 m, and after 3 days they could potentially move beyond 1 m. The wet conditions that initiated movement in high-shore individuals persisted for more than 4 days and few high-shore snails were observed to move farther than 1 m from their home mark. In contrast, N. versicolor were observed to make lengthy migrations, likely as a result of relieved stress in the mid-shore zone brought on from daily tidal inundation.

As demonstrated in this study, the movement behavior of snails occupying emersed habitats on tropical, platform-like rocky shores is primarily a response related to the intensity of desiccation. Because movement, and presumably foraging, only occurs when snails are on moist substrata, opportunities for foraging decrease with increasing shore height. Thus, midshore species such as N. versicolor forage for much greater proportion of time than high-shore species such as T. antonii, which forage opportunistically when substrata are wetted either by rainfall, dew, or saltwater, and splash and spray.

TABLE 1. Site and shore level characterization among mid and
high-shore zones of rocky shores on the southeastern Yucatan
Peninsula, Mexico, May 12-25, 2002.

                      Mid    High    F-value    df       P
                     Shore   Shore

Rugosity (m)          1.25    1.19    1.03     1,34    0.318
Zone width (m)        3.97    4.97    2.85     1,34    0.105
Rock temperature
  ([degrees]C)       26.77   27.82    7.93     1,327   0.005
Sponge surveys (g)    0.27   -6.97    5.146    1,16    0.038

TABLE 2. Differential use of microhabitats by mid-shore Nerita
versicolor and high-shore Tectarius antonii on rocky shores of
the southeastern Yucatan Peninsula, Mexico, during periods of
normal (May 12-20, 2002) and increased (May 21-25, 2002) wave
activity.

                                                    Proportion
                                      Expected       of Snails
       Zone                          Proportion      Observed
  (wave activity)     Microhabitat    (points)        (number)

Mid shore (normal)      Surface      0.537 (206)    0.199 * (38)
                        Crevice      0.425 (163)    0.539 (103)
                        Pit          0.039 (15)    0.262 ** (50)
Mid shore (high)        Surface      0.537 (206)     0.443 (43)
                        Crevice      0.425 (163)     0.433 (42)
                        Pit          0.039 (15)      0.124 (12)
High shore (normal)     Surface      0.662 (254)    0.269 * (59)
                        Crevice      0.297 (114)   0.626 ** (137)
                        Pit          0.042 (16)      0.105 (23)
High shore (high)       Surface      0.662 (254)    0.287 * (25)
                        Crevice      0.297 (114)   0.563 ** (49)
                        Pit          0.042 (16)      0.149 (13)

       Zone                             95% Confidence Interval
  (wave activity)     Microhabitat      on Observed Proportions

Mid shore (normal)      Surface       0.025 [less than or equal to] p
                                      [less than or equal to] 0.373
                        Crevice       0.322 [less than or equal to] p
                                      [less than or equal to] 0.757
                        Pit           0.070 [less than or equal to] p
                                      [less than or equal to] 0.454
Mid shore (high)        Surface       0.227 [less than or equal to] p
                                      [less than or equal to] 0.660
                        Crevice       0.217 [less than or equal to] p
                                      [less than or equal to] 0.649
                        Pit          -0.020 [less than or equal to] p
                                      [less than or equal to] 0.267
High shore (normal)     Surface       0.076 [less than or equal to] p
                                      [less than or equal to] 0.463
                        Crevice       0.414 [less than or equal to] p
                                      [less than or equal to] 0.837
                        Pit          -0.029 [less than or equal to] p
                                      [less than or equal to] 0.239
High shore (high)       Surface       0.090 [less than or equal to] p
                                      [less than or equal to] 0.485
                        Crevice       0.347 [less than or equal to] p
                                      [less than or equal to] 0.780
                        Pit          -0.006 [less than or equal to] p
                                      [less than or equal to] 0.305

Arrows indicate a lower * or higher ** significant deviation from
expected proportions.

TABLE 3. Differential use of microhabitats by mid-shore Nerita
versicolor and high-shore Tectarius antonii on rocky shores of
the southeastern Yucatan Peninsula, Mexico, May 12-20, 2002
with respect to degree of substratum moisture.

             Substrate   Expected Proportion   Proportion of Snails
   Zone      Moisture         (points)          Observed (number)

Mid shore       W           0.779 (299)            0.750 (78)
                WL          0.000 (0)              0.067 (7)
                UW          0.175 (67)             0.135 (14)
                D           0.047 (18)             0.048 (5)
High shore      W           0.237 (91)             0.286 (62)
                WL          0.000 (0)              0.028 (6)
                UW          0.000 (0)              0.083 (18)
                D           0.763 (293)            0.604 (131)

             Substrate       95% Confidence Interval
   Zone      Moisture        on Observed Proportions

Mid shore       W         0.553 [less than or equal to] p
                          [less than or equal to] 0.947
                WL       -0.047 [less than or equal to] p
                          [less than or equal to] 0.181
                UW       -0.021 [less than or equal to] p
                          [less than or equal to] 0.290
                D        -0.049 [less than or equal to] p
                          [less than or equal to] 0.145
High shore      W         0.080 [less than or equal to] p
                          [less than or equal to] 0.491
                WL       -0.047 [less than or equal to] p
                          [less than or equal to] 0.102
                UW       -0.042 [less than or equal to] p
                          [less than or equal to] 0.208
                D         0.381 [less than or equal to] p
                          [less than or equal to] 0.826

Moisture categories: W, wet; WL, waterline; UW, underwater; D, dry.

TABLE 4. Statistical results of aggregation analyses for Nerita
versicolor on rocky shores of the southeastern Yucatan Peninsula,
Mexico, May 12-25,2002.

Grouping   Distance   Parameter   Observed   Expected   Probability

   1       5-10 cm    [n.sub.0]    64         61.59     P = 0.0774
                      [n.sub.1]    15         19.81
                      [n.sub.2]     4          1.59
                          N        83
                          p         0.139
   2       10-15 cm   [n.sub.0]    65         63.11     P = 0.1001
                      [n.sub.1]    13         16.77
                      [n.sub.2]     3          1.11
                          N        81
                          p         0.117
   3       15-20 cm   [n.sub.0]    65         64.34     P = 0.2903
                      [n.sub.1]     8          9.32
                      [n.sub.2]     1          0.34
                          N        74
                          p         0.068
   4       20-25 cm   [n.sub.0]    43         43.25     P = 0.2178
                      [n.sub.1]     7          6.51
                      [n.sub.2]     0          0.25
                          N        50
                          p         0.070

Parameters: [n.sub.0] represents the number of random quadrat pairs
(at the indicated spacing level) in which neither member was occupied
by a snail, [n.sub.1], represents the number of random quadrat pairs
in which only one member was occupied, [n.sub.2] represents the number
of random quadrat pairs in which both members were occupied. Expected
values were calculated as follows: [n.sub.0] = [(1-P).sup.2]N,
[n.sub.1] = 2p (1 - p)N, and [n.sub.2] = [p.sup.2]N, where
N = [n.sub.0] + [n.sub.1] + [n.sub.2] and p is estimated by
(2[n.sub.2] + [n.sub.1])/2N (Goodall 1974).

TABLE 5. Statistical results of aggregation analyses for Tectarius
antonii on rocky shores of the southeastern Yucatan Peninsula, Mexico,
May 12-25, 2002.

Grouping   Distance    Parameter   Observed   Expected   Probability

   1       5-10 cm     [N.sub.0]    50         43.78     P = 0.0167
                       [N.sub.1]    22         34.45
                       [n.sub.2]    13          6.78
                           N        85
                           p         0.282
   2       10-15 cm    [n.sub.0]    44         43.32     P = 0.4382
                       [n.sub.1]    29         30.36
                       [n.sub.2]     6          5.32
                           N        79
                           P         0.259
   3       15-20 cm    [n.sub.0]    37         34.93     P = 0.2541
                       [n.sub.1]    27         31.13
                       [n.sub.2]     9          6.93
                           N        73
                           p         0.308
   4       20-25 cm    [n.sub.0]    29         28.31     P = 0.4256
                       [n.sub.1]    18         19.37
                       [n.sub.2]     4          3.31
                           N        51
                           p         0.255

Parameters: [n.sub.0] represents the number of random quadrat pairs (at
the indicated spacing level) in which neither member was occupied by a
snail, [n.sub.1] represents the number of random quadrat pairs in
which only one member was occupied, [n.sub.2] represents the number
of random quadrat pairs in which both members were occupied. Expected
values were calculated as follows: [n.sub.0] = [(1-P).sup.2]N,
[n.sub.1] = 2p (1 - p)N, and [n.sub.2] = [p.sup.2]N, where N =
[n.sub.0] + [n.sub.1] + [n.sub.2] and p is estimated by (2[n.sub.2]
+ [n.sub.1])/2N (Goodall 1974).

ACKNOWLEDGMENTS

The authors thank Suzanne Dilworth (Texas A&M University-Corpus Christi, TAMUCC) and Bart German (retired) for assistance in the field. John W. Tunnell, Jr., Kim Withers, and David A. McKee of TAMUCC provided critical reviews of the manuscript. D.L. Hawkins of The University of Texas at Arlington provided assistance with the MIXED and CATMOD procedures in SAS.

LITERATURE CITED

Bingham, F. O. 1972. The mucus holdfast of Littorina irrorata and its relationship to relative humidity and salinity. Veliger 15:48-50.

Britton, J. C. 1992. Evaporative water loss, behavior during emergence, and upper thermal tolerance limits in seven species of eulittoral-fringe Littorinidae (Mollusca: Gastropoda) from Jamaica. In: J. Grahame, P. J. Mill, D. G. Reid, editors. Proceedings of the 3rd International Symposium on Littorinid Biology. London: The Malacological Society of London. pp 69-83.

Britton, J. C. 1995. The relationship between position on the shore and shell ornamentation in two size-dependent morphotypes of Littorina striata, with an estimate of evaporative water loss in these morphotypes and in Melarhaphe neritoides. Hydrobiol. 309:129-142.

Britton, J. C. & R. F. McMahon. 1992. Patterns of dispersion in four intertidal gastropods from the rocky shores of Tolo Harbour, Hong Kong. In: B. Morton, editor. The marine flora and fauna of Hong Kong and Southern China, III. Hong Kong: Hong Kong University Press. pp. 383-394.

Britton, J. C., R. F. McMahon & J. W. Hart. 1991. Relationships between topography, substratum composition and surface temperature, and the spatial distribution of intertidal fauna on rocky shores of south-western Australia. In: F. E. Wells, D. I. Walker, H. Kirkman & R. Lethbridge, editors. Proceedings of the International Marine Biological Workshop: The Marine Flora and Fauna of Albany, Western Australia. Western Australian Museum, Perth. vol. 2. pp. 521-540.

Chapman, M. G. 1994. Small-scale patterns of distribution and size-structure of the intertidal littorinid, Littorina unifasciata (Gastropoda: Littorinidae) in New South Wales. Aust. J. Mar. Freshwater Res. 45: 635-652.

Chapman, M. G. & A. J. Underwood. 1996. Influences of tidal conditions, temperature and desiccation on patterns of aggregation of the high-shore periwinkle, Littorina unifasciata, in New South Wales, Australia. J. Exp. Mar. Biol. Ecol. 196:213-237.

Emson, R. H., D. Morritt, E. B. Andrews & C. M. Young. 2002. Life on a hot dry beach: behavioral, physiological, and ultrastrnctural adaptations of the littorinid gastropod Cenchritis (Tectarius) muricatus. Mar. Biol. 140:723-732.

Forrest, R. E., M. G. Chapman & A. J. Underwood. 2001. Quantification of radular marks as a method for estimating grazing of intertidal gastropods on rocky shores. J. Exp. Mar. Biol. Ecol. 258:155-171.

Garrity, S. D. 1984. Some adaptations of gastropods to physical stress on a tropical rocky shore. Ecol. 65:559-574.

Garrity, S. D. & S. C. Levins. 1984. Aggregation of a tropical neritid. Veliger 27:1-6.

Gochfeld, D. J. & D. T. Minton. 2001. When to move and where to go: movement behavior of the tropical littorinid Cenchritis muricatus (Linnaeus, 1758). J. Shellfish Res. 20:447-452.

Goodall, D. W. 1974. A new method for the analysis of spatial pattern by random pairing of quadrats. Vegetation 29:135-146.

Haney, J. C. & A. R. Solow. 1992. Testing for resource use and selection by marine birds. J. Field Ornithol. 63:43-52.

Heath, D. J. 1975. Colour, sunlight and internal temperatures of the land snail Cepaea nemoralis (L.). Oecologia 19:29-38.

Kaplan, E. H. 1988. Southeastern and Caribbean seashores. New York: Houghton Mifflin Co.

Kensler, C. 1967. Desiccation resistance of intertidal crevice species as a factor in their zonation. J. Anim. Ecol. 36:391-406.

Lang, R. C., J. C. Britton & T. Metz. 1998. What to do when there is nothing to do: the ecology of Jamaican intertidal Littorinidae (Gastropoda: Prosobranchia) in repose. Hydrobiol. 378:161-185.

Lewis, J. R. 1964. The ecology of rocky shores. London: The English Universities Press.

McMahon, R. F. 1988. Respiratory response to periodic emergence in intertidal molluscs. Amer. Zool. 28:97-114.

McMahon, R. F. 1990. Thermal tolerance, evaporative water loss, air-water oxygen consumption, and zonation of intertidal prosobranchs: a new synthesis. Hydrobiol. 193:241-260.

McMahon, R. F. 1991. Upper critical temperature limits, tissue temperatures during emersion and evaporative cooling at elevated temperatures in Western Australian intertidal gastropods: Evidence for a physiological barrier at the high tide mark. In: F. E. Wells, D. I. Walker, H. Kirkman & R. Lethbridge, editors. Proceedings of the International Marine Biological Workshop: the marine flora and fauna of Albany, vol. 2. Western Australia. Perth: Western Australian Museum. pp. 661-673.

McMahon, R. F. 2001. Acute thermal tolerance in intertidal gastropods relative to latitude, superfamily, zonation and habitat with special emphasis on the Littorinoidea. J. Shellfish Res. 20:459-467.

McMahon, R. F. 2003. Acute hypo- and hypersaline activity responses relative to zonation of intertidal rocky shore and mangrove gastropods from the Burrup Peninsula, Western Australia. In: F. E. Wells, D. I. Walker & D. S. Jones, editors. The marine flora and fauna of Dampier, Western Australia. Perth: Western Australian Museum. pp. 131-146.

McMahon, R. F. & W. D. Russell-Hunter. 1977. Temperature relations of aerial and aquatic respiration in six littoral snails in relation to their vertical zonation. Biol. Bull. 152:182-198.

McMahon, R. F. & J. C. Britton. 1991. The relationship between vertical distribution, rate of evaporative water loss, behaviour during emergence and morphometrics in six species of rocky shore gastropods from Princess Royal Harbour, Western Australia. In: F. E. Wells, D. I. Walker, H. Kirkman & R. Lethbridge, editors. Proceedings of the international Marine Biological Workshop: the marine flora and fauna of Albany, Western Australia, vol. 2. Perth: Western Australian Museum. pp. 675-692.

Menge, B. A., J. Lubchenco, L. R. Ashkenas & F. Ramsey. 1986. Experimental separation of effects of consumers on sessile prey in the low zone of a rocky shore in the Bay of Panama: direct and indirect consequences of food web complexity. J. Exp. Mar. Biol. Ecol. 100:225-269.

Minton, D. & D. J. Gochfeld. 2001. Is life on a tropical shore really so hard?: the role of abiotic factors in structuring a supralittoral molluscan assemblage. J. Shellfish Res. 20:477-483.

Newell, R. C. 1970. Biology of intertidal animals. London: Logos Press.

Newell, R. C. 1979. Biology of intertidal animals, 3rd ed. Kent: Marine Ecology Surveys Limited.

Norton, T. A., S. J. Hawkins, N. L. Manley, G. A. Williams & D. C. Watson. 1990. Scraping a living: a review of littorinid grazing. Hydrobiol. 193:117-138.

Ohgaki, S. 1988. Rain and the distribution of Nodilittorina exigua (Dunker) (Gastropoda: Littorinidae). J. Exp. Mar. Biol. Ecol. 122:213-223.

Peckol, P., S. Guarnagia & M. Fisher. 1989. Zonation and behavioral patterns of the intertidal gastropods Nodilittorina (Tectininus) antoni (Philippi, 1846) and Nerita versicolor Gmelin, 1791, in the Bahamas. Veliger 32:8-15.

Pielou, E. C. 1964. The spatial pattern of two-phase patchworks of vegetation. Biometrics 20:156-167.

Raffaelli, D. G. & R. N. Hughes. 1978. The effects of crevice size and availability on populations of Littorina rudis and Littorina neritoides. J. Anita. Ecol. 47:71-83.

Rochette, R. & L. M. Dill. 2000. Mortality, behavior and the effects of predators on the intertidal distribution of littorinid gastropods. J. Exp. Mar. Biol. Ecol. 253:165-191.

Rooker, J. R., Q. R. Dokken, C. V. Pattengill & G. J. Holt. 1997. Fish assemblages on artificial and natural reefs in the Flower Garden Banks National Marine Sanctuary. Coral Reefs 16:83-92.

Ruwa, R. K. & V. Jaccarini. 1988. Nocturnal feeding migrations of Nerita plicata, N. undata and N. textilis (Prosobranchia: Neritacea) on the rocky shores at Mkomani, Mombasa, Kenya. Mar. Biol. 99:229-234.

Slemko, T. Weather data from virtual Mexico. Available at http:// www.virtualmex.com/cancun.htm. Accessed: Data downloaded 20 April 2002.

Stephenson, T. A. & A. Stephenson. 1950. Life between tide marks in North America, I. The Florida Keys. J. Ecol. 38:354-402.

Underwood, A. J. 1977. Movements of intertidal gastropods. J. Exp. Mar. Biol. Ecol. 26:191-201.

Underwood, A. J. 1979. The ecology of intertidal gastropods. Adv. Mar. Biol. 16:111-120.

Underwood, A. J. 2000. Experimental ecology of rocky intertidal habitats: what are we learning? J. Exp. Mar. Biol. Ecol. 250:51-76.

Underwood, A. J. & M. G. Chapman. 1985. Multifactorial analyses of directions of movement of animals. J. Exp. Mar. Biol. Ecol. 91:17-43.

Underwood, A. J. & K. E. McFadyen. 1983. Ecology of the intertidal snail Littorina acutispira Smith. J. Exp. Mar. Biol. Ecol. 66:169-197.

Vermeij, G. J. 1973. Morphological patterns in high-intertidal gastropods: adaptive strategies and their limitations. Mar. Biol. 20:319-346.

Voltolina, D. & C. F. Sacchi. 1990. Field observations on the feeding habits of Littorina scatulata Gould and L. sitkana Philippi (Gastropoda: Prosobranchia) of southern Vancouver Island (British Columbia, Canada). Hydrobiol. 193:147-154.

THOMAS W. BATES (1) AND DAVID W. HICKS (2), *

(1) Center for Coastal Studies, Texas A&M University-Corpus Christi, 6300 Ocean Drive, Corpus Christi, Texas 78412; (2) Department of Biological Sciences, The University of Texas at Brownsville, 80 Fort Brown, Brownsville, Texas 78520

* Corresponding author. E-mail: [e-mail]; Fax: [phone]