Restoration of bay scallop (Argopecten irradians (Lamarck)) populations in Florida coastal waters: planting techniques and the growth, mortality and reproductive development of planted scallops.

ABSTRACT

Bay scallops (Argopecten irradians [Lamarck]) are a culturally and economically important component of Florida's nearshore marine community. However, many of the local populations that compose the hay scallop metapopulation in Florida have virtually disappeared since the 1950s. This study reports the results of a 3-year effort to restore bay scallop populations at 4 sites along the west central coast of the state (Tampa Bay, Anclote, Homosassa and Crystal River). During late summer of 1997, 1998 and 1999, wild adult scallops were retrieved from each of those four target sites and induced to spawn in the laboratory. The resultant offspring were grown to at least 20-mm shell height in a nursery setting and then transplanted to cages deployed at the site where their parents were originally harvested. The growth, survival and reproductive development of the planted scallops were recorded on an approximate 6-week schedule. Results suggest that caged scallops generally grew more slowly than their wild counterparts and that at most of the planting sites mortality was high, especially during late summer. Reproductive development and spawning, although delayed in the caged scallops relative to their wild conspecifics, appears to have proceeded in an otherwise normal fashion. Approximately 1,100 scallops survived and spawned during the first year of the project, whereas -4,700 and 12,000 scallops survived to spawn in the second and third years of the project, respectively. Studies were also conducted to determine the optimal stocking density and the best placement of the cages. Results of the density study indicate that planting at lower densities increased growth and survival but did not necessarily result in more live scallops at the time of spawning. Results of the cage-placement study, which compared scallops planted in cages either inside or outside of a seagrass bed and either mounted on legs or placed directly on the sediment, revealed that scallops planted directly on the substrate within a seagrass bed suffered higher mortality and slower growth than did scallops planted in the other three treatment combinations. Overall results of this 3-year project suggest that planting cultured scallops in cages can be a successful strategy for increasing the local spawner stock density of bay scallops in depleted populations and, ultimately, for increasing larval supply to the metapopulation.

KEY WORDS: Argopecten irradians, restoration, growth, mortality, reproduction, metapopulation scallop

INTRODUCTION

Overharvest and habitat loss are prominent factors that exert negative pressure on the stability of many commercially or recreationally harvested marine species. Even when harvesting pressure is removed or when habitat loss is arrested or reversed, there is no assurance that affected populations will rebound, because as population density decreases below a threshold level, depensatory effects may render the population effectively sterile (e.g., Stoner & Ray-Culp 2000). In such cases, active intervention may be required to restore the population to reproductive viability. Several techniques have been applied during the 100+ year history of marine population-restoration efforts to rebuild reproductive viability in collapsed populations. Examples include the construction or rehabilitation of habitat with (Wallace et al. 1995) or without (Hoffman et al. 1982) the inclusion of seeding efforts, direct larval release (Preece et al. 1997, Arnold et al. 2002) and the introduction of cultured juveniles or adults (Tettelbach & Wenczel 1993, Peterson et al. 1996, Arnold et al. 2002). Results have been mixed. Analysis of a multidecade release of larvae into cod populations in Norway found no evidence of a contribution (Chan et al. 2003). In contrast, stocking larval striped bass in Chesapeake Bay contributed 20% to 30% to overall juvenile abundance in years of poor natural larval production (Secor & Houde 1998). Construction, rehabilitation and seeding of oyster (Crassostrea virginica [Gmelin]) reefs in Atlantic and Gulf of Mexico estuaries throughout the United States have long and successful histories (Wallace et al. 1995). Cultured and wild-caught individuals have been transplanted in efforts to restore or supplement scallop populations worldwide. In French coastal waters, more than 20% of seeded scallops were recaptured in the commercial fishery (Dao et al. 1994), but there was no clear indication of a successful contribution to future year classes from transplanted adult bay scallops (Argopecten irradians [Lamarck]) in a North Carolina study (Peterson et al. 1996), and extensive bay scallop restoration efforts in New York waters have been insufficient to overcome the effects of harmful algal blooms (Tettelbach & Wenczel 1993).

In the eastern Gulf of Mexico, the bay scallop once supported important commercial and recreational fisheries from Florida Bay to the Chandeleur Islands (Fig. 1). For example, in the early 1950s between 10 and 120 thousand pounds of bay scallop meats were landed annually in Pine Island Sound in southwest Florida (Murdock 1955), but commercial bay scallop landings from Florida waters decreased to essentially zero by the early 1960s. In recent years, bay scallop populations have been severely depleted over a substantial portion of the animal's range in Florida waters (Arnold & Marelli 1991), to the point that even the recreational fishery is threatened. Formerly abundant populations in areas such as Sarasota Bay and Tampa Bay essentially disappeared by the early 1960s (Haddad 1988). By the late 1980s few scallop populations were extant in areas as far north as Crystal River, and only two high-density populations remained in Florida waters: in the near-shore waters near Steinhatchee in northern peninsular Florida and in the enclosed waters of St. Joseph Bay in the Florida Panhandle (Arnold et al. 1998). Possible causes of the demise of bay scallops in Florida include habitat loss, diminished water quality, overfishing, recruitment failure, or a combination of these factors (Barber & Blake 1983, Arnold & Marelli 1991, Blake et al. 1993, Sargent et al. 1995).

[FIGURE 1 OMITTED]

To properly manage bay scallop populations in Florida, it is necessary to understand the biology and ecology of the organism. Bay scallops are short-lived animals; in Florida waters, scallops rarely live longer than 18 mo (Barber & Blake 1981). They are simultaneous hermaphrodites and generally undergo one primary spawning event during late fall or winter, although minor spawning events may occur during other seasons (Barber & Blake 1983, Bologna 1998). The gametes are shed sequentially rather than simultaneously, and fertilization is external. Once shed, the gametes (particularly the sperm) may be viable for only minutes to hours (Levitan 1995), so fertilization success is dependent on the density of the spawning stock. Successfully fertilized eggs initiate development almost immediately; proceed through trochophore, veliger and pediveliger stages during the next 7-14 days; and settle to the substratum at a size of about 180-200 [micro]m (Sastry 1965). Factors that influence successful spawning and recruitment may have a disproportionate influence on population abundance because the eggs and larvae tend to be more sensitive to environmental conditions than are the juveniles and adults (Tettelbach & Rhodes 1981). Additionally, bay scallops in Florida waters generally are dependent on seagrass as a substrate for larval settlement (Thayer & Stuart 1974, Pohle et al. 1991). Thus, three factors play a critical role in the successful recruitment of bay scallops: adequate adult density to ensure successful fertilization, biological and environmental factors suitable for the survival of gametes and larvae, and the availability of seagrass beds appropriate for the settlement and survival of recruiting larvae.

We hypothesize that the bay scallop population in Florida functions as a metapopulation (Hanski 1991), with local populations historically occupying discrete embayments and seagrass beds from West Palm Beach (Fig. 1) to the Chandeleur Islands in Louisiana (Waller 1969, Arnold et al. 1998). It is possible that what appears to be metapopulation structure is simply the structure of a population that was once continuous but is now in the throes of extinction (Hanski & Simberloff 1996); however, several lines of evidence argue against this alternative. First, along much of the Florida coastline, suitable bay scallop habitat is confined within bay systems such as Biscayne Bay, Florida Bay, Pine Island Sound and Tampa Bay in the south and St. Joseph Bay, St. Andrew Bay and Pensacola Bay in the north (Fig. 1). Similarly, the "Big Bend" region may function as a single large bay defined along its seaward edge by a coastal "jet" that flows between Tampa Bay and St. Joseph Bay (Yang et al. 1999, Li & Weisberg 1999). Even within the Big Bend, scallop populations are discretely distributed at sites such as Anclote, Homosassa and Steinhatchee (Arnold et al. 1998). Second, commercial fisheries landings and anecdotal information suggest that local populations have experienced large fluctuations in abundance, as would be expected in a short-lived animal such as the bay scallop. Thus, it appears that the bay scallop population in the eastern Gulf of Mexico has always been composed of discrete local populations and that extinction and recolonization are common features of those populations. These traits characterize a metapopulation (Levins 1969).

If habitat degradation has occurred, as we suspect it has in areas such as Pine Island Sound and Tampa Bay, then it is probable that directly enhancing scallop populations in those areas will be fruitless. However, if one or more "source" (Pulliam 1988) populations have been adversely affected by habitat degradation or other factors, with the result that other (neutral or sink) populations in otherwise suitable areas have been adversely affected, then it may be possible to rebuild multiple local populations by rebuilding the key source populations. Tampa Bay appears to have functioned as a key source population on the Florida west coast. Historically, Tampa Bay supported numerous seagrass beds, and each of those beds may have harbored a population of bay scallops. As the land area around Tampa Bay was developed, dredge-and-fill and bridge-building activities radically altered the seascape, causing degradation of water quality and loss of seagrass. In recent decades, efforts to restore Tampa Bay have resulted in a considerable increase in water quality (Blake 1996, 1998) and a concomitant increase in the habitat available for scallops. This provides an ideal opportunity to test the feasibility of bay scallop restoration in Florida.

In areas north of Tampa Bay, including Anclote, Homosassa and Crystal River, the loss of scallop populations may be the result of several factors acting individually or in concert. Harmful algal blooms (HABs) occasionally invade this area of Florida and appear to be inimical to the survival of scallops (Cosper et al. 1987, Summerson & Peterson 1990). Additionally, seagrass habitat at sites such as Anclote and Homosassa may be degraded to the point that scallop populations are no longer supported. However, although residential and commercial development continues along the west coast of Florida, the area north of Anclote still appears to be relatively pristine and the seagrass beds vast and abundant. Finally, it may be that each local population between Anclote and Crystal River simply experienced collapse in the normal course of its life history. As with any annual species, population fluctuations tend to be drastic even in apparently healthy populations. During years of low population abundance, reproduction in the local population may no longer be successful (Levitan 1995, Marelli et al. 1999). In that circumstance, the self-seeding necessary for year-to-year maintenance of the population (Arnold et al. 1998) may fail, and the population will not recover without allochthonous larval inputs. Such allochthonous inputs and subsequent population recovery may require many years (Peterson & Summerson 1992), and in the case of Florida bay scallops, recovery may never occur if neighboring local populations have also collapsed.

Our strategy for bay scallop restoration on the Florida west coast exploits the earlier mentioned considerations to test the bay scallop metapopulation hypothesis and to try to rebuild local populations within that framework. We hypothesized that restoration of Tampa Bay, Anclote, Homosassa and Crystal River scallop populations would require the input of locally generated recruits. The plan we implemented involved capturing wild scallops from the target site and spawning them in the bivalve hatchery at the University of South Florida College of Marine Science (Lu & Blake 1997). We raised the offspring in local marine waters until they were -20-30 mm shell height (SH = maximum distance from umbo to ventral margin) and then planted them in cages at the target site from which the parents were originally collected. Our goal was to create dense concentrations of spawners that would produce large numbers of fertilized gametes. Some proportion of the resultant offspring would settle in the surrounding seagrass beds (Arnold et al. 1998), thereby providing the spawner stock for future generations. We report here on our planting methods and on basic biological parameters (growth, survival and reproductive development) of the planted scallops. We also report the results of a study comparing planting methods and densities.

METHODS

During preliminary trials, we used 0.6-m L x 0.6-m W x 0.16-m H cages constructed of 25.4-mm plastic-coated wire mesh to house planted bay scallops. However, the 25.4-mm mesh size proved inadequate to deter predators such as stone crabs (Menippe spp.), which apparently were able to work their claws into the cages and kill the scallops. For this study, we used cages of identical dimensions but constructed of 12.7-mm plastic-coated wire mesh. Use of the smaller mesh not only reduced predatory losses but also allowed us to plant the scallops at a smaller size without loss through the mesh, thereby enabling us to plant the scallops during spring rather than summer. This approach minimized temperature-and salinity-related mortality in our nursery facilities during June and especially July.

Study sites were located within Tampa Bay and in the coastal waters near Anclote, Homosassa and Crystal River (Fig. 1). The specific location of each study site within each of those general areas was determined based on the presence of seagrass and anecdotal and survey information that indicated the selected areas supported an existing scallop population. The Anclote, Homosassa and Crystal River sites were chosen based on our previous research experience in those areas (Arnold et al. 1998). In Tampa Bay, shrimp boat captains reported that scallops were captured during trawling operations in the area between the Sunshine Skyway Bridge and Tierra Verde (Fig. 1). Within that area, several locations are designated as "no-motor" zones for boating activity to provide protection for the extensive shallow-water seagrass beds in those locations. We selected those locations for our study sites.

Year One-1998 Planting

Tampa Bay

We originally planted 300 scallops into each of five cages at our Tampa Bay study site on June 17, 1998 (Fig. 1). We also deployed an additional cage at Station 5 to house scallops that were to be collected for reproductive analyses. At three of the five stations, the scallops had either suffered complete mortality (Stations 2 and 5) or the cages could not be relocated (Station 4) by the beginning of July. At Station 4, it was evident that the cages had been stolen because we were able to relocate the anchor blocks but not the cages.

In response to the loss of scallops and cages suffered at our June 17 planting sites, on July 20 we reorganized our planting sites in Tampa Bay in anticipation of a second planting attempt. First, we abandoned Station 2 and renumbered Station 3 as Station 2. As a result, we now had two stations at Indian Key, each of which contained scallops that were originally planted on June 17 and exhibited good survival through the middle of July. Second, we deployed additional cages at six sites within the no-motor zone surrounding Tarpon Key, and we deployed cages at two sites in the vicinity of Madeline Key (Fig. 1). Finally, on July 21 we planted various numbers of scallops in each of the cages at the stations that we had established on July 20. Because of the limited number of Tampa Bay offspring available for this second planting, we planted fewer scallops in each cage and the scallops were offspring of two separate parental groups (Table 1).

Following our reorganization of the Tampa Bay cage deployments, we suffered one final act of vandalism. The cages planted at Station 2 (originally Station 3 at Indian Key) were stolen sometime between our July 17 and August 20 sample dates. As a result, we report growth and mortality from only nine stations in Tampa Bay during 1998.

Growth, mortality and reproductive development of scallops planted in Tampa Bay were monitored approximately every 3 weeks beginning on August 20, 1998 and continuing through February 8, 1999. To estimate mortality, all live scallops were counted from one randomly selected cage at each station. An exception to this was that paired and unpaired valves were counted if few mortalities were noted, in which case the total number of unpaired valves was divided by two to estimate mortality. Mean scallop SH was estimated by measuring 30 scallops (or all remaining scallops if <30 remained alive) from the same cage from which mortality was estimated. Scallops were measured to a precision of 1 mm using a lined ruler. Statistically significant differences in the number of surviving scallops or in the mean shell height of the surviving scallops, among dates, were determined using the Ryan-Einot-Gabriel-Welsch Q test, a multiple-stage test that controls the maximum experimentwise error rate (Day & Quinn 1989, SAS Institute Inc. 1999).

For analysis of gonad developmental stage, 15 scallops were collected on each sample date from the extra cage at Station 7 and returned alive to the laboratory. There, the gonad was dissected from the remaining visceral mass and fixed in a 5% formalin/ seawater solution for at least 24 h. To ensure osmotic balance during the fixation process, seawater for the solution was collected from the vicinity of Station 7 at the time of sampling. Following fixation, each gonad was thoroughly rinsed with tap water and preserved in 70% ethanol in anticipation of histologic preparation.

Histologic processing was initiated by dehydrating the gonad in 95% ethanol for a minimum of 3 hours. Gonads were embedded in JB4 glycomethacrylate resin, beginning with an exposure to 50% JB4 dissolved in 95% ethanol for several days and followed by two sequential 3-day exposures to 100% JB4. At least two 3.5-[micro]m sections were then cut from the embedded sample by using a microtome mounted with a glass knife, maintaining a minimum separation of at least 60 [micro]m between sections. Thin sections were stained with hematoxylin and eosin and mounted on prelabeled glass slides for analysis. The resultant slides were examined at x 200-400, and each sample was assigned to a reproductive stage following a classification scheme (Table 2) modified from the work of Loosanoff (1937), Jaramillio et al. (1993) and Walker and Heffernan (1994).

An essentially identical methodology was followed for monitoring growth, survival and reproductive development at the other study sites during all 3 years of the study. Sampling efforts were conducted approximately every 6 weeks unless otherwise noted.

Anclote

We were unable to plant scallops at the Anclote study site during 1998 because of the lack of success in culturing scallops from broodstock originally collected from that site.

Homosassa

We planted 50 scallops in each of five cages at each of our 10 restoration stations in Homosassa (Fig. 1), and we planted an additional cage of 300 scallops at Station 5 for provision of reproductive samples (Table 1). Growth, mortality and reproductive development of scallops planted at the Homosassa study site were monitored beginning on August 12, 1998 and continuing through January 21, 1999.

Crystal River

We planted 50 scallops in each of five cages at each of our 10 restoration stations in coastal waters near Crystal River (Fig. 1), and we planted an additional cage of 300 scallops at Station 1 for provision of reproductive samples (Table 1). Unfortunately, the cages at only one station (Station 10) could be relocated and all of the scallops in those cages were dead, so the Crystal River effort was terminated.

Aquaculture Raft 

Scallops from a variety of spawns were planted, at a density of 200 scallops per cage, in 15 cages on an aquaculture demonstration raft anchored in 5 m of water off Crystal River (Fig. 1; Table 1). The scallop raft measured 8 m x 8 m and supported an extruded aluminum pen that extended 1 m below the bottom surface of the raft. Scallops were planted into 12.7-mm-mesh cages identical to the ones used at the bottom planting sites described earlier, and the cages were placed inside the pen. Scallops were periodically monitored for growth by measuring 50 scallops from a single preselected cage on each sample date, and mortality was estimated for that same cage by removing and counting the number of empty shells. Samples (n = 15 per date) for gonadal analysis were collected from various cages on each of three dates: August 18, September 15 and October 29, 1998.

Year Two-1999 Planting

Tampa Bay

We were not able to plant scallops at the Tampa Bay study site because we were unable to collect enough scallops from Tampa Bay during summer 1998 to serve as parental stock.

Anclote

We planted 300 offspring from the 98A5C brood in each of five cages at each of our 10 restoration stations at Anclote (Fig. 1), and we planted an additional cage of 300 scallops at Station 5 for provision of reproductive samples (Table 1). At the time of planting, we detected a significant difference in mean SH among the 10 cages (Nested ANOVA, F = 4.49, P < 0.001, n = 150 scallops per station). Mean SH ranged from a minimum of 27.3 mm at Station 2 to a maximum of 29.9 nun at Station 8. We initially monitored the growth, mortality and reproductive development of those scallops 3 weeks after deployment (May 27) and then at approximate 6-week intervals through April 24, 2001.

Homosassa

We planted 300 scallops in each of five cages at each of our 10 restoration stations in Homosassa, and we planted an additional cage of 300 scallops at Station 5 for provision of reproductive samples. At the time of planting, mean SH varied significantly among stations (Nested ANOVA, F = 7.07, P < 0.0001, n = 150 scallops per station) and ranged from a minimum of 25.7 mm at Station 7 to a maximum of 29.1 mm at Station 3. We monitored the growth, mortality and reproductive development of those scallops beginning on April 20, 1999 and continuing through December 20, 1999.

Crystal River

We planted 300 scallops in each of five cages at each of our 10 restoration stations in Crystal River, and we planted an additional cage of 300 scallops at Station 5 for provision of reproductive samples. At the time of planting, mean SH varied significantly among stations (Nested ANOVA, F = 13.47, P < 0.0001, n = 150 scallops per station) and ranged from a minimum of 23.7 mm SH at Station 2 to a maximum of 28.8 mm SH at Station 4. We monitored the growth, mortality and reproductive development of those scallops beginning on April 20, 1999 and continuing through October 26, 2000.

PLANTING DENSITY STUDY

During July 1999, we initiated a study comparing growth, survival and gonadal development of bay scallops planted at various densities in cages at our Anclote study site. Offspring from parental group 98A5C were planted at densities of 50, 150 and 300 animals per cage. Three replicates of each density were deployed in a random fashion within the cage array on July 20, 1999 and were monitored through July 5, 2000.

Sampling was conducted on an approximate 6-week schedule but varied depending on weather and other logistical constraints. During each sampling event, all living scallops were counted in each cage. Shell height was measured for a subset of 10 scallops from each treatment replicate, ensuring that a total of at least 30 scallops were measured from each treatment combination (or that all scallops were measured if <30 remained alive). Beginning in November 1999, five scallops were harvested from each cage on each sample date and returned to the laboratory for histologic preparation according to our previously described methods. However, we were unable to process samples collected from the 300-scallop treatment on January 13 because of excessive grit in the tissue.

PLANTING HABITAT STUDY

On September 3, 1999 at our Anclote study site, we initiated a study comparing various schemes for deploying cages in the field. We tested four treatment combinations of cages deployed either within or outside of a seagrass bed and either directly on the substrate or elevated 20 cm above the substrate on 12.7-mm-diameter PVC legs. All treatments were deployed in triplicate, and offspring from parental group 98A5C were planted at a density of 50 scallops per cage. Samples for survival and growth analysis were collected approximately every 6-weeks beginning October 11, 1999 and continuing through April 6, 2000. During each sampling event, all living scallops were counted in each cage. Shell height was measured for a subset of 10 scallops from each treatment replicate, ensuring that at least 30 scallops were measured from each treatment combination (or that all scallops were measured if <30 remained alive).

Year Three-2000 Planting

Based on the results of the planting habitat study that we conducted at our Anclote study site in 1999, all cages deployed for our year 2000 planting efforts were mounted on 20-cm-long PVC legs. Similarly, based on the results of the planting density study that we conducted at Anclote in 1999, we reduced our planting density to 200 scallops per cage at most sites (100 per cage at our Tampa Bay sites).

Tampa Bay

We planted 100 scallops in each of five cages at each of our 10 restoration stations in Tampa Bay, and we planted an additional three cages containing 100 scallops each at Station 5 for provision of reproductive samples. At the time of planting, mean SH ranged from a minimum of 32.1 mm at Station 2 to a maximum of 33.9 mm at Station 3 but did not differ significantly among stations (Nested ANOVA, F = 1.02, P = 0.4205, n = 50 scallops per station). We monitored the growth, mortality and reproductive development of those scallops beginning on August 21, 2000 and continuing through July 3, 2001.

Anclote

We planted 200 scallops in each of five cages at each of our 10 restoration stations at the Anclote study site, and we planted an additional cage containing 200 scallops at Station 5 for provision of reproductive samples. At the time of planting, mean SH differed significantly among stations (Nested ANOVA, F = 71.12, P < 0.0001, n = 100 scallops per station) and ranged from a minimum of 22.6 mm at Station 10 to a maximum of 30.5 mm at Station 8. We monitored the growth, mortality and reproductive development of those scallops beginning on August 22, 2000 and continuing through August 30, 2001.

Homosassa

We planted 200 scallops in each of five cages at each of our 10 restoration stations at the Homosassa study site, and we planted an additional cage containing 200 scallops at Station 5 for provision of reproductive samples. At the time of planting, mean SH differed significantly among stations (Nested ANOVA, F = 27.78, P < 0.0001, n = 150 scallops per station) and ranged from a minimum of 28.2 mm at Station 2 to a maximum of 33.1 mm at Station 9. We monitored the growth, mortality and reproductive development of those scallops beginning on July 25, 2000 and continuing through July 17, 2001.

Crystal River

We planted 200 scallops in each of five cages at each of our 10 restoration stations at the Crystal River study site, and we planted an additional cage containing 200 scallops at Station 5 for provision of reproductive samples. At the time of planting, mean SH differed significantly among stations (Nested ANOVA, F = 34.39, P < 0.0001, n = 150 scallops per station) and ranged from a minimum of 27.5 mm at Station 10 to a maximum of 33.2 mm at Station 6. We monitored the growth, mortality and reproductive development of those scallops beginning on July 24, 2000 and continuing through July 19, 2001.

RESULTS

Year One-1998 Planting

Tampa Bay

Because of the confounding of mortality with the theft of scallops planted in mid-June, we report mortality on the July 1 sample date only for the scallops planted at original Stations 1 and 3. The reported mortality would have been much greater if we had included the loss of scallops from Stations 2, 4 and 5. Nevertheless, mortality was substantial throughout the summer but especially in July (Fig. 2A). From the starting value of approximately 92% survival at the Indian Key stations (new Stations 1 and 2) at the beginning of July, mortality increased to approximately 68% at the remaining nine stations by August 20, 1998. Although we detected no significant change in the number of dead scallops between August 20 and September 4, we recorded a significant increase in scallop mortality between September 4 and September 18; by September 18, approximately 90% of the scallops were dead. Subsequent mortality was minimal, and by February 8, 1999, approximately 4% of the scallops we originally planted remained alive.

[FIGURE 2 OMITTED]

Shell growth of scallops planted in Tampa Bay during 1998 was steady throughout the summer and fall. Scallops originally planted at ~30 mm SH during June and July had increased in mean SH by almost every sample date between August and October (Fig. 2A). Following our October 2 sample date, growth rate leveled off, and mean scallop SH remained in the 50-55 mm range for the remainder of the study. Shell height of naturally occurring scallops in Florida waters usually peaks at about 60-65 mm (Barber & Blake 1983), so the scallops that we grew in our Tampa Bay cages remained about 10 mm smaller than their natural counterparts.

The timing of reproductive development and spawning of the planted scallops also appeared to have occurred several months later than that of their naturally occurring conspecifics (Fig. 3A). Scallops from the natural population inhabiting Florida waters generally initiate gonadal development in late spring, are ripe by September, and initiate spawning in October and November (Barber & Blake 1983). Scallops planted in cages in Tampa Bay during 1998 also appear to have experienced gonadal development during spring and early summer, but most of the scallops that we sampled did not become ripe and initiate spawning until late fall or early winter. We collected ripe animals as late as February 1999. We estimate that approximately 10% (870) of the scallops that we originally planted in Tampa Bay remained alive to spawn during late fall and early winter and that at least 70% (610) of those scallops spawned.

[FIGURE 3 OMITTED]

Homosassa

The scallops that we planted at Homosassa in 1998 were not sampled as frequently as those planted in Tampa Bay. We sampled to estimate survival and reproductive condition on three dates postplanting, and we collected shell measurements as an estimate of growth on two dates postplanting. Fewer than 25% of the scallops that we planted in early July had died by August 12, but by October 30, almost 75% of the caged scallops were dead (Fig. 2B). The mortality rate decreased considerably between October 30, 1998 and January 21, 1999, at which time approximately 16.5% of the originally planted scallops remained alive.

Scallops planted in Homosassa grew more slowly than those planted in Tampa Bay and achieved a final mean SH of only 43.3 mm on January 21, 1999 (Fig. 2B). The delayed growth pattern recorded for Homosassa scallops appeared to have influenced the timing of gonadal development (Fig. 3B). Approximately 40% of the scallops we sampled in September were still developing, and the remainder had not yet initiated gonad development. Even in January, some scallops remained in the developing or ripe stage. Based on 25% of the scallops surviving until October and 80% of those surviving scallops spawning at that time, we conservatively estimate that 500 of the scallops planted in Homosassa (excluding those planted in the extra cage) spawned during October 1998.

Aquaculture Raft

Survival of scallops planted on the aquaculture raft was much better than we recorded for scallops planted at our Homosassa study site during 1998. Scallop mortality on the raft was essentially nil through July 22, and as late as October 9 we had recorded only 5.5% mortality. By October 30, 22.5% of the scallops originally planted on the aquaculture raft were dead, substantially below the 73% cumulative mortality we recorded at the Homosassa site on that same date.

Shell growth of scallops planted in off-bottom culture on the aquaculture raft at Crystal River was similar to that recorded for scallops grown in bottom-planted cages. Scallops planted at 21 mm SH on June 11 reached 30 mm SH by July 15 and 40 mm SH by October 2 (Fig. 2C). Those scallops were ~7 mm larger than scallops grown at our Homosassa study site and sampled 10 days later. However, the scallops on the aquaculture raft showed little subsequent growth, so that by the October 30 termination of the raft study, the mean SH for those scallops had increased only another 0.6 mm.

Most of the scallops that we collected from the aquaculture raft on August 18 were either developing or ripe, but only a small percentage had initiated spawning (Fig. 3C). By the time of our September 15 collection date, the majority of scallops were spawning. Scallops collected on October 29 were evenly split between a ripe and early-spawning condition. This may be indicative of gonadal redevelopment but is more likely a reflection of the range of variability in gonadal development among scallops during the spawning season.

Year Two-1999 Planting

Anclote

The Anclote 1999 scallop cohort suffered considerable mortality during the first summer and fall following planting (Fig. 4A), and by October 6, 1999, 79% of the scallops were dead. Mortality rate decreased substantially following the October 6 sampling, and by January 6, 2000, we had lost only an additional 2.5% of the planted scallops. On average, ~60 of the original 300 scallops that we planted in each cage during May 1999 survived the fall 1999 spawning season. Because the remaining scallops appeared to be functioning quite well, we decided to leave them in the cages for as long as there were survivors. Those remaining scallops suffered an average of 1% mortality per sampling episode between January 2000 and April 2001, and approximately 1.5% of the scallops that we planted at the Anclote study site in May 1999 remained alive in April 2001.

[FIGURE 4 OMITTED]

Scallops planted at Anclote during 1999 showed steady growth throughout the May 1999 to April 2001 deployment period (Fig. 4A). Scallops planted at a mean SH of 28.5 mm during May 1999 had achieved a mean SH of 40.2 mm by October 6, 1999. Growth slowed during the winter and early spring, and by March 2000 mean SH averaged 42.6 mm. Mean SH increased to ~49 mm by June 2000, the second June for these scallops. That shell size is more typical of naturally occurring scallops experiencing their first June (e.g., Barber & Blake 1983), although naturally occurring scallops rarely survive to a second June (Rhodes 1991), especially in Florida waters (Barber & Blake 1983). Ultimately, the scallops that we planted in Anclote during spring 1999 and that survived until spring 2001 achieved a SH of ~61 mm at an age of 2.5 y.

Consistent with the delayed shell growth of scallops planted at the Anclote study site, the reproductive cycle of the planted scallops was also delayed. Throughout fall 1999, most planted scallops remained in an undifferentiated or developing stage (Fig. 5A), and the majority of scallops did not spawn until late March 2000. At that time, approximately 10% (1500) of the scallops that we originally planted at the Anclote study site remained alive. Those scallops appeared to have continued to spawn (or to have "spent" gonads) throughout the remainder of the summer and into the early fall. Although few animals remained alive in fall 2000, those animals that did remain alive apparently initiated a second cycle of spawning activity during late fall and early winter.

[FIGURE 5 OMITTED]

Homosassa

By the second sampling date (May 20, 1999) after the deployment of scallops at the Homosassa study site, we had lost only 12% of the 15,300 scallops that we originally planted (Fig. 4B). The rate of mortality increased substantially during the summer months, however, and by August 18, 1999 almost 97% of the scallops had died. Thus, at the beginning of the spawning season ~500 caged scallops remained alive at our Homosassa study site. That number of survivors was essentially maintained until the study was terminated on December 20, 1999.

Scallops planted at the Homosassa study site in 1999 showed slow shell growth, especially from the time of planting until late September (Fig. 4B). From an initial mean SH of 26.9 mm in March, the scallops grew only to 29.8 mm SH by September 29. Between September 29 and the December 20 termination date, scallops grew to a size of 35.5 mm SH. Therefore, for the 6-month period spanning late spring to early fall, mean SH of the planted scallops increased by approximately 11%, whereas for the 3-month period spanning the fall and early winter, mean SH of the planted scallops increased by approximately 19%. Nevertheless, by the end of the fall spawning season, scallops planted at Homosassa during 1999 averaged only 35.5 mm SH, a size more typical of scallops harvested in May from the natural population (Barber & Blake 1983).

The scallops that did survive the summer at the Homosassa study site appear to have followed a pattern of gonadal development and spawning that is typical of natural scallop populations occupying Florida coastal waters (Fig. 5B). Most scallops that we sampled from our cages had gonads that were either undifferentiated or developing during the summer, became ripe and were spawning by late September and were predominately spent by December 20. We therefore estimate that ~500 of the scallops that we originally planted in March 1999 had successfully spawned by December 20, 1999.

Crystal River

Mortality of scallops planted in cages at Crystal River in March 1999 was low through May 19 (Fig. 4C). Mortality increased substantially during the summer months, and by August 18, almost 75% of the scallops that we originally planted at Crystal River were dead. The mortality rate decreased between August 18 and January 31, 2000, and by the latter date, more than 20% of the scallops remained alive. Mortality increased during the spring of 2000, but approximately 12% of the scallops that we originally planted remained alive on April 24, 2000. We finally terminated the Crystal River 1999 planting on October 26, 2000, although approximately 0.5% of those second-year scallops remained alive.

From an initial planting size of 25.9 mm SH on March 29, 1999, the scallops planted in cages at Crystal River grew rapidly and achieved a mean SH of 39.6 mm by September 30. Between the end of September and the end of January 2000, no significant increase in mean SH was detected (Fig. 4C). Mean SH increased gradually throughout the spring and summer of 2000 and reached a peak in September just before the termination of the 1999 Crystal River planting. At that time, mean SH was ~50 mm, which is still about 10 mm less than would be expected for first-year wild scallops during fall.

Gonadal development and the timing of spawning at Crystal River appears to have followed a pattern typical of naturally occurring scallops in Florida waters (Fig. 5C). During spring 1999, scallop gonads were either undifferentiated or developing. During August, we detected the first animals that had achieved a ripe condition, and by September all of the scallops that we sampled were either ripe or were spawning. Spawning appears to have continued throughout the fall and early winter. By March 2000, the scallops were either spent or had initiated redevelopment of the gonad. The scallops continued to spawn, and apparently to redevelop, throughout summer and fall of 2000. We conservatively estimate that 20% (3000) of the scallops that we originally planted survived through fall 1999 and that at least 90% (2700) of those scallops spawned successfully. Additionally, it appears that a small percentage of those scallops survived until fall 2000 and spawned at least one more time.

Planting Density Study

On our first sample date (August 25, 1999) after the initial planting, scallop mortality was highest in the high density treatment and lowest in the low density treatment, but the number of survivors differed significantly among treatments (Fig. 6A). By the October 7 sample date, the 300-scallop treatment averaged only 127 scallops per cage (58% mortality), the 150-scallop treatment averaged 94 scallops per cage (37% mortality), and the 50-scallop treatment averaged 46 scallops per cage (7% mortality), although those differences in percent mortality were not significant (REGW MRT, P = 0.05). We recorded data from only two of the three replicate 300-scallop cages on November 18 (n = 3 for the other two treatment combinations), but differences in mean scallop abundance among the three density treatments remained nonsignificant. Nevertheless, planting at higher density (at least 150 scallops per cage) resulted in more scallops alive during the fall spawning season than was realized in the 50-scallop treatment, even though the rate of mortality in the two high-density treatments was greater than that recorded for the 50-scallop treatment. Note that subsequent survival data are confounded by collection of five scallops from each cage for analysis of reproductive stage. We collected on five dates (Fig. 7), so we collected 25 scallops from each cage. That data represents a harvest of 8.3% of the scallops originally planted in the 300-scallop treatment, 16.7% of the scallops originally planted in the 150-scallop treatment and 50% of the scallops originally planted in the 50-scallop treatment.

[FIGURES 6-7 OMITTED]

Growth rate differed significantly between treatments on most dates (Fig. 6B). On the initial planting date, scallops planted into the 150-scallop treatment were significantly larger than were the scallops planted in the 50-scallop and 300-scallop treatments. However, by the August 25 sample date, we detected no significant difference in SH between the 50-scallop and 150-scallop treatments, although scallops planted at each of those densities were significantly larger than were the scallops planted at 300 per cage. That trend continued until the final (July 5, 2000) sample date, at which time we detected no significant difference in SH among any of the treatment groups. For all sample dates after November 18, differences in growth rate among treatments may have been influenced by the differential removal of scallops from cages for reproductive analyses.

We noted few differences in the pattern of gonadal development among the various density treatments. We first collected samples for gonadal analysis on November 22, 1999, and at that time, all of the scallops that we sampled were in some stage of spawning or were spent (Fig. 7). On subsequent dates, we noted an increase in the number of scallops in an undifferentiated or developing stage, but there were few differences in the distribution of developmental stages among treatments.

Planting Habitat Study

The survival of scallops planted outside of the seagrass bed was not significantly different from the survival of scallops planted in cages on legs within the seagrass bed, regardless of whether the cage was deployed directly on the substrate or on PVC legs (Fig. 8A). However, on several sampling dates we detected significantly lower survival of scallops planted in cages within the seagrass bed directly on the substrate than in any other treatment. An average of fewer than 20 of the scallops deployed in cages planted directly on the substrate in the seagrass bed survived until the end of the study, whereas an average of more than 30 scallops per cage survived in the other treatments.

[FIGURE 8 OMITTED]

Because the scallops used for this study were randomly selected from a holding cage deployed at the study site, we did not obtain their sizes when they were initially planted. Instead, our initial size estimates were obtained on the first sample date (October 11, 1999), and at that time, a significant size hierarchy was already established (Fig. 8B). Scallops planted in cages deployed on legs outside of the seagrass bed were generally larger than scallops planted in any other treatment combination at the beginning of the experiment and at the end. Scallops planted in cages deployed directly on the substrate inside the seagrass bed were significantly smaller than scallops planted in any other treatment combination, and that relationship held throughout the duration of the experiment. However, the rate of growth did not significantly differ among treatments (ANCOVA, F = 1.29, P = 0.2757), indicating that the significant differences in SH reflected initial planting size or growth differences realized during the first few weeks after planting (September 3, 1999 to October 11, 1999).

Year Three-2000 Planting

Tampa Bay

As has been typical of our scallop plantings at every site during the summer of each planting year, the rate of mortality was high in Tampa Bay during summer 2000. By August 21, we had lost only 17% of the scallops that we originally planted on July 7 (Fig. 9A), but 68% of the scallops were dead by September 14. Mortality continued at a lesser pace throughout the fall and leveled off during January and February, but we observed a significant increase in mortality during spring. By May 2001, almost 97% of the scallops that we originally planted were dead, and the experiment was terminated on July 3, 2001.

[FIGURE 9 OMITTED]

After our initial planting, scallops grew rapidly and reached a SH of 51.3 mm by August 21, 2000 (Fig. 9A). We also recorded significant increases in SH of scallops collected on the October 5 and November 14 sample dates relative to those collected on the previous sample date, although those increases were much less than the increase observed between the initial planting date and the August 21 sample date. Nevertheless, by September, the scallops that we planted in Tampa Bay had achieved a SH of almost 57 mm, a size typical of naturally occurring scallops collected during the same time of year from Florida coastal waters (Barber & Blake 1983). The surviving scallops continued to grow throughout the winter and spring, albeit at a slow rate, and they ultimately achieved a SH of-61 mm by May 2001.

Scallops planted in Tampa Bay during 2000 were ripe and spawning by August 21 (Fig. 10A). By early October most of the scallops were either redeveloping, ripe or again in some stage of spawning, and that pattern continued throughout the winter and spring. We estimate that ~3,700 of the scallops that we planted in Tampa Bay during 2000 spawned during August 2000 and that many of those scallops continued to spawn throughout the remainder of 2000 and into 2001.

[FIGURE 10 OMITTED]

Anclote

Initial mortality was severe at the Anclote study site, and by October 11, 2000, approximately 82% of the scallops that we originally planted on July 13 were dead (Fig. 9B). The rate of mortality decreased substantially after our October 11 inspection and we detected no significant increase in mortality until April 24, 2001. We estimate that ~1,750 of the 10,000 scallops that we originally planted remained alive on October 11 and that ~1,100 of those scallops remained alive on April 24, 2001.

Shell growth of the scallops that we planted at Anclote in 2000 was continuous and relatively rapid throughout the summer and fall of 2000 and the winter of 2001 (Fig. 9B). Mean SH of scallops increased significantly on every sample date from planting through March 8, 2001, although no significant increase in SH was subsequently recorded. During the growth period, mean SH of scallops increased 3-5 mm per sample date. However, the maximum SH that the scallops achieved was ~46 mm, so even at their peak SH, these scallops remained 10-15 mm smaller than their natural conspecifics.

None of the scallops planted at Anclote in 2000 showed any evidence of reproductive development when sampled on August 22, but by October 11, approximately 90% of the scallops were spawning and the remainder were ripe (Fig. 10B). The scallops appear to have continued a cycle of spawning and gonad redevelopment throughout the remainder of the deployment. Even in August 2001, a substantial number of the surviving scallops were ripe or spawning, and most of the remainder were in a spent condition. We estimate that by October 11, 2000, ~90% (1,575) of the 1,750 remaining scallops were spawning. Spawning activity appears to have continued at least through April 2001, when 1,100 scallops remained alive.

Homosassa

Scallop mortality was typically high during the first few months after deployment (Fig. 9C). By September 11, 2000, 72% of the scallops that we originally planted were dead, but we detected no significant increase in the number of dead scallops from that date to May 21, 2001. By February 5, 2001, approximately 20% (2,000) of the 10,000 scallops that we originally planted remained alive.

Shell growth of the scallops planted at Homosassa was slow relative to what would be expected from the natural population, but was steady throughout the fall and winter. We detected a significant increase in SH on every sampling date from planting through February 5, 2001 (Fig. 9C). At that time, the scallops had achieved a SH of ~45 mm and they maintained that size throughout the remainder of their life.

Gonadal development and spawning were delayed in the caged Homosassa scallops relative to what would be expected in the natural population. Many scallops remained in the developing or ripe stage in October (Fig. 10C), and less than 40% of the scallops had initiated spawning by December 11. The proportion of scallops that were spawning increased steadily during the winter, and by May 21 all of the scallops that we sampled from the cages were reproductively spent. We estimate that ~2,000 scallops remained alive in cages at the Homosassa study site on February 5 and that approximately 45% of those scallops (900) had spawned by that date. Additionally, approximately 15% (1500) of the scallops that we originally planted at Homosassa remained alive on May 21, and 100% of those scallops had spawned by that date.

Crystal River

Considering the high rate of mortality recorded for most plantings during this study, mortality of scallops planted in cages at the Crystal River study site during 2000 was remarkably low (Fig. 9D). By February 6, 2001, 50% of the scallops that we originally planted at this study site remained alive. We did not record a significant increase in the number of dead scallops until April 3, and even then more than 40% of the scallops that we originally planted remained alive. However, mortality was severe during summer 2001 and was essentially complete by July 19, 2001.

Despite or perhaps because of the low rate of mortality experienced by caged scallops at the Crystal River study site, shell growth was slow and essentially nonsignificant from sample date to sample date (Fig. 9D). Scallops planted at a mean SH of ~30 mm achieved a SH of only 42 mm by May 22, 2001. That is approximately the size that would be expected of first-year scallops on that same date.

Gonadal development and spawning of the scallops planted at Crystal River appear to have proceeded in a normal fashion. All of the scallops that we sampled on July 25, 2000, were immature, but by September most were developing and by October most were ripe (Fig. 10D). The scallops were in some stage of spawning by our December 12 sampling, and during the winter months they showed continued spawning and redevelopment of gonads. By May 21, 2001, all remaining scallops were spent. We estimate that approximately 52% (5,200) of the scallops that we originally planted at the Crystal River study site remained alive on December 12, and 100% of those scallops had spawned by that date.

DISCUSSION

The bay scallop has become a popular target for restoration activities throughout the eastern United States, and restoration efforts have been ongoing since, at least, 1979 (Morgan et al. 1980). The methods used in the various efforts differ, but two basic approaches have been applied. The first approach involves culturing scallops in the laboratory and planting the resultant seed in the field under various conditions. The culture and field-growout techniques on which this approach is based were developed from the work of Loosanoff and colleagues (e.g., Loosanoff & Davis 1963), Sastry (1963), Castagna and colleagues (Castagna & Duggan 1971, Duggan 1973) and Rhodes and Widman (1980), but even those developments were predicated on previous published observations of bay scallop biology and ecology (e.g., Belding 1910, Gutsell 1930). The second approach involves harvesting scallops from the wild population and replanting them at a site targeted for enhancement. This approach circumvents the expensive culture phase but remains dependent on the availability of natural stocks, either of settling larvae or postset animals (e.g., Turner & Soares 1994, Peterson et al. 1996). Moreover, when transplanting postset animals, conflict may arise among alternative user-groups because one scallop population is being depleted to some degree in an effort to restore another population.

Planting of cultured scallops has been the most popular approach. The first such bay scallop restoration effort that we are aware of was reported by Morgan et al. (1980). Those authors released almost 50,000 cultured scallops of various sizes into an eelgrass bed near Waterford, Connecticut. Survival of the seeded scallops was size dependent: most scallops less than 22 mm SH disappeared within 1 day, although even scallops larger than 45 mm SH generally survived for only a few weeks at most. Similar results have been reported for other restoration efforts that relied on planting cultured scallops into the natural environment without some sort of protection from predators and constraint to dispersal (e.g., Tettelbach & Wenczel 1993). However, protection is of more or less importance depending on the size of the scallops that are transplanted. Whereas Tettelbach and Wenczel (1993) reported generally poor survival of 20-mm seed scallops planted without protection into Long Island, New York estuaries, Peterson et al. (1996) reported negligible mortality of adult bay scallops transplanted from Back Sound to Bogue Sound in coastal North Carolina.

Replanting scallops harvested from the natural population was conducted in the Westport River, Massachusetts, using an approach that included juvenile and adult scallops (Turner & Soares 1994). In the first phase of this study, adult scallops were transplanted from various sites into floating rafts, where the scallops were allowed to grow and spawn in the hope that their offspring would replenish nearby suitable habitat. Those researchers also deployed recruit collectors in the vicinity of the floating rafts (Tammi et al. 1994). Juvenile scallops harvested from the collectors were then planted into similar floating rafts and allowed to grow to adult size and ultimately spawn. Quantitative estimates of success were not reported, but the authors mentioned that the efforts "culminated in a substantial harvest" in 1996 (Tammi & Turner 1998). In North Carolina, Peterson et al. (1996) transplanted several hundred thousand adult bay scallops between coastal sounds during 1992 to 1994 as a means of increasing local spawner density in an otherwise collapsed population. Pre versus posttransplant monitoring of recruitment at the receiving site indicated a 568% increase in the posttransplant rate of recruitment relative to the recruitment rate measured at that site prior to transplantation.

Restoration of bay scallop populations is only practical if the conditions that originally led to the demise of the population have been ameliorated. For the North Carolina study, it seems that the loss of scallops from the Back Sound study site followed a localized harmful algal bloom (HAB) event that instigated recruitment failure in the local population (Summerson & Peterson 1990). That event seems to have been non destructive to scallop habitat as attested by the rapid recovery of scallop populations in nearby sounds that were also affected by this HAB event (Peterson & Summerson 1992). By restoring the local scallop population in this otherwise hydrologically isolated embayment, it proved possible to reestablish adequate larval supply and subsequent recruitment (Peterson et al. 1996). In contrast, a persistent HAB composed of Aureococcus anophagefferens Hargraves et Sieburth has decimated scallop populations in the Peconic Bay system of Long Island, New York. There, restoration efforts have not yet been successful (Tettelbach & Wenczel 1993) because the scallops appear incapable of overcoming these HAB events.

In Florida, various factors contributed to the loss of scallop populations, and those factors seem to have differed from site to site. For example, in the Pine Island Sound area, channelization of the Caloosahatchee River and construction of a causeway to Sanibel Island may have contributed to the collapse of that population by redirecting freshwater flow onto the scallop beds. Loss of the Tampa Bay scallop population can be attributed at least in part to a greater than 80% reduction (Haddad & Harris 1985) in essential seagrass habitat (Thayer & Stuart 1974). In the pristine coastal waters between Anclote and Crystal River, where considerable seagrass habitat remains, local reductions in the density of adult scallops coupled with a lack of allochthonous larval inputs (caused by the collapse of neighboring scallop populations such as Tampa Bay) appears to have inhibited population regrowth despite the presence of suitable habitat patches (Arnold et al. 1998, Arnold, 2001).

Because of the different factors that have contributed to population collapse of bay scallops in various Florida estuaries, the success of a restoration effort may be site-specific. For example, it will be difficult to restore scallop populations in areas such as Pine Island Sound and Sarasota Bay, where red tide (Karenia brevis [Davis] Hanson and Moestrup) blooms continue to be a common feature of the seascape. In contrast, water quality has improved in Tampa Bay to the point where scallop restoration may be feasible in at least some areas (Blake 1996, Blake 1998), and in the Anclote-Crystal River area habitat does not seem to be an issue. Because scallop stocks are so depleted in peninsular Florida, a source for the large numbers of scallops needed to supply a transplant operation is not available. We therefore chose to implement the two-stage, culture-based approach described herein. The few adult scallops that we were able to harvest from the natal population were used as broodstock with which to create large numbers of seed scallops that were then planted in each of the targeted restoration sites to serve as parent stock for future generations (Blake 1998).

Our field planting methods were influenced by aquaculture rules and regulations in Florida. To minimize interference with navigation, only near-bottom aquaculture is allowed in Florida coastal waters. We were therefore prevented from using techniques such as pearl nets hung from longlines, stacked cages or other approaches that extend more than ~0.5 m above the sediment-water interface. The only exception to this rule pertained to the aquaculture raft in Crystal River, which was anchored within an aquaculture lease site and was permitted by a management use agreement that allowed access to the entire water column. Otherwise, we were required to limit our cage structures to the near-bottom zone. At the initiation of the study, we did not consider this to be a significant impediment to success because the bay scallop generally is a benthic organism, but our results indicate that scallops planted higher in the water column survived better than did their bottom-dwelling counterparts, even though growth rates were similar. Duggan (1973) also reported higher mortality of caged scallops grown within 1 m of the bottom than of those grown at a height 2 m above the bottom.

The growth of planted scallops lagged behind that of their naturally occurring counterparts during most experiments. Naturally occurring scallops in Florida coastal waters generally achieve 50 mm SH by the first June following birth and achieve a maximum SH of ~60 mm during September (Barber & Blake 1983). Only in Tampa Bay did our planted scallops exceed 50 mm SH during the first fall after planting. Scallops that we planted at Anclote in spring 1999 exceeded 50 mm SH, but not until the fall of the following year when they were almost 2 years of age. This is not surprising considering that few scallop restoration efforts have reported terminal scallop size exceeding 50 mm SH. Castagna and Duggan (1971) reported that a mean SH of 50-55 mm was achieved during late fall for scallops held in floats (mesh-lined boxes), but that height appears to have been asymptotic. Those same authors reported that scallops achieved a larger size when grown in pens rather than floats (57.4 mm SH in November), but it is not clear whether that was a mean or maximum size (Castagna & Duggan 1971). In most other studies involving the growth of cultured scallops under confined conditions in open waters, terminal SH does not exceed 50 mm (e.g., Duggan 1973, Morgan et al. 1980, Heffernan et al. 1988, Walker et al. 1991, Oesterling & DuPaul 1994, Smith & Tettelbach 1996, Goldberg et al. 2000). For studies conducted in the northeastern United States (e.g. Morgan et al. 1980, Smith & Tettelbach 1996, Goldberg et al. 2000), the terminal SH reported by the authors represents the SH achieved at the end of the first growing season prior to the onset of winter. Naturally occurring scallops in northeastern waters also reach ~50 mm SH during fall of their first year, but wild and cultured scallops gain an additional 10 mm or more in SH during their second year (Hickey 1977, Bricelj et al. 1987, Goldberg et al. 2000). These results suggest that in the northeastern United States, cultured scallops planted into local waters grow at a rate similar to their natural conspecifics under most conditions, regardless of whether they are free-planted or planted under confinement. In southern locales, SH at the end of the first growing season essentially represents maximum SH. Even the small proportion of scallops that live through the winter do not appear to reinitiate shell growth as waters warm during the following spring (Kirby-Smith 1970, this study). Thus, in southeastern waters, the growth rate and terminal SH of cultured scallops lag behind that of naturally occurring scallops under most field-growout conditions.

The timing of reproductive development for our planted scallops varied from site to site, even among years within sites and does not seem to be coincident with planting date or the size of the scallops at planting. In Tampa Bay during the first study year, we detected a small percentage (~10%) of spawning scallops on our July, August, October and November sample dates, but we did not detect a substantial spawning event until December. From December through March, spawning was essentially continuous. In contrast, more than 80% of the scallops planted in Homosassa during the first year were spawning in October, despite the fact that these scallops were planted later and at a smaller size than those planted in Tampa Bay. Scallops planted on the aquaculture raft also showed evidence of considerable spawning activity during their first fall, and it seems that these scallops initiated spawning earlier than did those planted at the nearby Homosassa site during the same year. Scallops planted in Homosassa and Crystal River during the second year of the study appear to have developed and spawned during late fall, consistent with what has been reported for feral scallops in the eastern Gulf of Mexico (Barber & Blake 1983), whereas those planted at Anclote showed no evidence of spawning until January 2000 and do not appear to have experienced a major spawn until March. The Anclote scallops were not planted until May 1999 but they were planted at a large (28.5 mm SH) average size. We have some evidence that a second cohort of scallops may exist in west Florida waters at least during some years (Arnold, unpub, data), and off-peak spawning has been reported for bay scallops from New York waters (Tettelbach et al. 1999). Although such off-peak spawning may not match in magnitude the effort realized from a peak spawning event, the recruits contributed from an off-peak spawn may be important for long-term population maintenance (Moyer & Blake 1986). During the third study year, scallops in Tampa Bay initiated spawning in August 2000 and continued spawning through May 2001. In contrast, scallops planted at Anclote, Homosassa and Crystal River had a spawning pattern typical of wild conspecifics. The early and continuous spawning observed in Tampa Bay scallops during 2000 is difficult to explain, especially relative to the scallops planted that same year at the other sites. Tampa Bay scallops were planted on about the same date as the Anclote scallops and at about the same size as the Homosassa and Crystal River scallops. The parental groups did differ, potentially implicating a genetic basis for the difference in spawning pattern among sites. If that is the case, it is not a consistent genetic difference because Tampa Bay offspring planted in 1998 spawned later than we expected based on literature reports (Barber & Blake 1983).

Mortality rates were high in our cages during most experiments, especially from July through September when water temperature peaks in Florida. Castagna and Duggan (1971) also reported maximum mortality rates during July-September in Virginia waters and attributed that high mortality to senescence and overcrowding in their enclosures. Virginia populations of bay scallops spawn earlier in the summer relative to Florida populations, so the high late-summer mortality of the scallops we planted was more likely caused by stress associated with the shifting of stored energy reserves to gametogenesis (Barber & Blake 1981). Although our overall mortality rates were high, they were comparable to mortality rates reported for many other scallop planting efforts. Walker et al. (1991) reported a mortality rate exceeding 90% in 140 days for scallops planted at one site near Skidaway Island, Georgia. Complete mortality of caged scallops has been reported from studies conducted in Long Island (Tettelbach & Wenczel 1993) and Florida (Blake 1998) waters, although the causative factors seem to have differed. In the Tettelbach and Wenczel (1993) study, mortality was attributed to starfish predation and cold water, whereas Blake (1998) attributed instances of 100% mortality to Karenia brevis exposure. Mortality rates were further exacerbated in our study because we are working at what is essentially the southern distributional limit of the bay scallop (Waller 1969). In Florida Gulf of Mexico waters, the cost of reproduction is higher for bay scallops than for other molluscs (Barber & Blake 1986), and gonadal development requires the depletion of energy reserves during the summer months as first lipid from the digestive gland and then carbohydrate from the adductor muscle is routed to egg production (Barber & Blake 1981, Barber & Blake 1985). This combination of high water temperature (>30[degrees]C) and reduced energy reserves in bay scallops inhabiting Florida coastal waters during summer increases the susceptibility of this organism to stress-related mortality.

Summer mortality seems to be related to planting density. Scallops planted into cages during June 1999 experienced higher rates of mortality when planted at higher density. Between June and September, the mortality rate was significantly greater for scallops planted at 300 per cage (equivalent to 833 scallops per [m.sup.2]) than for scallops planted at either 150 (417 [m.sup.-2]) or 50 (139 [m.sup.-2]) per cage, and scallops planted at 150 per cage suffered significantly greater mortality than those planted at 50 per cage. By September 1999, and for the duration of the density study, there were no additional differences in mortality rate despite the fact that density in the 150-and 300-scallop treatments remained more than twice the density of scallops remaining in the 50-scallop treatment. Scallops in the lower-density treatments also remained larger than those in the 300-scallop treatment on most sample dates. Moreover, the summer mortality rates we recorded were generally high despite the fact that we planted at lower densities than the 1,000-10,000 scallops per [m.sup.2] reported from previous studies (e.g., Rhodes & Widman 1980, Karney 1991, Walker et al. 1991, Oesterling & DuPaul 1994).

For our approach to bay scallop restoration, the currency of concern is gamete production because we are using an indirect rather than direct strategy to reinvigorate scallop populations. At least during the first 2 days of life, larger scallop eggs survive better than smaller eggs (Kraeuter et al. 1982), but we have no evidence that the eggs produced by our planted scallops were on average smaller than those produced by wild scallops. The number of eggs produced by a bay scallop does seem to be proportional to the size of the scallop (Orensanz et al. 1991), so our planted scallops were probably not optimal producers during any single spawning event. However, the slow growth of our planted scallops may have been beneficial for overall egg production, because the planted scallops appear to have survived and been reproductively active for a much longer period of time than their wild conspecifics. For scallops, senescence seems to be size- rather than age-related (Gutsell 1930, Orensanz et al. 1991), so the slower growth of the caged scallops may have worked to the advantage of our overall goal of increasing egg production and, ultimately, larval supply to depleted bay scallop habitat.

In summary, we have developed techniques to capture adult bay scallops from preselected sites in Florida Gulf of Mexico nearshore waters, to spawn those animals in a laboratory setting, to successfully raise their offspring in a nursery setting (Blake 1998) and to plant the resultant offspring in cages and raise them to adulthood. Our approach attempts to take full advantage of the reproductive potential of adult bay scallops. Rather than directly releasing large numbers of cultured scallops into suitable habitats, we implemented a two-stage approach that involves planting juvenile scallops in protective cages and allowing them to grow to adulthood, at which time they spawn at high density. The resultant offspring, dispersing in much greater numbers than we could achieve by any production strategy, then settle to coastal seagrass habitats and thereby contribute to rebuilding the local population. Our efforts to assess the success of this restoration strategy will be reported in future publications.

ACKNOWLEDGMENTS

Technical support was essential to all aspects of this project, and we extend our appreciation for the many contributions of the following people: Candice Way and Chris Canning were indispensable, overseeing all aspects of bay scallop spawning and nursery culture; Frank Courtney, Buck Dennis, Bill Halstead, James Kipp, Bill Sessler, Greg Vermeer and Chris Young assisted with nursery culture and with the safe and efficient transport of scallops from the nursery to the field; field sampling support was provided by Joel Bickford, Keri Ferenc, Lisa Gentile, Kate Hagner, Philip Hoffman, Micah Humphrey, Tracy Idocks, Chris Matterson, Cindy Meyer and Winnie White; Janessa Cobb assisted with graphics; and Noretta Perry and Charlie Plybon assisted with histologic preparation of gonad samples. Gabriel Delgado, Judy Leiby, Jim Quinn and two anonymous reviewers provided valuable suggestions on earlier drafts of the manuscript. This study was funded in part by grant NA76FK0426 from the National Oceanic and Atmospheric Administration. The views expressed herein are those of the authors and do not necessarily reflect the views of NOAA or any of its sub-agencies.

LITERATURE CITED

Arnold, W. S. 2001. Bivalve enhancement and restoration strategies in Florida, USA Hydrobiologia 465:7-19.

Arnold, W. S. & D. C. Marelli. 1991. Assessment of bay scallop populations on the west coast of Florida. Internal Report, Florida Department of Environmental Protection, Florida Marine Research Institute. 19 pp.

Arnold, W. S., D. C. Marelli, C. P. Bray & M. M. Harrison. 1998. Recruitment of bay scallops Argopecten irradians in Floridian Gulf of Mexico waters: scales of coherence. Mar. Ecol. Prog. Ser. 170:143-157.

Arnold, W. S., D. C. Marelli, M. Parker, P. Hoffman, M. Frischer & J. Scarpa. 2002. Enhancing hard clam (Mercenaria spp.) population density in the Indian River Lagoon, Florida: a comparison of strategies to maintain the commercial fishery. J. Shellfish Res. 21:659-672.

Barber, B. J. & N. J. Blake. 1981. Energy storage and utilization in relation to gametogenesis in Argopecten irradians concentricus (Say). J. Exp. Mar. Biol. Ecol. 52:121-134.

Barber, B. J. & N. J. Blake. 1983. Growth and reproduction of the bay scallop, Argopecten irradians (Lamarck) at its southern distributional limit. J. Exp. Mar. Biol. Ecol. 66:247-256.

Barber, B. J. & N. J. Blake. 1985. Intra-organ biochemical transformations associated with oogenesis in the bay scallop, Argopecten irradians concentricus (Say), as indicated by [.sup.14]C incorporation. Biol. Bull. 168: 39-49.

Barber, B. J. & N. J. Blake. 1986. Reproductive effort and cost in the bay scallop, Argopecten irradians concentricus. Int. J. Invert. Repro. and Develop. 10:51-57.

Belding, D. L. 1910. A report on the scallop fishery of Massachusetts, including the habits, life history of Pecten irradians, its rate of growth, and other facts of economic value. Special Report of the Commonwealth of Massachusetts Commission on Fisheries and Game, Boston, MA. 150 pp.

Blake, N. J. 1996. Demonstration of large-scale reintroduction of the southern bay scallop to Tampa Bay, Florida. Final Report, Tampa Bay National Estuary Program. pp. 1-28.

Blake, N. J. 1998. The potential for reestablishing bay scallops to the estuaries of the west coast of Florida. Trans. 63rd N. Am. Wildl. Natur. Resour. Conf. 63:184-189.

Blake, N. J., Y. Lu & M. Moyer. 1993. Evaluation of Tampa Bay waters for the survival and growth of southern bay scallop larvae and juveniles. Final report, Tampa Bay National Estuary Program. Tech. Pub. 04-93. 18 pp.

Bologna, P. A. X. 1998. Growth, production, and reproduction in bay scallops Argopecten irradians concentricus (Say) from the northern Gulf of Mexico. J. Shellfish Res. 17:911-917.

Bricelj, V. M., J. Epp & R. E. Malouf. 1987. Comparative physiology of young and old cohorts of bay scallop Argopecten irradians irradians (Lamarck): mortality, growth, and oxygen consumption. J. Exp. Mar. Biol. Ecol. 112:73-91.

Castagna, M. & W. Duggan. 1971. Rearing the bay scallop, Aequipecten irradians. Proc. Natl. Shellfish. Assoc. 61:80-85.

Chan, K.-S., N. C. Stenseth, M. O. Kittilsen, J. Gjosaeter, K. Lekve, T. Smith, S. Tveite & D. Danielssen. 2003. Assessing the effectiveness of releasing cod larvae for stock improvement with monitoring data. Ecol. Appl. 13:3-22.

Cosper, E. M., W. C. Dennison, E. J. Carpenter, V. M. Bricelj, J. G. Mitchell, S. H. Kuenstner, D. Colflesh & M. Dewey. 1987. Recurrent and persistent brown tide blooms perturb coastal marine ecosystems. Estuaries 10:284-290.

Dao, J.-C., J. Barret & M. L. Muzellec. 1994. Seed-recapture of the scallop, Pecten maximus, by commercial dredging in the Bay of Brest during 1991 to 1992. In: N. F. Bourne, B. L. Bunting & L. D. Townsend, editors. Proceedings of the 9th international pectinid workshop, Nanaimo, B.C., Canada, April 22 to 27, 1993. Vol. 2. Can. Tech. Rep. Fish. Aquat. Sci. 1994. pp. 16-22.

Day, R. W. & G. P. Quinn. 1989. Comparisons of treatments after an analysis of variance in ecology. Ecol. Monogr. 59:433-463.

Duggan, W. P. 1973. Growth and survival of the bay scallop, Argopecten irradians, at various locations in the water column and at various densities. Proc. Natl. Shellfish. Assoc. 63:68-71.

Goldberg, R., J. Pereira & P. Clark. 2000. Strategies for enhancement of natural bay scallop, Argopecten irradians irradians, populations; a case study in the Niantic River estuary, Connecticut, USA. Aqua. Int. 8: 139-158.

Gutsell, J. S. 1930. Natural history of the bay scallop. Bull. U.S. Bureau of Fish 46:569-632.

Haddad, K. 1988. Habitat trends and fisheries in Tampa and Sarasota Bays. NOAA estuary of the month series # 11, Tampa and Sarasota Bays. pp. 113-128.

Haddad, K. D. & B. A. Harris. 1985. Assessment and trends of Florida's marine fisheries habitat: an integration of aerial photography and thematic mapper imagery. In: S. K. Mengel & D. B. Morrison, editors. Eleventh International Symposium, machine processing of remotely sensed data with special emphasis on quantifying global processes: models, sensor systems, and analytical methods. West Lafayette, Indiana: Purdue University laboratory for applications of remote sensing. pp. 130-138.

Hanski, I. 1991. Single-species metapopulation dynamics: concepts, models and observations. Biol. J. Linn. Soc. 42:17-38.

Hanski, I. & D. Simberloff. 1996. The metapopulation approach, its history, conceptual domain, and application to conservation. In: I. A. Hanski & M. E. Gilpin, editors. Metapopulation biology: ecology, genetics, and evolution. San Diego, CA: Academic Press. pp. 5-26.

Heffernan, P. B., R. L. Walker & D. M. Gillespie. 1988. Biological feasibility of growing the northern bay scallop, Argopecten irradians irradians (Lamarck, 1819) in coastal waters of Georgia. J. Shellfish Res. 7:83-88.

Hickey, M. T. 1977. Age, growth, reproduction and distribution of the bay scallop, Aequipecten irradians irradians (Lamarck), in three embayments in eastern Long Island, New York, as related to the fishery. M.S. thesis, C.W. Post College, Long Island University, New York. 101 pp.

Hoffman, W. E., M. J. Durako & R. R. Lewis, III. 1982. Habitat restoration in Tampa Bay. In: S.-A. F. Treat, J. L. Simon, R. R. Lewis III & R. L. Whitman Jr., editors. Proceedings Tampa Bay area scientific information symposium. Florida Sea Grant Report #65, pp. 636-657.

Jaramillio, R., J. Winter, J. Valencia & A. Rivera. 1993. Gametogenic cycle of the Chiloe scallop (Chlamys amandi). J. Shellfish Res. 12:59-64.

Karney, R. C. 1991. Ten years of scallop culture on Martha's Vineyard. In: S. E. Shumway & P. A. Sandifer, editors. An international compendium of scallop biology and culture: a tribute to James Mason. World Aquaculture Workshops, Number 1. pp. 308-312.

Kirby-Smith, W. W. 1970. Growth of the scallops, Argopecten irradians concentricus (Say) and Argopeeten gibbus (Linne), as influenced by food and temperature. M.S. thesis, Duke University, Durham, North Carolina. 127 pp.

Kraeuter, J. N., M. Castagna & R. van Dessel. 1982. Egg size and larval survival of Mercenaria mercenaria (L.) and Argopecten irradians (Lamarck). J. Exp. Mar. Biol. Ecol. 56:3-8.

Levins, R. 1969. Some demographic and genetic consequences of environmental heterogeneity for biological control. Bull. Entomological Soc. Am. 15:237-240.

Levitan, D. R. 1995. The ecology of fertilization in free-spawning invertebrates. In: L. McEdward, editor. Ecology of marine invertebrate larvae. Boca Raton, FL: CRC Press. pp. 123-156.

Li, Z. & R. H. Weisberg. 1999. West Florida shelf response to upwelling favorable wind forcing: kinematics. J. Geophys. Res. 104:13507-13527.

Loosanoff, V. L. 1937. Seasonal gonadal changes of adult clams, Venus mercenaria (L.). Biol. Bull. 72:406-416.

Loosanoff, V. L. & H. G. Davis. 1963. Rearing of bivalve mollusks. Advan. Mar. Biol. 1:1-136.

Lu, Y. & N. J. Blake. 1997. The culture of the southern bay scallop in Tampa Bay, an urban Florida estuary. Aqua. Int. 5:439-450.

Marelli, D. C., W. S. Arnold & C. Bray. 1999. Levels of recruitment and adult abundance in a collapsed population of bay scallops (Argopecten irradians) in Florida. J. Shellfish Res. 18:393-399.

Morgan, D. E., J. Goodsell, G. C. Matthiessen, J. Garey & P. Jacobson. 1980. Release of hatchery-reared bay scallops (Argopecten irradians) onto a shallow coastal bottom in Waterford, Connecticut. Proc. World Maric. Soc. 11:247-261.

Moyer, M. A. & N. J. Blake. 1986. Fluctuations in calico scallop production (Argopecten gibbus). Proc. 11th annual tropical and subtropical fisheries conference of the Americas. Texas agricultural extension service, Texas A & M University department of animal science, pp. 45-58.

Murdock, J. F. 1955. Investigation of the Lee County bay scallop fishery. The Marine Laboratory, University of Miami, Report 55-13. pp. 1-10.

Oesterling, M. J. & W. D. DuPaul. 1994. Shallow water bay scallop, Argopecten irradians, culture in Virginia. Proceedings of the 9th international pectinid workshop, Nanimo, BC., Canada, April 22-27, 1993, vol. 2. Can. Tech. Report of Fish. Aquatic Sci. pp. 58-63.

Orensanz, J. M., A. M. Parma & O. O. Iribarne. 1991. Population dynamics and management of natural stocks. In: S. E. Shumway, editor. Scallops: biology, ecology, and aquaculture. Developments in aquaculture and fisheries science, vol. 21. Amsterdam: Elsevier. pp. 625-713.

Peterson, C. H. & H. C. Summerson. 1992. Basin-scale coherence of population dynamics of an exploited marine invertebrate, the bay scallop: implications of recruitment limitation. Mar. Ecol. Prog. Ser. 90: 257-272.

Peterson, C. H., H. C. Summerson & R. A. Luettich, Jr. 1996. Response of bay scallops to spawner transplants: a test of recruitment limitation. Mar. Ecol. Prog. Ser. 132:93-107.

Pohle, D. G., V. M. Bricelj & Z. Garcia-Esquivel. 1991. The eelgrass canopy: an above-bottom refuge from benthic predators for juvenile bay scallops Argopecten irradians. Mar. Ecol. Prog. Ser. 74:47-59.

Preece, P. A., S. A. Shepherd, S. M. Clarke & J. K. Keesing. 1997. Abalone stock enhancement by larval seeding: effect of larval density on settlement and survival. Moll. Res 18:265-273.

Pulliam, H. R. 1988. Sources, sinks and population regulation. Am. Nat. 132:652-661.

Rhodes, E. W. 1991. Fisheries and aquaculture of the bay scallop, Argopecten irradians, in the eastern United States. In: S. E. Shumway, editor. Scallops: biology, ecology and aquaculture. Developments in aquaculture and fisheries science vol. 21. Amsterdam: Elsevier. pp. 913-924.

Rhodes, E. W. & J. C. Widman. 1980. Some aspects of the controlled production of the bay scallop (Argopecten irradians). Proc. World Maric. Soc 11:235-246.

SAS Institute, Inc. 1999. SAS/STAT User's Guide, Version 8. Cary, NC: SAS Institute, Inc. 384 pp.

Sargent, F. J., T. J. Leary, D. W. Crewz & C. R. Kruer. 1995. Scarring of Florida's seagrasses: assessment and management options. Florida Department of Environmental Protection FMRI Technical Report TR-1. pp. 1-43.

Sastry, A. N. 1963. Reproduction of the bay scallop, Aequipecten irradians Lamarck. Influence of temperature on maturation and spawning. Biol. Bull. 125:146-153.

Sastry, A. N. 1965. The development and external morphology of pelagic larvae and post-larval stages of the bay scallop, Aequipecten irradians concentricus (Say) reared in the laboratory. Bull. Mar. Sci. 15:417-435.

Secor, D. H. & E. D. Houde. 1998. Use of larval stocking in restoration of Chesapeake Bay striped bass. ICES J. Mar. Sci. 55:228-239.

Smith, C. F. & S. Tettelbach. 1996. Bay scallop restoration, western Peconic Bay. Project report to the Peconic Bay estuary program. 44 pp.

Stoner, A. W. & M. Ray-Culp. 2000. Evidence for allee effects in an over-harvested marine gastropod: density-dependent mating and egg production. Mar. Ecol. Prog. Ser. 202:297-302.

Summerson, H. C. & C. H. Peterson. 1990. Recruitment failure of the bay scallop, Argopecten irradians concentricus during the first red tide Ptychodiscus brevis, outbreak recorded in North Carolina. Estuaries 13:322-331.

Tammi, K. A. & W. H. Turner. 1998. Mind your E's and C's! An effective approach to shellfish restoration. J. Shellfish Res. 17:1314.

Tammi, K. A., S. J. Soares, W. Turner & M. A. Rice. 1994. Settlement and recruitment of bay scallops, Argopecten irradians (Lamarck, 1819), to artificial spat collectors in the Westport River estuary, Westport, Massachusetts. United States Environmental Protection Agency, Office of Water Proceedings, 1994 annual meeting of the National Shellfisheries Association, EPA 842-F-98-004. pp. 8-25.

Tettelbach, S. T. & E. W. Rhodes. 1981. Combined effects of temperature and salinity on embryos and larvae of the northern bay scallop Argopecten irradians irradians. Mar. Biol. 63:249-256.

Tettelbach, S. T. & P. Wenczel. 1993. Reseeding efforts and the status of bay scallop Argopecten irradians (Lamarck 1819) populations in New York following the occurrence of "brown tide" algal blooms. J. Shellfish Res. 12:423-431.

Tettelbach, S. T., C. F. Smith, R. Smolowitz, K. Tetrault & S. Dumais. 1999. Evidence for fall spawning of northern bay scallops Argopecten irradians irradians (Lamarck 1819) in New York. J. Shellfish Res. 18:47-58.

Thayer, G. W. & H. H. Stuart. 1974. The bay scallop makes its bed of seagrass. Mar. Fish. Rev. 36:27-30.

Turner, W. H. & S. J. Soares. 1994. The bay scallop restoration project in the Westport River. United States Environmental Protection Agency, Office of Water Proceedings, 1994 Annual meeting of the National Shellfisheries Association, EPA 842-F-98-004. pp. 35-44.

Walker, R. L. & P. B. Heffernan. 1994. Gametogenic cycle of the Carolina marshclam, Polymesoda caroliniana (Bosc, 1801) from coastal Georgia. Am. Malacolog. Bull 11:57-66.

Walker, R. L., P. B. Heffernan, J. W. Crenshaw, Jr. & J. Hoats. 1991. Mariculture of the southern bay scallop, Argopecten irradians concentricus (Say, 1822), in the southeastern US. In: S. E. Shumway and P. A. Sandifer, editors. An international compendium of scallop biology and culture: a tribute to James Mason. World Aquaculture Workshops, Number 1. pp. 313-321.

Wallace, R. K., K. Heck & M. Van Hoose. 1995. Oyster restoration in Alabama. In: M. W. Luckenbach, R. Mann & J. A. Wesson, editors. Oyster reef habitat restoration: a synopsis and synthesis of approaches. Proceedings from the symposium, Williamsburg, Virginia. pp. 101-106.

Waller, T. R. 1969. The evolution of the Argopecten gibbus stock (Mollusca: Bivalvia), with special emphasis on the tertiary and quaternary species of North America. J. Paleont. 43:1-125.

Yang, H., R. H. Weisberg, P. P. Niiler, W. Sturges & W. Johnson. 1999. Lagrangian circulation and forbidden zone on the west Florida shelf. Continental Shelf Res. 19:1221-1245.

WILLIAM S. ARNOLD, (1), * NORMAN J. BLAKE, (2) MELISSA M. HARRISON, (1) DAN C. MARELLI, (1,4) MELANIE L. PARKER, (1) SARAH C. PETERS (1) AND DON E. SWEAT (3)

(1) Florida Fish and Wildlife Conservation Commission, Fish and Wildlife Research Institute, 100 Eighth Avenue SE, St. Petersburg, Florida 33701; (2) College of Marine Science, University of South Florida, 140 Seventh Avenue South, St. Petersburg, Florida 33701; (3) Florida Sea Grant College Program, 830 First Street South, St. Petersburg, Florida 33701 (4) present address: Academic Diving Program, Florida State University, 315 Stadium Drive, Building 428, Tallahassee, Florida 32306

* Corresponding author. E-mail: [e-mail]

TABLE 1.
Planting sites, dates, and quantities for bay scallop (Argopecten
irradians) restoration on the west coast of Florida.

                                     # of
Year   Site               Stations   Cages   #/Cage   Brood

1998   Tampa Bay #1          1-5       26      200    97T3-1
       Tampa Bay #2          1-2       10      300    97T3-1
                             3-8       31      130    97T3-10
                            9-10       10      180    97T3-1
       Anclote
       Homosassa            1-10       51       50    97C5D
       Crystal River        1-10       51       50    97C5D
       Aquaculture Raft      N/A       15      200    Various
1999   Tampa Bay
       Anclote              1-10       51      300    98A5C
       Homosassa            1-10       51      300    98C2
       Crystal River        1-10       51      300    98C2
2000   Tampa Bay            1-10       53      100    99T1
       Anclote              1-10       51      200    99A3
       Homosassa            1-10       51      200    99CR3A
       Crystal River        1-10       51      200    99CR3A

                          Measure    Plant           Plant
Year   Site                date       Date           size

1998   Tampa Bay #1       6/2/98    6/17/98    25.9 [+ or -] 2.6
       Tampa Bay #2       6/2/98    6/17/98    25.9 [+ or -] 2.6
                          7/21/98   7/21/98    29.8 [+ or -] 3.4
                          7/21/98   7/21/98    24.6 [+ or -] 3.0
       Anclote
       Homosassa          6/23/98   7/6-7/98   24.6 [+ or -] 2.5
       Crystal River      6/23/98   7/6-7/98   24.6 [+ or -] 2.5
       Aquaculture Raft   6/11/98   6/11/98    21.0 [+ or -] 0.4
1999   Tampa Bay
       Anclote            5/4/99    5/4/99     28.5 [+ or -] 3.0
       Homosassa          3/30/99   3/30/99    26.9 [+ or -] 2.8
       Crystal River      3/29/99   3/29/99    25.9 [+ or -] 3.1
2000   Tampa Bay          7/7/00    7/7/00     33.2 [+ or -] 3.8
       Anclote            7/13/00   7/13/00    27.3 [+ or -] 5.4
       Homosassa          5/22/00   5/25/00    30.4 [+ or -] 4.6
       Crystal River      5/22/00   5/25/00    30.8 [+ or -] 4.9

Year   Site                            Comments

1998   Tampa Bay #1       Total Mortality at Stations 2 & 5;
                            Station 4 lost;
                            Repro cage at Station 5
       Tampa Bay #2       Indian Key, Station 2 = previous
                            Station 3
                          Tarpon Key
                          Madeline Key
       Anclote            No Activity
       Homosassa          Repro station 5
       Crystal River      Repro station 1
       Aquaculture Raft   Repro samples from various cages
1999   Tampa Bay          No Activity
       Anclote            Repro station 5
       Homosassa          Repro station 5
       Crystal River      Repro station 5
2000   Tampa Bay          Repro station 5
                            (3 cages of 100 each)
       Anclote            Repro station 5
       Homosassa          Repro station 5
       Crystal River      Repro station 5

TABLE 2.
Qualitative reproductive staging criteria for bay scallops
(Argopecten irradians) collected from Florida waters.

Stage            Description

No Data          Unable to read slide
Inactive         Gonadal tissue undifferentiated
Developing       Follicle wall expanded and lumen contains
                   spermatocytes or oocytes
Ripe             Follicular areas filled with either dense or
                   radiating bands of sperm or ova
Early Spawning   Central lumen of either male or female follicles
                   beginning to empty
Spawning         Lumen of follicles more extensively emptied
Spent            Follicle lumen emptied of ova and spermatids; many
                   amoebocytes and cellular debris often present