Forty Years of Field and Laboratory Experiments with Hatchery-Reared Queen Conch: the Case for Conserving Natural Stocks

Reviews in Fisheries Science & Aquaculture

ISSN: 2330-8249 (Print) 2330-8257 (Online) Journal homepage: https://www.tandfonline.com/loi/brfs21

Forty Years of Field and Laboratory Experiments with Hatchery-Reared Queen Conch: The Case for Conserving Natural Stocks

Allan W. Stoner

To cite this article: Allan W. Stoner (2019): Forty Years of Field and Laboratory Experiments with Hatchery-Reared Queen Conch: The Case for Conserving Natural Stocks, Reviews in Fisheries Science & Aquaculture, DOI: 10.1080/23308249.2019.1628705 To link to this article: https://doi.org/10.1080/23308249.2019.1628705

Published online: 27 Jun 2019.

Submit your article to this journal

View Crossmark data

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=brfs21 REVIEWS IN FISHERIES SCIENCE & AQUACULTURE https://doi.org/10.1080/23308249.2019.1628705

Forty Years of Field and Laboratory Experiments with Hatchery-Reared Queen Conch: The Case for Conserving Natural Stocks

Allan W. Stoner Community Conch, Lopez Island, Washington, USA

ABSTRACT KEYWORDS Rehabilitating overfished species through releases of cultured juveniles depends upon Stock restoration; stock growth and survival of those animals in the field. Releasing cultured queen conch (Strombus enhancement; hatchery gigas) was proposed 40 years ago to rebuild over-harvested populations. Hatchery culture culture; Strombus gigas; has been perfected and thousands of juveniles can be produced with meticulous hus- cost-benefit analysis bandry. Many field experiments have been conducted with hatchery-raised queen conch since the early 1980s, but the results cast doubt regarding stock enhancement. Queen conch demonstrates high natural mortality rates, and cultured conch can have morphological and behavioral deficiencies that diminish survival. Deficiencies can be reduced with perfect culture conditions and prerelease field conditioning; and survival can be improved by releasing large juveniles, but all of these measures add to cost of seed stock. Also, release requirements are problematic. Habitat selection, release timing, and release patterns are all critical for reducing predation losses. While all of these challenges have been considered, no field release has resulted in survival rates that are encouraging. The cost of stock restoration through release of cultured queen conch will be enormously expensive, if successful at all, and every effort should be made to conserve wild populations. Hatchery production for stock restoration should be considered a last resort.

Introduction difficult to evaluate, and the risks and contributions of hatchery rearing for release have been debated Restocking overharvested species (Blankenship and Leber 1995; Molony et al. 2003; Hatcheries have been built around the world with the Waples and Drake 2004; Lorenzen 2005; Johnston goal of either enhancing or rehabilitating stocks of et al. 2018). In the concluding chapter of their com- economically important marine invertebrates and prehensive treatment of stock enhancement experience fishes in the natural environment (Leber et al. 2004; with marine invertebrates, Bell et al. (2005) noted Bell et al. 2005). The overarching principle is to that: “Restocking and stock enhancement are intui- reduce or eliminate high mortality that occurs in the tively appealing strategies; hatchery technology is now earliest life stages (e.g., pelagic larvae and smallest widely available, and there have been some success juveniles), and release juveniles into the field that are stories. However, managers and resource owners most likely to survive and grow to sizes that will con- should not adopt the premise that, given sufficient tribute either to the fishery or reproductive stock. research, these technologies can be made to work to Some hatchery systems, such as those for salmonids, solve their problems. Rather, objective questions need have been in place for decades (Kennedy 1988; Heard to be asked about whether investments in release pro- 2012) while others continue to be experimental. Some grams will add value to other forms of management.” of the programs aimed at enhancement have been This review is aimed at exploring those questions with documented as successful or at least promising (e.g., regard to queen conch. Leber et al. 2004; Davis et al. 2005; Roberts et al. Queen conch, Strombus gigas is a large gastropod 2007; Heard 2012) and others have failed or shown which is an iconic symbol of nature in the Caribbean questionable success for one reason or another (e.g., region. The species is harvested for its meat and shell, Bannister and Addison 1997; Svåsand et al. 2002; Ellis making it one of the most important demersal fishery et al. 2015). Successful enhancement is sometimes resources in the region. The queen conch is so

CONTACT Allan W. Stoner [email protected] ß 2019 Taylor & Francis Group, LLC 2 A. STONER important to the economy and culture of The et al. 1993), but Berg was the first to culture conch Bahamas and the Turks and Caicos Islands that it through metamorphosis to benthic juvenile stages. appears on the coats of arms for both nations. In the Interest in culturing queen conch toward the goals of Florida Keys, the queen conch was once so prominent stock restoration and direct grow-out for food that people living in the Keys are still called “conchs;” increased rapidly after that (see Siddall 1983, 1984), however, because of depletion, the Florida fishery has and by the early-1980s cultures of queen conch to been closed entirely since 1985 with only modest juvenile stages had been successfully accomplished in recovery since that time (Glazer and Delgado 2003; Venezuela (Brownell 1977; Laughlin and Weil 1983), Delgado and Glazer 2007; Florida Fish and Wildlife Puerto Rico (Appeldoorn and Ballantine 1983; Conservation Commission, unpubl. data). Hatchery Ballantine and Appeldoon 1983), the Turks and production of queen conch has been pursued in all of Caicos Islands (Davis and Hesse 1983; Davis et al. these locations and several others (see below). 1987; Davis and Dalton 1991), The Bahamas (Heyman Queen conch fisheries over the entire Caribbean et al. 1989), Florida, Mexico, Bonaire and the Virgin region are under ever increasing threats caused by Islands (Jory and Iversen 1985). While it took dedi- over-exploitation and losses of shallow-water habitats cated efforts in several laboratories to advance cultures critical for the species (Appeldoorn et al. 2011). Early from beakers to hatchery scale, and many of the warnings about declining queen conch populations research programs listed above ended for lack of were advanced by Adams (1970) and Berg (1976), but funding, it is now possible, with a supply of field-col- it took until 1992 for queen conch to be listed in lected egg masses, clean seawater, algal cultures for Appendix II of the Convention on International Trade food and meticulous husbandry, to reliably produce in Endangered Species (CITES) whereby international thousands (or millions) of larvae ready for metamor- trade in queen conch products is monitored under phosis in 2 and 3 weeks in relatively small land-based CITES because of its threatened status. Regulations on systems. Thousands of juveniles with shell lengths conch fishing have become stricter with time but vary (SL) of 20–40 mm were produced in each of the labo- substantially over the more than 30 nations where ratories listed above. Megan Davis, one of the pioneer conch are distributed (Appeldoorn 1994; Theile 2001; culturists, provides excellent guidance on the proce- 2005; Boman et al. 2018), ranging from no regulation dures and systems needed for a commercial-scale whatsoever to complete moratoria, closed seasons, hatchery (Davis 1994, 2000a, 2005). closed areas, catch and export quotas, size and age Given early successes in culture Goodwin (1983) limitations for harvest, and gear restrictions designed suggested that an alternative to traditional fishery to provide for sustainable fisheries. Despite the vari- regulation might be to supplement natural production ous fisheries management efforts continued reductions through mariculture. This included both seeding in queen conch stocks, decreasing catches per unit hatchery-reared juvenile conch into depleted areas effort, and changing age structures throughout the and developing methods for the direct production of region resulted in a review by the National Oceanic conch for market on both commercial and “cottage and Atmospheric Administration (U.S. Department of industry” scale. He added that “Because recent pro- Commerce) to consider queen conch for listing under gress in conch mariculture is both exciting and the Endangered Species Act. Threatened species status encouraging, we need to insure that we do not was declined in 2014; however, the species continues encourage unrealistic expectations among potential to be severely overfished in many parts off its distri- beneficiaries or among prospective investors.” After bution, and the decision remains under appeal in more than 40 years of experimentation with queen 2019. Long-term and repeated surveys led Stoner et al. conch culture and releases, Goodwin’s statement pro- (2019) to conclude that queen conch populations in vides the impetus for the present assessment. The Bahamas are showing signs of serial depletion. There is no question that hatcheries can produce large numbers of queen conch intended as seed for grow-out and/or field release, although culture contin- Hatchery production of juvenile queen conch ues to be dependent upon field-collected eggs. By the Berg (1976) proposed that stock enhancement and end of the 1990s, the Caicos Conch Farm in rehabilitation of queen conch might be possible Providenciales, Turks and Caicos Islands, was produc- through hatchery rearing. D’Asaro (1965) had cul- ing millions of hatchery-cultured juvenile conch per tured queen conch through larval stages and provided year in a commercial-scale system (Davis 2005; Davis detailed descriptions of the veligers (see also Davis et al. 2010). The next crucial challenge involves REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 3 successfully transferring hatchery-reared conch into mortality rates of queen conch observed in the field. the field and achieving acceptable rates of survival. As All of these topics are addressed in the sections with the topic of hatchery culture, field experiments that follow. with juvenile conch have spanned several decades, but the demonstrable successes have been fewer. Seed stock quality Nevertheless, discussions of queen conch farming and the possibility of restoring queen conch populations Growth and shell form through hatchery production of juveniles continue on Shell strength and long lateral spines of juvenile queen a regular basis. The purpose of this review is to sum- conch (Figure 1) are the primary lines of defense marize the successes and failures in growing out against the many predators that either crush shells or juvenile queen conch, and to provide the relevant peel them at the aperture. Consequently, any culture- bibliography of primary literature for those contem- related conditions that compromise shell morphology plating their own cultures for research, a business can impact survivorship, and differences have been model, or fisheries management. reported widely. Jory and Iversen (1988) were first to point out potential deficiencies in queen conch pro- Challenges associated with releasing queen conch duced in hatchery environment. They found that the into the field resistance of conch shells to breaking forces generally Achieving acceptable rates of growth and survival are increased with shell length as expected, but they the primary challenges facing a biologist interested in observed that shell strengths varied between different restoring or enhancing a field population of queen hatcheries in Venezuela, Mexico, and their laboratory conch. When cultured juvenile conch are moved into in Miami, and that the strengths of cultured conch the field most of the controls on environment offered were generally lower than those of wild conch with by hatchery setting are lost (Creswell and Davis 1991). comparable shell length. In fact, hatchery-reared Decades of experience with juvenile conch in the field juvenile queen conch often develop shells that are have shown that selecting habitats which promote lighter than wild conch, and they have shorter lateral high growth rates in young conch can be challenging. spines, longer spires, and narrower overall shell width An even more complex challenge is reducing field (Marshall 1992; Stoner and Davis 1994; Ray and mortality (see Table 1). To achieve acceptable survival Stoner 1995a; Spring 2008) (Figure 1). Shell weight rates: (1) Seed stock must have characteristics that was about 40% lower in hatchery-reared conch than allow them to survive at rates equivalent to wild wild conch measured by Stoner and Davis (1994), and conch. (2) Release sites must be chosen to give the Spring (2008) observed that morphological deficits best possible opportunities for seed stock to feed, increased with the length of time spent in hatchery grow, and survive the gantlet of predators that occur environment. naturally. (3) Once a site is chosen, release methods The variation in shell morphology observed appropriate to the specific site must be developed to between wild and hatchery-reared queen conch, and optimize survival of the seed stock. (4) Assuming that between different hatcheries and different investiga- all of these requirements can be met the release pro- tions, stems from growth-related processes observed gram must be able to accept the high natural in the species. Field transplant experiments with

Table 1: Variables that influence the survival of juvenile queen conch released into the field, and references to experimental research that address these topics. Variable References Intrinsic variables Morphological deficits in hatchery-reared conch: shell form, shell strength 9, 19, 21, 22, 28, 30 Behavioral deficits in hatchery-reared conch: burial, predator & food recognition, etc. 8, 19, 28, 30 Shell size 2, 3, 4, 6, 9, 11, 13, 16, 18, 22, 23, 26, 27, 30 (and see Table 2) Extrinsic variables Habitat – sediment, macrophytes, currents, foods, etc. 1, 12, 14, 15, 17, 22, 23, 27, 29 Predator density & types 24, 29 Conch density 10, 14, 15, 17, 20, 23, 27, 29 Time – year, season, lunar phase, time of day 5, 15, 24, 25, 26 References (chronological order): (1) Appeldoorn and Ballantine (1983); (2) Jory and Iversen (1983); (3) Siddall (1983); (4) Creswell (1984); (5) Appeldoorn (1985); (6) Iversen et al. (1986); (7) Appeldoorn (1987a); (8) Coulston et al. (1987); (9) Jory and Iversen (1988); (10) Stoner (1989); (11) Creswell and Davis (1991); (12) Stoner and Sandt (1991); (13) Davis (1992); (14) Marshall (1992); (15) Stoner and Ray (1993); (16) Dalton (1994); (17) Ray and Stoner (1994); (18) Ray et al. (1994); (19) Stoner and Davis (1994); (20) Stoner and Lally (1994); (21) Martın-Mora et al. (1995); (22) Ray and Stoner (1995a); (23) Ray and Stoner (1995b); (24) Glazer and Jones (1997); (25) Stoner and Glazer (1998); (26) Glazer and Delgado (1999); (27) Ray-Culp et al. (1999); (28) Delgado et al. (2002); (29) Stoner (2003); (30) Spring (2008). 4 A. STONER

Figure 1. Shell morphology demonstrated by juvenile queen conch under different growth conditions (A & B) Typical shell form for slow-growing wild conch from nursery grounds in The Bahamas (C & D) Rapid-growing conch typical of those produced in culture. The shells of naturally slow-growing conch have higher weights, longer lateral spine, and higher breaking strengths than those growing rapidly to equivalent lengths (120 mm SL shown). Shell morphology varies with culture environment, and some cultured queen conch have short spire lengths, as well as stunted growth and other deficits (see text). juvenile queen conch (Martın-Mora et al. 1995) have more vulnerable to shell-breaking predators than shown that short lateral spines, long spires, smaller those that grow more slowly. This is highly relevant overall shell diameter, and low shell weights (and to hatchery production of queen conch because hatch- strength) are all associated with rapid growth, while eries are motivated to produce the largest conch in slow growth resulted in shell morphology more typical the shortest time possible, and the “push” for growth of the wild conch found in natural nursery grounds results in conch with the shell characteristics that are (i.e., long lateral spines, heavier shells, etc.). Therefore, vulnerable to shell-breaking predators. In fact, field rapid growth in shell length results in conch that are experiments in conch nursery grounds in The REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 5

Bahamas confirm that short-spined conch are killed at Growth and survival of hatchery-reared conch in significantly higher rates than long-spined conch (Ray the field and Stoner 1995a). Stoner and Davis (1994) found We know from field experiments that growth rates that the shells of hatchery-reared conch converged demonstrated by hatchery-reared queen conch can be with the morphology of wild conch after 7.5 months lower than those observed in wild conch. Stoner and when tagged and released in a well-studied nursery Davis (1994) reported significantly lower growth rates ground; however, the experiment was not designed to for cultured conch compared with wild conch in field determine whether convergence resulted from changes enclosures, on tethers, and with free-ranging conch in the shells as they grew or from losses of the conch monitored long-term in two different habitats. that had poor fitness qualities. Because only 9% of the Summer growth rates of wild conch were twice as hatchery-reared conch survived, and the average high as hatchery-reared conch, but rates were more hatchery conch grew just 10 mm (from initial release similar in the cooler season when growth was natur- at 101 mm SL), it is likely that differential mortality ally low. Differences in growth observed by Marshall explains the convergence. (1992) in relatively short-term experiments were apparent, but not significant statistically. The mecha- Observations on behavioral deficits nisms for these differences in growth are speculative Juvenile conch spend part of their time partially but could be related to hatchery conch being accus- buried in the sandy sediments associated with nursery tomed primarily to artificial foods, or differences in grounds. In contrast with wild conch, hatchery-reared feeding rates, activity levels, conversion efficiencies or individuals often demonstrate low burial propensity or other basic physiological deficits. capability (Coulston et al. 1987). This may be related Marshall (1992) provided the first direct comparison of field mortality in hatchery-reared and wild to the fact that most hatcheries use little or no sedi- conch. He tethered conch with comparable sizes ment in culture for reasons related to tank cleaning. (72–74 mm SL) in two nominally similar seagrass Marshall (1992) quantified burial frequencies in hatch- beds, one with resident conch and one without. Both ery-reared and wild conch that were tethered in the hatchery-reared and wild conch were lost at very field, finding a non-significant but somewhat lower rapid rates at the site with no resident conch, showing burial rate in hatchery conch. In contrast, Stoner and the critical density dependence of survival (see below). Davis (1994) observed that the proportion of free- At the site with resident conch, cumulative mortalities ranging hatchery-reared conch buried in sediment was of wild and hatchery-reared conch were not signifi- less than half that observed in wild conch in the same cantly different at the 33-week end of the experiment habitat, and the difference persisted over the 7.5- (75 and 82%, respectively), but hatchery-reared conch month duration of the experiment. The contrast died at an accelerated pace and the difference at observed between the two field experiments with 18 weeks was highly significant (hatchery ¼ 50%, wild untethered conch might be related to the smaller size ¼ 27%). Given the site differences observed in mortal- – of individuals tested in the later study (35 40 mm vs. ity rates, Marshall concluded that location was a more > 100 mm SL). Low burial rates and elevated rates of important determinant of mortality than the differen- – general activity were also noted for 30 40 mm SL ces observed between hatchery-reared and wild conch. conch reared in a Florida hatchery (Delgado et al. In another tether experiment, Ray et al. (1994) inter- 2002), but burial rates were significantly higher when spersed hatchery-reared and wild conch (75–-76 mm conch were exposed to a caged lobster for two weeks. SL) in a homogeneous seagrass meadow in The Other behavioral deficiencies have been speculated Bahamas and monitored mortality on a regular basis as possible explanations for low growth and survival for 54 days. As observed by Marshall (1992), mortality observed in hatchery-reared conch. These include a accumulated more rapidly for hatchery-reared than reduced ability to identify and consume natural foods for wild conch, and at the termination of the experi- after release (Stoner and Davis 1994) and reduced ment the overall mortality of hatchery-reared conch ability to avoid predators (Delgado et al. 2002) (see was 65% while wild conch suffered just 29% loss. below). Low responsiveness to predators has been Field trials with large numbers of free-ranging wild observed in cultured pinto abalone (Haliotis kamtscha- and hatchery-reared queen conch (both 100 mm SL) takana) and like queen conch the deficits increased were marked and released together showed that the with time in the hatchery (Hansen and hatchery-reared conch had significantly lower surviv- Gosslein 2016). als than wild conch (Stoner and Davis 1994) with 6 A. STONER mortality rates of hatchery-reared conch approxi- counterparts and developed shells that were thicker mately double the rates recorded for wild conch. This and denser. In subsequent unrestrained exposure to large experiment is discussed further under “Results lobsters, 50% of the preconditioned animals survived of Field Tagging Experiments,” as will the most recent predation while only 10% of the unconditioned conch experiment conducted by Spring (2008). Possible survived, and Delgado et al. concluded that the explanations for high mortality in hatchery-reared increased shell strength provided by slower growth conch include the shell deficiencies described above, improved survival. No comparisons with wild conch low burial rates leading to a reduced ability to avoid were made. predators through shelter seeking behavior, and higher It is possible that these and other improvements in general activity discussed by Delgado et al. (2002). culture techniques, such as adding sediment to the culture environment, may reduce deficiencies in hatchery- Remediation of deficits reared queen conch so that survivals can more closely Two field studies have addressed the need to reduce approach those of wild conch. Most of these improve- morphological and behavioral deficits observed in ments, however, involve slowing growth to natural hatchery-reared queen conch. First, Spring and Davis rates, greater exposure to natural environment includ- (2003) raised juvenile queen conch (31 mm SL) in five ing high water flow, natural foods, and predators; all different stocking densities and on three different sub- adding time and cost to the production of seed conch strata for four months. They concluded that growth in large enough to survive without protection. The fact culture was inversely related to stocking density and that differences between hatchery-reared and wild that low density (resulting in high growth rate) would conch in shell strength and other morphological and be preferred for land- or pen-based aquaculture. Slow behavioral characters increase rapidly with time in cul- growth, resulting from high-density culture, produced ture led Spring (2008) to conclude that hatchery-reared stronger shells and shell forms more appropriate for conch need to be moved into field environment at a field release. These results comport with the earlier size <40 mm SL. While conch at this size might survive field experiments on growth and shell form (Martın- in some locations or in covered pens, others have found Mora et al. 1995). Subsequently, Spring (2008) reared that larger size is critical for a cost-effective survival of queen conch juveniles under four different conditions seed stock (e.g., Stoner and Glazer 1998; Glazer and of seawater flow ranging from 0 to 30 cm/sec in tor- Delgado 1999; see “Determining seed stock size”). oid-shaped culture chambers. She found that high-flow environment resulted in heavier and stronger shells Choosing release sites more similar to wild conch than typical hatchery culti- vars. It is worth noting that queen conch nurseries in Predicting locations for good growth The Bahamas are usually located in areas with high Unlike some mariculture species, queen conch do not tidal current velocities (Stoner et al. 1996; Stoner grow rapidly. Brownell’s early studies in Venezuela 2003). In another experiment, where juvenile conch showed that average growth rate for early post-settle- were held in field enclosures for several months, conch ment conch is 0.2 mm/day (Brownell 1977), which developed shells that had intermediate measurements results in growth from 2 mm shell length at settlement between hatchery-reared and wild conch (Spring to 75 mm in 12 months after metamorphosis 2008). After 20 months in the field, hatchery-reared depending upon temperature. Early enthusiasm for juveniles had shell strengths that were equivalent to queen conch culture in the 1980s hinged in part upon those measured in wild conch. Based upon her field Berg’s(1976) suggestion that market size could be experience, Spring concluded that cultured conch need achieved in 2.5 years. Subsequently, Appeldoorn et al. to be moved to field environment at the soonest pos- (1987) evaluated size at age for a number of different sible time to reduce hatchery-related deficits in both locations in the Caribbean finding that age-1 conch shell strength and form as well as burial behavior. ranged 88–121 mm SL, age-2 from 114 to 170 mm, Delgado et al. (2002) conducted well-designed and age-3 from 148 to 225 mm, and it is now clear laboratory experiments on the possible benefits of that maximum length in shell length does not occur conditioning hatchery-reared conch with spiny lob- before 3.5 years. The characteristic flared shell lip sters (Panulirus argus). For two weeks 35–40 mm SL only begins to form at that time (Appeldoorn 1988a; hatchery-reared conch were held in enclosures with Stoner and Sandt 1992), and recent histological studies and without caged lobsters. Conch in the presence of show that sexual maturity occurs well after the shell lobster grew more slowly than the unconditioned lip flare is formed (Aldana-Aranda and Frenkiel 2007; REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 7

Stoner et al. 2012a; Foley and Takahashi 2017; Boman rates ranged from the typical 0.15–0.20 mm/day where et al. 2018). So, unless juveniles are the intentional conch were present to near zero in habitats with target for harvest, time to maximum meat weight is at seemingly similar habitats but without conch. In least 3 or 3.5 years. About five years of development is another transplant experiment, Ray and Stoner (1994) required for reproductively mature conch. To achieve found that seagrass cover was a poor predictor of growth rates resembling a natural stock the hatchery- growth rates in juvenile conch, and the presence of reared queen conch will need to be placed in ideal conspecifics was the best predictor. These findings, growth environments. showing variation in habitat suitability for queen Queen conch demonstrate substantial ontogenetic conch, illustrate the challenge of selecting appropriate shifts in habitat requirements from newly metamor- habitats for any stock restoration or enhancement phosed juvenile to adult stages (Sandt and Stoner efforts, and the need for preliminary experimental 1993; Ray and Stoner 1995b). Conch in their first year evaluation of prospective release or grow-out sites. of benthic life spend daytime hours buried in the sedi- Another variable critical to any queen conch release ment or sheltered within the branches of complex or grow-out effort in the field is determining the car- algae or benthic debris. Physical structure for shelter rying capacity of the habitat. Queen conch are herbi- becomes somewhat less important with conch size, vores during their entire life cycle (Randall 1964; particularly after 100 mm SL, and large juveniles are Brownell and Stevely 1981; Stoner and Waite 1991), found on a variety of substrata including hard-ground aggregations of juvenile conch can quickly consume covered with a thin algal turf, coral rubble, and mead- most of the food items in their surroundings, and the ows of turtlegrass (Thalassia testudinum) (Randall population often needs to move over a large space 1964; Alcolado 1976; Hesse 1979; Glazer and Berg when the population is dense because of their impacts 1994; Glazer and Kidney 2004). In The Bahamas, on the benthic environment (Stoner et al. 1995b). most queen conch nursery grounds occur in turtle- Carrying capacities have never been determined for a grass meadows with moderate shoot densities (Stoner queen conch nursery ground, but most nurseries 2003); however, statistical interpretation of juvenile rarely have densities more than two juveniles/m2 distribution over large seemingly similar seagrass (Stoner 2003), and enclosure experiments have shown meadows in the central Bahamas indicate that nom- that densities exceeding four juveniles per square inal habitat descriptors do not provide good predictive meter significantly inhibit growth rates and body con- power on suitability for queen conch. For example, dition under ideal nursery conditions in The Bahamas following a multivariate analysis for juvenile distribu- (Stoner 1989; see also Davis et al. 2010). tion including a wide range of environmental variables Spatially explicit habitat suitability models for (e.g., seagrass biomass, sediment qualities, algal prod- juvenile queen conch (incorporating variables of uctivity, and hydrography), Stoner et al. (1995a) concluded that “Conch nursery grounds were associated depth, seagrass cover, sediment qualities, and tidal with specific combinations of food production and flow pattern, etc.) have been developed for seagrass- shelter not immediately obvious within large mead- based nursery grounds in The Bahamas (Jones 1996), ows. Most seagrass beds are probably unsuitable for and conceptual syntheses for habitat suitability have aggregations of juvenile queen conch; therefore, stock been made (Stoner 2003; Appeldoorn et al. 2011). The management programs will need to identify and con- productivity of macroalgal foods at a site is crucial to centrate on the unique nursery habitats.” the quality of habitat for juvenile queen conch, and Experimental analysis of habitat quality has resulted particularly difficult to evaluate. Despite a general pat- in conclusions similar to those resulting from statis- tern of habitat characteristics appropriate to queen tical analyses of distribution. Enclosure experiments conch, each site will have unique features, and it is with both wild and hatchery-reared queen juvenile very likely that each has a specific carrying capacity conch have shown that there is enormous spatial vari- that varies with time. Those planning for juvenile ation in growth rates observed in habitats that, on the conch releases will need to have a broad understand- basis of seagrass cover, sediment type, and depth, ing of their prospective release sites for a reasonable would appear to provide appropriate and similar nur- expectation that seed conch will thrive. In lieu of sery characteristics. For example, Stoner and Sandt long-term experience with a natural nursery ground, (1991) enclosed conch in a well-studied nursery preliminary enclosure experiments (see Stoner 1994) ground and other seemingly identical habitats outside should always be conducted prior to any release or long-term nursery grounds. They found that growth grow-out endeavor. 8 A. STONER

Predicting locations for high survival densities proven to be the ideal structure to protect Minimizing predator-induced mortality is the most juvenile conch from predators, resulted in highly vari- challenging problem for those who wish to release able mortalities for conch in two size classes (mean hatchery-reared queen conch into the field with the shell lengths ¼ 70 mm and 100 mm). The predation goal of enhancing a fishery or rebuilding a reproduct- rates for those tethered outside the nursery sites by ive population. The list of animals that consume just 300–500 m were 3–11 times higher than rates juvenile queen conch is long. For example, Ray-Culp inside the nurseries. The problem of predicting sur- et al. (1997) identified 28 species of abundant small vival is compounded by the fact that mortality in invertebrates in queen conch nurseries that can con- juvenile queen conch is inversely density dependent sume newly settled queen conch (<10 mm SL) includ- (Marshall 1992; Ray and Stoner 1994). That subject is ing a wide range of polychaetes, shrimps, and crabs. discussed further in the next section. It is clear that Even juvenile queen conch one or two years old (to certain structural elements of habitat such as specific 170 mm SL) are heavily preyed upon by large crabs, seagrass shoot densities and high tidal current provide spiny lobsters, octopus, predaceous snails, a long list the best refuge for juvenile conch in The Bahamas of fishes with strong teeth, and sea turtles. Randall (Stoner and Ray 1993, Ray and Stoner 1995b; Stoner (1964) identified 22 important predators through dir- 2003), but because of the high degree of spatial vari- ect observations on feeding or by examination of ation in predation-induced mortality only a small part stomach contents, and Jory and Iversen (1983) listed of nominally similar seagrass habitat appears to be 10 known predators and 27 likely predators on juven- suitable for the survival of juvenile queen conch. ile queen conch beyond those listed by Randall. Some Jones (1996) estimated that only 1% of the seagrass of the larger predators are particularly devastating to meadows with appropriate shoot densities near Lee populations of juvenile conch. For example, sea turtles Stocking Island offered habitat that could support can kill hundreds of relatively large juvenile conch by juvenile queen conch. crushing in a few hours, and octopus can quickly kill Glazer and Jones (1997) have suggested that small or large conch, even those in enclosures, in a releases of hatchery-reared conch should be made in short period without harming the shells (author, per- areas with low predator density. In fact, it is unknown sonal observations). Given extremely high rates of whether historically important nursery grounds have predation-induced mortality in small queen conch naturally low concentrations of predators, the physical (Appeldoorn 1984; Stoner and Glazer 1998) most structures of the nurseries present especially good ref- projects will need to avoid releasing the most vulner- uge for conch, or the high currents associated with able early juvenile stages. By the end of the 1980s, it some nursery areas make predators less effective. was generally accepted that releases would need to be While some of the physical structures of suitable habi- made with juvenile conch that were 40–60 mm SL tat can be assessed for prospective conch releases, (Creswell and Davis 1991) or even much larger. The evaluating densities of highly motile predators such as topic is discussed further under the heading “Size lobster, octopus, crabs, certain fishes, and sea turtles is at release.” difficult. Given that, the only practical way to evaluate Since the earliest ecological studies of juvenile a site for likely predation pressure is to conduct queen conch (e.g., Alcolado 1976; Berg 1976), includ- small-scale field experiments with conch enclosed in ing those with cultured conch (Appeldoorn and low fences or on tethers (Stoner 1994; Stoner and Ballantine 1983), it has been recognized that predation Glazer 1998). New experiments with underwater cam- rates are highly variable in both time and space. eras may offer new insights with regard to the identity Experimental studies with tethered conch were con- and abundance of conch predators at different loca- ducted later with the goal of understanding the locations (A. Kough, Shedd Aquarium, unpubl. data); tions of historically important nursery grounds and however, predation-related mortality can be highly for predicting where survivals of hatchery-reared variable with time as well as space making it queen conch might be highest. Tethering experiments extremely difficult to predict likely survival of seed conducted by Marshall (1992) revealed that mortality conch. In fact, highly variable field survival rates have rates in known nurseries were half the rates observed been reported for other hatchery-reared gastropods. outside those nurseries even when the different loca- For example, Carson et al. (2019) released 11,000 tions appeared to have similar physical characteristics. pinto abalone (Haliotis kamtschatkana) over several A similar tether experiment (Ray and Stoner 1994), years at 10 sites carefully selected in Puget Sound for focused on seagrass sites with intermediate shoot ideal abalone habitat. Detection rates at the sites, REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 9 some separated by only 200 m, ranged from 0 to 23% Table 2, and expanded discussion under the heading in the first year after release, and the spatial patterns “Experience with free-ranging conch”). Producing were consistent over both site and year. As with queen conch this large with a minimum of hatchery-associ- conch, they speculated that differences in survival ated deficits is likely to be very costly. were affected by difficult-to-assess variation in the Some researchers have proposed releasing cultured predator types, abundance or behavior, and that larvae instead of juveniles to reduce the cost of release success would need to be determined by trial hatchery facilities and hatchery-associated deficits in and error. the seed stock (Arnold 2008). Molluscan examples include releases of larval bay scallops (Argopecten irradians) (Arnold et al. 2005) and abalone (Haliotis Perfecting release procedures fulgens and H. corrugata) (Searcy-Bernal et al. 2013) Size at release in the thousands or millions. While some survivals A large number of studies conducted in the 1980s and were indicated in both studies, results are equivocal 1990s provided insight into the effect of queen conch because it is difficult without genetic markers to size on their survival in field trials (Table 2). Without definitively identify hatchery survivors. Also, the exception, those concerned primarily with seed pro- Caicos Conch Farm released excess production of duction reported that costs rose rapidly after juveniles queen conch larvae from their hatchery into sur- reached 20 or 30 mm SL because of space require- rounding waters beginning in the 1980s, undoubtedly ments, density-dependent growth, and rapidly increas- totaling many millions over the long term (Julie ing volumes of feed required (Siddall 1983; Creswell Davis, grow-out manager at the Caicos Conch Farm, 1984; Dalton 1994). By the end of the 1980s, the con- 2004–2007; pers. commun.), but there was no effort sensus was that releases of conch <10 mm shell length to track their fate. would result in unacceptable losses in the field, and that conch no smaller than 40 to 60 mm SL would be Release timing needed to insure “reasonable” survivorship (Creswell Achieving high survivals in queen conch seed stock and Davis 1991). Jory and Iversen (1988) found that requires determining the best time for releases into an important inflection point in the strength of shells the field. It is generally believed that mortality rates in was reached at 55 mm SL, and conch smaller than juvenile queen conch are highest in the warmest sum- that were highly vulnerable to predators that chipped mer months. For example, Glazer and Jones (1997) or crushed the shells of their prey. We know now that released hatchery-reared juveniles in the Florida Keys size for optimal survival needs to balance hatchery at 75–85 mm SL in tag-recovery experiments running costs, likely survival in the field and increasing hatch- for 12 to 13 weeks; they reported survival >40% in ery-induced morphological and behavioral deficits the fall period, and rates between 4.9 and 11% in each that accumulate with time in culture. Direct metrics of the other three seasons. A similar seasonal pattern on survival in field experiments with either tagged or with lower overall survivals (21.8% in fall) was tethered conch have all resulted in the conclusion that reported for slightly smaller (55–65 mm SL) seed conch seeded into the wild without protection will conch (Glazer and Jones 1997). Studies with free- need to be larger, between 75 and 150 mm SL (see ranging conch described in the following section show

Table 2. Suggested minimum sizes for release of hatchery-reared queen conch into the field and the basic criteria used to make that judgement. Reference Size for release (mm) Primary criteria for conclusion Siddall (1983) 20 Hatchery cost Dalton (1994) 30 Hatchery cost Spring (2008) 40 Hatchery cost and field survival Creswell (1984) 50 Hatchery cost Jory and Iversen (1988) 55 Survival – shell crush strength Creswell and Davis (1991)40–60 Hatchery cost and field survival Stoner and Glazer (1998) 70 Field survival Ray et al. (1994)75–90 Field survival Glazer and Delgado (1999) 100 Field survival and cost analysis Jory and Iversen (1983) 120–150 Field survival Three of the studies considered primarily the cost for hatchery production, while most of the later studies were based upon survivorship observations in the field suggesting larger sizes for release. Longer culture time, however, increases the incidence of hatchery-induced behavioral and morphological deficits as well as cost to produce seed stock (see text). 10 A. STONER that the seasonal patterns of mortality over different and Iversen 1983; Glazer and Delgado 1999), but locations including Puerto Rico (Appeldoorn 1985), this is not the typical distribution observed where the Florida Keys (Glazer and Jones 1997) and The wild conch are abundant, and all of the relevant Bahamas (Stoner and Davis 1994; Stoner and Glazer field experiments indicate that releases at natural 1998) have not been consistent. While summer is densities will yield the highest survivals. Spring probably the poorest time for release, optimal seasons (2008) concluded that 1 juvenile/m2 would be the are likely to vary at different potential release sites ideal density for her releases but the observed sur- depending upon seed stock size and the predator vivals in a two-week-long experiment were very low suites in those locations. (see below). Other temporal patterns in queen conch survivor- Other questions related to optimizing releases of ship have been identified. Glazer and Jones (1997) cultured queen conch include how best to scale found that daily losses to predation in short-term releases over both time and space. How many conch experiments during the period approaching a new need to be released at one location at one time? moon in January were three times higher than the Would releases in multiple aggregations result in losses near the following full moon period (see below, higher survivals per capita than one or two large and Table 4). Consequently, Spring (2008) made her releases? Would repeated releases provide a hedge initial field releases near the full moon. Releases at against temporal variables in predation? None of these particular times of day might also improve the early, questions have been studied for queen conch. The short-term survival of queen conch released into the general conclusion is that release strategy for queen field. Small juveniles are most active at night (Randall conch has not been perfected and it is likely that the 1964; Sandt and Stoner 1993) and this might be the best strategies will vary with location, time of year, best time for releases. Given the importance of accli- and size of the seed stock. A summary of the many mation to field environment, it has been suggested factors that affect survival of cultured queen conch is that releases of hatchery-raised queen conch should be provided in Table 3. made after a minimum of two weeks held in field enclosures where the conch are protected from preda- Experience with free-ranging conch tors, and even longer periods may be crucial (see Spring 2008). Mortalities observed in field tagging experiments Numerous investigations have been designed to Density and distribution of released juveniles explore the survival of free-ranging juvenile queen Observations on the distribution of juvenile queen conch in the field. Such studies began with hatchery- conch in the field indicate that conch less than three reared conch at the University of Puerto Rico in the years old are highly aggregated (e.g., typically 1–2 early 1980s, continued into the 1990s, and the most 2 juveniles/m in seagrass meadows of The Bahamas), recent experiment was conducted near the Caicos and probably benefit on a per capita basis from safety Conch Farm in 2007 as part of a PhD dissertation in numbers. Field experiments with tethered (Marshall (Spring 2008). Tables 4 and 5 provide a summary of 1992; Ray and Stoner 1994; Stoner and Lally 1994) the primary tag and recovery experiments. The Tables and tagged, free-ranging juveniles (Stoner and Ray do not include every treatment for every study, but do 1993) show that mortality of young conch is signifi- include results where at least 50 tagged conch were cantly lower in aggregations than in the surrounding released at the beginning of a trial. For economy of areas and inversely density dependent. Controlled space, some treatments with very low returns (high manipulations of conch density in large open-topped mortality) were excluded from the Tables but are enclosures (Stoner 1989) have shown that both growth mentioned where appropriate. and survival are density dependent with peak per- Two indices for mortality which standardize for formance at a density near 2 juveniles/m2. Regardless different numbers of animals marked and different of the exact mechanisms, success in releasing hatch- intervals of tag recovery were used to explore patterns ery-reared queen conch will depend upon making of mortality in free-ranging juvenile queen conch. The those releases in the most appropriate habitats with first is simply the observed loss of conch attributed to sufficient densities and total numbers to swamp the mortality divided by the number of days since the effects of predators. Some have argued that predation release date, giving a percent of population lost per rates might be lowest if conch are widely dispersed at day. The second is the standard index of instantan- low density so that predators do not find them (Jory eous mortality (M) applied in earlier investigations for REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 11

Table 3: Critical factors that affect survival of hatchery-reared queen conch released into the field. All of these variables, choices, and procedures must be optimized to achieve survival rates for “successful” stock enhancement or rehabilitation. Critical Factor Effects Pre-release variables (hatchery conditions) Associated effects on seed quality Water quality Conch size Substratum quality Shell form and strength Food quality and quantity Growth potential Water flow Adaptations for feeding, predator recognition, and avoidance Conch density Survival in the field Behavioral conditioning Added cost to produce seed stock Culling or not culling during culture Plan for genetic integrity (not discussed here) Site selection for releases Associated effects on conch after release Water quality Availability of appropriate diet Water flow Acceptable growth rates Depth Continued development of appropriate shell form and strength Sediment quality Ability to avoid predation via sheltering Food quality and productivity Increased survival rate Presence of conspecifics Predator abundance Shelter Competitor abundance Release methods Associated short- and long-term effects Size of conch at release Reduce predation during most vulnerable sizes Stocking density Achieve “safety in numbers” Total numbers released Avoid periods of highest predation and lowest growth potential Repetition of releases Hedge bets on survival with multiple releases Timing of releases time of day, lunar phase, tidal Provide seed conch with a protected acclimation period to phase, season habitat and foods Increase survival rate Temporary protection/conditioning in enclosure Cost to enhance to specific stage or size Although site selection can be made with preliminary transplant experiments, many unknowns remain with regard to necessary pre-release behavioral conditioning and optimal release strategies. Costs to produce seed stock and cost-per-individual to enhance will decrease in inverse proportion to survival rate. queen conch (e.g., Stoner and Glazer 1998) unprotected, into the natural environment does not ÀÁ appear to be a practical way to support such a fishery. M ¼ log n1– log n2 =y e e This conclusion is tentative, being based on prelimin- where n1 and n2 are the numbers of individuals at ary data, and does not take into account specific time 1 and 2 (e.g., number at release and recovered measures that could be developed to enhance juvenile later), and y is the fraction of a year between the two survival.” They went on to state that strong manage- observations. ment measures are critical for a sustainable fishery of While both indices were calculated for each of the queen conch. Ironically, instantaneous rates of mortal- studies explored, the first index is most useful (and ity (M) calculated for the Appeldoorn and Ballantine intuitive) where the observation intervals were short study are among the lowest recorded to date for free- and relatively similar (13–22 days) immediately fol- ranging queen conch juveniles in the first 2-3 weeks lowing releases of conch (Table 4). Instantaneous after release (Table 4). mortality is less intuitive, but compensates well for As discussed earlier, mortality of queen conch can differences in the duration of field experiments. vary substantially with size at release, release location, One of the earliest unprotected releases of hatch- season, and effects of hatchery-rearing. All of these ery-reared conch into the field was conducted by effects are apparent in Tables 4 and 5. Very high val- Appeldoorn and Ballantine (1983). They released ues for M (e.g., M > 4.0) have been observed in every 150 conch in a size range of 20–50 mm SL and experiment, long or short term, with conch smaller tracked survival for two months (Tables 4 and 5). than 50 mm SL. Mortalities for conch >50 mm SL, Based upon observations of 1.3–2.4% mortality per and especially those >90 mm in long-term experi- day, mostly caused by crustaceans, they concluded: ments, have been substantially lower in certain cases, “At the above rate of mortality, the survival of 1 mil- but results have been highly variable depending upon lion half-year-old conchs would be on the order of the location and time of year, and other circumstances 0.1% by age 2.5 [years]. The seeding of small conch, (Table 5). For example Glazer and Jones (1997) 12 A. STONER

Table 4. Survivals and mortality rates (see text) observed for hatchery-reared and wild juvenile queen conch in short-term field experiments with tagged free-ranging individuals. Initial shell Conch Initial Duration Survival Mortality/ Instantaneous Reference month released length (mm) type number (days) (%) day (%) mortality Appeldoorn and Ballantine (1983) April 20–50 (41) H 56 14 73 1.93 8.20 October 20–50 (41) H 97 14 78 1.57 6.48 Appeldoorn (1984) January 22–35 (29) H 112 17 68 1.88 8.28 January 35–57 (40) H 112 17 83 1.00 4.00 Appeldoorn (1985) January 22–55 (35) H 234 17 82 1.08 4.34 March 20–54 (33) H 102 13 52 3.75 18.92 July 23–45 (34) H 77 14 15 5.86 50.73 Sandt and Stoner (1993) January 35–54 w 500 21 50 2.38 12.05 Glazer and Jones (1997) April 55–65 H 76 22 62 1.72 7.97 July 55–65 H 80 12 77 1.93 8.25 October 55–65 H 73 18 84 0.89 3.64 April 75–85 H 84 22 77 1.03 4.25 July 75–85 H 80 12 63 3.08 14.30 October 75–85 H 84 18 75 1.38 5.83 January 75–85 H 161 16 63 2.33 10.64 Full moon – January 65–75 H 60 13 88 0.92 3.48 New moon – January 65–75 H 60 11 68 2.91 12.63 Spring (2008) all started in March 40–50 (43) H 324 15 44 3.73 19.73 70–80 (74) H 324 15 52 3.20 15.98 40–50 (43) HC 195 15 23 5.13 36.23 70–80 (74) HC 236 15 48 3.47 18.14 40–60 (46) w 100 15 43 3.80 20.54 70–90 (77) w 100 15 36 4.27 24.86 Results are shown for tag recoveries made at 22 days or less. Shell lengths are shown as ranges and (means) as possible from the primary sources. Conch types are hatchery-reared without conditioning (H), hatchery-reared plus two weeks of field conditioning (HC), and wild conch (w). showed that conch released in the Florida Keys in the Glazer 1998; Stoner 2003) (i.e., site selection, season, fall had relatively low mortalities, but average mortal- lunar phase, replicated and random block design, ities for conch released in winter, spring, and summer conch marked with metal tags for re-location with seasons ranged from 89 to 94% in trials running metal detectors, conch density, etc.). The experiment <3 months. The only values for M < 2.0 have been was designed to compare three different conch treat- reported for conch >100 mm SL; all of those have ments: wild conch, hatchery-reared conch conditioned been for wild conch (Iversen et al. 1986; Stoner and with two weeks spent in large enclosures in a natural Ray 1993; Stoner and Davis 1994), excepting one 3- seagrass habitat, and hatchery-reared conch with no month-long interval for hatchery-reared conch (Stoner field conditioning. More than 600 conch (total) in and Davis 1994). each of two size classes (40–50 mm, 70–90 mm) were Only two studies have compared the survival of tagged and released in replicated seagrass plots near hatchery-reared and wild queen conch in a free-rang- the Caicos Conch Farm known to have served as nur- ing experiment. In the first and largest study to date, sery habitat for queen conch in the past. It was Stoner and Davis (1994) released >5000 conch in expected that the larger size class would have lower each class in an experiment running for 7.5 months. vulnerability to predators than the small size class, Mortality in both wild and hatchery-reared conch was and that conditioned conch might survive better than high during the first two months (in summer), and cultured conch without conditioning. Interestingly, then decreased in both conch types (Table 5); how- Spring found no significant differences in mortality ever, mortality rates consistently ran from 0.8 to 5 among any of the treatments, either conch type or times higher in hatchery-reared conch than wild size class (see Table 4). Based upon those results, conch in fully interspersed populations. More recently, Spring concluded that “The short-term goal of Spring (2008) conducted one of the most rigorous restocking, survival of the hatchery-reared organisms short-term release experiments at a field site near the approximately two weeks after release, has been Caicos Conch Farm in 2007, attempting to follow the achieved using methods which minimize predation very best release strategy based upon earlier recom- after release, including site selection, stocking density, mendations (Glazer and Jones 1997; Stoner and timing of release, and release techniques.” While that REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 13

Table 5. Survivals and mortality rates (see text) of hatchery-reared and wild juvenile queen conch in long-term field experiments with tagged free-ranging individuals. Initial Shell Hatchery Mortality/Day Instantaneous Reference Month released length (mm) or wild Initial number Duration (days) Survival (%) (%) Mortality Appeldoorn and Ballantine (1983) April 20–50 (41) H 41 42 5 2.26 26.24 October 20–50 (41) H 47 46 36 1.39 8.06 Appeldoorn (1984) January 22–35 (29) H 76 39 9 2.33 22.32 January 35-57 (40) H 93 39 23 1.99 13.93 Appeldoorn (1985) January 22–55 (35) H 183 39 13 2.23 19.01 March 20–54 (33) H 52 43 17 1.93 14.88 July 23–45 (34) H 11 41 27 1.78 11.57 Iversen et al. (1986) (all started in March) 70–89 w 48 180 10 0.50 4.59 90–109 w 120 180 37 0.35 2.03 110–129 w 20 180 60 0.33 1.03 130–169 w 36 180 86 0.078 0.30 Coulston et al. (1987) February 40–60 H 65 90 5 1.05 12.47 Stoner and Ray (1993) Repeated tag recoveries September (103) w 1012 30 74 0.87 3.66 October (106) w 749 85 58 0.49 2.34 January (114) w 434 80 56 0.55 1.56 April (133) w 308 185 71 0.16 0.67 Stoner and Davis (1994) Repeated tag recoveries April (101) H 5092 64 21 1.23 8.90 June (105) H 963 100 63 0.37 1.70 September (112) H 575 66 79 0.32 5.07 April (101) w 5033 64 42 0.91 4.95 June (108) w 1992 100 81 0.19 0.76 September (117) w 1587 66 86 0.21 0.84 Glazer and Jones (1997) April 55–65 H 76 82 7 1.13 12.11 July 55–65 H 80 84 15 1.01 8.24 October 55–65 H 73 72 22 1.08 7.69 April 75–85 H 84 82 5 1.15 13.55 July 75–85 H 80 84 11 1.05 9.49 October 75–85 H 84 72 40 0.83 4.58 January 75–85 H 161 86 8 1.07 10.68 Results are shown for tag recovery intervals running for four weeks or longer. Some of these results represent extensions from short earlier intervalsof tag recovery reported in Table 4. Shell lengths are shown as ranges and (means) as possible from the primary sources. conclusion may be correct statistically the study has of conch size. Values for M near the mean rate some serious shortcomings. First, only 25 wild conch observed in this experiment (22.5) mean that >80% and 81 hatchery-reared conch were placed in each of all conch would be lost in one month after release, replicate grid, and variation in survival among the and the optimistic conclusion that releases of small replicates was very large (e.g., survival of large hatch- hatchery-reared queen conch can be used to enhance ery-reared conch ranged 12–78%). This level of vari- and restore depleted stocks seems premature. ation, of course, caused the lack of statistical As noted earlier, there is some evidence that queen difference among treatments. Second, while mortality conch mortality varies with season (Tables 4 and 5). has always been highest in the earliest post-release For example, Appeldoorn and Ballantine (1983) periods (Tables 4 and 5), the mortality rates reported observed higher rates of mortality in hatchery-reared by Spring (2008) are among the highest ever reported conch in October than in April, and high mortalities in a field experiment with free-ranging queen conch have been associated with warm summer months (Table 4). Even the lowest value for M (15.98) was when predators might be most active or abundant higher than all values for earlier experiments, except (Appeldoorn 1985; Glazer and Jones 1997); however, two calculated for experiments conducted by very high values for M have also been reported in Appeldoorn (1985) with much smaller conch long-term tag recovery intervals in January (20–54 mm SL). Also, it is curious that wild conch (Appeldoorn 1984, 1985), and low values have been had average instantaneous mortality rates higher than reported in summer months (Stoner and Davis 1994) hatchery-reared conch without conditioning and (Table 5). With the available data on seasonal vari- unlike all earlier studies there was no observed effect ation in mortality it is difficult to make any 14 A. STONER generalizations about optimal time for releasing hatch- for hatchery-reared conch was 1.70 during one 100- ery-reared conch. day interval. These conch, however, were large (aver- Experiments with free-ranging queen conch, like age ¼ 105 mm SL), the conch were released among experiments with either tethered or enclosed conch, abundant native conch of similar size, and over the show that survival is critically dependent upon loca- entire 7.5-month experiment the population of hatch- tion, and the mechanisms for those differences are ery-reared conch was reduced to 9% of the original not well understood. In addition to physical factors release. During the same period, the tagged wild discussed earlier, conch density and abundance appear conch population was reduced to 27%. This release to have large effects on likely survival. Stoner and Ray experiment resulted in one of the very best survivals (1993) released >1000 tagged juvenile wild conch in of hatchery-reared conch. Unless the cost of seed each of three locations with seemingly equal charac- stock can be kept low, hatchery-related deficits in the teristics of depth, tidal flow, seagrass cover and food seed can be eliminated, and release methods can be availability in and around a well-studied nursery perfected, mortality rates in hatchery-reared queen ground near Lee Stocking Island, The Bahamas, and conch will continue to be the primary obstacle to tracked those releases for over one year. Lowest mor- effective stock enhancement. talities occurred in the center of the nursery ground Other more anecdotal observations provide insight where many thousands of juvenile conch were into the likely success or failure of releasing hatchery- observed in densities of 1–2 individuals/m2. M values reared queen conch. First, tens of thousands of hatch- ranged 3.66–0.67 (Table 5), while rates at the other ery-reared juvenile queen conch, mostly larger than two sites, less than 1.5 km distant from the nursery 100 mm shell length, were released into Florida Keys center, were more than twice the rates observed in the nursery grounds in the 1990s (the hatchery was oper- nursery (Stoner and Ray 1993). As discussed earlier, it ated from 1991 to 1998); and only one individual with is clear from this and other experimental studies that a flared shell lip was observed subsequent to release survival in queen conch juveniles is density-dependent (G. Delgado and R. Glazer, Florida Fish and Wildlife (Ray and Stoner 1994). Conservation Commission, pers. commun.). They note that other conch likely reached that stage, but they were not observed. At the Caicos Conch Farm Impacts of observed mortality rates and cost- excess production of queen conch juveniles (mostly to-enhance 50–80 mm SL) were released into adjacent shallow Instantaneous mortality rates (M) observed for free- habitats between 2002 and 2007 with numbers esti- ranging queen conch juveniles (20–169 mm SL) span- mated at one half million conch per year (Spring ning a wide variety of locations and seasons vary 2008); however, most of these conch were culled from from <0.5 to >50 (Table 4 & 5). These values need to culture because of slow growth or other defects (J. be interpreted with regard to stock enhancement and Davis, pers. commun.), and it is completely unknown rehabilitation goals, considering both ecological and if any of these releases contributed to the local popu- economic feasibility: lation of adults. Similarly, millions of larvae at various First, values for M observed in field experiments stages were released from the Farm, with no know- can be translated to more intuitively understandable ledge of their survival. Unfortunately, the entire values. For example, the lowest value for M in Table 4 understanding of hatchery releases is confined to the (3.48) translates to 25% loss of a released population handful of experiments discussed above. in the first 30 days after release while high values near At the end of the queen conch hatchery program 20 mean that >80% of releases would be lost in the in the Florida Keys in 1998, Glazer and Delgado same period. As can be expected, observations on (1999) evaluated the cost of various release strategies larger conch over longer periods yielded lower M val- over the growth range that they observed empirically. ues. Wild conch released at 90–133 mm SL yielded To insure 50,000 individuals surviving to 145 mm many M values <2.4 (Table 5). An M value equal to SL, the most cost-effective release strategy was to 2.0 means that only 15% of the released conch will be release juveniles at 105 mm SL. The cost per survivor lost in 30 days. In fact, low mortalities have been under that scenario would be $9.06 ($13.96 in 2019). observed for hatchery-reared conch in particular cir- Released at 40 mm SL, the cost per survivor was esti- cumstances. For example, in the long-term field mated to be $76.84 ($118.43 in 2019) because of release experiment conducted by Stoner and Davis high mortality at the smaller size. It can be noted (1994) in a well-studied nursery ground, the M value that 145 mm is 50-75 mm smaller than typical REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 15 market size and at least two years short of sex- Another tag and recovery experiment (Stoner and ual maturity. Davis 1994) beginning with 101 mm SL wild conch in Ten years later, Spring (2008) revisited the topic of a known nursery ground resulted in 27% survival cost, considering improvements to husbandry and with conch averaging just 129 mm SL after release strategies. She suggested that a good way to 7.5 months. While survival rates can be expected to look at cost is to consider first the desired outcome in increase with increasing size of conch beyond the sizes terms of survival to a particular age or size whereby: observed, these empirical studies with large numbers Cost to restock ¼ CT=S of conch and long-term tracking, in ideal habitats, surrounded by ambient populations of wild juvenile where C is the cost per individual at release, T is the conch suggest that Spring’s (2008) likely costs to total number of individuals that need to survive, and restock are wildly underestimated for any conch S is the decimal proportion of survival. Spring used approaching market size. Cost for sexually mature an example of $1.00 seed stock cost, a goal of 10,000 conch would be substantially higher. Naturally, if a conch surviving, and a 20% survival rate. In this case, stock has been completely decimated, extreme cost for the cost for survivors (10,000 from 50,000 seed stock) restoration may be the only solution. is $50,000 or $5.00 per survivor. She notes that any The issue of production cost versus size and field improvement in survival will yield lower cost and that survival remains one of the most difficult problems 50% survival might be possible with the highest qual- associated with any stock enhancement or rehabilita- ity seed stock. While short-term survivals often exceed tion effort. It is now abundantly clear that finding an 20% (see Table 4), empirical, long-term survival data acceptable balance point between cost of culturing are not available for queen conch reaching market queen conch to a specific size and size-specific field size. The longest tag and recovery experiment with survival is difficult to achieve. The challenge is exacer- highest observed survivals ran for 380 days for >1000 bated by the fact that hatchery-reared queen conch wild conch starting at 103 mm SL in ideal nursery set- tend to accumulate culture-related deficits in both ting surrounded by other free-ranging wild conch morphology and behavior as discussed above. Cost- (Stoner and Ray 1993). The survival rate over that related concerns voiced earlier about the economic one-year period, with conch averaging 145 mm SL at feasibility of conch aquaculture and stock enhance- the end of the experiment was 18%. This is not far ment (Appeldoorn et al. 1987; Iversen et al. 1987; below Spring’s (2008) hypothetical survival rate; how- Stoner and Glazer 1998) remain relevant today, par- ever, it is likely that seed conch >100 mm SL would ticularly when queen conch products are still being cost more than $1.00 (e.g., Glazer and Delgado (1999) exported from a handful of nations with substan- calculated a cost of $3.39 each for 75 mm seed conch tial fisheries. two decades ago) and the juvenile conch observed in the experiment (Stoner and Ray 1993) were still far Future prospects for culturing conch from market size or maturity. Given the costs discussed above, some evaluation Fisheries enhancement and business- of current market value for conch meat is warranted. related prospects In a survey of 10 retail markets in the southeast Several possible goals associated with hatchery pro- United States, the average price for cleaned queen duction of queen conch are outlined in Table 6. The conch meat without skin was $21 per pound (range = original goal for most of the programs culturing $17–26 ($46 per kg; range = $37–57). This means that queen conch in the 1970s and 1980s was to rehabili- at least four large conch (100 g of meat/individual) tate and enhance rapidly declining populations would yield about $21, or about $5 each, and if the observed throughout the Caribbean region. (One conch were small (e.g., 50 g meat weight; see below) exception was the Caicos Conch Farm which was the retail market value would be half that value. focused on culturing conch directly for marketable When fishers receive less than half the retail price for products.) To achieve successful population restoration their catches, an individual wild conch, before clean- (Goal 1), seeded conch need to grow and survive at ing, has a value of just $1.25–2.50 each, and it is diffi- acceptable rates and reach sexual maturity in densities cult to envision how cultured conch can compete with sufficient for reproduction. To date, there has been no this price while wild fisheries exist unless the cultured successful demonstration of a free-ranging hatchery- conch can be marketed as an expensive spe- reared conch achieving reproductive status in the cialty product. field. Stock enhancement (Goal 2) is slightly less 16 A. STONER

Table 6. Advantages and challenges associated with different goals intended for cultured queen conch. Intended Goal Advantages Challenges 1. Population restoration Land-based costs are limited to Need the highest seed stock quality producing high-quality seed stock at greatest cost Possible community-based support Site selection & release strategy is and participation critical for some proportion of the Intended preservation and seed stock reaching reproductive conservation of native stock and age local fishery-based income Need genetic compatibility with the native stock if extant No control of predators Documenting contribution of the seed stock to the population. No control of predators Cost per spawner could be very high Same challenges as with natural fisheries management Government subsidy likely to be needed

2. Fisheries stock enhancement Land-based costs are limited to Same challenges as above (seed producing high-quality seed stock stock quality, predation, genetic Possible community-based support concerns), but with less concern and participation that seed stock contribute Intended local fishery-based income to the reproductive stock Cost per harvested conch could be very high Most of the same challenges associated with natural fisheries management Government subsidy likely to be needed

3. Ranching for direct harvest Land-based costs limited to Same as above – need for high- producing adequate seed stock quality seed, predation, costs per Possible community-based support harvested conch. and participation Acquiring rights for seed placement No need for conch to reach sexual may be more complex than above maturity if harvest is proprietary Possible poaching

4. Field grow-out in pens Land-based costs less than intensive Site selection and lease requirements culture for pastures Partial control of predation and diets High labor requirement Possible reduction of feed cost Large costs for field infra-structure Need high-quality seed stock Limited control of predators

5. Land-based grow-out Diet and predation are controlled Large costs for infrastructure, feeds, High growth and survival rates utilities, labor Likely need for specialty markets Each of the five strategies is possible with very high financial backing and tolerance for variable success rates. Costs for infrastructure increase down the list; however, the highest quality seed stock is required for all of the first three goals, which may increase the cost per individual for seed. See text for a discussion of actual successes and failures. complex than stock restoration because a fishery sup- consideration particularly when hatchery-reared indi- plemented with hatchery-reared queen conch might viduals are expected to intermingle with wild stock function primarily on a “put-and-take” basis. The seed (not discussed here, but see Allendorf and Ryman stock, in such a case, does not necessarily need to 1987 for a general discussion, and Delgado et al. reach sexual maturity. Both stock restoration and 2004 and Delgado and Glazer 2007 for direct refer- stock enhancement can help provide long-term sus- ence to queen conch.) tainability of a fishery if well managed and such an It is unlikely that stock restoration could occur effort might garner substantial community support; without substantial financing from public sources, and however, both goals require the very highest quality the same might be true for stock enhancement. of seed stock. Hatchery-related deficits must be Beyond the cost of seed stock, the challenges of man- minimized and release strategies need to be per- aging supplemented stocks are the same as those for fected for acceptable survival rates and cost effec- the sound management of natural stocks. As will be tiveness. Also, genetic integrity needs careful clear from the previous discussion, the challenges REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 17 associated with establishing breeding field populations almost impossible to exclude entirely from pens. Costs with cultured juveniles are very large for queen conch associated with the construction and maintenance of and completely unproven. enclosure structures and predator removals can be Ranching under more controlled environmental high, along with costs for feed. Survivals in enclosures, conditions (Goal 3) is another potential strategy for even over short-term experiments discussed above, growing out cultured queen conch. In this scenario have been substantially lower than 60–90%. hatchery-reared juveniles would be placed in carefully The only commercial-scale enclosure system was selected, perhaps leased, nursery grounds for grow-out developed over several decades at the Caicos Conch to market size. The project could be private or com- Farm. Davis et al. (2010) described the field grow-out munity based, and the stakeholders would have exclu- system at the Farm, and reported that they could pro- sive rights to the harvest. The challenges associated duce 160 mm SL conch in 42 months, and those conch with ranching conch are nearly the same as those for yielded 43 g meats marketed as “Caribbean Queen.” stock enhancement except that the ranched stock Survival rates in the enclosures have never been would not need to reach sexual maturity. The require- reported, and the meat weights reported are very small. ments for very high seed quality remain, and site Wild conch typically grow to 200 mm SL in 3.5 years selection and understanding carrying capacity would and have meat weights close to 80 g (calculated from be critical so that the seed stock thrive and remain equations in Appeldoorn et al. 2017). In some loca- accessible to the “ranchers.” Given that ranching tions, subadult queen conch could be substantially would probably be pursued as a profit-making larger (see Randall 1964; Appeldoorn 1988a; Stoner endeavor, the challenge of cost per harvested conch and Schwarte 1994), yielding 100 g of meat weight. would be paramount, with the added challenges of The Caicos Conch Farm also explored land-based, acquiring exclusive rights to the grow-out area and intensive culture of queen conch juveniles (Goal 5) in the risk of poaching. With the present market value of the mid-2000s. These systems, described by Davis queen conch such an endeavor seems unlikely to be et al. (2010), were shallow concrete ponds 185 m2 in successful unless a specialty market for high-cost, surface area. Culture in pools eliminates most of the high-value ranched products can be developed. A pre- problems associated with predators, allows for strict liminary ranching project was initiated with wild control of water quality and diets, and can help to juvenile conch at Santa Catalina Island, Colombia, in optimize both growth and survival rates. Davis et al. 2006 (Shawl et al. 2007), but no follow-up results have proposed that these ponds could produce 75–80 mm been reported. SL conch for a small-meat product in 20 months, Enclosure-based grow-out of queen conch (Goal 4) along with smaller conch for field seeding and aquar- has been proposed as an alternative to releasing cul- ium trade. This approach, however, requires large, tured conch into a free-ranging population (Davis expensive seawater systems, including waste-water 2005; Davis et al. 2010). The desired purpose for enclo- treatment, high maintenance, and conch growth is sures is to reduce predation on the seed stock, to con- entirely dependent upon artificial feeds. Other prob- trol diet, and to retain the conch on the farm. lems encountered were difficulty with temperature Behavioral and morphological deficits in seed stock control, stunted growth, and susceptibility to disease might not be as critical in an enclosure-based maricul- and mass mortalities observed in 2007. Survivals and ture system as it is for free-ranging conch, and with economic analyses for the Caicos Conch Farm’s heavy feeding it might be possible to push the cultures enclosure- and land-based grow-out systems (Goals 4 for faster growth of meats instead of shell material. & 5) have never been made public, and commercial- Davis (2005) summarized culture methods, including a scale operations at the Farm ceased except for small hypothetical field grow-out system of enclosures, and demonstration projects in about 2008. projected that “Survival in the sea pasture should be It is possible that some of the problems encoun- 60–90%, depending on the effectiveness of the cages or tered in pen- or land-based queen conch culture fences, the availability of feed, and the success of the might be overcome with further research on improved predator trapping program.” She discussed predator or different holding systems, diets and other hus- control methods that can be used in and around large bandry methods. Long ago, Jory et al. (1984) proposed enclosures, including fence structures and predator that it might be possible to grow out juvenile conch traps; however, it is clear that an enclosure-based grow- in floating raft systems. This has never been tried, but out system requires routine and diligent removals of such a system requires another form of infrastructure, predators such as octopus, crabs and lobsters that are high feeding and maintenance costs, and concerns 18 A. STONER about environmental impacts near a large raft system. behavior, Davis et al. (1984) maintained 50–200 adults The fact remains that a supply of wild conch meat in a 1600 m2 field enclosure that produced a steady continues to be available at a “reasonable” cost (see supply of eggs over two breeding seasons for the above) from a handful of better managed fisheries in Caicos Conch Farm. Also, Weil and Laughlin (1984) the Caribbean region such that an aquaculture prod- placed reproductive age queen conch in a large pen, uct will be difficult to justify in business terms at pre- observing that a single female could produce 25 egg sent. As demand for product and/or value grows, it is masses per season at their site in Venezuela. possible that conch aquaculture might become finan- Furthermore, it appears likely that cultured conch are cially viable; however, with some signs that fisheries capable of reproduction. Julie Davis (pers. commun.) management for natural stocks of queen conch might reported that mating was observed among hatchery- be improving, at least in certain nations, the prospect reared conch in large field enclosures. Those conch for profit from aquaculture might lie with other more were placed in the enclosures 27 months earlier at promising seafood species. 70 mm SL, and had achieved well-formed shell lips. While mating was observed with some regularity, egg- laying was never observed. Also, Shawl and Davis Research value of queen conch culture (2004) reported that one cultured queen conch female The prospects for economically feasible stock held in a large laboratory tank with three cultured enhancement for queen conch seem dim and conch males (all with no field experience whatsoever) pro- aquaculture may not make sense from a business per- duced four egg masses. Shawl and Davis cultured lar- spective at present; however, there are important rea- vae from one of the egg masses through sons for continuing research on queen conch culture. metamorphosis indicating that cultured conch can Cultured conch have been critical for both laboratory produce viable offspring. This represents significant and field research designed to understand the basic progress in the direction of aquaculture for queen life history and ecology of the species. For example, conch, but the problem of survival in the field for five laboratory-produced veligers have been used to inves- years prior to reproduction remains a significant chal- tigate relationships between larval growth and tem- lenge for fisheries restoration. perature and salinity (Davis 2000b), for exploring Lastly, some form of culture might be necessary in trophic relationships (Davis et al. 1996; Patino-S~ uarez the future to preserve particular genetic stocks that and Aldana-Aranda 2000) and for both laboratory are threatened with extinction. For example, hatchery and field experiments aimed at understanding patterns production is a critical component of the recovery of vertical migration, larval transport and settlement plan for (Haliotis sorenseni) on the west coast of (Barile et al. 1994; Davis and Stoner 1994; Noyes North America (Rogers-Bennett et al. 2016). As of 1996). Also, because newly settled juveniles are very 2019, only about 3,000 white abalone exist in the wild difficult to locate and collect in the field, many of the and the captive breeding program is probably the only studies on the distribution (Ray et al. 1994; Ray and mechanism preventing total extinction (L. Rogers- Stoner 1995b), trophic relationships (Ray-Culp et al. Bennett, California Department of Fish and Wildlife, 1997, 1999; Moreno de la Torre and Aldana-Aranda pers. commun.). To date viable populations of queen 2007), and behavior (Sandt and Stoner 1993) of the conch remain in the Caribbean region, and other earliest post-settlement stages have depended upon management actions, discussed below, are more likely the availability of cultured conch. Cultured queen to be successful and practical in the near term. conch larvae and juveniles may be particularly import- Nevertheless, experience with cultures will prove valu- ant in new physiological and ecological studies related able if populations decline to levels where natural to current topics such as ocean acidification and cli- recovery is impossible. mate change (e.g., Chavez-Villegas et al. 2017). Cultures of queen conch require a steady source of Population restoration approaches not requir- egg masses. Until recently, availability of eggs gathered ing culture from the wild has not been a major problem, but as Translocation of sexually mature queen conch might populations of adult conch continue to decline in be a logical approach for restoring depleted popula- some areas it will become increasingly difficult to find tions, without the cost and unproven hatchery- and to justify collecting the necessary number of eggs dependent approach to stock restoration. Taniguchi for culture. Toward the goal of acquiring eggs for cul- et al. (2013) reported some success in transplanting ture and making observations on reproductive pink abalone (Haliotis corrugata) to replenish REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 19 reproductive populations, but less so with green aba- laying in their relatively short-term experiments. They lone (H. fulgens).The primary benefit of translocation also observed that conch translocated from areas with is concentrating adult queen conch to densities high adult densities had higher mating frequencies required for mate-finding and egg laying (56–100 than those from low-density environments and sug- adults/ha) (see Stoner and Ray-Culp 2000; UNEP gested that this was associated with delayed functional 2012). Delgado et al. (2004) moved 96 adult conch maturity at low density (see Appeldoorn 1988b, 1995 from two inshore areas near the Florida Keys (where for a discussion of density-dependent oogenesis). poor water quality was effecting reproduction) to two Therefore, it is possible that carefully designed con- clean offshore locations. The gonads of translocated centrations of adults could lead to increased repro- conch developed and egg-laying was observed within ductive behavior in queen conch. The questions three months. A larger translocation experiment was remaining are: Can habitats suitable for reproduction conducted in 2001 (Delgado and Glazer 2007) and be predicted reliably? Will larvae produced in a con- conch were monitored with acoustic tags. Based upon centrated population be retained and recruit to the the results of their experiments, Delgado and Glazer desired nursery or fishing ground or will they be lost concluded that moving adult conch to healthy spawn- from the system? Can an “egg farm” be used to ing grounds is a more dependable and lower cost restore or enhance a target population? strategy for restoring a conch population than releas- Rather than concentrating adult conch, Laughlin ing hatchery-reared juveniles and there were no and Weil (1983) moved 2,000 egg masses per year adverse effects on the native conch population. from a protected sector of Islas Los Roques Concentration of large juveniles (e.g., >200 mm SL) (Venezuela) to overfished areas in the upstream direc- might also be effective in certain cases where adults tion. The goal was to increase the numbers of larvae cannot be collected and where the transplanted indi- hatching in an area where they would be retained in viduals will survive and not be harvested. the archipelago rather than being lost downstream. In Concentrating adults in no-take marine protected this way, the area closed to fishing served as an egg areas might be a good strategy if the critical densities source for the fished area. While the concept was there have already declined. Of course, whether or not good in principle, there was no indication of whether the eggs and larvae produced in an enhanced adult or not egg translocation enhanced juvenile or adult population contribute to the intended population will populations in Los Roques. Also, with ever decreasing depend heavily upon the transport of those larvae and numbers of queen conch around the Caribbean subsequent recruitment to the benthos and ultimately region, there are probably few locations where egg to the fishery. The success of such a rehabilitation masses could be harvested for transplanting without strategy will be very difficult to evaluate and require having other negative impacts. With regard to moving long-term monitoring. either eggs or mature adults, it would be important to Transplanting large queen conch poses certain “first, do no harm”. risks. In the early 1990s, biologists moved 300 adults The benefits of areas completely closed to queen and large juveniles from a shallow fishing ground to a conch fishing have been discussed (Stoner and Ray large enclosure (10,000 m2) created in a well-studied 1996; Baqueiro-Cardenas and Aldana-Aranda 2009; reproductive ground offshore from Lee Stocking Kough et al. 2017, 2019; Tewfik et al. 2017), and Island (Bahamas) (Stoner and Sandt 1992). While Stoner et al. (2019) suggested that a large network of foods in the enclosure (primarily red and green mac- closed areas is probably the best management tool roalgae) were abundant at the beginning of the experi- that can be applied for restoration of conch popula- ment, the conch migrated to one side of the enclosure tions in The Bahamas. Closed areas protect concen- (for unknown reasons), they exhausted the food sup- trated populations of the largest and most fecund ply in the critical area, and many died along the fence. adult queen conch necessary for highest reproductive No mating and no egg masses were observed during potential. Of course, even a large closed area like the the subsequent reproductive season (author, unpub- Exuma Cays Land and Sea Park in The Bahamas may lished data). In contrast with this failed experiment, depend upon an upstream source of queen conch lar- Gasciogne and Lipcius (2004) concentrated small vae and be susceptible to overharvesting in the sur- numbers of adult conch in enclosures in shallow- rounding waters (Stoner et al. 2019). Hence, all of the water pens near Lee Stocking Island to track individ- traditional fisheries management tools including ual records of maturity and reproductive behavior. closed seasons, limited entry, gear restrictions, min- They recorded an abundance of mating but no egg imum size or age limits for harvest, fishing quotas, 20 A. STONER and other controls on fishing effort and harvest are behavior necessary for near-natural probability of probably needed to restore severely depleted stocks. survival will continue to be very high. 2. The habitat requirements for juvenile queen conch are very specific, and transplant experi- Conclusions ments show that even similar appearing seagrass Hatchery-based culture of queen conch has been opti- meadows within a few hundred meters of one mized for more than two decades. Queen conch have another can yield very different results in terms very high fecundity, and millions of larvae can be car- of both growth and survival. Despite dozens of ried through to metamorphosis and thousands of studies on conch habitat and movements, site small conch can be produced in modest hatchery set- selection for releases of juvenile conch remains tings so long as egg masses can be found, a good sea- challenging without preliminary experiments. water source is provided, and the established Experience in field enclosures shows that growth protocols are followed meticulously. The real difficulty is density dependent and crowding can become a in enhancing or rehabilitating a queen conch popula- significant problem even when prepared foods are tion with releases of hatchery-reared juveniles begins added to supplement natural foods. Also, queen when cultured conch are taken to the field. The over- conch have few natural defenses from predators arching problem in any release to a free-ranging except for their shells, burial observed in small queen conch population is that natural mortality rates juveniles, and sheltering in complex structures in young queen conch are very high (Appeldoorn that inhibits at least some types of predators. All 1988c; Stoner and Glazer 1998; this analysis), and the of these habitat-related variables, along with goal is to achieve survival rates in released juvenile observed ontogenetic shifts in habitat preferences, conch that match or exceed natural mortality. The are critical in the selection of appropriate habitats relatively slow growth of juvenile conch exacerbates for stocking hatchery-reared queen conch into the mortality problem. the field. There are several factors making hatchery releases 3. Releases need to be conducted with methods of queen conch, especially difficult if not totally designed for maximum likelihood of survival, par- impractical: ticularly in the earliest post-release period. Assuming that hatchery-induced deficits can be 1. Producing juvenile queen conch with an unim- mediated with proper culture conditions and paired probability of survival has been a difficult behavioral training, releases need to be made in challenge. Spring (2008) offered hope for produc- the best possible locations, at optimal time (sea- ing seed conch with natural shell strength and son, tidal phase, and time of day), with appropri- overall form through very specific husbandry ate conch size, density and dispersal pattern, and methods including high water flow, perfect diets with adequate repetition of releases and total and controlled growth rates. Furthermore, Spring numbers. Also, the seed conch will probably need and others (e.g., Delgado et al. 2002) have pro- to be acclimated for some period of time under posed that natural burial and other behavioral protection in an enclosure for acclimation to the attributes can be improved with prerelease expos- local foods and environment. While field experi- ure to natural habitats and predators; however, ments have been conducted on some of these deficits in both behavior and morphology increase subjects, many unknowns remain regarding best rapidly with time in culture and it has been possible release methods. Furthermore, different argued that juvenile conch need to be moved into locations will undoubtedly require different strat- the field at 40 mm SL. Unfortunately, this is com- egies because of different carrying capacities, food pletely incompatible with field experiments and qualities, and predator types or patterns of abun- mortality models that demonstrate the need for dance. It is unlikely that any of these variables are substantial size in seed stock (75–100 mm SL, or static on an inter-annual basis. even larger) to survive in the field. Acceptable 4. Early enthusiasm for releasing hatchery-reared survival rates for seed stock are the major obstacle conch into the field was based, in part, upon the to the cost effectiveness of releasing hatchery- conclusion that market size in queen conch could reared juveniles for stock recovery, stock enhance- be reached at 2.5 years. We know now that legal ment, or even grow-out in field enclosures. The market size in most nations requires three to four cost for rearing conch with size, morphology and years of growth and sexual maturity requires five REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 21

years or more depending upon location. Only one shore-based nursery facilities and field enclosures for queen conch female reared in a hatchery has been grow-out. It can be argued that the numbers of juven- observed laying eggs (in a laboratory tank), and ile conch with characteristics suitable for survival in no hatchery-reared queen conch has ever been the field have never been adequate to conduct a truly observed laying eggs in the field. Fecundity and robust experiment to test the hypothesis that released reproductive potential in queen conch is high, but juveniles can result in stock enhancement or restor- the time to reproductive maturity is long, and ation. Each of the experiments discussed above have substantial densities of mature individuals are involved only hundreds to a few thousand hatchery- needed for mating. Successful reproduction reared individuals, and larger releases were not moni- would, of course, be critical where the goal is to tored. The cost for producing a rigorous test of rehabilitate a lost or severely depleted queen hypothesis, however, will be very large, and only a conch population. It seems likely, based upon one government-sponsored program such as that in the laboratory observation, that cultured conch will Florida Keys or very generous philanthropic donations be able to reproduce but the concept of restoring could fund a full-scale experiment. a reproductive population with cultured individu- There is a long record of restocking efforts with als has not been proven and it would take many hatchery-reared abalone (Haliotis spp.), and because years to document any successes. of life history similarities with queen conch, results 5. Even if all of the challenges in rearing and release with abalone provide useful background for those methods mentioned above can be overcome, the interested in conserving queen conch populations. A fact remains that natural mortality rates observed broad review of the restocking topic led Prince (2004) for queen conch juveniles are extremely high. to conclude that stock enhancement is technically pos- Summaries of survival and mortality of free-rang- sible for Haliotis spp. but unlikely to be profitable ing conch juveniles provided here and in earlier because of slow growth, high predation rates on juve- analyses (Appeldoorn 1988c; Stoner and Glazer niles, 5–10 years to maturity, and density-dependent 1998) show that instantaneous mortality (M) for reproduction, all mirrored in the life history of queen 100 mm conch is likely to be at least 2.0–4.0 conch. Hamiguchi and Kitada (2008) were somewhat under best possible circumstances. This means more optimistic in their assessment of abalone releases that 15–28% of a population placed in the field in Japan spanning more than four decades. They con- can be expected to be lost to predation in the first cluded that survivals are highly variable by site and 30 days after release. As surviving conch grow, year, but that hatchery-reared juveniles could augment mortality rates decline, but at least two years is fisheries in some cases. The consensus, however, is needed for 100 mm SL seed conch to grow to that hatchery-releases of abalone should be limited to market size and a much longer period is needed last-resort recoveries for species that are severely for sexual maturity as discussed above. In any depleted or threatened with extinction, and all other case, with current knowledge of queen conch cul- management practices have been exhausted (Bell et al. ture, grow-out methods, and highest possible 2005). As an example, the most threatened abalone in expectations for field survival, restoration of a North America is Haliotis sorensens (white abalone). queen conch population would be enor- The species is currently listed under the Endangered mously expensive. Species Act of the United States, and a captive breeding program is a critical part of the restoration plan There have been no significant advances and no (Rogers-Bennett et al. 2016). To date, several hundred new attempts to release hatchery-reared conch into individuals have been cultured and released into the the field in the last decade. The long-term endeavor field, but it remains unknown whether the enor- of culturing and releasing conch in the Florida Keys mously expensive captive breeding program can was abandoned in 1998 and replaced with other stock restore populations. (Approximately $10 million dol- rehabilitation endeavors including transplantation of lars has been spent on white abalone culture since the adults and continued total moratorium on conch fish- stock recovery plan was developed in 2002, and ing. The rationale for this decision was discussed by 30,000 juveniles are currently in laboratory holding; Delgado et al. (2004) and Glazer and Delgado (2003). M. Neuman, NOAA, Long Beach, California; Also, the Caicos Conch Farm collapsed financially in pers. commun.). about 2007 after more than 30 years of queen conch Haliotis kamtschatkana (pinto abalone) is another culture and many millions of dollars investment in severely depleted species being cultured for stock 22 A. STONER rebuilding after complete closure of the fishery in the questioned the cost effectiveness of such an endeavor State of Washington. Carson et al. (2019) released and were concerned that any financial benefits would 11,000 pinto abalone from cultures over several go to foreign investors and not to Bahamians (TNC- years in Puget Sound. They reported detection rates Nassau, unpubl. report). After another three decades (not survivals) ranging from 0 to 23% over 10 sites in of culture and field experiments, Appeldoorn’s caution the first post-release year, and abalone in several of regarding the likely role of hatchery production of the sites grew to reproductive size and survived in suf- queen conch for stock rehabilitation remains valid. ficient numbers to create viable spawning populations. Recently, great progress has been made on topics Based upon this success, the pilot program is making directly relevant to fisheries management for natural a transition to hatchery production. stocks of queen conch. For example, we have an One of the largest abalone experiments was carried increasing knowledge of reproductive biology (Stoner out two decades ago. Rogers-Bennett and Pearse et al. 2012a; Boman et al. 2018), the function of mar- (1998) released 50,000 cultured red abalone (Haliotis ine protected areas specific to queen conch (Kough rufescens) at six sites in northern California and <1% et al. 2017; Tewfik et al. 2017; Kough et al. 2019), were recovered two years later. They suggested that fishing impacts on population structure (Stoner et al. survival may have been higher than that reported 2012b, 2019), and regional analysis of molecular gen- because of extreme crypsis and proposed that genetic etics (Truelove et al. 2017) as well as biophysical mod- analyses should be made to determine whether els that help elucidate population connectivity (Kough releases of hatchery-reared abalone contribute to the et al. 2019). These new results combined with trad- adult and breeding stocks. To date, no such analysis itional population models (e.g., Appeldoorn 1987b; has been made for any release of abalone or queen Valle-Esquivel 2003; Ehrhardt and Valle-Esquivel conch. Furthermore, given the lack of demonstrable 2008) and a regionally harmonized management plan field survivals in cultured queen conch and mixed (see Prada et al. 2017) should make it possible, along success reported for Haliotis spp., a close life-history with sufficient political will, to conserve natural popu- analog, we should not translate high success in hatch- lations of queen conch. Wise managers will realize ery production directly into likely success in stock res- that it will be easier to conserve a self-sustaining nat- toration. The conclusion drawn by Bell et al. (2005) ural stock of queen conch than to restore a collapsed for Haliotis spp. seems directly applicable to queen population, and that it is a very large mistake to use conch: “Ultimately, restocking and stock enhancement hatchery-production as an excuse to delay the best should only be used where comprehensive analysis of possible conservation measures for natural fisheries. alternative models for management of self-recruiting Releases of hatchery-reared queen conch should be populations demonstrates that releases of juveniles considered the very last resort for restoring a popula- will deliver greater benefits than other interventions.” tion that has been closed entirely. Long ago, after several years of culture and field experimentation, Appeldoorn et al. (1987) stated that Acknowledgments stock enhancement with queen conch was not yet The author wishes to acknowledge the many collaborators feasible. They stated further that even if enhancement who participated in the research leading up to this review. through culture became practical it would be expen- In particular, these include long-term colleagues Melody sive and “As such, it would only be recommended as Ray-Culp, Nikhil Mehta, Megan Davis, Richard a last resort, for restoring a once productive area that Appeldoorn, and Robert Glazer with whom the author has had collapsed and not recovered after several years of been discussing topics related to queen conch conservation for nearly 30 years. The review also profited from conversa- closure.” Ten years later Iversen and Jory (1997) tions with Julie Davis, Laura Rogers-Bennett, Melissa expressed significant reservations about the feasibility Neuman, and Hank Carson. Colleagues associated with the of stock enhancement with cultured queen conch. non-profit organization Community Conch including Nevertheless, discussion of conch farming and Martha Davis, Catherine Booker, and Karl Mueller contrib- enhancement via hatchery production continues on a uted significantly to thoughts on queen conch biology, management, and population restoration. Gabriel Delgado, regular basis. For a recent example, The Nature Megan Davis, and Martha Davis all offered constructive Conservancy convened a meeting of Bahamian conch criticism on the manuscript. fishers in Nassau in April 2018 to solicit ideas and opinions on management options for the queen conch fishery. One of the topics discussed was the possibility Disclosure statement of farming queen conch. To their credit, the fishers No potential conflict of interest was reported by the author. REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 23

References resources of the lesser antilles region. Rome: FAO Fisheries Report No. 383; pp. 36–106 Adams JE. 1970. Conch fishing industry of Union Island, Arnold WS. 2008. Application of larval release for restock- Grenadines, West Indies. J Trop Sci. 12:279–288. ing and stock enhancement of coastal marine bivalve Alcolado PM. 1976. Crecimiento, variaciones morfologicas populations. Fish Sci. 16(1–3):65–71. de la concha y algunos datos biologicos del cobo, Arnold WS, Blake NJ, Harrison MM, Marelli DC, Parker Strombus gigas L. (Mollusca: Mesogastropoda). Ser ML, Peters SC, Sweat DE. 2005. Restoration of bay scal- Oceanol. 34 Academia de Ciencias de Cuba 36p. lop (Argopecten irradians Lamarck) populations in Aldana-Aranda D, Frenkiel L. 2007. Lip thickness of Florida coastal waters: planting techniques and the Strombus gigas (Mollusca: Gastropoda) versus maturity: a growth, mortality and reproductive development of management measure. Proc Gulf Carib Fish Inst. 58: planted scallops. J. Shellfish Res. 24:883–904. 431–442. Ballantine DL, Appeldoorn RS. 1983. Queen conch culture Allendorf FW, Ryman N. 1987. Genetic management of and future prospects in Puerto Rico. Proc Gulf Carib hatchery stocks. In: Ryman N, Utter F, editors. Fish Inst. 35:57–63. Population genetics and fishery management. Bannister CA, Addison JT. 1997. Enhancing lobster stocks: Washington, DC: Washington Sea Grant; pp. 141–159. a review of recent European methods, results, and future Appeldoorn RS. 1984. The effect of size on mortality of prospects. Bull Mar Sci. 62:369–387. small juvenile conchs (Strombus gigas Linne and S. costa- Baqueiro-Cardenas E, Aldana-Aranda D. 2009. Histories of tus Gmelin. J Shellfish Res. 4:37–43. success for the conservation of populations of queen Appeldoorn RS. 1985. Growth, mortality and dispersion of conch (Strombus gigas). Proc Gulf Carib Fish Inst. 62: juvenile laboratory-reared conchs, Strombus gigas and S. 306–312.. costatus, released at an offshore site. Bull Mar Sci. 37: Barile PJ, Stoner AW, Young CM. 1994. Phototaxis and ver- 785–793. tical migration of the queen conch (Strombus gigas Appeldoorn RS. 1987a. Assessment of mortality in an off- Linne) veliger larvae. J Exp Mar Biol Ecol. 183:147–162. shore population of queen conch, Strombus gigas L., in Bell JD, Rothlisberg PC, Munro JL, Loneragan NR, Nash southwest Puerto Rico. Fish Bull. 85:797–804. WJ, Ward RD, Andrew NL. 2005. Stocking and Stock Appeldoorn RS. 1987b. Practical considerations in the Enhancement of Marine Invertebrate Fisheries. Adv Mar 49 assessment of queen conch fisheries and population Biol. :, 374p. dynamics. Proc Gulf Carib Fish Inst. 38:307–324. Berg CJ. Jr. 1976. Growth of the queen conch Strombus Appeldoorn RS. 1988. Age determination, growth, mortality gigas, with a discussion of the practicality of its maricul- 34 – and age of first reproduction in adult queen conch, ture. Mar Biol. (3):191 199. Strombus gigas L., off Puerto Rico. Fish Res. 6(4): Blankenship HL, Leber KM. 1995. A responsible approach – to marine stock enhancement. Amer Fish Soc Symp. 15: 363 378. – Appeldoorn RS. 1988b. Fishing pressure and reproductive 167 175. Boman EM, Graaf MD, Nagelkerke LAJ, Stoner AW, potential in strombid conchs: is there a critical stock Bissada CE, Avila-Poveda OH, Baqueiro-Cardenas ER, density for reproduction? Mem Soc Natur LaSalle. 48: Smaal AC. 2018. Variability in size at maturity and 275–288. reproductive season of queen conch, Lobatus gigas Appeldoorn RS. 1988c. Ontogenetic changes in natural mor- (Gastropoda: Strombidae) in the wider Caribbean region. tality rate of queen conch, Strombus gigas (Mollusca: Fish Res. 201:18–25. Mesogastropoda). Bull Mar Sci. 42:159–165. Brownell WN. 1977. Reproduction, laboratory culture and Appeldoorn RS. 1995. Potential depensatory mechanisms growth of Strombus gigas, S. costatus and S. pugilus in operating on reproductive output in gonochoristic mol- Los Roques, Venezuela. Bull Mar Sci. 27:668–680. luscs, with particular reference to strombid gastropods. 199 – Brownell WN, Stevely JM. 1981. The biology, fisheries, and ICES Mar Sci Symp. :13 18. management of the queen conch. Strombus Gigas. Mar Appeldoorn R, Baker N, Ojeda E, Ruiz H. 2017. Assessing Fish Rev. 43:1–12. shell-based legal-size compliance when queen conch are Carson HS, Morin DJ, Bouma JV, Ulrich M, Sizemore R. not landed in the shell: a Puerto Rico case study. Proc 2019. The survival of hatchery-origin pinto abalone 69 – Gulf Carib Fish Inst. :248 256. Haliotis kamtschantkana released into Washington Appeldoorn RS, Ballantine DL. 1983. Field release of cul- waters. Aquatic Conserv Mar Freshw Ecosyst. 29: tured queen conch in Puerto Rico: Implications for stock 424–441. doi:10.1002/aqc.3004. 35 – restoration. Proc Gulf Carib Fish Inst. :89 98. Chavez-Villegas JF, Enrıques-Dıaz M, Aldana-Aranda D. Appeldoorn RS, Castro-Gonzalez ER, Glazer RA, Prada M. 2017. Efecto de temperature y la acidificacıon en larvas 2011. Applying EBM to queen conch fisheries in the de Strombus gigas (Mesogastropoda: Strombidae). Rev Caribbean. In: Fanning L, Mahon R, McConney, P. edi- Biol Trop. 65:505–515. tors. Towards marine ecosystem-based management in Coulston ML, Berey RW, Dempsey AC, Odum P. 1987. the wider Caribbean. Amsterdam: Amsterdam Univ. Assessment of the queen conch (Strombus gigas) popula- Press; pp. 177–186 tion and predation studies of hatchery reared juveniles in Appeldoorn RS, Dennis GD, Monterrosa-Lopez O. 1987. Salt River Canyon, St Croix, U.S. Virgin Islands. Proc Review of the demersal resources of Puerto Rico and the Gulf Carib Fish Inst. 38:294–306. Lesser Antilles region. In: Mahon R. editor. Report and Creswell LR. 1984. Conch mariculture in the Caribbean proceedings of the expert consultation on shared fishery region and its potential for Bermuda., pp. 133–141. In: 24 A. STONER

Sleeter TD, editor. Assessment of the potential for aqua- Davis M, Stoner AW. 1994. Trophic cues induce metamor- culture in Bermuda. Ferry Reach, Bermuda: Bermuda phosis of queen conch larvae (Strombus gigas Linnaeus). J Biological Station for Research. Exp Mar Biol Ecol. 180:83–102. Creswell LR, Davis M. 1991. Queen conch: the well-bred Delgado GA, Bartels CT, Glazer RA, Brown-Peterson NJ, queen of the Caribbean. World Aquacul. 22:28–38. McCarthy KJ. 2004. Translocation as a strategy to D’Asaro CN. 1965. Organogenesis, development, and meta- rehabilitate the queen conch (Strombus gigas) population morphosis in the queen conch, Strombus gigas, with notes in the Florida Keys. Fish Bull. 102:278–288. on breeding habits. J Shellfish Res. 4:23–30. Delgado GA, Glazer RA. 2007. Interactions between translo- Dalton A. 1994. Mariculture of queen conch (Strombus gigas cated and native queen conch Strombus gigas: evaluating L.): development of nursery and grow-out techniques. In: a restoration strategy. Endanger Spec Res. 3:259–266. Appeldoorn RS, Rodriquez BQ, editors. Queen conch Delgado GA, Glazer RA, Stewart NJ. 2002. Predator- biology, fisheries and mariculture. Caracas: Fundacion induced behavioral and morphological plasticity in the Cientifica Los Roques, 356p; pp. 253–260. tropical marine gastropod Strombus gigas. Biol Bull. Davis M. 1994. Mariculture techniques for queen conch 203(1):112–120. (Strombus gigas L.): egg mass to juvenile stage. In: Ehrhardt NM, Valle-Esquivel M. 2008. Conch (Strombus Appeldoorn RS, Rodriquez BQ. editors. Queen conch gigas) stock assessment manual. San Juan, Puerto Rico: biology, fisheries and mariculture. Caracas: Fundacion Caribbean Fisheries Management Council, 128p. Cientifica Los Roques, 356 p.; pp. 231–252 Ellis CD, Hodgson DJ, Daniels CL, Boothroyd DP, Davis M. 2000a. Queen conch (Strombus gigas) culture tech- Bannister RCA, Griffiths AGF. 2015. European lobster niques for research, stock enhancement and growout stocking requires comprehensive impact assessment to 27 – markets. In: Fingerman M, Nagabhushanam R. editors. determine fishery benefits. ICES J Mar Sci. :i35 i48. Recent Advances in Marine Biotechnology. Vol 4. Enfield Foley JR, Takahashi M. 2017. Shell lip thickness is the most reliable proxy to sexual maturity in queen conch (NH): Science Publishers, Inc; pp. 127–59. Davis M. 2000b. The combined effects of temperature and (Lobatus gigas) of Port Honduras marine reserve, Belize; informing management to reduce the risk of growth salinity on growth, development, and survival for tropical overfishing. Front Mar Sci. 4:179. doi:10.3389/fmars.2017. gastropod veligers of Strombus gigas. J Shellfish Res. 19: 00179. 883–889. Gascoigne J, Lipcius RN. 2004. Conserving populations at Davis M. 2005. Species profile: queen conch, Strombus gigas. low abundance: delayed functional maturity and Allee Southern Regional Aquaculture Center, Publ. no. 7203, effects in reproductive behaviour of the queen conch pp. 12. Strombus gigas. Mar Ecol Prog Ser. 284:185–194. Davis M, Bolton CA, Stoner AW. 1993. A comparison of Glazer RA, Berg CJ. 1994. Queen conch research in Florida: larval development, growth, and shell morphology in 36 – an overview. In: Appledoorn, RS, Rodriquez B. editors. three Caribbean Strombus species. Veliger. :236 244. Queen conch biology, fisheries and mariculture. Caracas: Davis M, Dalton A. 1991. New large-scale culturing techni- Fundacion Cientıfica Los Roques; pp. 79–95. ques for Strombus gigas post larvae in the Turks and Glazer RA, Delgado GA. 2003. Towards a holistic strategy 40 – Caicos Islands. Proc Gulf Carib Fish Inst. :257 266. to managing Florida’s queen conch (Strombus gigas) Davis J, Dyer C, Boettcher A. 2010. Optimizing offshore population. In: Aldana-Aranda, D., editor. El Caracol growout conditions for commercial scale culture of queen Strombus gigas: Concimiento Integral Para su Manejo 41 – conch. World Aquacul. :31 33. Sustentable en el Caribe. Yucatan, Mexico: CYTED, Davis M, Hesse C. 1983. Third world level conch maricul- Programa Iberoamericano de Ciencia y Technologıa para ture in the Turks and Caicos Islands. Proc Gulf Carib el Desarrollo; pp. 73–80 35 – Fish Inst. :73 82. Glazer RA, Delgado GA. Part I – optimizing size at release Davis M, Hesse C, Hodgkins G. 1987. Commercial hatchery of hatchery-raised queen conch outplants: a cost-benefit produced queen conch, Strombus gigas, seed for the approach. Part II – modifications of behavior and research ad grow-out market. Proc Gulf Carib Fish Inst. morphology in hatchery-reared queen conch., 55pp. U.S. 38 – :326 335. Fish and Wildlife Service, Final Report for Project P-1. Davis M, Hodgkins GA, Stoner AW. 1996. A mesocosm File code: FO 633–98-99-F. 1999. system for ecological research with marine invertebrate Glazer R, Jones R. 1997. Temporal factors influencing sur- larvae. Mar Ecol Prog Ser. 130:97–104. vival of hatchery-raised queen conch outplants. 56pp., Davis M, Mitchell BA, Brown JL. 1984. Breeding behavior U.S. Fish and Wildlife Service, Final Report for Project of the queen conch Strombus gigas Linne held in a nat- P-1. File code: FO 507–95-F. ural enclosed habitat. J. Shellfish Res. 4:17–21. Glazer RA, Kidney JA. 2004. Habitat associations of adult Davis JLD, Young-Williams AC, Hines AH, Zohar Y. 2005. queen conch (Strombus gigas L.) in an unfished Florida Assessing the potential for stock enhancement in the case keys back reef: applications to essential fish habitat. Bull of the Chesapeake Bay blue crab (Callinectes sapidus). Mar Sci. 75:205–224. Can J Fish Aquat Sci. 62(1):109–122. Goodwin MH. 1983. Overview of conch fisheries and cul- Davis M. 1992. Predation of hatchery-reared juvenile queen ture. Proc Gulf Carib Fish. Inst. 35:43–45. conch, Strombus gigas (L.) by juvenile spiny lobsters, Hamiguchi K, Kitada S. 2008. The enhancement of abalone Panulirus argus (L.). M.S. thesis, Florida Institute of stocks: lessons from Japanese case studies. Fish Fish. 9: Technology, Melbourne, 37p. 243–260. REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 25

Hansen SC, Gosselin LA. 2016. Are hatchery-reared abalone Leber KM, Kitada, H. L. Blankenship S, Svåsand T. 2004. naïve of predators? Comparing the behaviours of wild Stock enhancement and sea ranching: developments, pit- and hatchery-reared northern abalone, Haliotis kamt- falls and opportunities. 2nd ed. Oxford: Blackwell schatkana (Jonas, 1845). Aquac Res. 47(6):1727–1736. Publishing Ltd. 562 p. Heard WR. 2012. Overview of salmon stock enhancement Lorenzen K. 2005. Population dynamics and potential of in southeast Alaska and compatibility with maintenance fisheries stock enhancement: practical theory for assess- of hatchery and wild stocks. Environ Biol Fish. 94(1): ment and policy analysis. Philos Trans R Soc Lond, B, 273–283. Biol Sci. 360(1453):171–189. Hesse RC. 1979. Movement and migration of the queen Marshall LA. Jr. 1992. Survival of juvenile queen conch, conch Strombus gigas in the Turks and Caicos Islands. Strombus gigas, in natural habitats: impact of prey, preda- Bull Mar Sci. 29:303–311. tor and habitat features. Ph.D. dissertation. Williamsburg Heyman WD, Dobberteen RA, Urry LA, Heyman AM. (VA): College of William and Mary, 144p. 1989. Pilot hatchery for the queen conch, Strombus gigas, Martın-Mora E, James F, Stoner A. 1995. Developmental shows potential for inexpensive and appropriate technol- plasticity in the shell of the queen conch, Strombus gigas. ogy for larval aquaculture in the Bahamas. Aquaculture. Ecology. 76:981–994. 77 – (2-3):277 285. Molony BW, Lenanton R, Jackson G, Norriss J. 2003. Stock Iversen ES, Jory DE. 1997. Mariculture and enhancement of enhancement as a fishery management tool. Rev Fish wild populations of queen conchs, Strombus gigas, in the Biol Fish. 13(4):409–432. 60 – Bahamas. Bull Mar Sci. :929 941. Moreno de la Torre R, Aldana-Aranda D. 2007. Iversen ES, Jory DE, Bannerot SP. 1986. Predation on queen Crecimiento y eficiencia alimenticia del caracol rosado conch (Strombus gigas) in the western Atlantic. Bull Mar Strombus gigas (Linnaeus, 1758) juvenil, alimentado con 39 – Sci. :61 75. nueve alimentos balanceados conteniendo diferentes Iversen ES, Rutherford ES, Bannerot SP, Jory DE. 1987. niveles de proteına y energıa. Proc Gulf Carib Fish, Inst. Biological data on Berry Islands (Bahamas) queen conchs, 58:463–468. Strombus gigas, with mariculture and fisheries manage- 85 – Noyes KH. 1996. Ontogeny of settlement behavior in queen ment implications. Fish Bull. :299 310. conch (Strombus gigas) larvae. M.S. thesis. Melbourne: Johnston FD, Allen MS, Beardmore B, Riepe C, Pagel T, Florida Institute of Technology, 56p. Huhn D, Arlinghaus R. 2018. How ecological processes Patino-S~ uarez V, Aldana-Aranda D. 2000. Ingestion and shape the outcomes of stock enhancement and harvest digestion of seven species of microalgae by larvae of regulations in recreational fisheries. Ecol Appl. 28(8): Strombus gigas (Mesogastropoda: Strombidae. Rev Biol 2033–2054. Trop. 48:87–93. Jones RL. 1996. Spatial analysis of biological and physical Prada MC, Appeldoorn RS, Van Eijs S, Perez MM. 2017. features associated with the distribution of queen conch, Regional queen conch fisheries management and conser- Strombus gigas, nursery habitats. M.S. thesis, Florida vation plan. 70 pp. FAO Fisheries and Aquaculture Institute of Technology, Melbourne. 98p. Jory DE, Carriker MR, Iversen ES. 1984. Preventing preda- Technical Paper No. 610. Rome: Fisheries and tion in molluscan aquaculture. J World Maricult Soc. Agriculture Organization 15(1–4):421–432. Prince JD. 2004. The decline of global abalone (genus Jory DE, Iversen ES. 1983. Queen conch predators: not a Haliotis) production in the late twentieth century: is there roadblock to mariculture. Proc Gulf Carib Fish Inst. 35: a future? In: Leber, KM, Kitada, S, Blankenship, HL, 108–111. Svåsand, T. editors. Stock enhancement and sea ranching: Jory DE, Iversen ES. 1985. Molluscan mariculture in the dvelopments, pitfalls and opportunities. 2nd ed. (Oxford: – greater Caribbean: an overview. Mar Fish Rev. 47:1–10. Blackwell Publishing Ltd.; pp. 427 443. Jory DE, Iversen ES. 1988. Shell strength of queen conch Randall J. 1964. E. Contributions to the biology of the 14 – (Strombus gigas L.): aquaculture implications. queen conch, Strombus gigas. Bull Mar Sci. :246 295. Aquaculture Res. 19(1):45–51. Ray-Culp M, Davis M, Stoner AW. 1997. The micropreda- Kennedy GJA. 1988. Stock enhancement of Atlantic salmon tors of settling and newly-settled queen conch (Strombus 16 – (Salmo salar L.). In: Mills D, Piggins D. editors. Atlantic gigas Linnaeus. J Shellfish Res. :423 428. Salmon. Dordrecht: Springer; pp. 345–372. Ray-Culp M, Davis M, Stoner AW. 1999. Predation by xan- Kough AS, Belak CA, Paris CB, Lundy A, Cronin H, thid crabs on early post-settlement gastropods: the role of Gnanalingam G, Hagedorn S, Skubel R, Weiler AC, prey size, prey density, and habitat complexity. J Exp Stoner AW. 2019. Ecological spillover from marine pro- Mar Biol Ecol. 240(2):303–321. tected area replenishes an overexploited population across Ray M, Stoner AW. 1994. Experimental analysis of growth an island chain. Conserv Sci Pract. doi:10.1111/csp2.17. and survivorship in a marine gastropod aggregation: bal- Kough AS, Cronin H, Skubel RS, Belak CA, Stoner AW. ancing growth with safety in numbers. Mar Ecol Prog 2017. The efficacy of an established marine protected Ser. 105:47–59. area at sustaining a queen conch Lobatus gigas popula- Ray M, Stoner AW. 1995. Growth, survivorship, and habitat tion during three decades of monitoring. Mar Ecol Prog choice in a newly-settled seagrass gastropod, Strombus Ser. 573:177–189. gigas. Mar Ecol Prog Ser. 123:83–94. Laughlin RA, Weil E. 1983. Queen conch mariculture and Ray M, Stoner AW. 1995. Predation on a tropical spinose restoration in the Archipielago de Los Roques: gastropod: the role of shell morphology. J Exp Mar Biol Preliminary results. Proc Gulf Carib Fish Inst. 35:64–72. Ecol. 187(2):207–222. 26 A. STONER

Ray M, Stoner AW, O’Connell SM. 1994. Size-specific pre- Stoner AW, Davis MH, Booker CJ. 2012. Abundance and dation of juvenile queen conch, Strombus gigas: implica- population structure of queen conch inside and outside a tions for stock enhancement. Aquaculture. 128(1/2): marine protected area: repeat surveys show significant 79–88. declines. Mar Ecol Prog Ser. 460:101–114. Roberts RD, Keys EF, Prendeville G, Pilditch CA. 2007. Stoner AW, Davis MH, Kough AS. 2019. Relationships Viability of abalone (Haliotis iris) stock enhancement by between fishing pressure and stock structure in queen release of hatchery-reared seed in Marlborough, New conch (Lobatus gigas) populations: synthesis of long-term Zealand. J Shellfish Res. 26(3):697–704.2.0.CO;2] surveys and evidence for overfishing in The Bahamas. Rogers-Bennett L, Aquilino KM, Catton CA, Kawana SK, Rev Fish Sci Aquacul. 27(1):51–71. Walker BJ, Ashlock LW, Marshman BC, Moore JD, Stoner AW, Glazer RA. 1998. Variation in natural mortality: Taniguchi IK, Gilardi KV, Cherr GN. 2016. implications for queen conch stock enhancement. Bull Implementing a restoration program for the endangered Mar Sci. 62:427–442. white abalone (Haliotis sorenseni) in California. J Stoner AW, Lally J. 1994. High-density aggregation of Shellfish Res. 35(3):611–618. queen conch, Strombus gigas: formation, patterns, and Rogers-Bennett L, Pearse JS. 1998. Experimental seeding of ecological significance. Mar Ecol Prog Ser. 106:73–84. hatchery-rared red abalone in northern California. J Stoner AW, Lin J, Hanisak MD. 1995a. Relationships Shellfish Res. 17:877–880. between seagrass bed characteristics and juvenile queen Sandt VJ, Stoner AW. 1993. Ontogenetic shift in habitat by conch (Strombus gigas) abundance in the Bahamas. J early juvenile queen conch, Strombus gigas: patterns and Shellfish Res. 14:315–323. potential mechanisms. Fish Bull. 91:516–525. Stoner AW, Mueller KW, Brown-Peterson NJ, Davis MH, Searcy-Bernal R, Anguiano-Beltran C, Espinoza-Montes JA, Booker CJ. 2012a. Maturation and age in queen conch Carpizo Ituarte E. 2013. Restocking of abalone popula- (Strombus gigas): urgent need for changes in harvest cri- tions (Haliotis spp.) in Mexico. J Shellfish Res. 32(1): teria. Fish Res. 131:76–84. 189–195. Stoner AW, Pitts PL, Armstrong RA. 1996. Interaction of Shawl AL, Davis M. 2004. Captive breeding behavior of physical and biological factors in large-scale distribution four Strombidae conch. J Shellfish Res. 23:157–164. of juvenile queen conch populations in seagrass mead- Shawl A, Prada M, Casro E, Taylor E, Davis M. 2007. ows. Bull Mar Sci. 58:217–233. Queen conch ranching and educational outreach as part Stoner AW, Ray M. 1993. Aggregation dynamics in juven- of the Seaflower Biosphere Reserve, Colombia. Proc Gulf ile queen conch (Strombus gigas): population structure, Carib Fish Inst. 59:79–83. mortality, growth, and migration. Mar Biol. 116(4): Siddall SE. 1983. Biological and economic outlook for 571–582. hatchery production of juvenile queen conch. Proc Gulf Stoner AW, Ray M. 1996. Queen conch (Strombus gigas)in Carib Fish Inst. 35:45–52. fished and unfished locations of the Bahamas: effects of a Siddall SE. 1984. Synopsis of recent research on the queen marine fishery reserve on adults. Fish Bull. 94:551–565. conch Strombus gigas Linne. J Shellfish Res. 4(1):1–3. Juveniles, and Larval Production. Spring A, Davis M. 2003. Improving the culture conditions Stoner AW, Ray-Culp M. 2000. Evidence for Allee effects in of juvenile queen conch (Strombus gigas Linne) for grow an over-harvested marine gastropod: density-dependent out purposes, pp. 147–153. In: Aldana-Aranda, D. editor. mating and egg production. Mar Ecol Prog Ser. 202: El Caracol Strombus gigas: Concimiento Integral Para su 297–302. Manejo Sustentable en el Caribe. Yucatan, Mexico: Stoner AW, Ray M, Waite JM. 1995. Effects of a large CYTED, Programa Iberoamericano de Ciencia y herbivorous gastropod on macrofauna communities in Tecnologıa para el Desarrollo tropical seagrass meadows. Mar Ecol Prog Ser. 121: Spring AG, 2008. Increasing restocking survival through 125–137. behavioral and morphological improvements of hatchery- Stoner AW, Sandt VJ. 1991. Experimental analysis of habitat reared queen conch, Strombus gigas. Ph.D. dissertation. quality for juvenile queen conch in seagrass meadows. Melbourne: Florida Institute of Technology, 226p. Fish Bull. 89:693–700. Stoner AW. 1989. Density-dependent growth and the graz- Stoner AW, Sandt VJ. 1992. Population structure, seasonal ing effects of juvenile queen conch (Strombus gigas L.) in movements and feeding of queen conch, Strombus gigas, a tropical seagrass meadow. J Exp Mar Biol Ecol. 130(2): in deep-water habitats of the Bahamas. Bull Mar Sci. 51: 119–127. 87–300. Stoner AW. 1994. Significance of habitat and stock pre-test- Stoner AW, Schwarte KC. 1994. Queen conch, Strombus ing for enhancement of natural fisheries: experimental gigas, reproductive stocks in the central Bahamas: distri- analyses with queen conch. J World Aquaculture Soc. bution and probable sources. Fish Bull. 92:171–179. 25(1):155–165. Stoner AW, Waite JM. 1991. Trophic biology of Strombus Stoner AW. 2003. What constitutes essential nursery habitat gigas in nursery habitats: diets and primary food sources for a marine species? A case study of habitat form and in seagrass meadows. J Mollus Stud. 57(4):451–460. function for queen conch. Mar Ecol Prog Ser. 257: Svåsand T, Kristiansen TS, Pedersen T, Salvanes AGV, 275–289. Engelsen R, Naevdal G, Nødtvedt M. 2000. The enhance- Stoner AW, Davis M. 1994. Experimental outplanting of ment of cod stocks. Fish Fish. 1(2):173–205. juvenile queen conch, Strombus gigas: comparison of wild Taniguchi IK, Stein D, Lampson K, Rogers-Bennett L. 2013. and hatchery-reared stocks. Fish Bull. 92:390–411. Testing translocation as a recovery tool for pink (Haliotis REVIEWS IN FISHERIES SCIENCE & AQUACULTURE 27

corrugata) and green (Haliotis fulgens) abalone in south- (Accessed January 12, 2018) Available from: www.cep. ern California. J Shellfish Res. 32(1):209–216. unep.org/content/about-cep/spaw/additional-output-on- Tewfik A, Babcock EA, Gibson J, Burns-Perez VR, the-clme-project/queen-conch-expert-workshop.pdf. Strindberg S. 2017. Benefits of a replenishment zone Valle-Esquivel M. 2003. Aspects of the population dynam- ’ revealed through trends in focal species at Glover s Atoll, ics, stock assessment, and fishery management strategies 580 – Belize. Mar Ecol Prog Ser. :37 56. of the queen conch, Strombus gigas, in the Caribbean. Theile S. 2001. Queen conch management in the Caribbean. Ph.D. dissertation, University of Miami, Coral Gables, Technical Report to the CITES Secretariat A-2000/01. FL. 338 p. Traffic, Europe.:91. p. Waples RS, Drake J. 2004. Risk/benefit considerations for Theile S. 2005. Status of the queen conch Strombus gigas marine stock enhancement: a pacific salmon perspective, stocks: management and trade in the Caribbean: a CITES – review. Proc Gulf Carib Fish Inst. 56:675–694. pp. 260 306 In: Leber, KM, Kitada S, Blankenship HL, Truelove N, Box SJ, Aiken CJ, Blythe-Mallett A, Boman Svåsand T, editors. Stock enhancement and sea ranching: EM, Booker CJ, Byfield TT, Cox CE, Davis MH, Delgado developments, pitfalls and opportunities. 2nd ed. Oxford: GA, et al. 2017. Isolation by distance and spatial Blackwell Publishing Ltd. genetic structure in an overharvested international fishery. Weil ME, Laughlin RA. 1984. Biology, population dynamics, Divers Distrib. 2017:1–9. doi:doi: 10.1111/ddi.12626. and reproduction of the queen conch Strombus gigas UNEP (2012). Report of the Queen Conch Expert Linne in the Archipielago de Los Roques National Park. J Workshop. May 22–24, Miami, FL, USA, pp.16. Shellfish Res. 4:45–62.