The association between the coquina clam Donax fossor say and its epibiotic hydroid Lovenella gracilis clarke.
KEY WORDS: Donax fossor, Lovenella gracilis, epibiosis
Along the Atlantic coast of North America, the hydroid Lovenella gracilis is a facultative epibiont on intertidal bivalves in the genus Donax (Ruppert & Fox 1988). Although epibiosis is common in rocky intertidal and subtidal habitats, it rarely has been documented in exposed sandy beaches (but see Manning & Lindquist 2003) or muddy habitats (Stachowitsch 1977, Gili et al. 1993, Jarms & Tiemann 1996, Patil & Anil 2000). Few infaunal or epifaunal macroinvertebrates occur in wave-swept sandy beaches due to the dynamic nature of the substrate. Thus, this habitat is largely devoid of macrofauna to serve as basibionts. However, the valves of living Donax spp. often serve as substrate for a variety of epibionts in this habitat (Ansell 1983). In addition, most hydroid species are subtidal and rarely occur in the intertidal (Jackson 1977, Gili & Hughes 1995). Consequently, the range of habitats for L. gracilis is broadened by the presence of Donax spp. in the swash zone. Although D. fossor is a common and highly abundant species and its association with L. gracilis is highly visible, few studies focus on this association, and no one has quantified the temporal variation of the distribution of L. gracilis on D. fossor.
Clearly L. gracilis benefits from its association with Donax spp. by using the clam as a substrate. Large colonies of L. gracilis are never found on the plethora of abiotic hard substrata (e.g., broken shells, pebbles, etc.) in the swash zone (personal observation). Rather, L. gracilis colonizes the posterior margins of the valves of living Donax spp. (Fig. 1A). The clam provides an anchor in the swash zone and acts as a stabilizing structure maintaining the hydrorhizae at the surface-water interface (Fig. 1B). This predictable orientation of the clam shell allows the hydroid colony to receive the full impact of the swash. Furthermore, Donax spp. actively maintain their position in the swash zone by migrating with the tide, a behavior known as "swash-riding" (Ellers 1995). Lovenella gracilis benefits from the tidal migratory behavior of the clam thus avoiding desiccation and overheating while constantly being exposed to food and dissolved gases in the swash zone.
[FIGURE 1 OMITTED]
Although Donax spp. are highly abundant on sandy beaches on the northwest Atlantic, its association with L. gracilis has been seldom documented and rarely quantified. Only Manning and Lindquist (2003) addressed the question of costs and benefits of the epibiotic hydroid on Donax spp. In a series of elegant experiments they showed that L. gracilis enhanced predation rates from some predators and decreased rates for another. They also quantified the role of the hydroid on the clam's abilities to orient in the swash, burrow, and to migrate in the backwash.
In this study, we examined whether the Donax-Lovenella association is a recent phenomenon by examining museum specimens of D. fossor. We also examined the role of the hydroid on clam predation by a gastropod. The moon snail, Neverita duplicata (formerly Polinices), which preys upon D. variabilis in Texas (Loesch 1957) and Florida (Mikkelsen 1978), appears to be a significant predator of D. fossor along beaches in southern New Jersey. We also examined the ability of the clam to swash-fide both with and without a hydroid colony. Finally, we quantified the frequency, abundance, and seasonal occurrence of L. gracilis on D. fossor in southern New Jersey.
Studies were performed on two exposed beaches located along the barrier islands of southern New Jersey (Fig. 2). The Sea Isle City site, 81st Street (39[degrees]12.9'N, 74[degrees]70.7'W), is just north of Hereford Inlet and 66th Street in Avalon (39[degrees]07.3'N, 74[degrees]74.0'W), is midway along Seven Mile Island.
[FIGURE 2 OMITTED]
To eliminate the possibility that the Donax-Lovenella association is a recent occurrence, we examined the dry specimen collections of D. fossor at the Delaware Museum of Natural History (DMNH) and the Academy of Natural Sciences of Philadelphia (ANSP) for hydroid presence. Some specimens were incorrectly identified as D. fossor and were excluded from this study. All potential occurrences of the hydroid were examined under a dissecting microscope to confirm that the valves exhibited remnant hydrorhizae and not debris.
We performed a choice experiment to determine if L. gracilis affects snail preference for clams with and without hydroids. Donax fossor were collected by hand from the swash zone at both study sites during late summer and fall 2003. Clams were transported to Villanova University and kept in a plastic bucket in a recirculating seawater system. Clams were pulse-fed numerous species of microalgae, and hydroids were fed brine shrimp each day. Neverita duplicata (n = 4) were obtained from Marine Biologic Laboratory (Woods Hole, MA). Prior to experiments, snails were held in a plastic container filled with beach sand in a recirculating seawater system. Snails were kept without food for 24 h prior to the start of the experiment.
Each snail was placed in a plastic bucket (diameter = 13.0 cm) fitted with a secure lid and filled with 2.5 cm of clean, sieved beach sand that was collected from 66th Street. Clams were oriented in the sand to mimic their natural orientation in the swash zone, with the anterior end facing downward and the posterior end at the sand-surface interface. Each bucket contained 20 clams (10 clams with hydroids and 10 clams without hydroids) arranged in an alternating pattern around the periphery of the bucket with the snail in the center. This pattern ensured that the snail had an equal probability of first encountering a clam in either treatment group. We recorded predation daily and partially drilled clams were treated as drilled. Both partially and completely drilled clams were replaced with live clams of the appropriate treatment and the initial clam arrangement was reestablished each day. Feeding bouts continued for 25 days. The response variable, the number of clams drilled in each treatment group, was analyzed using a [chi square] ([alpha] = 0.05).
We performed a swash-riding experiment to determine the effect of the hydroid colony on shoreward passive transport of clams by wave swash. The experiment was performed on a flood tide (when clams migrate shoreward) in the swash zone at 66th Street in Avalon on October 5, 2002. Donax fossor with robust hydroid colonies and those lacking hydroid colonies were collected by hand from the swash zone and kept in a bucket with oxygenated seawater. Each clam with a hydroid was paired with a clam (of similar size) without a hydroid. Unequal clam size differences were balanced between the two groups (with and without hydroids) so that there was no net bias. Clams were painted with bright nail polish and numbered to distinguish experimental clams from the natural population.
Just prior to an incoming wave, a pair of clams (one with and the other without a hydroid) was placed side by side on the sand with the posterior end of the clam facing shoreward and the anterior end facing seaward. This position mimics the natural orientation they assume when emerging from the sediment in response to an incoming wave. The starting point was marked with a flag, and the clams were carried shoreward by an incoming wave. Two assistants followed the clams and marked the point of furthest shoreward transport for each clam in the pair. Our attempts to track clams after shoreward transport were unsuccessful because we lost the clams in the backwash. Therefore, we were unable to quantify net transport (starting point to the location of burrowing after passive seaward transport in the backwash). The linear distance from the starting point to the farthest distance traveled before being carried seaward in the backwash was recorded (n = 46).
We used the Shapiro-Wilk test for normality on the distribution of the difference variable (distance traveled by clams with hydroids --distance traveled by clams without hydroids). We used a 2-tailed, paired t-test to test for any difference between the two groups.
Monthly or bimonthly samples of D. fossor were collected at low tide from the two study sites from September 2001 through September 2003. Brief surveys of the foreshore were made using a garden spade to locate D. fossor populations. After identifying patches of D. fossor, cores (10 cm or 15 cm deep, depending on preliminary surveys) were extracted with an aluminum clam gun (diameter = 10 cm) until approximately 200 clams were collected from each field site. This protocol was used for the majority of the sampling period (January 2002 to September 2003) and maximized sampling efficiency.
We used a variation of the standard protocol for the initial samples in the study (September 2001 to December 2001). Three replicate cores were collected at three different levels in the foreshore. The most seaward sample was collected in the swash zone, the most shoreward sample was collected at the highest point that D. fossor occurred and the third site was equidistant from the seaward and shoreward sites. We abandoned this stratified sampling technique after the first three field surveys because it proved inefficient; however, both protocols yielded density estimates (n/ [m.sup.2]) and size distributions.
We sieved cores using a 1-mm vinyl mesh screen to isolate clams. The postsieving matrix consisting of D. fossor, other invertebrates, shell hash, sand, and pebbles was immediately preserved in 70% ethanol.
Large clams were isolated by hand from the preserved matrix and were measured (length and height) using knife-edge calipers. Small volumes (~25 mL) of the matrix were examined in a plankton counting wheel using a dissecting microscope to identify smaller clams in the matrix. The microscope was fitted with an ocular micrometer, and clam length and height were recorded. All clams were scored for the presence or absence of the hydroid and the length of the largest hydrocaulus of the colony was measured.
The matrix was subsampled to generate density estimates when too many cores were extracted to reach the n = 200 target or when large volumes of matrix were present in the samples. To subsample, the matrix was homogenized and a fixed volume was isolated. Clams in the subsample were also measured and scored for the presence of the hydroid. Density and percentage of clams with a hydroid colony on each sampling date were adjusted to account for subsampling volumes.
Two out of 27 lots at the DMNH and 3 lots out of 30 at ANSP had valves with hydroids (Table 1). Although intact hydrocauli were absent, remnant hydrorhizae were observed on some specimens.
In all replicates, moon snails preferred clams without hydroids over clams with hydroids. Over the 25-day period, snails drilled more clams that lacked a hydroid colony, however, only snails 3 and 4 were significant (Table 2). When replicate snails were pooled (Heterogeneity [chi square] = 4.01, P = 0.135), N. duplicata drilled significantly fewer clams with hydroids than without hydroids (Table 2, P < 0.0001).
The difference variable (distance traveled by clams with hydroids --distance traveled by clams without hydroids) was normally distributed (W = 0.998, P = 0.993). The average distances traveled ([+ or -] SD) by clams with and without hydroids were 186 [+ or -] 104 cm and 181 [+ or -] 101 cm, respectively, and were not significantly different (Table 3).
There was a marked seasonal cycle of D. fossor occurrence at the two study sites (Fig. 3). Overall, D. fossor populations were present from late spring through fall. Only a few survived the winter in the swash zone. Peak clam density for Sea Isle occurred in September 2002, with 1127 individuals/[m.sup.2]. In Avalon, the highest abundances were observed in September with densities exceeding 2000 individuals/[m.sup.2] both in 2001 and 2002. As in Sea Isle, except for a few individuals, D. fossor was largely absent from the swash zone in Avalon during winter and spring 2002 and 2003.
[FIGURE 3 OMITTED]
Lovenella gracilis also exhibited a marked seasonal cycle. The three major occurrences of L. gracilis were September to December 2001, July to November 2002, and June to September 2003 (Fig. 4). At the onset of sampling, 11% of the clams sampled supported a hydroid colony (Fig. 4). In November and December 2001, the percentage of clams with hydroids fell, and L. gracilis was not seen again until March 2002. Although L. gracilis was detected in early spring 2002 (March to May), only five animals throughout that period supported a hydroid colony (Fig. 4). Well-established hydroid colonies were observed from July to November 2002, with the maximum percentage observed in August. Although there was a decline in hydroid abundance in September and October 2002 from its maximum in August 2002, L. gracilis rebounded in November 2002. For the 2003 season, L. gracilis was first detected in early June 2003, although only a minimal proportion of the population at both sites supported the hydroid through July (Fig. 4). Peak abundances of L. gracilis were observed in August 2003, when 75% of clams sampled had a hydroid colony. Although there was a decrease in September, 59% of the clam population supported hydroid colonies.
[FIGURE 4 OMITTED]
The proportion of the D. fossor population supporting a hydroid colony varied with clam size, months, years, and between the two study sites. Size-frequency distributions of D. fossor at both sites are presented in Figures 5, 6, 7, 8, and 9 for the months of hydroid occurrence.
[FIGURES 5-9 OMITTED]
At the start of the survey (September 2001), the hydroid was much more abundant in Avalon than in Sea Isle (Fig. 5). The majority of clams at both sites were <5 mm in length, suggesting a recent recruitment event. Large clams were much more abundant in Avalon than in Sea Isle and most of these animals supported a hydroid colony. No large clams were found at either site in November 2001 (Fig. 5). However, a small percentage of the juveniles (<5 mm) that were present had a hydroid colony. In December 2001, only one large clam was found (Fig. 5) and therefore, hydroids were restricted to smaller clams (1-6 mm).
In July 2002, small clams lacked L. gracilis, but larger clams (8-12 rom) supported hydroid colonies (Fig. 6). The mode of clams in the 2-3 mm size classes in Sea Isle in July suggested a recent recruitment event. In Avalon, clams in this size class were not found until August. Unlike the pattern observed in 2001, the hydroid was more abundant in Sea Isle than in Avalon, and this pattern persisted for the remainder of the 2002 season. In early August 2002 the hydroid appeared on small clams at both sites, but as in September 2001, it was more frequently found on larger individuals (8-15 mm) (Fig. 6). A large proportion of the population in Avalon was 1-3 mm, suggesting that recruitment had recently occurred. Similar patterns were observed in late August (Fig. 6).
In September and October 2002, hydroids were found more frequently on larger clams than on juveniles in Sea Isle (Fig. 7). In November 2002, hydroids were not found on smaller clams, but were found on larger clams in Avalon (Fig. 7). In Sea Isle, although some small clams had hydroids, most hydroid-bearing clams were >10 mm in length.
Only one clam in Sea Isle and four clams in Avalon had hydroids in early June 2003 (Fig. 8). Similar patterns were observed in late June 2003. In early July 2003, only one clam at each site had a hydroid (Fig. 8).
In late July 2003, the hydroid was all but absent (Fig. 9); no clams in Sea Isle had hydroids, whereas seven clams (10-13 mm in length) in Avalon had hydroids. Larger clams that were present in July 2003 were nearly absent from the swash zone in August and September 2003 (Fig. 9). During those months, the majority of the clams were <5 mm in length (Fig. 9), suggesting a possible second recruitment event in early August. This month had the greatest proportion of clams with hydroids in the 2-y sampling period.
Hydroid length varied throughout the sampling period (Fig. 10). In September 2001, L. gracilis colonies were profuse on D. fossor such that the hydroid identified clam location in the swash zone by protruding above the sand-water interface (see Fig. 1B). Hydroids as small as 0.40 mm were observed on recently settled juveniles. In November and December 2001, average hydroid length declined from September, as nearly every hydroid observed was on a recently settled clam. Although a few clams had hydroids in early spring 2002, these hydroids were quite small. Robust colonies were abundant from July 2002 through November 2002. The largest hydroids observed during the 2-y sampling period were in July 2002. These colonies identified clam presence in the sand as in September 2001. Throughout 2003, hydroids remained undetectable by the unaided eye. Thus, despite a large proportion of the D. fossor population supporting a hydroid in August and September 2003, the hydroids did not identify clam presence in the swash as in previous months.
[FIGURE 10 OMITTED]
Overall, there was a size-specific relationship between clam size and hydroid presence (Fig. 11). Larger clams were more likely to support a hydroid than smaller clams ([chi square] = 1031, P = <0.0001). However, larger clams were far less abundant than smaller clams.
[FIGURE 11 OMITTED]
The Donax-Lovenella association has been in existence >100 y (Table 1), but it is not surprising that it has been understudied and not well documented. Relatively few of the museum lots (and fewer specimens within the lots) displayed evidence of the association (Table 1). These data should not be interpreted as representing the relative occurrence of clams with hydroids from the time periods sampled for two reasons. First, these samples are death-assemblages and the shells were likely to have been subjected to surface-abrasion in the swash zone prior to collection. Any hydroids that had been present could have been removed prior to collection. Second, these clams do not represent random samples. There is an element of "shell-collector bias" associated with malacological museum samples. Collectors screen samples before deposition into the museum collection and select specimens that best represent shell morphology, not necessarily any epibionts associated with the clam. In addition, some shell collectors could have removed hydroids prior to deposition. The fact that some of the shells showed unequivocal evidence of the hydroid is sufficient to establish that the association has been around for at least 100 y and probably longer.
The results of our choice experiment suggest that the epibiotic hydroid Lovenella gracilis can modify predator-prey interactions between D. fossor and N. duplicata. Donax fossor with hydroids were less likely to be drilled than those without hydroids (Table 2). One explanation for this pattern involves cnidocytes. Neverita duplicata must manipulate the bivalve into the appropriate position for drilling, which may take several hours or days to complete. During this process, the snails may be continually exposed to discharging nematocysts. Lovenella gracilis is equipped with microbasic mastigophores (Calder 1971), which can penetrate animal tissues and inject venom (Mariscal 1974, Purcell & Mills 1988). Although this type of nematocyst provides little defense for some hydroid species (e.g., Tridentata marginata, T. turbinata, Aglaophenia latecarinata), in other hydroid species it may produce more toxic venoms (Stachowicz & Lindquist 2000). Donax fossor appears to profit from the defenses of its epibiont that typically lives in a different habitat.
In addition, the clam may be more difficult to manipulate as a consequence of hydroid presence. This phenomenon has been observed in other bivalve-epibiont associations. For example, predators have a harder time gripping, manipulating, and capturing Chlamys spp. with epibiotic sponges than those without epibionts (Bloom 1975, Forester 1979, Chernoff 1987).
Investigations with other bivalves also suggest that epibionts reduce the risk of predation, particularly by seastars. Vance (1978) showed that removal of a variety of epibionts (erect bryozoans, coralline algae, hydrozoans) substantially increased mortality by Pisaster giganteus on the jewel box clam Chama pellucida. Starfish predation was impeded by sponges colonizing the spines on the spiny oyster Spondylus americanus (Feifarek 1987). Feifarek (1987) suggested that spines may have evolved solely to attract epibionts. The blue mussel Mytilus edulis experiences reduced predation pressure from Asterias rubens and the crab Carcinus maenus (Wahl et al. 1997, Enderlein et al. 2003) by supporting the hydrozoan Laomedea flexuosa.
In some cases, hydroids enhance predation. Hydractinia enhanced predation by the calico crab Hepatus epheliticus (Brooks & Mariscal 1985). Brooks and Mariscal (1985) suggested that this crab might be accustomed to stinging cnidocytes because anemones (which are also equipped with cnidocytes) are often found on its carapace. Buckley and Ebersole (1994) found that hermit crabs with hydroids are more likely to have shell-degrading worms, which makes the shell more vulnerable to predation by the blue crab Callinectes sapidus. In addition to anemones and hydroids, epibiotic bryozoans also provide protection. The whelk Bunupena papyracea covered in the bryozoan Alcyonidium nodosum was found to be resistant to lobster attacks (Barkai & McQuaid 1988).
The Donax-Lovenella association could be considered mutualism, since predation pressure on D. fossor is reduced by the presence of the hydroid (Table 2). Further evidence for this categorization comes from predation studies involving other taxa. The role of L. gracilis in protecting Donax spp. is not unique to gastropods; the hydroid is capable of deterring vertebrate predators as well. Manning and Lindquist (2003) demonstrated that L. gracilis defends D. variabilis against a common Donax predator in North Carolina, the Florida pompano Trachinotus carolinus. Only 0.87% of the clams consumed in that study had hydroids. When nematocysts were deactivated, the pompano failed to discriminate between clams with and without hydroids, suggesting that nematocysts are responsible for deterring this predator.
Lovenella gracilis is not an effective predator deterrent in all consumer interactions of Donax spp. Instead, the hydroid facilitates predation by the portunid speckled crab Arenaeus cribrarius and the semiterrestrial ghost crab Ocypode quadrata (Manning & Lindquist 2003), although one other investigation showed that L. gracilis had no effect on ghost crab predation rates (Wolcott, pers. comm.). Manning and Lindquist (2003) suggested that the hydroid (which can project above the surface) indicates clam location in the sediment as the crab moves along the surface of the sand.
Predation pressure of N. duplicata, T. carolinus, A. cribrarius, on Donax spp. may vary throughout clam's range on the Atlantic coast. Although pompano prey upon D. fossor in New Jersey (McDermott 1983), N. duplicata is likely the more important predator. Shells drilled by N. duplicata frequently can be observed along beaches in southern New Jersey (personal observation). Although A. cribrarius and O. quadrata occur in New Jersey (Williams 1984), their relative importance as predators of D. fossor is unknown.
The Donax-Lovenella epibiosis shifts from mutualism to parasitism, depending on the relative importance of predators in specific geographic regions. Lovenella gracilis protects Donax spp. from some predators (moonsnail and pompano) and attracts others (ghost crab and speckled crab). Rarely are interactions of animals living in close association so simple to be deemed as strictly parasitic, mutualistic, or commensal.
Lovenella gracilis had no effect on the shoreward passive transport of D. fossor, suggesting that the hydroid has no effect on the ability of D. fossor to swash-ride. Manning and Lindquist (2003) performed a different in situ swash experiment in North Carolina. They compared the passive seaward transport of D. variabilis with and without hydroids in the backwash. In their study, they quantified the distance traveled from where the clams were dropped in the backwash to the site where burrowing initiated. They found that clams without hydroids traveled significantly farther in the backwash than those with hydroids.
We cannot compare this experiment to the one Manning and Lindquist (2003) performed. They examined transport in the backwash (seaward transport), whereas we examined transport in the swash (shoreward transport). Although we attempted to measure the net distance traveled by clams, it was not feasible. However, assuming that clams are affected in the backwash as they are in the swash, our experiment would predict that L. gracilis does not affect passive transport of Donax spp., which is contrary to the findings of Manning and Lindquist (2003). Their results indicate that clams without hydroids would travel farther in the swash. This may be attributable to differences in wave velocity and/or beach morphology between the two study sites.
The results of our field surveys indicate that the Donax-Lovenella association is seasonal in occurrence, because neither species over winters in the swash zone. Similar patterns of seasonal occurrence have been documented in other Donax-hydroid associations. Clytia bakeri displays a seasonal pattern of abundance on D. variabilis in Texas and D. gouldii in California. Johnson (1966) documented the occurrence of a commensal hydroid, Clytia bakeri, on Donax gouldii on the west coast of North America. During the fall and winter, hydrocauli were present on only few clams, however, by August nearly every clam supported a "plume" of C. bakeri, which itself supported a brown alga (Johnson 1966). Loesch (1957) reported similar findings on the occurrence of Clytia on D. variabilis on Mustang Island, Texas.
Lovenella gracilis exhibited a seasonal occurrence throughout the sampling period. Lovenella gracilis is active only (as indicated by presence of hydranths) from April to October in Virginia (Calder 1990). This hydroid becomes active when water temperatures reach 15[degrees]C at the beginning of the season and ceases activity at 20[degrees]C at the end of the season (Calder 1990).
Our observations of the seasonal occurrence of L. gracilis on D. fossor is consistent with cycles of activity and dormancy previously reported. Decreasing temperatures cause the tissue of hydroids to regress into a small section of hydrocaulus or hydrorhizae, which regrow when favorable temperatures return (Gili & Hughes 1995). Surface water temperature along barrier islands in southern New Jersey reaches this temperature during the first 2 wk of June (NOAA), which is consistent with observed recruitment in 2002 and 2003 (Fig. 4).
Although annual patterns of hydroid activity are determined by temperature, L. gracilis cannot exist in the swash zone without a suitable substrate. Favorable temperatures are not necessarily indicative of hydroid presence. For example, temperatures were favorable in July 2003 and there was a general lack of hydroids compared with 2002.
Furthermore, the proportion of the D. fossor population supporting a hydroid colony varies among clam size, months, years, and between the two study sites. Although timing of hydroid recruitment was consistent between sites, the proportion of the D. fossor population with hydroids varied between sites. A greater proportion of the population in Avalon supported a hydroid colony during fall 2001 compared with Sea Isle (Fig. 5). In contrast, a greater proportion of the population in Sea Isle supported a hydroid colony during summer 2002 and fall 2003 compared with Avalon (Fig. 6, Fig. 9). Wave action, local water currents, variation in the density of clams between the two study sites, along with stochastic events are likely factors influencing these patterns.
The Donax-Lovenella association seemed atypical in 2003 compared with observations in the 2 previous years. Clam density never reached the peak abundances (except for September 2003 Avalon) that it had in previous years. Large clams that were present in July were not present in August or September. Furthermore, in 2003 hydroid colonies were small and nearly invisible to the unaided eye from June through September. In the previous 2 y, large hydroids were easily detected in the swash zone. However, the percentage of the D. fossor population with L. gracilis reached a maximum for the sampling period in fall 2003.
Overall, there was a clear size-specific relationship with larger clams more likely to support hydroids than smaller clams (Fig. 10). Larger clams have more available surface area for hydroid settlement. Furthermore, larger clams are older and the settlement area of shell is exposed for a longer period of time to potential hydroid larvae. Both time and surface area are likely factors contributing to the size-specific pattern in Figure 11.
Lovenella gracilis takes a risk by settling on a biotic substrate. For example, survival of the hydroid is dependent upon survival of the clam. If the clam dies, the hydroid likely dies as well. In general, neither the clam nor the hydroid survives the winter. Because hydroids produce gonangia quite quickly in warm water, L. gracilis needs Donax spp. for only a brief period. The seasonality of both species makes them ideal partners for epibiosis.
TABLE 1. Summary of dry specimen collections of Donax fossor with remnant hydroids. Museum Accession Locality Year Month [N.sub [N.sub .T] .w] ANSP 79623 Wildwood, 1900 ND >100 4 NJ ANSP 81550 Sea Isle 1901 July >100 1 City, NJ ANSP 53033 Atlantic -1886 ND 34 1 City, NJ DMNH 44079 Avalon, NJ 1962 September 14 1 DMNH 222232 Indian 1965 August 11 2 River Inlet, DE Museum is either the Academy of Natural Sciences Philadelphia (ANSP) or Delaware Museum of Natural History (DMNH). Accession is the accession number of the lot. Locality is collection site. Year is the year of collection or year of accession if the year of collection was not available. Month is the month when the specimens were collected. ND = No data available. [N.sub.T] is the total number of valves in lot. [N.sub.w] is the number of valves with the hydroids in the lot. TABLE 2. Chi-square analysis of clams drilled by Neverita duplicata. [N.sub.w] [N.sub.w/o] df Snail 1 8 13 1 Snail 2 7 16 1 Snail 3 1 15 1 Snail 4 4 14 1 Total [[chi square] .sub.calc] 3 Pooler [[chi square] .sub.calc] 20 58 1 Heterogeneity [[chi square].sub.calc] 2 [chi square] P Snail 1 1.19 0.275 ns Snail 2 3.53 0.0606 ns Snail 3 12.3 0.000465 *** Snail 4 5.56 0.0184 Total [[chi square] .sub.calc] 22.5 <0.0001 *** Pooler [[chi square] .sub.calc] 18.5 <0.0001 *** Heterogeneity [[chi square].sub.calc] 4.01 0.135 ns [N.sub.w] is the number of clams with hydroids drilled during the experiment, [N.sub.w/o] is the number of clams without hydroids drilled during the experiment, df are the degrees of freedom, [chi square] is the [chi square]-value of the test, and P is the significance level. TABLE 3 Two-tailed, paired t-test of passive shoreward transport clams. [D.sub.w] [D.sub.w/o] df t-stat P-value (cm) (cm) Mean 186 181 45 -0.86 0.40 ns [[sigma] 10,771 10,196 .sup.2] [D.sub.w] is the distance traveled by clams with hydroids, [D.sub.w/o] is the distance traveled by clams without hydroids, df are the degrees of freedom, t-stat is the t-value of the test, and P is the significance level.
The authors thank K. Wieder and D. Miller; for their constructive comments, the Gambino family and C. Cherry; who provided logistic support in the field while P. Dougherty, A. Smolock, A. Nolan, A. Palamo, K. Frank, J. Beck, S. Narayanan, and J. Campbell assisted with the lab and field work. This research was generously supported by the Lerner-Gray Fund for Marine Research, a Conchologists of America Research Grant, and a Grant-in-Aid of Research from Sigma-Xi, and The Scientific Research Society awarded to J.R.D. This work formed the basis of J. R. D.'s MS thesis at Villanova.
Ansell, A. D. 1983. The biology of the genus Donax. In: A. McLachlan & T. Erasmus, editors. Sandy beaches as ecosystems. The Hague: Dr. W. Junk Publishers. pp. 607-635.
Barkai, A. & C. McQuaid. 1988. Predator-prey role reversal in a marine benthic ecosystem. Science 242:62-64.
Bloom, S. A. 1975. The motile escape of a sessile prey: a sponge-scallop mutualism. J. Exp. Mar. Biol. Ecol. 17:311-321.
Brooks, W. & R. Mariscal. 1985. Protection of the hermit crab Pagurus pollicaris Say from predators by hydroid-colonized shells. J. Exp. Mar. Biol. Ecol. 87:111-118.
Buckley, W. J. & J. P. Ebersole. 1994. Symbiotic organisms increase the vulnerability of a hermit crab to predation. J. Exp. Mar. Biol. Ecol. 182:49-64.
Calder, D. R. 1971. Hydroids and hydromedusae of southern Chesapeake Bay. VA. Inst. Mar. Sci. Spec. Pap. Mar. Sci. 1.
Calder, D. R. 1990. Seasonal cycles of activity and inactivity in some hydroids from Virginia and South Carolina. USA. Can. J. Zool. 68: 442-450.
Chernoff, H. 1987. Factors affecting mortality of the scallop Chlamys asperrina (Lamarck) and its epizooic sponge in South Australian waters. J. Exp. Mar. Biol. Ecol. 109:155-171.
Ellers, O. 1995. Behavioral control of swash-riding in the clam Donax variabilis. Biol. Bull. 189:120-127.
Enderlein, P., S. Moorthi, H. Rohrscheidt & M. Wahl. 2003. Optimal foraging versus shared doom effects: interactive influence of mussel size and epibiosis on predator preference. J. Exp. Mar. Biol. Ecol. 292:231-242.
Feifarek, B. P. 1987. Spines and epibionts as antipredator defenses in the thorny oyster Spondylus americanus Hermann. J. Exp. Mar. Biol. Ecol. 105:39-56.
Forester, A. J. 1979. The association between the sponge Halichondria panicea (Pallas) and the scallop Chlamys varia (L.): a commensal-protective mutualism. J. Exp. Mar. Biol. Ecol. 36:1-10.
Gili, J., P. Abello & R. Villanueva. 1993. Epibionts and intermoult duration in the crab Bathynectes piperitus. Mar. Ecol. Prog. Ser. 98:107-113.
Gili, J. & R. G. Hughes. 1995. The ecology of marine benthic hydroids. Oceanogr. Mar. Biol. Annu. Rev. 33:351-426.
Jackson, J. B. C. 1977. Competition on marine hard substrata: the adaptive significance of solitary and colonial strategies. Am. Nat. 111:743-767.
Jarms, G. & H. Tiemann. 1996. On a new hydropolyp without tentacles, Microhydrula limopsicola n. sp., epibiotic on bivalve shells from the Antarctic. Sci. Mar. 60:109-115.
Johnson, P. T. 1966. On Donax and other sandy-beach inhabitants. The Veliger 9:29-30.
Loesch, H. C. 1957. Studies of the ecology of two species of Donax on Mustang Island, Texas. Pub. Inst. Mar. Sci. Univ. Tex. 4:201-227.
Manning, L. & N. Lindquist. 2003. Helpful habitant or pernicious passenger: interactions between an infaunal bivalve, an epifaunal hydroid and three potential predators. Oecologia 134:415-422.
Mariscal, R. 1974. Nematocysts. In: L. Muscatine & H. Lenhoff, editors. Coelenterate biology. New York: Academic Press.
McDermott, J. J. 1983. Food web in the surf zone of an exposed sandy beach along the mid-Atlantic cost of the United States. In: A. McLachlan & T. Erasmus, editors. Sandy beaches as ecosystems. The Hague: Dr. W Junk Publishers. pp. 529-538.
Mikkelsen, P. S. 1978 A comparison of intertidal distribution, growth rates and shell polychromism between two Florida populations of the coquina clam, Donax variabilis Say, 1822 (Bivalvia: Donacidae), M.Sc. thesis. Florida Institute of Technology. 110 pp.
Patil, J. S. & A. C. Anil. 2000. Epibiotic community of the horseshoe crab Tachypleus gigas. Mar. Biol. 136:699-713.
Purcell, J. & C. Mills. 1988. The correlation between nematocyst types and diets in pelagic hydrozoa. In: D. Hessinger & H. Lenhoff, editor. The biology of nematocysts. San Diego: Academic Press.
Ruppert, E. & R. Fox. 1988. Seashore animals of the Southeast University of South Carolina Press, Columbia, SC.
Stachowicz, J. J. & N. Lindquist. 2000. Hydroid defenses against predators: the importance of secondary metabolites versus nematocysts. Oecologia 124:280-288.
Stachowitsch, M. 1977. The hermit crab microbiocoenosis: The of role of mobile secondary hard bottom elements in a North Atlantic benthic community. In: B. Keegan, P. Ceideigh & P. Boaden, editor. Biology of benthic organisms. Oxford: Pergamon.
Vance, R. R. 1978. A mutualistic interaction between a sessile marine clam and its epibionts. Ecology 59:679-685.
Wahl, M., M. E. Hay & P. Enderlein. 1997. Effects of epibiosis on consumer-prey interactions. Hydrobiologia 355:49-59.
Williams, A. 1984. Shrimps, lobsters, and crabs of the Atlantic coast of the Eastern United States, Maine to Florida Smithsonian, Washington, DC.
J. R. DOUGHERTY AND MICHAEL P. RUSSELL *
Biology Department, Villanova University, Villanova, PA 19085
* Corresponding author. E-mall: email@example.com; Fax: +1-610-519-7863
|Printer friendly Cite/link Email Feedback|
|Author:||Russell, Michael P.|
|Publication:||Journal of Shellfish Research|
|Date:||Jan 1, 2005|
|Previous Article:||Short-term effects of rapid salinity reduction on seed clams (Mercenaria mercenaria).|
|Next Article:||Commercial harvest and population structure of a northern quahog (Mercenaria mercenaria linnaeus 1758) population in St. Mary's Bay, Nova Scotia,...|