JOURNAL OF SHELLFISH RESEARCH VOLUME 3, NUMBER 1 JUNE 1983 The Journal of Shellfish Research (formerly Proceedings of the National Shellfisheries Association) is the official publication of the National Shellfisheries Association Editor Dr. Robert E. Hillman Battelle New England Marine Research Laboratory Duxbury, Massachusetts 02332 Managing Editor Dr. Edwin W. Cake, Jr. Gulf Coast Research Laboratory Ocean Springs, Mississippi 39564 Associate Editors Dr. Jay D. Andrews Virginia Institute of Marine Sciences Gloucester Point, Virginia 23062 Dr. Anthony Calabrese National Marine Fisheries Service Milford, Connecticut 06460 Cornell University Ithaca, New York 14853 Dr. Richard A. Lutz Nelson Biological Laboratories Rutgers University Piscataway, New Jersey 08854 Dr. Kenneth K. Chew College of Fisheries University of Washington Seattle, Washington 98195 Dr. Gilbert Pauley College of Fisheries University of Washington Seattle, Washington 98195 Dr. Paul A. Haefner, Jr. Rochester Institute of Technology Rochester, New York 14623 Dr. Daniel B. Quayle Pacific Biological Laboratory Nanaimo, British Columbia, Canada Dr. Herbert Hidu Ira C. Darling Center University of Maine Walpole, Maine 04573 Dr. Louis Leibovitz New York State College of Veterinary Medicine Dr. Aaron Rosenfield National Marine Fisheries Service Oxford, Maryland 21654 Dr. Frederic M. Serchuk National Marine Fisheries Service Woods Hole, Massachusetts 02543 Journal of Shellfish Research Volume 3, Number 1 ISSN: 00775711 June 1983 Journal of Shellfish Research, Vol. 3, No. 1, 1-9, 1983. PREDATION OF JUVENILES OF THE HARD CLAM MERCENARIA MERCENARIA (LINNE) BY THE SNAPPING SHRIMP ALPHEUS HETEROCHAELIS SAY AND ALPHEUS NORM ANNI KINGSLEY BRIAN F. BEAL 1 The University of North Carolina at Chapel Hill Institute of Marine Sciences Morehead City, North Carolina 28557 OCT 5 1984 ABSTRACT Two species of snapping shrimp, Alpheus heterochaelis and A. normanni, collected near Beaufort, North Carolina, during June 1982, and then held in the laboratory, used their major chelae to crush and consume juveniles of the hard clam Mercenaria mercenaria. Snapping shrimp (19.1 to 39.4 mm in total body length (TL] ) ate clams in the largest size-class (15.1 to 20.0 mm in shell length), but preferred smaller clams when offered equal numbers in this large size-class and in each of three smaller size-classes. Female snapping shrimp, regardless of species, exhibited a statistically higher predation rate than males when the results of five separate experiments were combined. The major chelae of the females of specimens of A. heterochaelis (>32.0 mm TL) were smaller than those of equal size males. Alpheus heterochaelis (19.1 to 27.2 mm TL) had a larger major chela for a given body length than did specimens of A. normanni; however, predation rates of the two species were not significantly different. The number of clams crushed was related to both the size of the major chelae and total body length for A. normanni, but not for A heterochaelis. Alpheus spp. inflict two types of shell damage which are identical to those caused by blue crabs. These results imply that previous studies may have overestimated the importance of crab predation and underestimated or ignored the importance of predation by snapping shrimp. KEY WORDS: Alpheus, snapping shrimp, predation, Mercenaria, hard clams INTRODUCTION The hard clam or northern quahog Mercenaria mercen- aria (Linne) is distributed along the Atlantic coast from the Gulf of St. Lawrence to the northern Gulf of Mexico and occurs intertidally down to 1 5 m (Menzel 1970). This species is harvested commercially throughout most of its range; e.g., during 1981 and 1982 in North Carolina, the hard clam fishery ranked third in importance of all commercial fisheries based on a dockside dollar value of $5.4 million and $6.6 million, respectively (Street 1982). A progression of predators follows the growth of the hard clam from the earliest planktonic (Loosanoff 1959, Carriker 1961), post-settlement (Hunt 1981), and juvenile stages (Carriker 1951, Goodwin 1968, Whetstone and Eversole 1978) through adulthood (Carriker 1951, MacKenzie 1977, Greene 1978, Peterson 1982). As M. mercenaria increases in size so does its predators; because large predators are more commonly recognized in the field and have been studied extensively in the laboratory, their importance in regulating hard clam population sizes may have been over- emphasized. Investigations of predation on natural or hatchery-reared juvenile hard clams by blue crabs (Callinectes sapidus Rathbun) (Carriker 1951, Menzel and Sims 1964, Castagna and Kraeuter 1977), mud crabs (various xanthid genera) (Landers 1954, MacKenzie 1977, Whetstone and Eversole 1978), and miscellaneous species (Menzel et al. 1976) imply that those predators are responsible for the Present address: The University of Maine at Orono, Cooperative Extension Service, 5 Cooper St., P.O. Box 189, Machias, ME 04654 majority of natural post-settlement mortality of hard clams. Resource managers and fishermen who operate commercial bottom leases should be aware of the potential effectiveness of these predators in reducing hard clam populations. I conducted a series field experiments near Beaufort, NC, from August 1981 through April 1982, in which juveniles of M. mercenaria (6.0 to 1 5.0 mm in length) were maintained in cages designed to exclude large (> 6.4 mm) epibenthic predators (Beal, unpublished data). Because numerous snapping shrimp were observed inside the field cages, which also contained several crushed juvenile hard clams, they were suspected of being an important additional consumer of juvenile clams. As a result of these field investigations, I performed several laboratory experiments that clearly showed that two species of snapping shrimp, Alpheus heterochaelis Say and Alpheus normanni Kingsley, should be added to the list of known hard clam predators. Here I demonstrate that both species will crush and consume juvenile hard clams under laboratory conditions and provide field observations that indicate they do so in nature as well. Several factors are also examined: 1. Is size of snapping shrimp correlated with its preda- tion rate? 2. Do shrimp show a size preference within the size- classes of clams they are able to crush? 3. Does sex or species of snapping shrimp affect predation rate? 4. Can clam mortality, caused by blue crabs, be distin- guished from that inflicted by snapping shrimp on the basis of shell damage? ^ BEAL MATERIALS AND METHODS Snapping shrimp and shell debris were obtained from two oyster rocks (reefs) near Beaufort, NC, on 18 and 26 June 1982, using a suction dredge. Shell debris (hash) was the substrate used in all experiments and consisted of dead and fragmented oysters and clams greater than 3.0 mm (the smallest size the dredge efficiently captured). Juvenile hard clams were purchased from a commercial dealer and given a color dot (Mark-Tex Corp., paint) on both valves (near the umbo) which distinguished them from any dead clams within the shell debris. Snapping shrimp and shell debris were brought to the laboratory on the same day they were collected. Shrimp were placed in glass finger bowls where they were given crushed hard clams as food. Bowls were placed in large tanks (75 X 75 X 30 cm) supplied with unfiltered seawater. No snapping shrimp were held longer than four days in the pre -experimental setting. Shell debris was sieved through a 3.2-mm mesh to remove all fine sediments and small benthos at the beginning of each experiment. Any large animals were also removed before the shell debris was placed in finger bowls (20.0 cm dia;6.5-cm depth) to a depth of 4 cm. Forty marked clams were placed at a depth of 1 cm before one snapping shrimp was added to each bowl in each experiment. Nylon window screening (1.2-mm mesh) was placed over the top of each bowl and secured by an elastic band to ensure that the shrimp remained inside the bowl during the course of the experiment. Controls were employed to separate all types of shrimp-caused mortality from all other sources of mortality. The controls were treated identically to the other clams placed in finger bowls except they received no snapping shrimp. Each tank held nine finger bowls and in experiments where more than one tank was used, treatment and control bowls were randomly assigned to tanks. The nylon tops were cleaned daily using hands to brush away accumulated silt; the bowls were not removed from the tanks. Snapping shrimp were removed from each bowl and the contents of the bowls were sieved through 1-mm mesh after one week. Recovered clams were separated into three categories: living, dead (empty, undamaged shells), or dead (crushed). Table 1 shows the experimental interval, the number of replicate Alpheus spp. used, and the number of controls for each experiment. Experiments A through C were designed to test whether A. heterochaelis could crush and consume juvenile hard clams. The same two snapping shrimp were used in both experiments A and B. Replication was increased in experiments D and E because of the large variability in crushing rates of the snapping shrimp. The major chela (MC) of each snapping shrimp was measured from the distal end of the dactylus longitudinally to the proximal end of the propodus, and total body length (TL) was measured from the rostrum to the telson after every experiment. These two morphological traits were measured to test whether the relationship between size of the MC and TL differed between sexes of large specimens of A. heterochaelis and between species of smaller snapping shrimp. In addition, I tested whether predation rate was related to either morphological trait. TABLE 1. The experimental interval, number of Alpheus, and number of controls for each experiment. Number of Alpheus Number of Experiment Dates spp. treatments Controls A 18 June to : 25 June 2 (A. heterochaelis) 2 B 25 June to 2 July 2 (.4. heterochaelis) 2 C 26 June to 3 July 4 (A. heterochaelis) 2 D 29 June to 6 July 14 (A. heterochaelis) 3 E 30 June to 7 July 12 (A. heterochaelis) 8 (A. normanni) 3 Four size-classes of juveniles of M. mercenaria (6.0 to 8.0, 8.1 to 10.0, 10.1 to 15.0, and 15.1 to 20.0 mm in shell length [SL, the greatest anterior to posterior measurement] ) were used to test if shrimp preferred clams within a certain size. Ten clams from each size category were placed in each bowl. A total of 20 large specimens of A. heterochaelis (mean TL = 34.1 mm ± 2.5 SD) was used in these experi- ments. To determine the effects of sex of snapping shrimp on predation rate, the nonparametric Wilcoxon two-sample test on total number crushed by individual snapping shrimp was used. Data from experiments A through D were combined because ( 1 ) the time interval for each experiment was identical (7 days); (2) there was no apparent effect of time on predation rate; and (3) size categories of juvenile hard clams, as well as number of clams used within each size category, were held constant. Mean total numbers crushed by individual shrimp were used from experiments A and B because the same shrimp were used in both trials. Total counts were used for individual shrimp in experi- ments C and D. Morphometric data from experiments A through D were combined and the lines expressing TL to MC for the 1 1 male snapping shrimp (Y = 2.99 + 0.487X; r 2 = 0.74) and 9 female snapping shrimp (Y = 5.03 + 0.323X; r 2 = 0.69) were compared using multiple regression analysis. In experiment E, individuals of both species were smaller than those specimens of A. heterochaelis used in the previous experiments. Twelve specimens of A. heterochaelis (mean TL = 23.4 mm ± 2.6 SD) and eight specimens of A. normanni (mean TL = 24.0 mm ± 1.9 SD) were used. Clams from only two size-classes (4.5—8.0 mm and 8.1— 10.0 mm) were used because of the small size of these snapping shrimp. Twenty clams from each size category were placed in each bowl. A Model I 2-way analysis of variance (ANOVA) was performed on numbers crushed to test the effects of species and sex of snapping shrimp on Predation of Juvenile clams by snapping Shrimp predation rate. Numbers crushed (Y) were first transformed with In (Y + 1 ) and a Bartlett's test (Sokal and Rohlf 1969) was performed to determine whether the transforma- tion helped eliminate variance heterogeneity. Morphometric data from male snapping shrimp were pooled with data from female shrimp for each species in experiment E to determine whether the two species differed in their relation between TL and MC for the 1 2 specimens of A. heterochaelis (Y = 2.71 + 0.382X; r 2 = 0.37) and the 8 specimens of A.normanni{y = 3.55 + 0.493X;r 2 = 0.56). Again, multiple regression analysis was used to compare lines. Five specimens of A. heterochaelis were placed in isopropyl alcohol within 12 hours after feeding to test whether shell fragments pass through the cardiac stomachs of snapping shrimp. After one hour the cardiac stomach of each shrimp was excised and examined. Temperature and salinity were monitored daily within each tank. Tanks never differed by more than 0.7°C or 1 ppt S on any given day. The temperature range for the entire experimental interval (18 June to 7 July) was 24.3 to 27.5°C. The salinity range for the same time interval was 32 to 34 ppt S. Four blue crabs, Callinectes sapidus Rathbun (carapace widths: 146.9, 136.7, 74.8 and 59.7 mm), were placed in separate seawater tanks (25 X 25 X 30 cm) without sedi- ment but containing 40 juvenile hard clams (10 from each size category used in experiments A through D) to compare shell damage inflicted by Alpheus spp. with that described for crabs (Venneij 1978). The crabs were used to test whether it is possible to correctly assign clam mortality to the proper predator on the basis of shell damage. Crabs remained in the tanks until at least 50% of the clams had been crushed. This took 3 days for the smallest blue crab and 3 hours for the largest. RESULTS Experiments A through D (Table 2) No clam mortalities occurred in the control bowls, but a total of 77 clam deaths occurred in those bowls containing the snapping shrimp A. heterochaelis; in each case a chipped or crushed clam shell was recovered. This clearly demon- strates that snapping shrimp crush juvenile hard clams; furthermore, body tissues were removed from each valve indicating that the clams were eaten. I observed a female of A. heterochaelis (35.2-mm TL) crush and consume a juvenile hard clam (~ 8.0-mm SL) in a small finger bowl (10-cm, dia; 5-cm depth) containing no shell, other substrate, or other clams. The snapping shrimp grasped the clam near the umbo with the minor chelae, then lifted the clam several millimeters off the bottom. With the dactylus cocked, the snapping shrimp raised its major chela so that the clam was wedged (anterior to posterior and 2 to 3 mm ventral of the umbo) between the propodus and dactylus with its umbo and dorsal margin straight up. The dactylus closed quickly fracturing most of the clam, leaving only a small portion of the umbo intact. Initially, the mantle held the fractured pieces of clam together, but after the shrimp used its minor chela to tear the mantle from the valve remnants, the small fragments of shell became separated. The shrimp then tore off pieces of body tissue and used its minor chela and pereiopods in feeding. The cardiac stomach of each snapping shrimp examined contained shell fragments and, in one case, the painted portion of the clam. Female snapping shrimp accounted for 92% of all clams crushed in experiments A through D; however, this was not statistically significant (P = 0.09). Snapping shrimp showed a statistical preference for smaller juvenile clams in a chi- square (X 2 ) test (X 2 = 34.8. df = 3, P < 0.001); 49% of all the clams crushed and consumed belonged to the smallest (6.0 to 8.0-mm SL) size-class. Clams were eaten in all size-classes including the largest (15.1 to 20.0-mm SL). The variances around the straight lines relating TL to MC for the 1 1 males and 9 females of A. heterochaelis (Figure 1) were not significantly different. The lines were parallel (P > 0.75), but not coincident (P < 0.001 in partial F-test). Analysis of covariance (ANCOVA) demonstrated that, even though females had a greater mean TL (35. 41 mm) than males (32.95 mm), males had a larger MC for a given TL than females (P < 0.001). Because of the apparent effect of sex on predation rate in experiments A through D, sexes were not combined when I tested whether predation rate could be explained by either morphological trait. No significant relationships existed between TL (r . = 0.48, n = ll:r 9 = 0.14, n = 9) or size of MC (r d =0.52, n= 11; ro = 0.43, n = 9) and predation rate. Experiment E (Table 3) One 5.6-mm SL clam died in a control bowl as a result of natural causes; however, 31 clams died as a result of crushing in bowls containing A. heterochaelis and 38 clams were crushed in bowls containing A. normanni. All 31 clams eaten by specimens of A. heterochaelis in experiment E belonged to the smaller size-class (4.5-8.0 mm); none were eaten in the larger size-class (8.1-10.0 mm) as were crushed and consumed by larger specimens of A. heterochaelis in experiments A through D. Similarly, 95% of those clams crushed and consumed by A. normanni came from the smaller size category. Bartlett's test demonstrated that the logarithmic transformation reduced variance heterogeneity and the Model I 2-way ANOVA resulted in no species X sex interaction (P > 0.50) or effect of species (P > 0.75). The 15 female snapping shrimp ate 67 of the 69 (97%) clams; the remaining 2 crushed clams were eaten by one of the five male shrimp. This was not statistically significant (P = 0.065 ). The straight lines relating TL to MC (Figure 1 ) from experiment E had equal variances (P > 0.05) and were parallel (P > 0.75), but not coincident (P < 0.001 in a partial F-test). Application of ANCOVA yielded a significant difference (P < 0.001) in the adjusted MC lengths between BEAL TABLE 2. Results of Experiments A through D in which Alpheus heterochaelis was exposed for 7 days to 10 clams in each of four size categories. Sex TL* (mm) MCf (mm) Number Crushed With in a Size Category (mm) Total Crushed Experiment 6.0-8.0 8.1-10.0 10.1-15.0 15.1-20.0 Number Mive A M 34.8 20.1 40 F 38.4 17.6 6 7 2 1 16 24 Control 1 40 Control 2 40 B M 34.8 20.1 1 1 39 F 38.4 17.6 5 2 3 10 30 Control 1 40 Control 2 40 C M 32.4 18.7 40 M 34.0 19.9 40 F 34.0 15.9 40 F 35.2 16.4 9 7 8 1 25 15 Control 1 40 Control 2 40 D M 29.9 17.2 40 M 30.0 17.2 40 M 30.4 18.0 40 M 30.9 19.0 40 M 34.4 19.9 1 1 39 M 34.6 19.9 1 1 39 M 34.9 18.5 2 1 3 37 M 35.9 21.1 40 F 32.1 16.2 4 1 5 35 F 33.5 15.6 40 F 35.0 16.3 1 1 39 F 35.3 15.4 1 1 39 F 35.8 16.6 40 F 39.4 18.1 9 3 1 13 27 Control 1 40 Control 2 40 Control 3 40 Total number of controls 360 Total number of males 4 1 1 6 474 Total number of females 34 21 14 2 71 329 *TL = total body length fMC = length of major chela species. Alpheus heterochaelis in the size range 19.1 to 27.2 mm had a larger MC for a given TL than A. normanni There was no significant (P > 0.05) relationship between either TL (r = -0.24, n = 12) or length of MC (r = -0.05, n = 12) and number of clams crushed by A. heterochaelis; however, predation rate was significantly (P < 0.05) corre- lated for TL (r = 0.76, n = 8) and MC size (r = 0.77, n = 8) for A normanni. Effect of Sex on Predation Rate Fischer's technique of combining probabilities from independent tests of significance (Sokal and Rohlf 1969) was applied to test the effect of sex of snapping shrimp on predation rate from all experiments. This test resulted in a significant (P = 0.04) overall effect of sex implying that females had a greater crushing rate over all experiments. The effect of sex in experiment E included information from both species; however, because there was no species X sex interaction, this test was justified over all experiments. The size distributions of males and females used in all experiments were compared because size of snapping shrimp may influence predation rate. Size of snapping shrimp was statistically independent of sex (X 2 = 9.38;df =6;P=0.195) over all experiments. Shell Damage Inflicted by Snapping Shrimp (Figure 2) and Blue Crabs Two types of shell damage caused by snapping shrimp were distinguished by visual inspection. In the first (Type I) Predation of Juvenile Clams by Snapping Shrimp 20 E E CD O E cr CD ° a A. normanni t <■ A. heterochaelis 15 10 20 25 30 Total length (in mm) 35 40 Figure 1. Relationships between total body length (TL) and size of major chela (MC) for snapping shrimp used in all experiments. Capital letters refer to experiment. Open circles: Alpheus normanni; closed circles: Alpheus heterochaelis. at least one of the valves remained intact. Shell chips or fractures were restricted along the posterior edge and often both valves had symmetrical chips. Where both valves were not chipped identically, one valve was chipped along the posterior edge while damage to the other valve ranged from restricted ventral margin fractures to an extensively broken valve having only the umbo region intact. Valves exhibiting damage of the second type (Type II) had been completely crushed and only the immediate area around the umbo was left intact and held together by the hinge ligament (Figure 2). To learn if shell damage inflicted by snapping shrimp and blue crabs was distinguishable, crushed shells from experiments A through D and from the blue crab experi- ment were collected and separated by size-class into damage types. Both predators caused Type I and Type II damage in each size-class. Sixty clams were crushed by A. hetero- chaelis in the size range 6.0 to 10.0 mm; 92% exhibited Type II damage, whereas 53% of the crushed clams between 10.1 and 20.0 mm exhibited Type I damage. Type II damage occurred in 70% of the juvenile hard clams (6.0 to 10.0 mm SL) crushed by blue crabs, whereas 8% of the crushed clams between 10.1 and 20.0 mm suffered Type I damage from blue crabs. DISCUSSION Experiments A through E demonstrate that two species of snapping shrimp, A. heterochaelis and A. normanni, can crush and consume juveniles of M. mercenaria and can also discriminate between sizes of prey when offered a choice. It is sometimes difficult to relate laboratory experiments to field experiments because the number of variables permitted to vary in each is different (Dayton and Oliver 1981); however, two observations from my caging study in the field suggested that snapping shrimp do indeed prey on juvenile hard clams (6.0 to 15.0 mm SL) in nature. Snapping shrimp were found inside complete 1 m 2 cages (6.4-mm mesh; see Beal [1983] for a detailed cage description) designed to keep large, epibenthic predators from preying on juvenile hard clams (Beal, unpublished data). When the contents of these cages were sieved in November 1981 and in April 1982, I found live clams as well as shell fragments which were identical in appearance to those clams crushed and consumed by Alpheus spp. in this study. No other predators or signs of predators were observed inside complete cages. Female snapping shrimp exhibited a higher predation rate than did males over all experiments; however, the mechanism for this behavior was not investigated. Elner and BEAL TABLE 3. Results of Experiment E in which Alpheus heterochaelis and Alpheus normanni were exposed for 7 days to 20 clams in each of two size categories. Species Sex TL* (mm) MCf (mm) Number Crushed Within a Size Category (mm) 4.5-8.0 8.1-10.0 Total Crushed Number Alive Alpheus heterochaelis M 21.4 10.0 M 25.4 13.7 M 26.9 12.4 2 Total males 2 F 19.1 10.7 5 F 20.3 9.1 F 20.8 11.2 10 F 22.9 12.7 F 23.2 12.5 3 F 23.5 11.5 3 F 24.2 9.5 2 F 25.6 14.2 2 F 26.9 12.1 4 Total females 29 Total A. heterochaelis 31 Alpheus normanni M 22.5 7.2 M 23.4 7.6 Total males F 21.6 8.1 7 F 23.1 7.8 F 23.4 7.6 F 23.8 8.3 F 26.1 8.1 13 F 27.2 11.1 16 Total females 36 Total A. normanni 36 Control 1 1 Control 2 Control 3 *TL = total body length f MC = length of major chela 1 1 2 2 40 40 2 38 2 118 5 35 40 10 30 40 3 37 3 37 2 38 2 38 4 36 29 331 31 449 40 40 80 7 33 40 40 40 14 26 17 23 38 202 38 282 39 40 40 Hughes (1978) examined the diet of the shore crab Carcinus maenus (Linnaeus) and, to avoid potential biases caused by sexual differences in morphology and predatory behavior, used only male crabs. Here both sexes were used and, at least for larger specimens of Alpheus heterochaelis, females had a smaller major chela than did males of a similar body length. Because the major chela is used in crushing juvenile hard clams, males should have had the highest predatory rate. Ennis (1973) found a difference in the feeding activity between sexes of the American lobster Homants americanus Milne Edwards; females continued to feed at a higher level longer into the winter than did males. Ennis (1973) suggested that this may have been caused by greater physio- logical demands on the female due to gonadal development. If an energetic explanation were true for snapping shrimp, similar experiments using females with developing versus developed gonads or, perhaps, immature (juvenile) versus mature females as well as males would be needed. Accounts of snapping shrimp as predators are rare. Hazlett (1962) determined that a species of Alpheus from Bermuda was omnivorous. Goldberg (1971) studied a species of Synalpheus in the Florida Keys which preyed upon the gastropod Coralliophila caribaea Abbott without crushing it. The shrimp lifted the flexible operculum with its major chela exposing the gastropod while the minor chela tore off pieces of the foot. I am unaware of any account of predation by either A. heterochaelis or A. normanni on a bivalve mollusc. Previous investigations concerning the role that the major chela plays in the behavior and ecology of these snapping shrimp suggest that it is used agonistically during intra- and interspecific interactions (Nolan and Salmon 1970, Schein 1977). Conover and Miller (1978) described the importance of the major chela in determining the success of a shrimp in competing for shelter. Glynn (1976) described a species of snapping shrimp off the Pacific coast of Panama Predation of Juvenile Clams by snapping shrimp Figure 2. The size range of the five size-classes of juvenile hard clams and the shell damage caused by Alpheus heterochaelis from experi- ment A through E. Damage in the smaller size-classes was similar for both species. Each tick mark represents 1 mm. 8 BEAL which repulsed the crown-of-thorns sea star and prevented it from preying on a branching coral. In this study the major chela of A. heterochaelis (29.9 to 39.4 mm TL) was smaller in females compared with equal size males. Nolan and Salmon (1970) noted this sexual dimorphism in both species. They showed that when a female approached a larger male, she was threatened and quickly retreated because of aggressive male snapping; if the TL of a female was greater than that of the male she approached, the encounter would continue until cues important in sexual discrimination could be exchanged. Whetstone and Eversole( 1978) investigated the predators of juvenile hard clams in a South Carolina sound. They collected 13 species of crustaceans from sub tidal and intertidal trays containing juvenile hard clams over a 19- month interval and examined their gut contents. They concluded, on the basis of shell fragments in the cardiac stomachs (as well as overall numbers collected), that the xanthid crab Panopeus herbstii Milne Edwards (1,465 collected from May 1975 through December 1976) was the most important predator of juvenile hard clams. Alpheus heterochaelis was the second most abundant crustacean found by Whetstone and Eversole ( 1 84 collected during that same time interval); nine specimens of A. normanni were also collected during that study. Whetstone and Eversole (1978) found no shell fragments in either species of Alpheus they examined and, on this basis, concluded that snapping shrimp were not hard clam predators; however, shell frag- ments were found in the cardiac stomachs of every snapping shrimp I examined. There may be several reasons why shell fragments were found in the cardiac stomachs of the snapping shrimp from this study and not in Whetstone and Eversole's (1978) investigation: 1. The snapping shrimp they collected may not have crushed any juvenile hard clams; Whetstone and Eversole (1978) used hard clams with a mean SL of 13 mm (however, 19% of the hard clams con- sumed in my experiments A through D were 10.1 to 15.0 mm SL [Table!]); 2. The snapping shrimp may have been collected or preserved after evacuation of the cardiac stomachs had occurred; or 3. The shell fragments may have dissolved in the 10% formalin solution they used as a preservative. The results presented in this paper suggest that Alpheus spp. may be an important predator of juveniles (< 20.0 mm SL) of M. mercenaria in South Carolina sounds. I have seen or heard snapping shrimp in a variety of areas in Bogue, Back, and Core sounds in North Carolina. These areas have several aspects in common. They either have muddy substrates with natural shelters such as living or dead oysters, or seagrass beds. Nolan and Salmon (1970) collected both species near Beaufort among clumps of oyster shells, as well as in eelgrass beds. Alpheus heterochaelis was more often found in muddy areas associated with clumps of oysters; A. normanni was found primarily in eelgrass beds. Hoff Stuart (National Marine Fisheries Service, Beaufort, NC, pers. comm.) found a mean of 6.1 adults of A. normanni and 1.1 adults of A heterochaelis (TL > 20.0 mm) per m 2 in a Back Sound eelgrass bed during 1975—1976. The mean number of clams consumed per snapping shrimp per day in my laboratory experiments was 0.72. This figure is indica- tive of clams < 15.0 mm SL because only two clams were consumed that were > 15.0 mm SL. Thus, if that rate is representative of their hard clam predation in nature, snapping shrimp of this size in that eelgrass bed may con- sume approximately 125 clams (4.5 to 15.0 mm SL) per m 2 per month. The type of shell damage inflicted by these snapping shrimp is typical of crabs (Vermeij 1978). Cake (1970) found that C. sapidus could open large specimens of the sunray venus clam Macrocallista nimbosa (Lightfoot) without breaking their shells by "inserting the finger and cutting the adductor muscles." That type of shell damage by Callinectes, which leaves behind minute scars of cheliped activity on the periostracum, was not observed in this study; in fact, both snapping shrimp and blue crabs inflict similar types of shell damage. The entire clam is either broken into bits leaving only the umbo region, or is marginally damaged with chips occurring around the posterior edge of at least one valve. According to the results of this study, past investigations in which clam mortalities were assigned a particular crushing predator based on shell damage may have overestimated the importance of crab predation and underestimated or ignored the importance of predation by snapping shrimp. Furthermore, commercial clam cultunsts need to be concerned about protecting seed clams from snapping shrimp as well as from crabs and other predators. The spatial distribution and abundance of the bottom- dwelling snapping shrimp, as well as their natural predation rates on small hard clams, must be determined to fully assess the importance of these findings. ACKNOWLEDGMENTS I am indebted to G. W. Safrit, Jr., who provided many of the snapping shrimp used in the laboratory experiments. K. Bowers, M. E. Colby and S. Smith assisted in the field and laboratory. R. J. Beal also helped in the laboratory. H. J. Porter aided in describing types of shell damage. H. E. Page took the photographs and V. Page prepared the figure. H. Stuart supplied density data from his dissertation work. F. J. Schwartz provided computer funds and S. R. Fegley dissected the snapping shrimp. Additionally, I thank W. G. Ambrose, Jr., D. R. Colby, P. B. Duncan. S. R. Fegley, C. H. Peterson, M. C. Watzin, and an anonymous reviewer for helpful suggestions on earlier drafts of this manuscript. Predation of Juvenile Clams by Snapping Shrimp D. R. Colby, P. B. Duncan, S. R. Fegley, and C. H. Peterson assisted with experimental design, statistical analyses, and writing. Financial support was provided by the Curriculum in Marine Sciences. University of North Carolina, Chapel Hill, NC, and the Institute of Marine Sciences, Morehead City, NC. Support was also provided by the Office of Sea Grant, NOAA, U.S. Department of Commerce under Grant No. NA81AA-D-0026, North Carolina Depart- ment of Administration to C. H. Peterson. REFERENCES CITED Beal, B. F. 1983. Effects of environment, intraspecific density, predation by snapping shrimp and other consumers on the popu- lation biology of Mercenaria mercenaria near Beaufort, North Carolina. Chapel Hill, NC: Univ. of North Carolina. 181 p. Thesis. Cake, E. W., Jr. 1970. Some predator-prey relationships involving the sunray venus clam, Macrocallista nimbosa (Lightfoot) (Pelecypoda: Veneridae), along the Gulf coast of Florida. Tallahassee. FL: Florida State Univ. 166 p. Thesis. Carriker. M. R. 1951. Observations on the penetration of tightly closing bivalves by Busycon and other predators. Ecology 32:73-83. . 1961. Interrelation of functional morphology, behavior, and autecology in the early stages of the bivalve Mercenaria mercenaria. J. Elisha Mitchell Sci. Soc. 77:168-241. Castagna, M. & J. N. Kraeuter. 1977. Mercenaria culture using stone aggregate for predator protection. Proc. Natl. Shellfish. Assoc. 67:1-6. Conover, M. R. & D. E. Miller. 1978. The importance of the large chela in the territorial and pairing behavior of the snapping shrimp, Alpheus heterochaelis. Mar. Behav. Physiol. 5:185-192. Dayton, P. K. & J. S. Oliver. 1981. An evaluation of experimental analyses of population and community patterns in benthic marine environments. Tenore, K. R. and B. C. Coull. eds. Marine Benthic Dynamics. Columbia, SC: Univ. of South Carolina Press, p. 93-120. Elner, R. W. & R. N. Hughes. 1978. Energy maximization in the diet of the shore crab, Carcinus maenas. J. Anim. Ecol. 47 : 103-116. Ennis, G. P. 1973. Food, feeding, and condition of lobsters. Homarus americanus. throughout the seasonal cycle in Bonavista Bay. Newfoundland./ Fish. Res. Board Can. 30:1905-1909. Glynn, P. W. 1976. Some physical and biological determinants of coral community structure in the Eastern Pacific. Ecol. Monogr. 46:431-456. Goldberg, W. M. 1971. A note on the feeding behavior of the snapping shrimp Synalpheus fritzmuelleri Coutiere (Decapod: Alpheidae). Crustaceana (Leiden) 21:318-320. Goodwin, W. F. 1968. The growth and survival of planted clams, Mercenaria mercenaria, on the Georgia Coast. Ga. Game Fish Comm.. Mar. Fish. Div., Contrib. Ser. No. 9:1-16. Greene, G. T. 1978. Population structure, growth and mortality of hard clams at selected locations in Great South Bay. New York. Stony Brook, NY: State Univ. of New York. 199 p. Thesis. Hazlett, B. A. 1962. Aspects of the biology of snapping shrimp (Alpheus and Synalpheus). Crustaceana (Leiden) 4:82-83. Hunt, J. H. 1981. The importance of adult-larval interactions in determining abundance patterns of soft-sediment infauna. Chapel Hill, NC: Univ. of North Carolina. 59 p. Thesis. Landers, W. S. 1954. Notes on the predation of the hard clam, Venus mercenaria, by the mud CTab.Neopanope taxana. Ecology 35:422. Loosanoff, V. L. 1959. Condylostoma-zn enemy of bivalve larvae. Science 129:147. Mackenzie. C. L., Jr. 1977. Predation on hard clam (Mercenaria mercenaria) populations. Trans. Am. Fish. Soc. 106:530-537. Menzel. R. W. 1970. The species and distribution of quahog clams Mercenaria. Proc. Natl. Shellfish. Assoc. 60:8 (abstract). , E. W. Cake, M. L. Haines. R. E. Martin & L. A. Olsen. 1976. Clam mariculture in northwest Florida: field study on predation. Proc. Natl. Shellfish. Assoc. 65:59-62. Menzel, R. W. & H. W. Sims. 1964. Experimental farming of hard clams, Mercenaria mercenaria. in Florida. Proc. Natl. Shellfish. Assoc. 53:103-109. Nolan, B. A. & M. Salmon. 1970. The behavior and ecology of snapping shrimp (Crustacea: Alpheus heterochelis and Alpheus normanni). Forma Functio 2:289-335. Peterson. C. H. 1982. Clam predation by whelks (Busycon spp.): Experimental tests of the importance of prey size, prey density and seagrass cover. Mar. Biol. (Berl.) 66:159-170. Schein, H. 1977. The role of snapping in Alpheus heterocliaelis Say, 1818, the big-clawed snapping shrimp. Crustaceana (Leiden) 33:183-188. Sokal, R. R. & F. J. Rohlf. 1969. Biometry: The Principles and Practice of Statistics in Biological Research. San Francisco, CA: W. H. Freeman and Co. Street, M. 1982. Trends in North Carolina's commercial fisheries, 1965-1981. NC Dep. Nat. Resour., Comm. Dev. Div. Mar. Fish. 17 p. Vermeij. G. J. 1978. Biogeography and Adaptation: Patterns of Marine Life. Cambridge, MA: Harvard Univ. Press. Whetstone, J. M. & A. G. Eversole. 1978. Predation on hard clams. Mercenaria mercenaria . by mud crabs, Panopeus herbstii. Proc. Natl. Shellfish. Assoc. 68:42-48. Journal of Shellfish Research, Vol. 3, No. 1, 11-17, 1983. SEASONAL GONADAL DEVELOPMENT OF YOUNG LABORATORY-SPAWNED SOUTHERN {MERCENARIA CAMPECHIENSIS) AND NORTHERN (MERCENARIA MERCENARIA ) QUAHOGS AND THEIR RECIPROCAL HYBRIDS IN NORTHWEST FLORIDA RODNEY DALTON 1 AND WINSTON MENZEL Department of Oceanography Florida State University Tallahassee, Florida 32306 ABSTRACT The seasonal gonadal development of laboratory-spawned southern and northern quahogs and their recipro- cal hybrids was investigated. All young clams were males and one or more stages of gametogenic activity were seen each month of the year. Winter spawning, which occurred in all pedigrees of quahogs, was considered abnormal and resulted from the unusually warm winter of 1 9 74- 7 5. Gonadal development of the hybrid 9 Mercenaria campechiensis X 6 Mercenaria mercenaria was similar to its southern parent; the reciprocal hybrid was similar to its northern parent. This may indicate maternal influence. Little or no spawning by M. campechiensis during warmer months was unlike that of the other three pedigrees. Temperature was the overall controlling factor in gonadal development and spawning, but genetic differences existed between the two species. KEY WORDS Genetics, gametogenesis, hybridization, hard clams, quahogs, Mercenaria spp. INTRODUCTION The seasonal gonadal development of the northern quahog clam Mercenaria mercenaria (Linne) has been studied from the New England area (Loosanoff 1937a.b), from Delaware Bay (Keck et al. 1975), from North Carolina (Porter 1964), and from South Carolina (Eversole et al. 1980). A closely related species, the southern quahog Mercenaria campechiensis (Gmelin). hydridizes readily with the northern quahog (Loosanoff 1954) and the hybrids are fertile (Menzel and Menzel 1965, Menzel 1968), hut the reproductive cycles of neither the southern nor the hybrids have been investigated. The present study is of the seasonal gonadal cycles of young, laboratory-spawned northern and southern quahogs and their reciprocal hybrids cultured in northwestern Florida. The results are compared with published reports from other areas. MATERIALS AND METHODS Southern quahogs, previously collected in the vicinity of Florida State University (FSU) Marine Laboratory, north- western Florida, were spawned by Dr. Charles Epifanio at the University of Delaware Center for Mariculture Research on 2 April 1974. Wild northern quahogs from Delaware Bay were also spawned. Besides making self-fertilizations of each species, reciprocal hybrids between the species were produced. The larvae were cultured to metamorphosis and grown to a size of 1 to 2 mm before shipment to Florida in late June 1 974. The clams were reared to a size of 4 to 8 mm at the FSU Marine Laboratory. On 4 October 1974, they were planted in 10-cm deep, sandfilled, screen-covered Present address: National Marine Fisheries Service, 9450 Roger Blvd., St. Petersburg, FL 33702 wooden boxes in Alligator Harbor, about 8 km from the laboratory. At mean low water 4 to 5 cm of water covered the clams. Ten clams of each pedigree were sampled on the 5th (± 1 day) of each month from 6 November 1974 through 5 November 1975, and additional samples were taken on the 20th (± 1 day) in December 1974, and in September and October 1975. The total sample included 660 clams from which 6,000+ follicles were microscopically examined. After February 1975, the stock of the hybrid 9 Mercenaria mercenaria X 6 Mercenaria campechiensis was depleted, primarily from crab predation. Additional clams of the same pedigree, planted as surplus in the same area, were sampled from May 1975 until the stock became exhausted by August 1975. Shucked clams were preserved in Bouin's fixative, trans- ferred to alcohol, imbedded in Paraplast®, sectioned at 8 nm, mounted on slides, and stained with Erlich's hemo- toxylin and erosin following standard histological proce- dures. Previous examinations showed that transverse mid- longitudinal sections gave a good representation of the gonad condition. All follicles in the most representative of 8 to 10 sections of each clam were used to determine gonadal condition. Determination of gonadal condition followed that of Ropes (1968) as modified by Haines (1976). As noted by Loosanoff (1937a). different follicles within the same clam and different clams within the same population were often in several stages of gonadal development. The gonadal stages are not illustrated because they have been reported pre- viously by Loosanoff (1937a), Porter (1964), Keck et al. (1975), and Eversole et al. (1980). Brief descriptions of each stage follow. 11 12 Dalton and Menzel Indifferent or Spent The lumen of the indifferent or spent follicles are usually conspicuously empty, although a few residual spermatozoa may be present (in spent follicles) and a few scattered spermatogonia occur around the membranes of the other- wise bare follicles. Early to Late Active Follicles in the early active stage are undergoing primary and secondary spermatogenesis, with a nearly continuous layer of cells forming around the follicle membrane. Later, the lumen fills with basophilic spermatids and a few sperma- togonia occur near the periphery. Early and late active stages were recorded separately but are presented as active stage only. Ripe The ripe phase is easily distinguished by a dense mass of spermatozoa, filling the follicles. Other types of gameto- genic cells may be present, but are not abundant. Partially Spent Partially spent follicles contain spermatozoa within the lumen of the follicle but these are substantially less abun- dant than in the ripe stage. Percentages of each gonadal stage for each pedigree at each sampling were graphed and the mean percentages of each stage of each pedigree were calculated and graphed to emphasize the similarities and differences between the four pedigrees. The first samples (November 1974) were not included in the mean calculations because no clams were mature enough to spawn and the results would be biased. Additionally, because of the smaller amount of data for the hybrid 9 Mercenaria mercenaria X d Mercenaria campechiensis, comparative data were recalculated using only samples collected in November 1974-February 1975, and May-August 1975. Water temperatures were taken at time of sampling at depths of 20 to 30 cm. These infrequent observations were supplemented with minimum and maximum air tempera- tures (mean of 6-day intervals) from local climatological data recorded at Apalachicola, FL (NOAA 1974a, 1975a). Although Apalachicola is about 50 km from Alligator Harbor, that coastal location has the same latitude and is considered representative for this study. In April 1976, when the clams were two years old, the remaining 19 southern, 4 northern and 11 hybrids (9 Mercenaria campechiensis X 6 Mercenaria mercenaria) were recovered and their sex was determined by the smear technique. RESULTS Both of the species and the hybrids were predominantly male. Two clams (0.3%) showed evidence of oogenesis. The follicles were in the early active stage, but no clams were observed with ripe female follicles. Occasionally, a few early stage female gamete cells occurred in otherwise male follicles, indicating a possibility for hermaphroditism. Game- togenesis had commenced by the first examination in Novem- ber 1974, when the quahogs were seven months old, but only 2 to 4 follicles were seen per histological section. Later, the number of follicles increased to 15 to 20 per section. Gametogenesis in one or more stages were seen throughout the entire period in all the samples and pedigrees. Differences in the seasonal occurrence and relative overall abundance of each stage occurred in each pedigree. A discussion of the seasonal occurrence of each gonadal stage and probable times of spawnings are given for each pedigree. Southern Quahog, Mercenaria campechiensis Indifferent or spent follicles were present in all the samples of the southern quahog (Figure 1 ) and were in the largest mean percentage, 54% (Figure 2A). Active stages were also seen in all the samples except that taken 5 April, but occurred in low percentages in December, May, and June, with values of 10, 5, and 6%, respectively (Figure 1). The mean percentage for the entire period was 23% (Figure 2A). The percentages of ripe stage follicles were highest in both samples taken in December (47% and 40%) and in January (43%). This stage decreased in February (10%), March (14%), and April (6%), and none or very low percentages occurred through the 20 September sample (6%). Ripe follicles were found in the remaining samples (9-14%) (Figure 1). The mean for the entire period was 13% (Figure 2 A). Partially spent stages were first seen in the sample taken 20 December (9%) and continued in relatively high percentages through the 5 April period (10—32%). This stage decreased by the May sample (6%) and was low until the following fall, increasing to 17% on 5 October (Figure 1 ). The mean was 10% (Figure 2A). Spawning, as indicated by comparison of ripe and partially spent stages, commenced after the 5 December sample and continued until 5 April, with a probable peak in March. Little or no spawning occurred during the summer months, but spawning commenced again after 5 September. Northern Quahog, Mercenaria mercenaiia Indifferent or spent follicles were present in all samples of Mercenaria mercenaria (Figure 1) but in considerably less abundance (X = 28%) than for the southern species (Figure 2A). Active stage follicles were also present in all samples (X = 58%) and in greater abundance than the southern species (Figure 2A). Ripe follicles occurred in all sampling periods, except the first on 6 November and those on 5 May and 20 September (Figure 1 ) (X = 10%) (Figure 2A). Partially spent follicles were seen in the samples taken 5 and 20 December, but not again until 5 March, when the highest percentage occurred (13%). This stage occurred on all the other sampling dates except that taken on 5 May (Figure 1 ). The mean was 4% (Figure 2A). Gonadal Di vi lopmtNt oi young Quahogs 13 Mercenaria campechiensis Mercenaria mercenaria DJFMAMJJAS ON N D JFMAMJJA S N ^Mercenaria campechiensis X ^Mercenaria mercenaria N D JFMAMJJA S CH Indifferent/Spent ES3 Active ^Mercenaria mercenaria X dMercernaria campechiensis ON N D J F Months Ripe ] Partially Spent m nAAA/ M J J A Figure 1. Reproductive cycles of southern and northern quahogs and their hybrids (660 total) shown as the percentage of follicles (males only) in each gonadal stage (period from 6 November 1974 through 5 November 1975). 14 Dalton and Menzel ru- 60- 50- c 40- 0) o w 0. 30- 20- 1 10- Mc Mm 9Mc x cfMm Indifferent/Spent ^SB Ripe C*3*l Active r^x^l Partially Spent Mm 9Mc x cfMm 9Mm x cfMc Figure 2. Mean percentages of follicle stages in southern (Mc) and northern (Mn) quahogs and their hybrids. (A) December 1974-November 1975: southern, northern and 9 southern X d northern. (B) December 1974-Februaiy 1975 and May-August 1975: southern, northern and reciprocal hybrids. The data for ripe and partially spent follicles indicate that spawning started by 5 December, but ceased from 20 December until after the 5 February sample. A peak of spawning occurred between 5 February and 5 May, with a probable high in March. Spawning resumed after 5 May and continued throughout the balance of the sampling period; a probable secondary peak occurred in September. Hybrid, 9 Mercenaria campechiensis X d Mercenaria mercenaria The sequences of follicle development stages in the hybrid 9 Mercenaria campechienses X 6 Mercenaria mercenaria are similar to the southern quahog parent. Indifferent or spent stages were found in all the samples (Figure 1 ) and. as in M. campechienses, had the highest mean (58%) (Figure 2 A). Active follicle stages were also present in all the samples, ranging from a high of 54% on 20 December to lows of 17% in April, June, and July (Figure 1 ); the mean for the entire period was 27% (Figure 2A). This hybrid was the only pedigree that had ripe follicles (21%) on the first sampling (6 November 1974). The highest percentages of the ripe stage occurred on 5 January (34%) and on 5 March (269! ). Ripe follicles were not seen in the 5 April samples but were observed in varying percentages for the balance of the sampling dates (Figure 1 ). The mean of the ripe follicles was 10%' (Figure 2 A). Partially spent stages were first seen 20 December and continued through the 5 March sample; none occurred on 5 April. This stage occurred in low percentages for the balance of the period, except for none on 20 December (Figure 1 ). The mean was 5% (Figure 2A). The data indicate that spawning commenced after 5 December and continued through March. The absence of both ripe and partially spent stages in the 5 April sample indicates a peak of spawning in March. Spawning resumed after 5 April and continued throughout the balance of the examinations, with probable peaks in May-July and again in September. Hybrid, 9 Mercenaria mercenaria X d Mercenaria campechiensis Unfortunately data for the hybrid 9 Mercenaria mercenaria X d Mercenaria campechiensis are incomplete, but those obtained show the sequences of follicle develop- ment to be similar to the northern quahog. Indifferent or spent stages were present in all the samples and ranged from a high of 40%' on 5 June to a low of 5% on 5 December (Figure 1) (X= 23%, Figure 2B). Ripe follicles (4%) first seen on 5 December, increased to a high of 26%' on 5 Janu- ary, and were found on all the other dates for which data are available; another high (28%) occurred on 5 July (Figure 1). The mean for the entire period was 15% (Figure 2B). Partially spent follicles were first observed on 20 December and were seen in all the other samples, except that on 5 May (Figure 1 );X= 7% (Figure 2B). Spawning commenced after 5 December and continued to at least 5 February. The absence of partially spent follicles on 5 May indicates that a peak of spawning occurred prior to this date. Spawning continued after 5 May to at least 5 August, the last date sampled. Sex could be determined for only 15 of the 34 two-year- old clams collected in April 1976. Of these clams. 13 were males and 2 were females (2 of 4 northern sampled). DISCUSSION This is the first study of the seasonal gonadal develop- ment of the southern quahog Mercenaria campechiensis and Gonadal Development of Young quahogs 15 its hybrids with the northern species Mercenaria mercenaria, with a comparison of laboratory-spawned clams of known age grown in the semitropical area of northern Florida. This study is not as thorough as those from more northern latitudes because observations were made for only one year and of male clams only. The spawnings that occurred in the winter period were undoubtedly atypical and are discussed in more detail below. Loosanoff (1937a) found that quahogs have a protandric development; almost all clams (98%) developed first as males, but eventually achieved an equal sex ratio as older clams. Eversole et al. (1980) also found a preponderance of males to females (9.5:1) in young quahogs and a 1:1 sex ratio in older animals. Our study confirms the protandric development in northern quahogs and documents the same type of development in the southern species and its hybrids. The samples of 2-year-old clams revealed that sex reversal to female was occurring, even though the sampling was very small. Large clams of both species and hybrids that were used in our spawning experiments over the past 20 years usually had a 1 : 1 sex ratio. Only 2 to 4 follicles were present in the first sample (6 November 1974) and were localized near the stomach ventral of the pericardial sinus. This was the same location reported by Loosanoff (1937a), but he found 6 to 8 follicles in clams of approximately the same size and probably of lesser age. The slighter gonadal development of quahogs grown in Florida was surprising, especially as growth rates have been reported to be greater than in more northern areas (Menzel 1961, 1962. 1977). One possible explanation is that the animals were laboratory reared and cultured in the natural habitat for only one month when first examined. Growth has always been less under our laboratory condi- tions than when planted in the open waters. Enough food may have been available for shell growth but not enough for gonadal development. Sastry (1966) stated that the bay scallop Argopecten irradians (Lamarck) "requires large amounts of food for gonad growth." Loosanoff and Davis (1950) found that Crassostrea virginica (Gmelin) did not mature sexually with poor glycogen reserve. Figures 1 and 2, especially 2, show a usually low per- centage of the partially spent stage in all the pedigrees. This probably indicates that once spawning is initiated in ripe clams, it is completed in a short period of time. If partially spent follicles occur for only a brief period, errors may have been made in deducing times of spawning, which were based on comparisons of ripe and partially spent clams at each examination (1 month inmost instances). Spawning throughout the year in marine invertebrates occurs most commonly in areas where there is little seasonal change, such as the tropics, polar regions, and deep sea (Goodbody 1965, Sanders and Hessler 1969). Northwestern Florida is subtropical, but warmer than normal tempera- tures occurred during the winter of 1975-75. Northern Florida experiences periods of air temperatures below freezing and water temperatures below 10°C;water tempera- tures in January-February 1958—61 were as low as 6 to 9°C (Menzel 1961). The lowest water temperature during the winter of 1974-75 was 1 1.5°C in early December and air temperatures at Apalachicola never dropped below freezing (Figure 3). Extended periods occurred during the winter of 1974—75 when air temperatures were above 20°C in December-February (Figure 3). Those periods coincided with minus spring tides of -5 to -40 cm during the hours of 0730—1700 (National Oceanic and Atmospheric Admin- istration, 1974b, 1975b). We have repeatedly observed in our laboratory that when alternating thermal stimulation is used to induce spawning, quahogs initiate spawning on the decreasing temperatures. Also, males usually spawn before females. The male quahogs in the boxes may, therefore, have been warmed to the critical spawning temperatures during the minus tides on warm days and stimulated to spawn when covered by the cooler incoming water at flood tide. All quahog pedigrees had ripe follicles during winter months. This is consistent with other observations. Chestnut (195 1) found that Mercenaria mercenaria often reach sexual maturity by mid-winter in North Carolina. Our thermal- induced laboratory spawning of both sexes has been most successful during the winter months. Winter spawnings are unusual in northern Florida. All wild quahogs have been found subtidally; a few may be uncovered by low tides of > —30 cm. Even if winter spawning does occur, it is unlikely that the gametes/larvae would survive in the relatively cold water. A larger percentage of the follicles may have been in the ripe condition during the winter months if normal temperatures had prevented spawning. Reproductive cycles in marine invertebrates vary with the latitude and modifications have been associated with differ- ences in temperature regimes (Orton 1920. Nelson 1928, Thorson 1950, Loosanoff and Nomejko 1951, Sastry and Blake 1971). The northern quahog ranges from Canada southward on the Atlantic coast and throughout the northern Gulf of Mexico (Abbot 1974) and thus experiences a wide range of temperatures. The spawning periods of the northern quahog have been documented for the areas ranging from Long Island Sound to South Carolina and now for northern Florida. The spawning periods in Florida, disregarding the winter spawning, showed bimodal spawning peaks in the spring and fall similar to that observed in the Carolinas (Porter 1964, Eversole et al. 1980); however, spawning began about a month (March) earlier and extended about a month (October) later than in the Carolinas. These northern clams were the progeny of clams native to Delaware Bay, where there is a single peak of spawning (Keck et al. 1975), similar to Long Island Sound (Loosanoff 1937b). Peak spawnings by southern and northern quahogs and the reciprocal hybrids were essentially the same. We noted that percentages of indifferent/spent and active stages of gonadal activity of the southern species and the 16 Dalton and Menzel 35 30 25 o 20 0> k- 3 ♦" O 15 a> Q. E a> f- 10 5- (MAX) (MIN) 1974 1975 Figure 3. Water temperatures (heavy line) at Alligator Harbor and maximum and minimum air temperatures (mean of 6-day intervals) at Apalachicola, Florida. hybrid 9 Mercenaria campechiensis X 6 Mercenaria mercenaria were very similar; whereas, the northern and the other hybrid were similar. Menzel (1962) has reported that hybrid quahogs in Florida grew faster than their northern parents and were more like the faster growing southern parent. The hybrid 9 M. campechiensis X 6 M. mercenaria had a slightly better growth rate than the reciprocal hybrid indicating the possibility of maternal influence. It would be interesting to determine the seasonal gonadal development of females of both species and hybrids in Florida. Previous observations in our laboratory have shown that it is virtually impossible to induce summer spawning of females of any pedigree after about March-April when the ambient water temperatures exceed 22 to 24°C. Active sperm appear in suspensions but few ripe ova occur in clams during the warmer months. Successful female spawnings have been induced during periods from October-March with no temperature conditioning. The seasonal gonadal development, therefore, may be different for female quahogs than reported here for young males. Also, it would be interesting to determine if quahogs of both species follow the pattern of gametogensis of the endemic population when transplanted to a colder latitude. Such observations might be difficult because the southern quahog and the hybrids lack a tolerance to low tempera- tures (Chestnut et al. 1956, Haven and Andrews 1956, Menzel 1977). Whether the northern quahog, native to warmer areas, would survive in cold winter regions is not known. Belding (1912) reported 70 years ago that tempera- ture is the controlling factor in quahog spawning. Based on the data of all the investigations, we believe that both species and the hybrids will have generally similar gamete development and spawning, regardless of their origin, within a specific area. Gonadal Development of Young Quahogs 17 REFERENCES CITED Abbott, R. T. 1974. American Seashells. New York, NY: Van Reinhold Company. 2nd edition. 663 p. Belding, D. L. 1912. A report upon the quahog and oyster fisheries of Massachusetts, including the life history, growth and cultiva- tion of the quahog (Venus mercenaria), and observations on the set of oyster spat in Well Fleet Bay, Boston. Boston, MA: Wright and Potter Print Co. 134 p. (Reissued: 1964. Mass. Dep. Nat. Resour. Div. Mar. Fish., Contrib. 12:134 p.) Chestnut, A. F. 1951. The oyster and other mollusks in North Carolina. Taylor, H.F., ed., Survey of Marine Fisheries of North Carolina. Chapel Hill, NC: Univ. N.C. Press; 141-190. , W. E. Fahy & H. J. Porter. 1956. Growth of young Venus mercenaria. Venus campechiensis, and their hybrids. Proc. Natl. Shellfish. Assoc. 47:50-56. Eversole, A. G., W. K. Michener & P. J. Eldridge. 1980. Reproductive cycle of Mercenaria mercenaria in a South Carolina estuary. Proc. Natl. Shellfish. Assoc. 70:22-30. Goodbody, I. 1965. Continuous breeding in populations of tropical crustaceans, Mysidium columbiae (Zimmer) and Emerita portori- censis (Schmidt). Ecology 46:195-197. Haines, M. L. 1976. The reproductive cycle of the sunray venus clam, Macrocallista nimbosa (Lightfoot, 1786). Proc. Natl. Shellfish. Assoc. 66:6-12. Haven, D. & J. D. Andrews. 1956. Survival and growth of Venus mercenaria, Venus campechiensis, and their hybrids in suspended trays and on natural bottoms. Proc. Natl. Shellfish. Assoc. 47:43-49. Keck, R. T., D. Maurer & C. H. Lind. 1975. A comparative study of the hard clam gonad developmental cycle. Biol. Bull. (Woods Hole) 148:243-258. Loosanoff, V. L. 1937a. Development of the primary gonad and sexual phases in Venus mercenaria Linnaeus. Biol. Bull. (Woods Hole) 72:389-405. . 1937b. Seasonal gonadal changes of adult clams, Venus mercenaria (L.). Biol. Bull. (Woods Hole) 72:406-416. . 1954. New advances in the study of bivalve larvae. A m. Sci. 43:607-624. & H. C. Davis. 1950. Conditioning Venus mercenaria for spawning in winter and breeding its larvae in the laboratory. Biol. Bull. (Woods Hole) 98:60-65. Loosanoff, V. L. &C. A. Nomejko. 1951. Existence of physiologically different races of oyster, Crassostrea virginica. Biol. Bull. (Woods Hole) 101:151-156. Menzel, R. W. 1961. Seasonal growth of the northern quahog, Mercenaria mercenaria and the southern quahog,M campechiensis, in Alligator Harbor, Florida. Proc. Natl. Shellfish. Assoc. 52: 37-46. . 1962. Seasonal growth of the northern and southern quahogs, Mercenaria mercenaria and M. campechiensis, and their hybrids in Florida. Proc. Natl. Shellfish. Assoc. 53:111-119. . 1968. Cytotaxonomy of species of clams (Mercenaria) and oysters (Crassostrea). Symp. Mollusca, Mar. Biol. Assoc. India. Part 1:75-84. . 1977. Selection and hybridization in quahog clams (Mercenaria spp.). Proc. World Maricult. Soc. 8:507-521. & M. Y. Menzel. 1965. Chromosomes of two species of quahogs and their hybrids. Biol. Bull. (Woods Hole) 129: 181-188. Nelson, T. C. 1928. On the critical temperatures for the spawning and for ciliary activity in bivalve molluscs. Science 67:220-221. National Oceanic and Atmospheric Administration. 1974a. Climato- logical Data, Florida. U.S. Dept. Commerce. 78. . 1974b. Tide Tables, East Coast of North and South America. U.S. Dept. Commerce. . 1975a. Climatological Data, Florida. U.S. Dept. Com- merce. 79. . 1975b. Tide Tables, East Coast of North and South America. U.S. Dept. Commerce. Orton, J. H. 1920. Sea temperature, breeding and distribution in marine animals. J. Mar. Biol. Assoc. U.K. 12:339-366. Porter, H. J. 1964. Seasonal gonadal changes of adult clams, Mercenaria mercenaria (L.) inNorth Carolina. Proc. Natl. Shellfish. Assoc. 55:35-5 2. Ropes, J. W. 1968. Reproductive cycle of the surf clam, Spisula solidissima. in offshore New Jersey. Biol. Bull. (Woods Hole) 135:349-365. Sanders, H. L. & R. R. Hessler. 1969. Ecology of the deepsea benthos. Science 163:1419-1424. Sastry, A. N. 1966. Temperature effects in reproduction of the bay scallop, Aequipecten irradians Lamarck. Biol. Bull. (Woods Hole) 130:118-134. & N. J. Blake. 1971. Regulation of gonad development in the bay scallop, Aequipecten irradians Lamarck. Biol. Bull. (Woods Hole) 140:274-283. Thorson, G. 1950. Reproductive and larval ecology of marine bottom invertebrates. Biol. Rev. Camb. Philos. Soc. 25:1-45. Journal of Shellfish Research, Vol. 3, No. 1, 19-27, 1983. EXPERIMENTAL PLANTINGS OF JUVENILES OF THE HARD CLAM MERCENARIA MERCENARIA (LINNE) IN THE WATERS OF LONG ISLAND, NEW YORK 1 PAUL J. FLAGG AND ROBERT E. MALOUF Marine Sciences Research Center Stare University of New York Stony Brook, New York 11794 ABSTRACT Planting of hatchery-reared seed of the hard clam Mercenaria mercenaria is a significant management tool in town-managed shellfisheries of New York. In the present study, seed planting techniques developed elsewhere were tested in New York waters. The objectives were to determine how seed survival was influenced by ( 1) seed size at the time of planting; (2) the presence, absence, and type of gravel aggregate; (3) the season planted; and (4) site selection. Site characteristics, particularly the types and abundance of predators present, were found to influence the results so strongly that general recommendations cannot be made. Mud crabs (Neopanope sayi [Smith] ) and whelks (Busy con carica [Gmelin] and B. canaliculatum [Linne]) were the most damaging predators at the sites tested. Gravel aggregate did not provide adequate protection for planted clams, and the use of large (25-mm) gravel appeared to have a negative impact on seed survival. Survival exceeded 10% only among clams that were at least 20 mm in length at planting; however, mortalities as high as 100% resulted from plantings of such seed (23 mm) at sites having significant populations of whelks. KEY WORDS: Hard clams, Mercenaria mercenaria, seed planting, predation INTRODUCTION The hard clam (or northern quahog) Mercenaria mercenaria (Linne) is the object of New York's most important shellfishery, accounting in recent years for about 50% of the total value of fishery products landed in the state (McHugh and Ginter 1978). Long Island's Great South Bay is the single most important producer of hard clams in the world. This 24,282-ha (60,000-acre) bay has historically produced about 90% of the New York harvest and 45% of the total United States harvest of hard clams. Since 1977, New York landings of hard clams have declined dramatically. For example, the 1976 reported Great South Bay landings were 24,684 m 3 (700,465 bu), but by 1981, the landings had dropped to 10,758 m 3 (305,287 bu) (National Marine Fisheries Service, Patchogue, NY, unpub- lished fishery statistics, 1982). Although stock assessment data are incomplete, declining harvests are perceived by many local fishery managers to represent a real drop in standing stocks (J. Kassner, Town of Brookhaven, NY, and Pieter Van Volkenburgh, NY Dept. Environm. Conserv., Stony Brook. NY, pers. comm.). Local management agencies, primarily the townships, have responded to declining landings by instituting programs intended to supplement natural hard clam reproduction. Among the most popular programs are those that involve the planting of seed clams. Nine Long Island townships, including all three of the townships that border Great South Bay. have carried out some type of seed clam planting program. Their efforts have ranged from trial plantings of a Contribution No. 378 of the Marine Sciences Research Center, State University of New York (SUNY) at Stony Brook. few thousand seed to annual plantings in excess of 1 million seed. Seed are purchased from a commercial hatchery, held in some type of nursery system, and eventually broad- cast onto the bay bottom without any protection. Nursery systems used include shore-based raceways and ponds, rafts, and gravel beds. The size of the seed at the time of release to the public fishery generally ranges from about 8 to 25 mm in shell length. There are no published studies of seed clam plantings in New York waters. In fact, some doubt has been expressed that the seed planting programs can possibly be of sufficient scale to significantly impact the fishery (McHugh 1981). The early work of Haven and Andrews (1957) showed that seed clams require some type of protection to ensure survival. Similarly, Menzel and Sims ( 1964) reported that seed clams planted in Florida required protection or had to be at least 12 mm in shell length to avoid very heavy predation losses. Castagna (1970) demonstrated that gravel aggregate helped prevent the loss of seed clams. Castagna and Kraeuter ( 1977) and Kraeuter and Castagna (1977) recommended the use of aggregate as part of a culture system that included baffles and fences. Their work and the work of Menzel et al. (1976) suggested that the use of stone aggregate alone affords planted seed clams some protection from predators. The use of stone aggregate would be particularly attractive for the extensive nursery plots that are required for large public fisheries because of its relative simplicity and low cost; it has been used on a limited basis for that purpose (Jeffrey Kassner, Town of Brookhaven, NY. pers. comm.). Eldridge et al. (1979) made the following recommenda- tions based on several years of seed clam planting in South Carolina: (1) select a physically suitable habitat, one that 19 20 Flagg and Malouf is free, for example, from extreme wave action; (2) cover the planting area with shell or stone aggregate; (3) plant seed clams in the fall when temperatures are 15 to 18°C; (4) plant seed of 12 to 15 mm shell length at a density of 300 m 2 ; and (5) harvest in the early summer of the second year. The authors pointed out that uncontrolled variables contribute to the uncertainty of such a planting as a private venture; however, they reported approximately 77% annual survival of 16- to 17-mm seed and 95% annual survival of 21- to 22-mm seed planted in this manner. Later work by Whetstone and Eversole (1981) also reinforced the case for fall plantings by demonstrating in laboratory studies that the activity of an important hard clam predator, the common mud crab Panopeus herbstii H. Milne-Edwards, was significantly reduced at temperatures below 17°C. The present study was part of an effort to test and refine a number of seed-clam planting techniques that have been developed elsewhere. The intention was to evaluate recommended planting procedures for possible application to a large public clam fishery. Specifically, the objectives were to determine in New York waters how the survival of three sizes of planted seed clams was affected: (1) by the size and shape of aggregate and sand substrate (Experiment I); (2) by the time (season) they were planted and recovered (Experiment II); and, (3) by site specific environmental differences within the same general location (Experiment III). materials and methods Experiment I was sited in a shallow cove, separated by a sand spit from Eastern Shinnecock Bay, Long Island, NY (designated as Site I, Figure 1 ). Mean low water depth at the site was approximately 0.5 m, and the tidal range averaged about 1.0 m. Sediments within the cove graded from coarse sand near the sand bar to soft mud near the northern edge of the cove. Eeel grass (Zostera marina Linnaeus) was present, but was relatively sparse through most of the planting area. A natural population of adults of Mercenaria mercenaria existed in the cove prior to our planting at a mean density of about 7 clams m" 2 . The seven substrates tested in this experiment consisted of sand and two shapes of gravel obtained in three sizes. The two shapes were (1) mechanically produced, crushed gravel having irregular shapes and jagged edges, and (2) more rounded, unbroken glacial gravel. Both gravel types were obtained in three nominal sizes: 6 to 10, 10 to 19, and 19 to 32 mm. The gravel was washed through wire screens to obtain the approximate size ranges given above. All gravel was obtained from Long Island glacial till and was washed thoroughly with fresh water during processing. Forty-two plastic, food-handling trays (Nestier® "Chill- tray 180") measuring 56.5 X 46.4 X 17.8 cm) were lined with 2-mm mesh plastic window screen. The trays were filled to a depth of approximately 8 cm with 20-mm gravel. They were then transported to the site, arranged in a 6X7 array, and hydraulically sunk (jetted) into the bottom so that approximately 3 cm of the tray edges protruded above the substrate. A 4-cm layer of one of the seven types of substrate was then added to the surface of each tray in a randomly generated pattern. Three sizes of seed clams used in the experimental plantings were obtained from Aquaculture Research Corp., Dennis, MA. At the time of planting (23 July 1980), the mean shell lengths and standard errors (n = 50) for clams of the size groups were 3.9 ± 0.06, 6.8 ± 0.08, and 28.7 ± 0.23 mm. Planting densities used were 1,241, 477, and 191 m" 2 for the small, medium, and large seed, respectively. Thus, a tray randomly received 325 small, 125 medium, or 50 large seed. The experimental design included two replicate plantings for each treatment. Because there was no differentiation of substrate shape for plantings in sand, for each clam size there were four replicate plantings in sand. Also, because they were in short supply, the largest seed clams were only planted in the three sizes of round gravel and in sand. The planting area was examined weekly to identify and count potential clam predators. The experiment was terminated on 20-22 October, when water temperatures in the area dropped below 10°C. The trays were lifted on board a small boat, and all remaining clams were removed and counted and their shell lengths were measured to the nearest millimeter. Empty shells and shell fragments were examined for evidence of predation, and any predators recovered with the trays were identified and counted. Growth and survival (recovery) data were statistically analyzed by analysis of variance ( ANOVA) following Sokal and Rohlf (1969). Shell length measurements were used to calculate growth in millimeters. Experiment II was initiated in the fall of 1980 at two locations (designated Sites IIA and IIB, Figure 1) in Eastern Long Island. Site IIA was located in Shinnecock Bay approximately 30 m east of the previously described site of Experiment I. Site IIA had a mean low water depth of approximately 0.35 m, and bare sandy sediments. Site IIB was located in Napeague Harbor, Long Island. Mean low water depth at the site was 1.0 m, and the tidal range was 0.9 m. Sediment at Site IIB consisted of a 3-cm-deep layer of sand over gravel and stones. The area was devoid of eel grass and macroalgal detritus. A sparse (< 1 irf 2 ) natural population of very large hard clams existed at Site IIB prior to our planting. Experiment II consisted of two replicate plantings of each of three clam sizes in two substrates types (sand and 1 cm crushed gravel) at two sites and at two planting times. The two planting times and ambient water temperatures at the two sites were: 30 September 1 980 ( 1 9°C) and 25 Novem- ber (8°C) for Site IIA, and 30 September 1980 (17°C) and 22 November 1980 (8°C) for Site IIB. Seed clams were again purchased from Aquaculture Research Corp. Mean shell lengths and standard errors (n = 50) for the three size Experimental Plantings of mercenaria mercenaria 21 CT. 10 n mi -I 5km .«&cP ATLANTIC OCEAN Shmnecock Bay meters I 1 1 1000 ATLANTIC OCEAN Nope ague Bay meters ATLANTIC OCEAN Figure I. Location of six sites used for experimental plantings of seed clams on the south shore of Long Island, New York. 22 FLAGG AND MAI.OUF classes in the September planting were 2.8 ± 0.17, 7.1 ± 0.10, and 22.7 ± 0.15 mm. Rapidly declining ambient water temperatures necessitated the planting of the November shipment immediately upon receipt. Therefore, although hatchery sorting through sieves was identical for the two shipments, shell measurements for the November shipment were not recorded. Tray handling, seed-planting procedures, and planting densities were as in Experiment I. The planting sites were inspected regularly for predator distribution and abundance. Final sampling of the trays was conducted 9 months after the planting date (15— 22 June and 23—27 August 1981 for the September and November plantings, respectively). Sampling procedures and data analysis were as in Experiment I except that no growth analyses were included in Experiment II. Experiment III consisted of plantings on prepared natural bottom without trays. Plantings were carried out at three sites (designated as Sites IIIA, IIIB, and IIIC, Figure 1 ) in one general location, Napeague Harbor, Long Island. Three sizes of seed clams (nominally, 3, 6, and 23 mm in length) were planted at each site, with and without gravel, during the summer of 1 98 1 . Site IIIA was located approximately 40 m east of Site IIB, described above. The site had a mean low water depth of 1.2 m and a tidal range of 0.9 m, and contained poorly sorted sand and gravel sediments. Site IIIB, in northeastern Napeague Harbor, had a mean low water depth of 0.4 m and a tidal range of 0.9 m. Sediments at the site consisted of coarse sand sparsely interspersed with rocks. The site was on the edge of an approximately 1 ha bare area in an eel grass flat. A dense (20 to 50 rrf 2 ) population of small adult hard clams existed at the site prior to our planting. Site IIIC was located on a large bare sand/mud flat in the southwestern part of the harbor. Mean low water depth was 0.4 m and tidal range was 0.9 m. Hard clams, predominately adults plus a few subadults, were moderately abundant (5 to 10 irf 2 ) prior to our planting. Seed clams were purchased from the same commercial source in the same three nominal sizes as used in the pre- viously described experiments (2 to 4, 6 to 8, and 22 to 28 mm length). Each of the three sites consisted of six 2- X 2-m subsites delineated by 30- wide X 15-cm-deep borders of 3-cm gravel. Each of the three seed clam sizes were randomly assigned to two subsites. One of the two subsites contained existing substrate, while the other contained a 2.5-cm-deep layer of 1 .0 cm gravel. On 20 May 1981, clams were planted at all sites at densities of 1,250, 675, and 260 m~ 2 for small, medium, and large clams, respectively. Surveys of predator abundance were conducted prior to planting (17-20 May 1981) and were repeated on 26—28 July and 13—14 September 1981. Sampling areas adjacent to each site (30 m 2 in May and July and 15 m 2 in Septem- ber) were raked with a clam rake lined with 1.3-cm Vexar®, and predators were collected, counted, and measured. Esti- mates of the abundance of the more mobile crabs (primarily Ovalipes ocellatus [Herbst] ) were subject to error because of the animals' mobility and are, therefore, not quantitative. Sampling to determine seed clam survival was conducted approximately two months after planting (26 July) and again at termination (14 September). For purposes of sampling, each subsite was divided into four 1 m 2 quadrats and each quadrat into nine equal parts (0.1 1 m 2 each). Two of the 0.1 1 m 2 areas were randomly selected from each of two randomly selected quadrats. A 0.10 m 2 sampling square was placed on a selected area, and substrate was removed to a depth of 15 cm. After being separated from the substrate, surviving clams were counted and returned to the sample area. Analysis of survival data was as described above. RESULTS Experiment I The most abundant clam predators observed in and around the trays following planting were Say's mud crabs (Neopanope sayi), calico crabs {Ovalipes ocellatus), channeled whelks (Busycon canaliculatwn), and oyster drills (Urosalpinx cinerea [Say] and Hupleura caudata [Say] . Other potential predators which were less frequently observed included blue crabs (Callinectes sapidus Rathbun), common mud crabs (Panopeus herbstii), and both winter and summer flounders (Pseudopleuronectes americans [Wal- baum] and Paralichthys dentatus [Linnaeus]), respectively. The abundance of the mud crab N. sayi was positively related to increased gravel size (Table 1). Those trays filled with 19- to 32-mm gravel contained numerous 0-year-class crabs. Up to 10 oyster drills (U. cinerea and E. caudata) per tray occurred during the summer, but no drills were found in the trays during the autumn sampling. Similarly, channeled whelks (B. canaliculatum) were visible at the substrate surface, and were most abundant during the first month (August) following planting. Few were observed later in the summer, and only two were recovered from the trays during sampling. Survival (recovery ) of planted seed clams was significantly influenced by their size at the time of planting (0.01 >?> 0.001). Mean survival rates for small, medium, and large clams were 4.0, 43.1, and 82.5%, respectively. The size of the gravel used also significantly affected clam survival (0.01 > P > 0.001). Further, the relationship between grain size, independent of shape, and clam survival appeared to be related to clam size (the interaction was significant; P < 0.01 ). The smallest seed clams planted (4 mm) did not survive well under any conditions. On the other hand, the survival of the 29-mm seed was high and was independent of grain size. The influence of grain size on clam survival at this site was most evident among the 8-mm seed, which showed declining survival with increased grain size (Table 1). Experimental Plantings of Mercenaria mercenaria 23 The shape of the gravel used had no significant effect (P > 0.05) on clam survival. TABLE 1. Experiment I, percent recovery (22 August - 22 October 1980) of three sizes of seed clams planted in three sizes of gravel and in sand. Also shown are the total number of mud crabs (Neopanope sayi) recovered from trays containing the four substrate types. Substrate Type Length of ■ — ■ — Seed at 6 to 10-mm 10 to 19-mm 19 to 32-mm Planting Sand Gravel Gravel Gravel Mean 3.9 mm (n = 4) 7.9 mm fn = 4) 28.8 mm 14.6 68.4 77.0 1.1 49.6 84.0 0.7 48.6 84.0 0.0 5.8 86.0 4.0 43.1 82.5 Total crabs recovered 24.0 (n= 10) 36.0 95.0 >306 Final mean shell lengths for the three clam sizes are given in Table 2. Effects of substrate size or shape on clam growth were not significant for 29-mm seed (P > 0.05). High mortality precluded an analysis of growth in the 4-mm clams. Increasing substrate size did have a significant nega- tive effect on the growth of 8-mm seed (0.01 >P>0.001). TABLE 2. Experiment I, final mean shell lengths (mm) with 95% confidence intervals (n =12, time = 85 days) for two sizes of seed clams planted in four types of substrate Length of Seed at Planting Substrate Type 6 to 10-mm 10 to 19-mm 19 to 32-mm Gravel Gravel Gravel Sand 3.9 mm * * * 7.9 mm 15.4 ±2.03 14.0 + 2.21 12.9 ±2.08 28.8 mm 31.8 ± 1.10 33.4 ±3.05 33.0 ±0.12 9.5 ±3.84 31.7 ± 1.48 'Survival was too low to calculate growth rates. Only a few shell fragments, indicative of crab predation, were found in the trays containing 4-mm seed. The shells of these clams were thin enough to be crushed and consumed by feeding crabs (Landers 1954; Whetstone and Eversole 1978, 1981). Many shell fragments were found in the trays containing the 8-mm seed. Laboratory studies indicated that clams of this size can be crushed and consumed by adult mud crabs, N. sayi (Landers 1954, Whetstone and Eversole 1978). Shells of dead clams of the larger (29-mm) seed were primarily paired, intact valves. Several shells had been cracked, possibly by a large calico crab (O. ocellatus) or blue crab (C. sapidus). A few shells had chipped or rasped shell margins suggesting predation by whelks, Busycon spp. (Carriker 1951, Peterson 1982). Oyster toadfish (Opsanus tau [Linnaeus] ) were observed burrowed along the outside edges of three of the trays throughout the summer and autumn. Three of the four trays of 4-mm clams planted in sand had survival rates of 3.0, 2.4, and 5.0%. The fourth tray, next to which a toadfish was burrowed, had a survival rate of 47.3%. Similarly, three of the four trays of 8-mm clams planted in 10- to 19-mm gravel contained a mean of seven mud crabs per tray and had clam survival rates of 48.0, 38.4, and 23.2%. The fourth tray, which had a toadfish beside it, contained no mud crabs and had a survival rate of 84.0%. A third toadfish was found beside a tray containing 29-mm clams. No mud crabs were found in this tray, but clam survival in that tray (82%) was not appreciably different from the mean for clams of that size (82.5%). From these observations, we hypothesize that the toadfish reduced the abundance of mud crabs and enhanced the survival of those seed sizes that were susceptible to mud-crab predation. Experiment II Predators observed at Site 1IA were essentially the same as those listed earlier for nearby Site I. The most abundant predators observed at Site IIB included calico crabs (Ovalipes ocellatus) and small knobbed whelks {Busycon carica). Mud crabs (Neopanope sayi) and small winter flounders (Pseudopleuronectes americanus) were present but not abundant. Significant interactions among the variables tested (size of seed planted, location, time of planting, and substrate type) indicated that unqualified general statements about any single variable cannot be valid (Tables 3 and 4); however, by considering some of the variables together, some important results may be noted. All of the variables tested had significant effects on survival (Table 4). Larger seed showed better survival than small seed, particularly at Site IIA. The September-to-June period resulted in better overall survival than the November-to-August period. Gravel was generally a better substrate than sand for the larger clams at Site IIA, but it did not appear to provide significant survival advantage at Site IIB (Table 3). As in Experiment I, mud crab colonization was greater in gravel than in sand. Experiment III Dominant predators observed during Experiment III included small (70- to 80-mm length) knobbed whelks (Busycon carica), adult (15- to 25-mm carapace width) mud crabs (Neopanope sayi), and adult (45-mm carapace width) calico crabs (Ovalipes ocellatus). Abundances of the two major predator species (B. carica and N. sayi) for which reliable counts could be made at Sites I IIA, IIIB, and IIIC are given in Table 5 for three observation dates. 24 Flagg and Malouf TABLE 3. Experiment II, percent recovery (time = 9 months) of three sizes of seed clams in replicate plantings at two sites in two types of substrate and at two times of the year. September Planting November Planting Site HA Site IIB Site HA Site IIB Clam Size Sand Gravel Sand Gravel Sand Gravel Sand Gravel 3 mm 3.6 0.6 0.0 0.3 0.0 0.0 0.0 0.0 5.7 0.9 0.0 0.0 0.0 0.0 0.0 0.0 Mean 4.7 0.8 0.0 0.2 0.0 0.0 0.0 0.0 7 mm 6.4 30.4 0.8 4.0 0.0 8.8 0.0 2.4 11.2 16.0 2.4 1.6 0.8 4.8 0.0 0.0 Mean 8.8 23.2 1.6 2.8 0.4 6.8 0.0 1.2 23 mm 68.0 94.0 24.0 48.0 48.0 50.0 10.0 18.0 68.0 96.0 20.0 42.0 34.0 66.0 26.0 14.0 Mean 68.0 95.0 22.0 45.0 41.0 58.0 18.0 16.0 TABLE 4. TABLE 5. Experiment II, four-way analysis of variance (ANOVA) of percent survival of three sizes of seed clams (3, 6, and 23 mm) planted in two types of substrate (sand and gravel) at two locations and at two times of the year (September and November). Experiment III, abundance of predators (m *) of the two numerically dominant predator species, the mud crab Neopanope sayi and the knobbed whelk Busycon carica. Source of Variation Mean Square d.f. F Ratio A = substrate type B = clam size C = time of year D = location AXB AXC AXD BXC BXD CXD AXBXC axbxd aXcxd BXCXD aXbxcxd Within Total 377.78 7,140.01 1,241.96 2,244.07 187.18 33.60 48.72 148.21 490.22 217.00 75.60 69.87 7.19 4.00 3.93 13.68 1 2 1 1 2 1 1 2 2 1 2 2 1 2 2 24 47 27.61* 521.76* 90.75* 163.99* 13.68* 2.46 n.s. 3.56 n.s. 10.83* 35.82* 15.86* 5.52f 5.11* 0.53 n.s. 0.29 n.s. 0.29 as. Sampling Date Neopanope sayi Site HIA IIIB IHC Busycon carica Site HIA IIIB IHC 20 May 1981 28 July 1981 14 September 1981 Mean 2.0 0.3 0.0 2.0 0.0 0.0 1.0 0.0 0.0 1.7 0.1 0.0 2.5 1.0 0.3 7.0 2.0 8.6 1.5 1.1 1.2 3.7 1.4 3.4 *significant at 0.01 fsignificant at 0.05 n.s. = not significant B>D>C>A In general, survival at Site IIIA was inversely related to seed size (Table 6). Overall survival was less than 2% even under the best conditions (3-mm seed in gravel). Only one of the 6- to 8-mm clams was recovered in July, and by the termination date (30 September) no clams of that initial size had survived. No larger seed clams were recovered in the July sampling. Within a week of planting, empty shells appeared on the substrate surface. Maximum recovery cf the 2- to 4-mm seed (in gravel) was 2.2% at Site IIIB. None of the 6- to 8-mm clams was recovered, and crushed and cracked shells appeared in the plots within two weeks of planting. Survival of seed planted at Site IIIB exceeded 50% only among the 22- to 28-mm seed clams. Note also in Table 6 that among the 22- to 28-mm seed there appeared to an initial survival advantage to clams planted in gravel compared to natural bottom, but by the time of the final sampling in September, survival rates were very similar in the two substrate types. Chipped shell margins and cracked shells indicated predation by whelks and crabs. At Site IIIC, survival of the small seed in sand, although still quite low, was somewhat better than that of the larger seed sizes (Table 6). By the end of the experiment none of Experimental Plantings of Mercenaria mercenaria 25 TABLE 6. Experiment III, percent recovery of three sizes of seed clams planted at three sites in two types of substrates. Clams were planted 20 May 1981. 3 mm 7 mm 23 mm Site Site Site IIIA llllt IIIC Sampling Date Sand Gravel Sand Gravel Sand Gravel IIIA 1MB IIIC IIIA IIIB IIIC Sand Gravel Sand Gravel Sand Gravel Sand Gravel Sand Gravel Sand Gravel 28 Jul 81 14 Sep 81 3.6 1.5 11.0 1.8 2.0 1.2 3.3 2.2 10.4 6.2 4.5 4.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 15.0 0.0 0.0 0.0 0.0 0.0 62.0 58.0 88.0 61.0 1.7 0.0 14.8 5.3 the 6- to 8-mm clams remained, and a few 23-mm seed survived only in gravel (5.3%). Heavy losses of the larger seed clams, the chipped or rasped shell margins of articu- lated, empty valves remaining in the planting areas, as well as the high densities of knobbed whelks (B. carica) at this site (Table 5) suggested that predation by whelks was an important cause of mortality. DISCUSSION The results of this study demonstrated that the character- istics of a given site, especially the types of predators present, had an important influence on the loss and presumed mortality of planted seed clams and on the degree of pro- tection afforded by recommended culture techniques. For example, we found that at sites such as Site IIIA where whelks (Busycon eanaliculatum and B. carica) were abundant, plantings of 25-mm seed clams suffered complete mortality despite the presence of gravel aggregate. At sites such as Site IIIB where mud crabs (Neopanope sayi) were the dominant predator, the smallest seed clams suffered high mortality, while the larger seed showed good survival. Clearly, the idea that seed clams having at least a 25-mm shell length are relatively immune from most predators (Menzel 1971, Eldridge et al. 1979) is valid only when the seed is planted at sites lacking significant populations of large predators. Existing literature has convincingly shown that the activity of some important predators such as mud crabs is significantly reduced by lower autumn temperatures (Whetstone and Eversole 1981); however, we found that autumn plantings eventually suffered the same high mortal- ities as the summer plantings, and the choice of planting season was inadequate protection against crab predation. The use of gravel aggregate at Site I, where the mud crab N. sayi was the dominant predator, gave inconsistent results in our Experiment I (Table 1). At that site, mortality among the smaller clams was complete and was independent of the presence or absence of gravel. On the other hand, mortality among the larger clams was very low, but it was again independent of substrate grain size. The survival of the medium size (7.9-mm) seed was inversely related to gravel size. The 6- to 10-, 10- to 19-, and 19- to 32-mm gravels, all of which are within the size range (10 to 30 mm) used by Castagna and Kraeuter(1977), were not consistently effective in enhancing seed clam survival (Table 1). Densities of the mud crab N. sayi were much higher in gravel beds than in the bare sand (Table 1). There is also evidence from our data (Table 2) of reduced growth rates among small seed clams planted in larger gravel compared to those planted in sand or small gravel. Gravel may be useful in preventing small clams from being carried away by currents, although our work offers no direct evidence for this. It is also possible that gravel and shell substrates offer more effective protection against larger crab species than against relatively smaller species such as N. sayi. Size-related differences in the food and space utilization of two sympatric xanthid crab species (Panopeus herbstii and Eurypanopeus depressus [Smith] ) were discussed by McDonald (1982). He noted that the larger of the two species (P. herbstii) was prevented by its size from entering narrow spaces between living oysters. This suggests that the lack of consistent results from seed plantings in gravel might be due in part to site-specific differences in the relative abundance of large and small crabs. Previous studies have shown that xanthid mud crabs (primarily N. sayi) are the most abundant clam predators in Long Island's Great South Bay (MacKenzie 1977). Their mean, baywide abundance is about 4.4 crabs m~ 2 . while that of Ovalipes ocellatus is about 0.2 crab m~ 2 (WAPORA, Inc. 1981). Mud crabs are capable of consuming 1.6 to 5 small (5- to 10-mm) hard clams each day (Landers 1954. MacKenzie 1977). Theoretically, mud crabs in Great South Bay could consume up to about 20 seed clams rrf 2 day" 1 . At this rate of loss, seed plantings of 200 to 500 clams m~ 2 would not survive long. Consequently, local seed planting efforts that do not somehow protect the young clams until they are large enough to avoid mud crab predation will probably be unsuccessful. Although seed hard clams are readily available from commercial hatcheries, their cost is relatively high. Costs for 3- to 5-mm seed range from $10 to $15/1,000 at the present time (J. Kassner, Town of Brookhaven, NY and 26 FLAGG AND Malouf S. Buckner. Town of Islip, NY, pers. comm.). Assuming that harvested littleneck clams have a dockside value of about $70 per bag of 500, then the survival and harvest of planted seed (initially costing $12/1,000) must exceed 9% of the number planted for the value of the harvest to exceed the cost of the seed alone. A typical Long Island town program might plant about 2 million seed and could require about 6 man-months of handling and planting time. If the costs of handling and planting are added to the cost of the seed itself, then the survival requirement might increase to about 15%. This estimated survival requirement is relatively low compared to other estimates for commercial culture (40% by Castagna and Kraeuter 1977, 50% by Menzel et al. 1976). It should be remembered that our estimated survival and harvest requirements are minimum values for seed planted in a public fishery. Existing programs involve relatively little handling and no maintenance or protection after planting on the bay bottom. If the costs of a nursery system (rafts, racks, etc.) were added to our estimate, the survival requirement for cost effectiveness would approach those given above for commercial systems. Our essentially unprotected plantings of 3- to 5-mm seed clams rarely resulted in survival rates as high as 10%, even in short-term experiments. Other work, summarized in Table 7, showed similar results with seed of this size. In fact, 0% survival was the most commonly encountered result of unprotected planting of small seed clams. Even when various types of protective measures were employed, mortality among small seed clams often exceeded 50% (Table 7). The relatively low expected survival rates contribute to the problem of scale in these programs (discussed by McHugh, 1981). For example, a survival rate of even 15% would leave only 300,000 clams available for harvest from a planting of 2 million seed. In the very unlikely event that all of these clams were harvested, this would yield only 21 m 3 (600 bu), or about 0.6% of each of the three Great South Bay towns' typical annual harvest. In fact, available data (Table 7) indicate that survival rates and consequentual harvest contributions might be much lower. ACKNOWLEDGMENTS The authors thank Dr. J. L. McHugh for his critical review of the manuscript and Charles DeQuillfeldt for his technical assistance during the study. The cooperation of Bradden Smith of Shinnecock Tribal Oyster Project, Emil Usinger of Blue Points Co., Inc., and the East Hampton Town Council is gratefully acknowledged. Support for this study was provided by the National Oceanic and Atmos- pheric Administration, Office of Sea Grant, through the New York Sea Grant Institute. TABLE 7. Published accounts of some trial plantings of seed clams (Mercenaria mercenaria) on the Atlantic coast of the United States. Reference Seed Size Planted (mm) Seed Size Recovered (mm) Duration (Months) Approximate Survival (%) Notes Menzel and Sims (1964) 33-44 - - 82-95 33-44 - - Godwin (1968) 18-22 - 10 18-22 35-37 10 50 18-22 - 10 18-22 50-52 10 51 18-22 36-37 10 36 Menzel (1971) 15-35 - - 90 Walne (1974) 9-13 17-21 6 88 Eldridgeet al. (1976) 12-13 16-25 4 64 16-25 29-45 12 76 Menzel et al. (1976) 7-10 - 11 0.6 7-10 - 11 2.3 7-10 - 11 10.1 7-10 - 11 58.6 Eldridge et al. (1979) 13 16-19 4 62 16-19 46-57 24 81 Castagna and Kraeuter (1977) 2 - 11 75 Kraeuter and Castagna (1977) 2 - 11 2 17 11 1- 3 2 17 11 10-22 Kraeuter and Castagna (1980) 32 39 4 94 32 39 4 9 Protection (fence, baited traps) No protection No protection No protection No protection Protection (wire mesh) Protection (wire mesh; loss due to "winter-kill") Protection (fence, traps) Protection (plastic mesh) Protection (covered trays) Protection (covered trays) No protection Protection (shell cover) Protection (gravel) Protection (wire mesh) Protection (covered trays) Protection (same planting as above) Protection (gravel, traps, baffles) No protection Protection (gravel only) Protection (gravel, baffles) Protection (pen, gravel, baffles) Protection (no pen, with gravel, baffles) EXPERIMENTAL PLANTINGS OF MERCENARIA MERCENARIA 27 REFERENCES CITED Carriker, M. R. 1951. Observations on the penetration of tightly closing bivalves by Busycon and other predators. Ecology 32:73-83. Castagna, M. 1970. Field experiments testing the use of aggregate covers to protect juvenile clams. Proc. Natl. Shellfish. Assoc. 60:2 (abstract). & J. Kraeuter. 1977. Mercenaria culture using stone aggre- gate for predator protection. Proc. Natl. Shellfish. Assoc. 67: 1-6. Etdridge, P. J., A. G. Eversole & J. M. Whetstone. 1979. Compara- tive survival and growth rates of hard clams Mercenaria mercen- aria, planted in trays subtidally and intertidally at varying densities in a South Carolina estuary. Proc. Natl. Shellfish. Assoc. 69:30-39. Eldridge. P. J., W. Waltz, R. C. Gracy & H. H. Hunt. 1976. Growth and mortality rates of hatchery seed c\ams,Mercenaria mercenaria, in protected trays in waters of South Carolina. Proc. Natl. Shell- fish. Assoc. 66:13-20. Godwin, W. F. 1968. The growth and survival of planted clams, Mercenaria mercenaria, on the Georgia coast. Georgia Game Fish Comm. Mar. Fish. Div. Contrib. Ser. No. 9. 16 p. Haven, D. & J. D. Andrews. 1957. Survival and growth of Venus mercenaria, Venus campechiensis, and their hybrids in suspended trays and on natural bottoms. Proc. Natl. Shellfish. Assoc. 47: 43-49. Kraeuter. J. N. & M. Castagna. 1977. An analysis of gravel, pens, crab traps, and current baffles as protection for juvenile hard clams (Mercenaria mercenaria). Proc. World Maricult. Soc. 8:581-592. . 1980. Effects of large predators on the field culture of the hard clam, Mercenaria mercenaria. U.S. Fish Wildl. Serv. Fish. Bull. 78(2):538-540. Landers, W. S. 1954. Notes on the predation of the hard clam Venus mercenaria by the mud crab. Neopanope texana. Ecology 35(3):422. Mackenzie, C. L. 1977. Predation on hard clam {Mercenaria mercen- aria) populations. Trans. Am. Fish. Soc. 106(6):530-537. McDonald, J. 1982. Divergent life history patterns in the co-occurring intertidal crabs Panopeus herbstii and Eurypanopeus depressus (Crustacea: Brachyura: Xanthidae). Mar. Ecol. Prog. Ser. 8: 173-180. McHugh, J. L. 1981. Recent advances in hard clam mariculture. J. Shellfish. Res. l(l):51-56. & J. J. C. Ginter. 1978. Fisheries. National Oceanic and Atmospheric Administration, Marine Ecosystems Analysis Program (MESA) New York Bight Atlas Monogr. No. 16. 129 p. Available from: NY Sea Grant Inst., Albany, NY. MenzeL R. W. 1971. Quahog clams and their possible mariculture. Proc. World Maricult. Soc. 2:23-36. , E. W. Cake, M. L. Haines. R. E. Martin & L. A. Olsen. 1976. Clam mariculture in northwest Florida: field study on predation. Proc. Natl. Shellfish. Assoc. 65:59-62. MenzeL R. W. & H. W. Sims. 1964. Experimental farming of hard clams, Mercenaria mercenaria, in Florida. Proc. Natl. Shellfish. Assoc. 53:103-109. Peterson, C. H. 1982. Clam predation by whelks {Busycon spp.): Experimental tests of the importance of prey size, prey density. and seagrass cover. Mar. Biol. (Berl.) 66:159-170. Sokal, R. R. & F. J. Rohlf. 1969. Biometry. San Franciso, CA: W. H. Freeman and Co. 776 p. Walne, P. R. 1974. Culture of Bivalve Molluscs, 50 Years' Experience at Conwy. Surrey, England: Fishing News (Books) Ltd. 173 p. WAPORA, Inc. 1981. Estuarine impact assessment (shellfish resources) for the Nassau-Suffolk streamflow augmentation alternatives, draft report on existing conditions. Available from: U.S. Environ. Protect. Agency, New York. 114 p. Whetstone, J. M. & A. G. Eversole. 1978. Predation on hard clams, Mercenaria mercenaria, by mud crabs, Panopeus herbstii. Proc. Natl. Shellfish. Assoc. 68:42-48. . 1981. Effects of size and temperature on mud crab, Panopeus herbstii, predation on hard clams, Mercenaria mercen- aria. Estuaries 4(2) : 1 53 — 156. Journal of Shellfish Research, Vol. 3, No. 1, 29-40, 1983. TRANSPORT OF BIVALVE LARVAE IN JAMES RIVER, VIRGINIA 1 J. D. ANDREWS Virginia Institute of Marine Science School of Marine Science College of William and Mary Gloucester Point, Virginia 23062 ABSTRACT For nearly 100 years, the James River has been the primary source of seed oysters for Virginia. A disease caused by Minchinia nelsoni (MSX) killed most oysters in high-salinity waters in the lower river in 1959 and 1960, and planting has not been resumed in these areas (Andrews 1983). Large populations of oysters on Hampton Bar and near the mouth of the river which served as broodstocks were destroyed. After 1960, setting declined drastically in regularity and intensity to about one tenth of that which occurred in the 1950's. Setting patterns suggest two types of seed areas in Chesapeake Bay: (1) high freshwater discharge, open or flushing estuaries with light spatfalls that decrease in intensity with distance from the river mouth; the James River is a typical example; and (2) low discharge, trap-type estuaries where intensive sets are heaviest near the head of the saline sector; examples are the Piankatank and Great Wicomico rivers in Virginia. Larval transport systems in the two estuarine types differ in quantity of larvae retained and regularity of spatfalls. Hourly plankton samples in the James River during 10 days in 1964 and 1965 revealed regular cyclic abundance of larvae with tidal stages. Larvae were 5 to 10 times more numerous during high-tide periods than at low-tide periods. Mostly early-stage larvae were distributed randomly throughout vertical columns of water. Larvae of other bivalve species exhibited similar distributions and fluctuations in abundance with tidal stages. Patterns of larval distribution were similar for all depths at five stations, both in the channel and over oyster beds, during 16 tidal cycles in 1965. Frequent recruitment of new larval broods and disappearance of most oyster larvae before ages of 3 to 5 days suggest losses due to physical disper- sion and predation. Only when larvae reached advanced umbo stages did they actively select deeper water strata in the channel which provided a transport system to carry them upriver. In the 1950's. spatfall occurred every week in the James River from 1 July to 1 October each year; since 1960, light, erratic setting has prevailed every year. If one assumes that predation, larval ecology, and physical transport systems have not changed, it appears that broodstocks have become inadequate, or that larvae were killed by toxic substances. KEY WORDS: Molluscs, bivalve larvae, transport, distribution, setting (or spatfall), James River, VA INTRODUCTION The James River has supplied seed oyster (Crassostrea virginica [Gmelin] ) for most private grounds in Chesapeake Bay for over 100 years (Andrews 1951, 1955, 1982a). The seed area is located in low-salinity waters (< 18 ppt in late summer) between the James River Bridge and the Deep Water Shoal (Figure 1). The horizontal salinity gradients in the James River are steep compared to those of other estuaries in Chesapeake Bay; salinity in the upper river seed beds ranges from ppt in late winter and spring to 10 or 12 ppt in late summer and fall. Consistent annual spatfalls of moderate intensity averaged 2.7 surviving spat per shell over 17 years from 1944 to 1960 (Andrews 1982a). During that period, 90% of surviving spat set on other oysters. Two to three million bushels (7.0 to 10.6 X 10 4 M 3 ) of seed oysters were harvested annually without depleting James River stocks. Oysters in the seed area were stunted in growth and storage of glycogen was low; therefore, they produced small quantities of spawn; but high-density populations were spread over large areas of natural shell beds; no management was applied except for limited harvesting by hand tongs. Good quality seed oysters with many single oysters and small clumps resulted from regular Contribution No. 1180, Virginia Institute of Marine Science spatfalls and low survival of initial sets (2 to 4% [Andrews 1949] ). Compared to high-salinity areas along the Atlantic coast of North America, those survival rates were high (Mackin 1946). Two types of seed areas are recognized in Chesapeake Bay based primarily on size of drainage areas and amount of freshwater discharge (Andrews 1979, 1982b). In the category of high-freshwater flow are the Susquehanna. Potomac and James rivers, but only the James permits recruitment of young oysters with enough regularity and intensity to be a seed area. Strong freshwater discharge provides the motive force in these estuaries to establish strong salinity gradients and a net counterflow of salty water upriver in the channel; it also produces high flushing rates to discharge the additional fresh water. The other category of estuaries, which I call trap-type seed areas (Andrews 1979), consists of low-discharge rivers with small drainage areas. Two examples of this type seed area which have been studied are the St. Marys River (Manning and Whaley 1954) for distribution and retention of larvae, and the Manokin River (Carter 1967) for circulation regimes. Other important seed areas in Chesapeake Bay which belong in this trap-type category are the Piankatank and Great Wicomico rivers in Virginia, and Broad Creek, a branch of the Choptank River in Maryland (Boicourt 1982). 29 30 ANDREWS OLD CHANNEL BURWELL BAY Figure 1. Map of James River seed area from Hampton Roads to last upriver seed bed at Deep Water Shoal. Sampling stations and associated oyster beds are designated in kilometers from mouth of the river. Transport of bivalve Larvae 31 The oyster setting patterns in these high-flushing and trap-type estuaries reflect differences in circulation patterns that result in dispersion or retention of larvae. The James River is the only flushing-type estuary in Chesapeake Bay with adequate spatfall to be a seed area. Spatfall was con- sistent annually, but from low to moderate in intensity; it exhibited a gradient of declining setting intensity from the mouth to upriver areas (Andrews 1982a). The gradient of setting was reversed in trap-type estuaries with highest spatfalls on the upriver beds (Manning and Whaley 1954, Andrew's data in Haven et al. 1978). For comparison, setting was consistent in intensity and regular by years in the James River; but intensity was much higher in trap-type estuaries and quite irregular by years with frequent failures. There was no change in the patterns of spatfall in trap-type estuaries following introduction of the disease caused by Minchinia nehoni (MSX) to Chesapeake Bay in 1959 (Andrews and Wood 1967); but in the James River there was a severe reduction in setting intensity and spatfall became erratic in distribution (Haven et al. 1978). All seed areas in Chesapeake Bay are in low-salinity (< 20 ppt) waters and usually not subject to MSX infections and mortalities; broodstocks were greatly reduced in the lower James River by MSX, but they were not in the trap-type seed areas which are located upbay and lay mostly above the endemic area for the disease. The geography and morphology of the two types of estuaries are probably significant factors with respect to dispersion and retention of larvae (Andrews 1979). The James River has a wide, deep channel, bordered by wide, shallow flats where oyster beds are located; it has few tributaries and limited marsh areas adjacent to the oyster- growing sector. The trap-type seed areas have meandering channels, numerous projecting points, very shallow flats, and many tributary creeks. Reduction and deflection of currents by boundary effects and morphometry in these tortuous estuaries probably aid in retention of larvae. The Great Wicomico River is an excellent example of the morphology of a trap-type estuary with its characteristics of infrequent but intensive spatfalls. Over 30 years, failures have been more frequent than successes in the Virginia trap-type rivers (Haven et al. 1978). The first study of larval transport in Chesapeake Bay was conducted in the James River in 1950 by the Virginia Fisheries Laboratory and the Chesapeake Bay Institute (CB1) (Pritchard 1953). An intensive study of physical and chemical hydrology was conducted by CBI (Pritchard 1952, 1955). Concurrently, bivalve larvae were sampled bi-hourly by Virginia biologists at three stations across the river at the Wreck Shoal (J 17) level (Andrews 1982c). Wreck Shoal is the largest and most productive oyster bed in the James River. The last period of sampling, from 30 August to 3 September, coincided with peak setting of oysters in that year with 40 spat per shellface per week on four replicate shell strings that were suspended off the bottom at Wreck Shoal (Andrews 195 1 ). Larvae were scarce at all stations and all sampling depths (3 depths in channel, 2 over beds). Primarily, straight-hinge larvae of less than 3 days of age were found, and many samples had no oyster larvae. Advanced larvae were encountered only rarely even when volume of plankton samples was increased from 100 to 500 C (Andrews 1982c). Preliminary data on larval densities were presented by Pritchard (1953) who calculated that only one mature larva per 100 C was needed to produce the observed spatfall. No conclusions were reached about distribution systems for larvae and for their retention in the seed area. The studies of Manning and Whaley (1954) in St. Marys River, Maryland, a trap-type estuary, were far more conclu- sive because advanced larvae were abundant and they moved upriver with wind-induced currents. Larvae in all stages were found and often 100 or more late-umbo larvae in 100-C samples. Densities of advanced stage larvae were much higher in deeper waters in the channel with peak counts of 900 late-umbo larvae per 100-2 sample. Manning and Whaley concluded that wind-induced convection currents moved surface waters landward in the lower-river sector with downriver flow in bottom layers. The typical characteristics of trap-type seed areas with tortuous geog- raphy and most intensive spatfalls near the head of the estuary are illustrated in Figure 1 of Manning and Whaley (1954). Carter (1967) conducted a physical study of hydrography of Manokin River on the Eastern Shore of Maryland using point release of dye to simulate physical dispersal of larvae. His conclusions were similar to those of Manning and Whaley (1954) that wind-induced convection currents carried larvae upstream. Freshwater discharge was almost negligible as in St. Marys River. Although the Manokin River is not a seed area, it could be according to Carter if enough brood oysters were planted in the lower river. Seliger and Boggs (1983 ) examined the physical hydrography of the Choptank River and its tributaries; they confirmed the physical regimes of trap-type estuaries but provided little information on larval biology from limited sampling, except that larvae were most abundant at the heads of saline river systems (creeks) where setting is known to be highest (Meritt 1977). More detailed studies of circulation in tributary creeks of the Choptank River were made by Boicourt (1982). Mechanisms of transport and setting of planktonic larvae in other estuaries are discussed by Ketchum (1954) in general, by Korringa (1952) for oysters in the Oosterschelde (Holland), and by Carriker (1951), Nelson (1957) and Haskin (1964) for oysters in New Jersey coastal bays and Delaware Bay. There is considerable literature on upstream movements of fish and crustaceans (e.g., Sulkin 1981), but larvae and juveniles of these groups make more positive responses to favorable strata and currents than do bivalve larvae. The most important bivalve larval studies of open 32 ANDREWS systems such as James River are those of Kunkle(1958) and Hidu and Haskin (1971) along the Cape May shore in Delaware Bay. In 1964—1965, mature and eyed-larvae were abundant in 200-C samples collected by the latter authors with 160-/im mesh plankton nets, and setting was intense. This area consistently had intense spatfalls (Nelson 1959), often far higher than any place in Chesapeake Bay. Delaware Bay is similar to James River in physical characteristics, but it has lower freshwater discharge than does Chesapeake Bay (Boicourt 1982). It has a tidal range of nearly 2 m, which is twice that of Chesapeake Bay (x = 0.72 m). Tidal- and wind-induced mixing in this wide, shallow bay, as in the James River, prevent much vertical density stratification in summer. By Pritchard's (1955) criteria for circulation regimes, both estuaries are type C in summer with lateral mixing; because of decreased river discharge and wide, shallow basins, salt balance is maintained by circular flow (Pritchard 1956). This report describes the patterns of larval transport in the James River and compares transport of larvae in the two types of estuaries. During 22 years (1946 to 1967) of intensive monitoring of spatfall in James River, the final distributions of larvae were determined (Andrews 1951, 1955, 1982a), but how they became distributed throughout the seed area is still obscure. The importance of large broodstock populations was shown after 1960, when setting rates declined to less than one-tenth the 1950's level; this followed cessation of private oyster planting in the lower river (Haven et al 1978, Andrews 1982a). High mortalities caused by MSX prohibited use of James River seed oysters in high-salinity waters of the lower river (Andrews 1983). Scarcity of oyster larvae during the 1960's, particularly of advanced stages, made studies of larval ecology difficult. Descriptions of the two types of seed areas are based primarily on patterns of spatfall that indicated wide differences in retention of larvae. Larval studies have not been made in trap-type estuaries in Virginia. Dye studies conducted in a physical model of James River at Vicksburg, Mississippi, suggested the probable extent of larval dispersion if transport were passive (Hargis 1966). Only field data collected in James River when sampling was most intensive in 1964 and 1965 are reported here. Data for earlier larval studies in James River are reported by Andrews (1982c). Some physical data collected during the 8 days of plankton samplingin the 1965 study were reported by Wood and Hargis (1971 ). MATERIALS AND METHODS Scarcity of larvae at Wreck Shoal in 1950 and recognition of higher spatfalls in the lower river resulted in selection of the Brown Shoal area for sampling in 1964 and 1965. Based on intensity of spatfalls over 20 years and preliminary plankton samples each year, a period near 1 September was chosen as the optimum time for sampling. This would not be true of any other estuary in Chesapeake Bay because the James River always has late setting. More emphasis was placed on sampling in the channel than over inshore oyster beds because deep-water currents are necessary for physical transport upriver. The channel is considered to be the primary transport route for upstream movement of larvae. Sampling was conducted hourly during night and day at four depths (0, 3, 6, 9 m) in the channel and at two depths over 3-m-deep beds for 2 days in 1964 and 8 days in 1965. After finding early-umbo larvae in the channel at Brown Shoal on 31 August 1964, stations were established at J33 in the channel and at Wreck Shoal (J33E) bed where sampling occurred for one tidal cycle on 3 September 1964. Three vessels were spaced 2 km apart and anchored in the channel in 1965, and two were anchored inshore over oyster beds opposite the central channel station above the James River Bridge. All plankton samples were taken synoptically on the hour with submerged pumps for each depth. Volume of water was measured by timing of calibrated pumps. Samples of about 300 C were pumped into plankton nets with 50-nm mesh submerged in watertight boxes. Surface and bottom samples were taken 1 m from interfaces with air and substrate to avoid boundary effects on currents and larvae. Plankton samples were preserved with 1% formalin buffered with an excess of NaHC0 3 or NaBr0 3 crystals. Counts of all species of bivalve larvae were made on Sedgwick-Rafter cells. In 1964, three or more 2-cm 3 aliquots were pipetted from magnetically stirred samples condensed to about 60 cm 3 . In 1965, entire samples were counted after excess fluid was decanted; sediments were swirled in 10-cm watch crystals to remove lighter periferal plankton and fecal pellets with pipettes. Several slides were counted for each swirl depending on the amount of sand and sediment; three or more swirls were made for each sample until larval counts declined rapidly. Early- stage larvae are lighter than advanced larvae, therefore they are more difficult to separate from other plankton by this swirling method. Total sample counts were necessary because of low density of larvae. All species were counted separately by stages of development; these were designated as straight-hinge, early-umbo, late-umbo, and mature or setting-size larvae (Chanley and Andrews 1971). Species and stages with low abundance were not summarized except as total bivalve larvae. Oysters comprised about one half of the bivalve larvae in most samples. RESULTS Brown Shoals was sampled hourly through one tidal cycle on 31 August 1964. A density of 10 to 40/8 of early-stage oyster larvae with some advanced larvae was encountered. A severe thunderstorm interrupted this field study at midnight, but a new operation during one daytime tidal cycle was carried out at J 19 and J33 on 3 September 1964. Counts of total bivalve larvae in the channel at J19 are shown in Table 1. Bivalve larvae were two to several Transport of Bivalve Larvae 33 times more abundant at 3- and 6-m depths than at and 9 m near surface and bottom boundaries. Larvae at 3 m depth had reached abundances of 30/2 at maximum flood tide and stayed high through high-slack water to maximum ebb. It is clear, however, that larvae were patchy in local distri- bution at various sampling times. A new group of early- stage larvae, 2 to 3 days old, had entered the Brown Shoal area on 3 September, and advanced larvae were less abun- dant than they had been on 3 1 August. TABLE 1. Total of bivalve larvae per 10 liters by depths in channel at Brown Shoal (J 19), James River, 3 September 1964* Bivalve Larvae by Depth (m) Time Tide 3 6 9 1000-1100 early flood 15 61 87 — 1100-1200 3 118 228 158 1200-1300 29 387 676 278 1300-1400 maximum flood 17 298 118 54 1400-1500 18 529 163 86 1500-1600 15 483 170 36 1600-1700 high slack 77 424 397 36 1700-1800 111 341 263 124 1800-1900 maximum ebb 189 640 222 168 Mean 47 328 233 105 *70% oyster larvae Samples at station J33 in the Wreck Shoal area on 3 September 1964 showed that advanced oyster larvae had moved upriver (Table 2). This table is arranged to show increasing densities of advanced-stage larvae with greater depths. Advanced larvae were much less abundant inshore over Wreck Shoal at station J33E in 3 m of water than in the channel. Again, patchiness of larvae was evident although some late-umbo larvae were found at all depths sampled. These counts were made by P. Chanley and the first 50 larvae were measured for size. This was the only one of 17 days sampled during full-tidal cycles over four years (1950, 1963, 1964, 1965) when significant numbers of advanced oyster larvae were found in James River. A light spatfall from these larvae occurred throughout the seed area in two subsequent weeks (Andrews 1982a). Hourly sampling around the clock from 5 and 3 stationary vessels, respectively, for 8 days (30 August to 3 September and 9 to 1 1 September) in 1965 showed bivalve larvae in regular cycles of abundance with tidal stages. High abun- dances occurred from maximal flood velocities through high-slack water to maximal ebb velocities, and low densities occurred during the other half of each tidal cycle. Combined totals for all bivalve larvae for four depths in the channel are shown for two stations (Figure 2). Most larvae of all species, including oyster larvae, were at straight-hinge stage (Andrews 1982c). Data for total bivalve larvae by four depths at one channel station exhibited similar patterns of cyclic abundance (Figure 3). Early-stage larvae were TABLE 2. Population densities of advanced oyster larvae (number per liter) by depths in channel at Wreck Shoal (J33), 3 September 1964. Oyster Larvae by Depths (m) and by Sizes (pm) 1 3.5-4.0 Time < 125 125-200 >200 < 125 125-200 >200 <125 7.0-8.0 125-200 >200 1125 1208 11 1227 1300 349 1325 1345 1359 429 1420 1442 1500 285 1522 1544 1600 177 1624 1646 1701 406 1725 1743 1800 202 Mean 266 32 40 11 48 41 34 29 16 14 1877 818 698 550 166 318 82 63 160 128 49 63 24 738 80 14 50 412 201 49 427 190 17 252 218 86 59 197 138 42 137 84 74 215 66 103 Stages of larvae by size are: straight-hinge = < 125 /im; early-umbo = 125 to 200 /Jm; late-umbo or eyed = > 200 )Jm. 34 ANDREWS T TIDAL VELOCITY 1 IOOO 1400 1800 2200 0200 0600 1000 1400 1800 2200 0200 0600 1000 September 9 September 10 September 11 Figure 2. Hourly densities of total bivalve larvae at four combined depths in channel, 9 to 11 September 1965. Two sampling stations designated by anchored vessels R/V LANGLEY and R/V PATHFINDER in channel 2 km apart. Total counts from 300-? samples at four depths adjusted to number per 100 5. Similar cycles of abundance occurred each tidal cycle at five stations over a period of 8 days between 30 August and 1 1 September 1965. Early-stage larvae predominated throughout the period. OT uj o o 300 a- 200 < > < UJ > < > O rr UJ 2 3 100 I METER DEPTH 4 METER DEPTH 7 METER DEPTH 10 METER DEPTH ~i r^ — I 1 1 1 1 1 1 1 1 r 1000 1400 1800 2200 0200 0600 1000 9 SEP 1 1 1 1 1 I 1 1 1 1 400 1800 2200 0200 0600 1000 10 SEP II SEP Figure 3. Cyclic abundance of bivalve larvae with tidal stage by depths in channel. Samples taken simultaneously with four submerged pumps at four depths at station J 19. TRANSPORT OK BIVALVE LARVAE 35 distributed throughout vertical columns of water with highest densities usually at 4 and 7 m. Data on bivalve larvae by species also showed highest densities from mid-flood to mid-ebb tidal velocities (Figure 4). Patchiness was evident, but peaks of abundance for oysters and other bivalves tended to occur near high- slack-water stage. Highest densities at high tides were 5 to 10 times as great as lowest densities at low tides. Oyster larvae were the most abundant of bivalve species, but peak densities tended to occur concurrently for all species. The cyclic abundance of larvae in shallow waters (< 3 m) over oyster beds is illustrated in Figure 5. High and low densities appeared at the same tidal stages as in the channel but tended to differ more widely in densities. DISCUSSION Oyster spawn is released at least weekly during summer from late June through September in the James River, but spatfall is most successful in late August and early September (Andrews 1955). Although spatfall occurred every week from 1 July to 1 October in the 1950's, 25 years of setting records indicate that conditions for survival and transport of larvae are most favorable in late summer (Andrews 1982a). This is a period of low-freshwater discharge and high salinities; therefore, stratification is minimal and net upriver movement of saline water in the channel at depths below 3 m is small and slow (Pritchard 1953, 1955). Nevertheless, in contrast to trap-type estuaries, the James River always has freshwater discharge which induces some stratification and mixing upriver in the seed area. Hampton Roads is nearly homogeneous for density of water in late summer, yet some saline water must move upstream in the channel to maintain salt balance in the seed area. Salinities increase gradually in the seed area as summer progresses. Dye releases near the mouth of the James River in the Vicksburg model showed that a 28.3-m 3 /s ( 1,000- ft 3 /s) dis- charge rate, which approximated salinity regimes observed in late summer of 1964 and 1965, resulted in higher concen- trations of dye at Burrells Bay after seven prototype days than a 90-m 3 /s (3,200-ft 3 /s) discharge (Hargis 1966). This suggests less importance of salt-balance transport upriver and greater effects of high-flushing rates that remove larvae from the river. If tidal dispersion is the primary factor or transport system regulating distribution of bivalve larvae, late-summer hydrographic regimes would be most favorable for retention of larvae in the river. Oyster larvae originate over shallow inshore flats and oyster beds in the James River. Early-stage larvae occur in the full vertical column of water over flats and in the channel; therefore, most larvae released in the seed area are probably carried downriver in shallow surface waters during their first days of planktonic life. Before MSX stopped the planting of seed oysters in Hampton Roads, a large oyster population near the river mouth supplied large quantities of spawn. In post-MSX years after 1960, most larvae originated in the seed area. The topography of the river below the James River Bridge delivers larvae off the extensive eastern shore seed beds into the channel of Hampton Roads where a deep-water column of 10 m or more is thoroughly mixed and available to allow vertical redistribution of larvae for (E UJ O O en UJ Q. < > < 200 w 100- > _j < > m z Crassostreo Mull ma Anomia Other Bivalves 1000 "i — i — i — i — f -\ — i — i 1 — i r — i — i — i — i 1400 1800 2200 0200 0600 1000 1400 1800 2200 0200 0600 1000 1 9 SEP 10 SEP I I SEP Figure 4. Cyclic abundance of bivalve larvae by species. Highest densities occurred between maximal flood and maximal ebb stages of tides. 36 ANDREWS k 300- 8 CC UJ a. UJ < > a. < UJ > < > CC UJ CD 2 3 200 100 1000 1800 9 SEP 2200 0200 0600 n r 1000 1400 10 SEP i 1 1 1 r 1800 2200 0200 0600 1000 I I SEP. Figure 5. Density of bivalve larvae at surface and bottom over Brown Shoal oyster bed. Abundance of larvae was lower over shoals but cyclic patterns with tidal stages were similar for species and depths. river ascent in the channel. Early-stage larvae appear to be recycled several times up the channel, out over the flats, and back down to Hampton Roads during their first days of pelagic life. Most larvae disappeared within less than 5 days; they were replaced by newly spawned larvae. Few larvae achieved advanced umbo stages during which they would have selected deeper layers of water thereby enabling them to ascend into the seed area. My data and concept of transport and dispersal of bivalve larvae apply primarily to early-stage larvae (Figure 6). The seed area provides the larvae and Hampton Roads is a deep- mixing zone which facilitates advection of larvae upriver in the channel. These are primary but not exclusive roles for the two river sectors shown in the diagram. It is apparent from plankton sampling and spatfall patterns that new groups of young larvae are being introduced every week, or more frequently. Larvae in waters discharged into Chesa- peake Bay are lost at an estimated flushing rate of 15% per tidal cycle (A. Kuo, Virginia Institute of Marine Science, Gloucester Point, VA; pers. coram.); this sums to 95% loss of larvae in 10 days or 20 tidal cycles, the shortest probable duration of larval life in nature. Data on larval abundance near the river mouth are not available, but it is presumed from the spatfall gradients that eventually setting-size larvae are at least as abundant as at Brown Shoals. Hourly sampling during 5- and 3-day physical and biological studies in a 13- day period in September 1965 showed the scarcity of advanced oyster larvae in the James River. Larvae were not surviving in the James River long enough to grow to umbo larvae (3 to 5 days) and, therefore, could not utilize the net upriver channel flow in waters greater than 3 m depth. There are no data on losses of bivalve larvae by predation in nature, although my assumption is that the same predators present in the 1950's are still equally active in the 1960's and 1970's. Many pelagic larvae, including fish fry, coelen- terates, ctenophores, as well as most adult bottom-living organisms with mucus and ciliary feeding mechanisms, capture bivalve larvae (Mileikovsky 1974, Andrews 1979). Most efficient as collectors are adult oysters on beds where mature larvae are most attracted by gregarious setting. Transport of Bivalve Larvae 37 TRANSPORT OF BIVALVE LARVAE IN THE JAMES RIVER FRESHWATER DISCHARGE FLATS FLATS MID-FLOOD I VESSEL STATIONS 1966 OR MID-EBB -V" MID- EBB / TO _l MID- FLOOD, POSITION OF MAJOR LARVAL BROODS JAMES RIVER BRIDGE TIDAL CURRENTS SALINE WATER INPUT SEED OYSTER BEDS (SPAWNING) SEED OYSTER AREA HAMPTON ROADS CHESAPEAKE BAY LARVAE LOST Figure 6. Diagram of a hypothesis of larva] transport in James River. Oyster beds and larval broods are located only symbolically. Channel transport is emphasized, but transport of larvae occurs throughout cross sections of the river. Width of arrows suggests intensity of transport system and density of larvae. A tidal excursion is about 1 1 km in channel. The bridge and Deep Water Shoal are 19 and 46 km, respectively, above the river mouth. 38 ANDREWS Figure 6 emphasizes the importance of channel waters for transport of larvae upriver. Tidal excursions average about 1 1 km in the channel; this means that larvae located at the bridge could be carried to Wreck Shoal in one flood tide, or downriver to the middle of Hampton Roads in one ebb tide. In three years (1963—1965) of late-summer sampling in the Brown Shoal area, oyster larvae were rarely absent; this indicates that one or more broods were dis- tributed at least 1 1 km above and below the bridge during a tidal cycle. The larval groups illustrated by ovals on Figure 6 are intended to suggest the location where larvae were most abundant at given tidal stages. The arrows suggest densities of larvae in the channel and at sites of dispersion over oyster beds. Most larvae carried upriver during flood tide appear to be carried back down the channel during ebb tide; a few must be trapped over shallow oyster beds or in meandering creeks by eddies and boundary effects (slowing of currents) of bottom and marginal features such as marshes. Apparently, advanced larvae at Wreck Shoal on 3 September 1964, which were abundant mostly in the channel, reached oyster beds in the seed area by slow advance in net upstream flow in deep channel currents. Wood and Hargis (1971) reported on a 24-hour period of sampling(l September 1965) during the same field study reported in this paper. Larvae showed the same patterns of abundance given in this report and also in the other days not reported by either of us. In their samples, oyster larvae were usually fewer than 100 per 300-8 sample, although early-umbo-stage larvae were relatively abundant. They reported physical data on circulation, salinity, temperature, and net flow based on seven complete tidal cycles of observation. These physical conditions apply equally well to plankton data presented in this paper for 9 to 1 1 September. The type C counter-clockwise circulatory pattern described by Pritchard (1955) prevails in the James River in late summer when freshwater discharge is low. Monthly river discharge averaged less than 28.3 m 3 /s (< 1,000 ft 3 /s) for the months of August and September 1964 and 1965. Net upriver flows are greater on the northeastern side of the channel, and discharge is greatest downriver on the south- western shore. Wood and Hargis (1971) contended that oyster larvae on the bottom responded to salinity stimulation during flood tides, but they provided no data that showed selective swimming or distribution of larvae by depths. Vertical salinity gradients in Hampton Roads where larvae originate with each flood tide were less than 1 ppt from surface to bottom. If larvae rested on the bottom during ebb and low tides, they could respond to increasing salinities during flood tides (Haskin 1964), but evidence that larvae rest on the bottom is inconclusive. Carriker (195 1) worked in high- salinity coastal bays where shallow water and strong pycno- clines prevented larvae from freely selecting strata for upriver transport. Both Carriker (1950) and Wood and Hargis (1971) support Nelson's hypothesis (Nelson and Perkins 1931) that oyster larvae ascend estuaries by resting on the bottom during ebb tides and by swimming during flood tides. Data of Wood and Hargis (1971) comparing coal particles with larvae seem irrelevant to me because it has been clearly established that bivalve larvae can move vertically by their own powers of swimming. Larvae were found during all tidal stages whereas coal particles were observed only during strong currents. Larvae were most often abundant at high-slack water and there was no evidence that larvae descended during periods of slack currents. Larvae were least abundant in samples taken near the bottom during strong tidal currents when large numbers of fecal pellets (primarily from oysters) and sand grains were found in samples. This leads me to believe that larvae are actually trapped on the bottom during strong currents by the roiling effects of bottom drag and constant pelting- even though all are being carried by slow bottom currents. Dirty samples taken too close to the bottom always con- tained few larvae. If distribution of larvae were completely passive, they would spend both high- and low-slack periods on the bottom just as coal particles and fecal pellets do, but feeding time would be reduced. Losses of larvae to smothering and predation on the bottom may be as great as those from dispersal and predation during planktonic life. Counts of larvae collected through 8 days (16 tidal cycles) show that the pattern of highest abundance from mid-flood to mid-ebb tides was regular and highly significant, but explanations of cyclic abundance vary in the literature. The important observations of the present study are: (1) total quantities of larvae at all stations before and after slack-high water were approximately equal; (2) persistence of early- stage larvae indicated that new broods were recruited fre- quently into the river; (3) older larvae were found most frequently in deeper waters and, therefore, in the channel; and (4) there was a noticeable decrease in density of larvae from the lower channel station to the upper one, only 4 km apart, at all tidal stages. Larval broods are three dimensional. The term swarm is inappropriate for there is no evidence that larvae remain together or aggregate horizontally. Advanced larvae choose deeper strata in the water column effectively. Passive physical transport probably far outweighs in significance any results from selective motion by larvae, particularly during the first 5 days of planktonic life. Larvae do respond to pheromones when setting is about to occur. It is not known whether they can respond to food or other stimuli. My scenario for the decline of setting in James River since 1960 assumes that loss of brood stocks to MSX disease in the lower river resulted in too few larvae to replenish oyster stocks in the seed area. It appears that broods of larvae are carried up and down the river several times with progressive thinning and dispersal of each brood. In the area sampled in 1965, near the James River Bridge, larvae probably moved up the channel and along the northeastern Transport of Bivalve Larvae 39 shallow flats, then back down the channel and over the southwestern flats to Hampton Roads (Wood and Hargis 1971). Most larvae were lost by dispersion and predation in 3 to 5 days before they were stimulated to swim in deeper strata. New broods replaced old ones repeatedly. Spring tides and storms that increase tidal amplitude over the mean 0.72 m may cause some larvae to be trapped inshore and result in spatfalls. Because the same circulatory patterns still exist in James River, regular spatfalls every week for 3 months in the 1950's may be attributed to much larger populations of brood oysters and greater abundance of larvae in that period. In the mid-1960's, Langley Wood (VIMS, Gloucester Point, Virginia, unpublished studies) constructed a vertical plexiglass cylinder about 2.5 m long and 0.3 m in diameter to study the swimming habits of oyster larvae. A strong light was mounted over the upper end and sampling ports were inserted at various levels. Larvae alternated between swimming upward in gyrals and falling slowly while resting for periods of a minute or so. When larvae bumped into one another they quickly retracted their velums. Pelagic larvae have two purposes: to distribute the species and to replenish adult stages (Galtsoff 1964). The velum provides a mechan- ism for swimming and feeding activities to meet these goals. Larvae must swim to eat. Resting for half of each tidal cycle on the bottom may require a doubling of the duration of larval life. In hatchery cultures, strong light causes swimming larvae to seek shade and curious distri- butional patterns visible to the naked eye are formed. In many estuaries, larvae are confronted with unfavorable natural conditions such as low temperatures or toxic com- pounds below surface waters (Quayle 1969). In these waters larvae are forced to swim continuously throughout their planktonic life regardless of dispersal effects. I conclude that bivalve larvae swim continuously during larval life and that their dispersal and ultimate fates are strongly dependent on current regimes and flushing rates of estuaries. The bottom is a hazardous place for larvae to rest: a host of sedentary filter feeders become predators or imprison larvae in mucous-wrapped fecal pellets (Cerruti 1941,Mileikovsky 1974). Siltation is a serious threat on the bottom in channels where currents are strong. Prolonged duration of larval life and exposure to predators are major threats to survival in the James River with its relatively high flushing rates. The trap-type estuaries with their rela- tively intensive setting rates provide physical transport regimes that allow greater retention of larvae. If oyster larvae can persist in an estuary long enough to reach umbo size, a preference for deeper waters prevails and, in the case of the James River, they should be able to ascend the deep channel currents more effectively than in the poorly stratified trap-type estuaries. Observations from setting records indicate that the opposite occurs and that they are less successful in remaining in strong flushing-type estuaries. This implies that passive physical transport predominates over larval reactions to physical and chemical stimuli to select favorable current strata. Presumably, more intensive oyster setting in Delaware Bay can be attributed to the large size of the estuary with lower freshwater-discharge rates and to its wide shallow flats; only the upper seed area sector exhibits type-C circulation in summer, and flushing rates in the widened lower sector (Hidu and Haskin 1971) are probably much lower than in James River. ACKNOWLEDGMENTS I acknowledge the dedicated support of Martha Eble. Sybil Lawler, Paul Chanley. Donna DeMoranville, and Ed Powell who counted larvae in many plankton samples during 1965 and 1966. REFERENCES CITED Andrews, J. D. 1949. The 1947 oyster strike in the James River. Proc. Natl. Shellfish. Assoc. (1948):61-66. . 1951. Seasonal patterns of oyster setting in the James River and Chesapeake Bay. Ecology 32:752-758. . 1955. Setting of oyster in Virginia. Proc. Natl. Shellfish. Assoc. 45:38-46. . 1979. Pelecypoda: Ostreidae. Giese, A. C. and J. S. Pearse, eds. Reproduction of Marine Invertebrates. Vol. 5. Molluscs: Pelecypoda and Lesser Classes. New York, NY: Academic Press. .198 2a. The James River public seed oyster area in Virginia. Va. Inst. Mar. Sci. Spec. Sci. Rep. Appl. Mar. Sci. OceanEng. 26 1 : 60 p. . 1982b. Reproduction of oysters in Virginia. Available from author on request: Virginia Institute of Marine Science, Gloucester Point, VA. (unpublished manuscript) . 1982c. Transport of the bivalve larvae in the James River, Virginia. Va. Inst. Mar. Sci. Spec. Sci. Rep. 1 1 1 : 75 p. . 1983. Minchinia nelsoni (MSX) infections of oysters in the James River seed area and their expulsion in spring. Estuarine Coastal Shelf Sci. 16:255-269. & J. L. Wood. 1967. Oyster mortality studies in Virginia. VI. History and distribution of Minchinia nelsoni, a pathogen of oysters in Virginia. Chesapeake Sci. 8:1-13. Boicourt, W. C. 1982. Estuarine larval retention mechnisms on two scales. Kennedy, V. S., ed. Estuarine Comparisons. New York, NY: Academic Press, p. 445-457. Carriker, M. R. 1951. Ecological observations on the distribution of oyster larvae in New Jersey estuaries. Ecol. Monogr. 21 : 19—38. Carter, H. H. 1967. A method for predicting broodstock require- ments for oyster (C. virginica) producing areas with application to the Manokin River. Chesapeake Bay Inst. Johns Hopkins Univ. Spec. Rep. 13: 37 p. Cerutti, A. 1941. Osservazioni ed esperimenti sulle cause di distru- zione delle larve d'ostrica nel Mar Piccole e nel Mar grande di Taranto. Arch. Oceanogr. Limno. 1:165-201. Chanley, P. & J. D. Andrews. 1971. Aids for identification of bivalve larvae of Virginia. Malacologia 11:45-119. Galtsoff, P. S. 1964. The American oyster Crassostrea virginica Gmelin. U.S. Fish Wildl. Serv. Fish. Bull. 64: 480 p. Hargis, W. J., Jr. 1966. Operation James River, an evaluation of physical and biological effects of the proposed James River navigation project. Va. Inst. Mar. Sci. Spec. Sci. Rep. Appl. Mar. Sci. Ocean Eng. 7: 73 p. 40 ANDREWS Haskin, H. H. 1964. The distribution of oyster larvae in Delaware Bay. Nanagansett, RI: Proc. Symp. Exp. Mar. Ecol., Occas. Publ. 2:76-80. Haven, D. S.. W. J. Hargis, Jr. & P. C. Kendall. 1978. The oyster industry of Virginia: its status, problems and promise. Va. Inst. Mar. Sci. Spec. Pap. Mar. Sci. 4: 1024 p. Hidu, H. & H. H. Haskin. 1971. Setting of the American oysters related to environmental factors and larval behavior. Proc. Natl. Shellfish. Assoc. 61:35-50. Ketchum, B. H. 1954. Relation between circulation and planktonic populations in estuaries. Ecology 35:191-200. Korringa. P. 1952. Recent advances in oyster biology. Q. Rev. Biol. 27:266-308,339-365. Kunkle, D. C. 1958. The vertical distribution of oyster larvae in Delaware Bay.Proc. Natl. Shellfish. Assoc. 48:90-91. Mackin, J. G. 1946. A study of oyster strike on the Seaside of Virginia. Va. Fish. Lab. Contr. No. 25: 18 p. Manning, J. H. & H. H. Whaley. 1954. Distribution of oyster larvae and spat in relation to some environmental factors in a tidal estuary. Proc. Natl. Shellfish. Assoc. 45:56-65. Meritt, D. W. 1977. Oyster spat set on natural cultch in the Maryland portion of the Chesapeake Bay (1939-1975). Cent. Estuar. Environm. Sci. Univ. MD 7: 30 p. Mileikovsky, S. A. 1974. On predation of pelagic larvae and early juveniles of marine bottom invertebrates by adult benthic invertebrates and their passing alive through their predators. Mar. Biol. (Berl.) 26:303-312. Nelson, T. C. 1957. On the reactions of oyster larvae in relation to setting on the cape shore of Delaware Bay, N.J. Available from Dept. Biology, Rutgers Univ., New Brunswick, NJ (unpublished manuscript) . 1959. Oyster seed production on Cape May's tidal flats. Cape May Geographic Soc. Ann. Bull. 13:12-16. & E. B. Perkins. 1931. Report of the Biology Department. NJAgric. Exp. Sta. Bull. 522:1-47. Pritchard, D. W. 1952. Salinity distribution and circulation in the Chesapeake Bay estuarine system./ Mar. Res. 11:106-123. . 1953. Distribution of oyster larvae in relation to hydrographic conditions. Proc. Gulf Caribb. Fish. Inst. 5:123-132. . 1955. Estuarine circulation patterns. Proc. Am. Soc. Civil Eng. 81:1-11. . 1956. A study of the salt balance in a coastal plain estuary. /. Mar. Res. 15:33-42. Quayle, D. B. 1969. Pacific oyster culture in British Columbia. Bull. Fish. Res. Board Can. 169: 34 p. Seliger, H. H. & J. A. Boggs. 1983. Physical-biological mechanisms for the transport of oyster larvae in the Chesapeake Bay. Mar. Biol. (Berl.) 71:57-72. Sulkin, S. D. (Convenor) 1981. Larval retention in estuaries. Abstracts for the Sixth Biennial International Estuarine Research Conference. Estuaries 4:238-240. Wood, L. & W. J. Hargis, Jr. 1971. Transport of bivalve larvae in a tidal estuary. Proc. Eur. Mar. Biol. Symp. 4:29-44. Journal of Shellfish Research, Vol. 3, No. 1, 41-44, 1983. BIOLOGICAL CONTROL OF FOULING ALGAE IN OYSTER AQUACULTURE CATHERINE ENRIGHT, DONNA KRAILO, LARRY STAPLES, MARIA SMITH, CARL VAUGHAN, DEBRA WARD, PAMELA GAUL, AND ELISABETH BORGESE Seafarm Venture, Ketch Harbour Nova Scotia, Canada BOJ 1X0 ABSTRACT The periwinkle (Littorina littorea Linne) provided excellent biological control of Ectocarpus sp., Entero- morpha sp., Ulva sp., and pennate diatoms, all of which foul oyster-rearing boxes. The addition of periwinkles (200/m*) to 1-mm mesh-covered rearing boxes containing juveniles of the European flat oyster Ostrea edulis Linnaes promoted a significantly higher oyster growth rate (t-test; p = 0.05). Examination of the means obtained from a 5-week study showed a 30% increase in oyster growth rate when periwinkles were added, in comparison to the unmanipulated control. There was no significant difference (t-test; p = 0.05) in oyster growth rates when the culture boxes were either brushed once a week or periwinkles were added. A density range of to 1,600 periwinkles/m of oyster-rearing surface was examined in culture boxes covered with 6-mm mesh. Similar oyster growth rates were obtained with densities between 300 and 1,600 peri- winkles/m of oyster-rearing surface. Isopods (Idotea balthica Pallas) at a density of 125/m of oyster-rearing surface were not effective as a biological control agent. KEY WORDS: biological control, oysters, periwinkles, algal fouling, Ostrea edulis. Littorina littorea, oyster culture INTRODUCTION Oyster-rearing boxes, trays, and lantern nets quickly foul with algae, mussels, bryozoans, sponges, and other marine organisms which restrict the flow of water and, consequently, the availability of phytoplankton to the oysters. Michael and Chew (1976) examined the effect of progressive fouling in off -bottom oyster culture in the state of Washington and correlated it with a decline in the growth rate of the Pacific oyster Crassostrea gigas Thunberg. The traditional methods of coping with fouling in oyster culture include routine manual scraping and brushing, air-drying, controlled burning, pesticides, and high-pressure spraying to remove fouling organisms (Arakawa 1980). Clime and Hamill (1979) found that high -pressure spraying with a portable 378.5 to 567.7-C/min (100 to 500-gal/min) capacity pump reduced marine fouling on oyster-culture gear in Maine. The cleaning schedules included bi-weekly treatments for small mesh enclosures and monthly cleaning for lantern nets and larger mesh enclosures during the height of the growing season. MacLeod (1974) investigated the use of a hot-water dip treatment for control of fouling organisms on oyster-culture gear. Huguenin and Huguenin (1982) examined the use of expanded metal mesh of a copper-nickel alloy in shellfish trays. Although these proce- dures are effective, they are both expensive and time consuming. Dr. E. Scura (Aquatic Farms, Hawaii, pers. comm.) estimated that 20% of the market price of inten- sively cultured oysters reflected the costs associated with reducing fouling organisms during the rearing stages. In Nova Scotia during 1983, the members of the Ostrea Edulis Cooperative Association Ltd. allocated more than half of the labor time associated with rearing oysters to cleaning of fouling from oysters and culture gear. Thus, fouling has traditionally been a costly problem in terms of equipment and labor costs as well as reduced oyster growth rates. An efficient, inexpensive means of ensuring maximum water flow about the oysters is greatly needed. Biological control is the utilization of natural or exotic species to control the density of undesirable organisms. Hidu et al. (1981) inadvertently enclosed a rock crab Cancer irroratus Say in a tray of over-wintering yearling European oysters and found that the typical thick mat of fouling organisms did not develop. By selecting crabs of a distinct size range, Hidu et al. (1981) demonstrated that the introduction of crabs to oyster culture may provide a means of biologically controlling the growth of fouling organisms. Movement by the crab was also believed to reduce silt accumulation on the oysters. While suitable for the culture of large oysters, crabs prey upon small oysters and can only be used with great care as a biological control agent with juvenile oysters. The fouling problem is more acute with juvenile oysters because they can not withstand the damage incurred by traditional cleaning methods. Also, the small- mesh screen needed to retain juvenile oysters fouls more quickly and accentuates the fouling problem. Because snails and isopods have demonstrated the ability to consume algae (Shaddock and Croft 1981, Steneck and Watling 1982), we investigated the usefulness of periwinkles and isopods as biological control agents in juvenile oyster culture. Bequaert (1943) noted that the herbivorous habits of L. littorea were sometimes used to keep oysters free of algal growth. We felt that such an application might be useful in oyster aquaculture. 41 42 ENRIGHT ET AL. MATERIALS AND METHODS Juveniles of the European oyster Ostrea edulis Linnaes were studied inSambro Harbour, Nova Scotia (44°28'5l"N, 63°34'2l"W). The water temperature range was 12 to 17°C and the salinity range was 29 to 3 1 ppt during the experi- mental period. The oysters were reared in boxes with wooden sides which were covered on the top and bottom with plastic screening. Two sets of three vertically suspended culture boxes were hung from a floating boom near each other. The top box in each set was approximately 20 cm beneath the water surface with subsequent boxes approxi- mately 25 cm apart. Oyster growth rate was assessed using change in volume or weight over the experimental period. An empty box with plastic screen was suspended between the experimental box sets. A small piece of mesh was clipped bi-weekly from this box for a microscopic examina- tion of the colonizing organisms throughout the experi- mental period. The fouling organisms were identified and the abundance of each was expressed as a percentage of the total fresh weight biomass of all fouling organisms. The first experiment was conducted from 7 July to 12 August 1981. The culture boxes were 83 X 60 X 6 cm and were covered with 1-mm plastic screening. Each of the six boxes was divided by wooden slats into four equal compartments, with each box receiving one of the following four treatments: the addition of 24 periwinkles (Littorina littorea) (200/m 2 ) approximately 2 cm in diameter; the addition of 13 isopods {Idotea balthicd) (125/m 2 ) approxi- mately 3 cm in length; weekly manual brushing of the screen mesh; and an unbrushed control. Juvenile oysters, approximately 5 mm in diameter, were stocked in the boxes at an initial "density" of 600 g/m 2 . The second experiment was conducted from 5 July to 3 October 1982. A similarly arranged culture unit was used with boxes measuring 30 X 30 X 6 cm and covered with 6-mm mesh plastic screen. The six boxes were divided into four equal compartments and suspended in two units, each with three boxes. The following series of treatments was replicated at each of the three-box positions (upper, middle and lower): weekly manual brushing of the mesh; (contol), 2, 5, 10, 15, 20 and 25 periwinkles in each compartment which corresponds to 0.01. 0.03, 0.05, 0.08, 0.10 and 01.3 periwinkJes/m 2 . The oysters used were approximately 2 cm in diameter and the oyster stocking "density" was 8,000 g/m 2 . RESULTS AND DISCUSSION Littorina littorea proved to be an excellent biological control agent for reducing algal fouling on the oysters and on the screens covering the oyster-rearing boxes. The addition of 200/m 2 periwinkles to 1-mm mesh-covered rearing boxes containing juvenile European oysters was shown to yield a significantly higher (t-test; p = 0.05) oyster growth rate (Table 1). Examination of the means obtained from a 5-week study showed an approximate 30% increase in oyster growth rate (Set I, 36%; Set II, 25%) when periwinkles were added compared with the unbrushed control (Table 1). The major fouling organisms were Ectocarpus sp. (90%), Enteromorpha sp. (3%), Ulva sp. (1%), and pennate diatoms (5%). Animal fouling accounted for less than 1% of the total fouling biomass. There was no apparent change in the species composition of the fouling organisms throughout the experimental periods. On the basis of visual inspections, the periwinkles kept the mesh cleaner than that obtained with a weekly manual scrubbing. There was no significant difference (t-test; p = 0.05) in oyster growth rates when the culture boxes were brushed once a week or periwinkles were added. Idotea balthica did not actively graze the fouling organisms which collected on the plastic screen, and the growth rate of the oysters TABLE 1. Increase in volume (m?) and the calculated growth rate (% volume increase day ) of Ostrea edulis cultured in boxes with unbrushed screens, with brushed screens, with periwinkles, and with isopods. The initial size of the oyster was approximately 5 mm in diameter and the experimental period was 5 weeks (7 July to 12 August 1981). Unbrushed Brushed With Periwinkles With Isopod s Box Position A Volume % day 1 A Volume % day ' A Volume % day ' A Volume % day" 1 Set 1 Upper 190 4.1 240 4.8 240 4.8 170 3.7 Middle 170 3.7 190 4.1 210 4.4 210 4.4 Lower 120 2.7 200 4.2 200 4.2 120 2.7 X 160 3.5 210 4.4 217 4.5 167 3.6 SD 36 0.7 26 0.4 21 0.3 45 0.8 Set II Upper 260 5.0 280 5.2 310 5.5 260 5.0 Middle 220 4.5 320 5.6 300 5.4 200 4.2 Lower 180 3.9 190 4.7 220 5.2 140 3.8 X 220 4.5 263 5.2 277 5.4 200 4.3 SD 40 0.6 67 0.5 49 0.2 60 0.6 Biological Control of Fouling algae 43 reared in such compartments did not differ significantly (t-test; p = 0.05) from that of the oysters in the unbrushed (control) compartments. Using a comparable isopod density. Shaddock and Doyle (1983) found that /. balthica vora- ciously grazed Ectocarpus sp., a brown seaweed which grows epiphytically on Chondrus crispus in tank cultures. Perhaps in the present experiment a higher isopod density would have negated the fouling rate in the oyster-rearing boxes. Oyster boxes suspended in the water column may not provide an adequate habitat for isopods; perhaps their feeding behavior is altered in that setting. From the data in Table 1, it is clear that higher oyster growth rates were obtained in box Set I compared to box Set II. The difference may have been the result of their relative position in the bay as box Set II was downstream from box Set I with respect to the food source. All other parameters were the same in each box set. An examination of a periwinkle density range from to 1 ,600/m 2 of mesh-rearing surface, when a 6-mm mesh size was used, indicated little change in oyster growth rates between 300 and 1 ,600 periwinkles/m 2 of screen (Figure 1 ). The optimal periwinkle density would be expected to vary as a function of the degree of fouling and with factors that influence the periwinkle grazing rate (e.g., temperature). There are many advantages to utilizing periwinkles for biological control of fouling organisms in juvenile oyster culture. Periwinkles are herbivors; therefore, they do not prey on oysters as do crabs and other organisms. Littorina littorea is extremely abundant in western Europe and in northeastern North America and locally exceed densities of 150 periwinkles/m 2 in the low intertidal zone. The periwinkle can completely withdraw its soft tissue into its shell, thus protecting itself against desiccation when the oyster boxes are removed from the water for data collec- tion or transportation. There was no evidence of erosion of the mesh fibers as a result of the periwinkles grazing along the plastic screens. The major advantage of using a biological control agent such as a periwinkle is the reduction in costs associated with cleaning algal fouling organisms. As water flow and phytoplankton availability are greatly 500- CD 5 co 10 o -♦— n i_ .c *— 1.0 5 o CJ5 Figure 1. Increase in weight (g fresh weight) and the corresponding calculated growth rate {% weight increase day" 1 ) of Ostrea edulis cultured with Littorina littorea at various densities and compared with a weekly, manual mesh-brushing treatment. The initial size of the oysters was approximately 2 cm in diameter and the mesh used on the rearing boxes was 6 mm. The experimental duration was 12 weeks (5 July to 3 October 1982). Standard deviations are shown (n = 3). 44 ENRIGHTETAL. enhanced for juvenile oysters cultured with periwinkles, the need to transfer oysters on to larger mesh sizes, as is presently the practice (Clime and Hamill 1979), is reduced. Such cost reductions will greatly improve the profitability of off-bottom oyster culture. ACKNOWLEDGM ENTS The financial assistance from the Nova Scotia Department of Development, Provincial Employment Pro- gram, is gratefully acknowledged. We thank P. Shacklock, S. Smith and J. Dale for their assistance on site. Sincere appreciation is expressed to Drs. J. Craigie, G. Newkirk, and H. Hidu for reviewing the manuscript. This study is dedicated to the memory of T. Moore, who assisted greatly in the initial stages of this project. REFERENCES CITED Arakawa, K. Y. 1980. Prevention and Removal of Fouling on Cultured Oysters: A Handbook for Growers. Translated from Japanese by R. Gillmore. Univ. Maine Sea Grant Tech. Rep. No. 56: 56 p. Bequaert, J. C. 1943. The genus Littorina in the western Atlantic. Johnsonia 7:1-28. Clime, R. & D. Hamill. 1979. Growing oysters and mussels in Maine. Golden, E., ed. Aquaculture Development Workshop; Bath, ME: Coastal Enterprises. Inc. 46 p. Hidu, H., C. Conary & S. R. Chapman. 1981. Suspended culture of oysters: biological fouling control. Aquaculture 22:189-192. Huguenin, J. E. & S. S. Huguenin. 1982. Biofouling resistant shell- fish trays. J. Shellfish Res. 2(l):41-46. Michael, P. C. & K. K. Chew. 1976. Growth of Pacific oysters. Crassostrea gigas, and related fouling problems under tray culture at Seabeck Bay, Washington. Proc. Natl. Shellfish. Assoc. 66:34-41. MacLeod, L. L. 1974. Controlling blue mussel (Mytilus edulis) fouling on oysters and oyster trays with hot water immersion. 8 p. Unpublished document. Available from: Nova Scotia Dep. fish.. Resour. Develop. Div. Fish. Train. Cen. Pictou, NS, Canada. Shacklock, P. F. & R. W. Doyle. 1983. Control of epiphytes in seaweed culture using grazers. Aquaculture 31:141-151. Shacklock, P. F. & G. C. Croft. 1981. Effect of grazers on Chondrus crispus in culture. Aquaculture 22:331-342. Steneck, R. S. & L. Watling. 1982. Feeding capabilities and limita- tion of herbivorous molluscs: a functional group approach. Mar. Biol. (Berl.j 68:299-319. Journal of Shellfish Research, Vol. 3. No. 1, 45-50, 1983. A STUDY OF GLUCOSE, LOWRY -POSITIVE SUBSTANCES, AND TRIACYLGLYCEROL LEVELS IN THE HEMOLYMPH OF CRASSOSTREA VIRGINICA (GMELIN) MARY L. SWIFT AND MOHAMMED AHMED Department of Biochemistry College of Medicine Howard University Washington, DC 20059 ABSTRACT Oysters, Crassosrrea virginica (Gmelin), were maintained in the laboratory under controlled conditions of temperature and salinity. Levels of several hemolymph constituents were analyzed. Average values of hemolymph glucose, Lowry-positive substances, and triacylglycerols were 8.83 ± 1.98 mg/100 mC (± SE), 11.0 ± 1.89 rag/mf (± SE), and 43.2 /Jg/100 mC, respectively. Hemolymph glucose values varied over a wide range. No deleterious effects of this variance (as judged by mortality rates) could be detected. Groups of animals with initial hemolymph glucose levels of 23.1 to 25.0 mg/100 m? survived as long as those with initial values of 5.3 to 8.4 mg/100 mC. Oysters held at constant water temperatures and salinities tended to maintain the concentration of their hemolymph glucose and Lowry-positive substances over a 27-day period of starvation; hence, some type of regulatory mechanism is involved in controlling the levels of these metabolites in oyster hemolymph. Extremes in environmental conditions appear to affect the concentrations of these metabolites in hemolymph. Groups of oysters maintained in sea water at a temperature of 4 C had significantly higher (p < 0.05) levels of hemolymph glucose and Lowry-positive substances than groups held at 20 C. Groups of oysters maintained at alow ambient salinity (12 ppt) had significantly lower (p ^0.05) levels of hemolymph glucose and Lowry- positive substances than groups kept in water of 18 ppt and 24 ppt salinity. KEY WORDS: oyster, Crassostrea virginica, hemolymph, glucose, regulation INTRODUCTION Traditionally, the physiological and nutritional condi- tions of oysters have been monitored by evaluating tissue glycogen content (Gabbott and Walker 1971. Willis et al. 1976). The deposition and utilization of not only glycogen but also lipid by the American oyster may be influenced by a number of factors. Seasonal variations in tissue glycogen and lipid content, which are keyed to the reproductive cycle, are well documented (Galtsoff 1964, Krishnamoorthy et al. 1979, Swift et al. 1980). The effects of starvation on these metabolic reserves in oysters have been examined (Riley 1976, Willis et al. 1976, Swift et al. 1980), as have environmental conditions which may also affect the rate of synthesis or utilization and, therefore, content of metabolic reserves. Several groups have investigated either the whole animal response or the response of selected excised tissues to changes in temperature and salinity. Ruddy et al. (1975) examined the growth rate of Crassostrea virginica (Gmelin) during exposure to a warm water temperature ( 14 to 19°C). Levels of each of the major classes of metabolites (carbo- hydrate, protein, and lipid) increased in these animals. At the same time gonadal development occurred four months earlier than usual. Similar increases in biochemical reserves have been observed in Crassostrea gigas (Thurnberg) and Ostrea edulis (Linne) (Mann 1979). Percy and Aldrich (1971), Percy et al. (1971), and Bass ( 1977) monitored the effect of changes in ambient water temperature and salinity on oxygen consumption of excised gills, mantle, and adductor muscle of C. virginica. These reports agree that, with increasing temperature or decreasing salinity, oxygen use increases. When subjected to extremes of temperature and salinity, these animals used more oxygen (Shumway and Koehn 1981). These data imply that the metabolic rate has increased and, thus, utilization of metabolic reserves has increased, resulting in a decrease in tissue content of glycogen and lipid. Despite the proven usefulness of data on tissue composi- tion, the processes required to obtain them are cumbersome and time consuming. In contrast, more complete information concerning the nutritional and physiological conditions of mammalian organisms may be obtained easily and rapidly by analysis of blood metabolites. Unfortunately little is known regarding the metabolite levels in the hemolymph of C. virginica. Hand and Stickle (1977) studied the effect of tidal-like fluctuations in salinity of ambient sea water on pericardial fluid composition of the oyster. Ion concentra- tions, except K\ were found to be isoionic to the various ambient salinity regimes: ninhydrin-positive substances ranged from 1 .5 to 6.0 mM. The lack of suitable data in the literature for establishing baseline values for hemolymph glucose, protein, and triacyl- glycerol levels in C. virginica prompted the following studies. Glucose," total Lowry-positive substances (LPS), and triacyl- glycerols were examined in hemolymph from groups of oysters subjected to: (1) starvation, (2) different ambient temperatures, and (3) different ambient salinities. 45 46 SWIFT AND AHMED MATERIALS AND METHODS Oysters (C. virginica), purchased commercially (Capt. White and Sons, Seafood, 110 Main Avenue, SW, Washing- ton, DC 20024), had been harvested two or three days before arrival in the laboratory. The height of the animals, measured as the distance from the hinge to the extreme ventral margin of the shell, ranged from 7 to 1 2 cm. Before any data were gathered the oysters were cleansed in tap water with the aid of a wire brush and acclimated to laboratory conditions for three days. Up to 20 unfed individuals were held in an aquarium in approximately 7 C of artificial sea water (Instant Ocean, Aquarium Systems Inc., 33208 Lakeland Blvd., Eastlake, OH 44094). The glass holding tanks were arranged so that the sea water was drawn off at the bottom of each tank, and then pushed up through a water-cooled condenser to the top of the holding tank by compressed air (Swift et al. 1975). A refrigerated bath and circulator was used to control the water tempera- ture. Sea water in the tank was changed every two days and the tank thoroughly rinsed at those times. Hemolymph was collected with a small syringe from die pericardial cavity of carefully opened oysters. The hemo- lymph was placed in an ice-cooled centrifuge tube. Cellular debris were separated from the hemolymph by centrifuga- tion at 1,000 X g for 20 minutes at 4°C. The supernatant liquid was transferred to a small vial and stored at — 10°C before glucose, total Lowry-positive substances (LPS), and triacylglycerol determinations were accomplished. Glucose was analyzed using the glucose oxidase method (Bergmeyer and Bernt 1974), total Lowry-positive substances were estimated according to Lowry (Lowry et al. 1951), and triacylglycerol was analyzed by the acetylacetone test (Fletcher 1968) with a slight modification. Hemolymph that was pooled from 3 to 4 oysters was extracted with n-heptane; 1 m2 of the upper layer was removed for analysis. After the aliquot was dried completely under a stream of air, 2.0 m2 of isopropanol were added. Thereafter the procedure was the same as described by Fletcher (1968). Hemolymph lipids were extracted by the Folch proce- dure (Folch et al. 1957). The chloroform layer, remaining after the aqueous NaCl wash, was evaporated to dryness under reduced pressure. The lipids were redissolved in a minimal quantity of 2:1 (v/v) chloroform :methanol and separated by thin-layer chromatography on silicic acid using n-hexane:diethyl ethenglacial acetic acid at a volu- metric ratio of 70:30:1 (Malins and Mangold 1960). The spots were visualized by iodine vapor retention or by ultraviolet fluorescence after spraying the chromatogram with 0.2% V :7'-dichlorofluorescein in 95% ethanol. To examine the effect of selected environmental condi- tions on the levels of metabolites in oyster hemolymph, groups of unfed animals were held in tanks for up to 27 days under the following conditions: (1) in 24 ppt sea water at temperatures of 4, 10, 15, or 20°C, and (2) in 12, 18, or 24 ppt sea water at 20° C or 15°C. Data were analyzed for significance (p < 0.05) by the Student's /-test. RESULTS Oysters obtained throughout the course of this study did not have significantly different initial levels of hemolymph glucose (Table 1). Overall hemolymph glucose concentra- tions averaged 8.83 ± 1.98 mg/100 mC (± SE) and ranged from 1.9 to 25.0 mg/100 ml. Hemolymph LPS levels averaged 11.0 ± 1.89 mg/m2 and ranged from 3.17 to 29.5 mg/mE. Hemolymph triacylglycerol values were quite low averaging 43.2 /ug/100 mC and ranged from 3.3 to 200 Mg/100 m8. TABLE 1. Initial hemolymph glucose, Lowry-positive substances (LPS) and triacylglycerol levels in groups of oysters. Glucose LPS Triacylglycerol Month N (mg/100 mC)* (mg/mS)* (Ag/lOOmC)** December 11 15.80 ±6.54 26.00 ±4.18 11.7 January 6 9.18 ±2.18 18.60 ±3.18 25.0 February 12 8.96 ±1.46 19.40 ±4.18 15.6 March 20 12.90 ±2.28 14.10 ± 1.30 — April 108 8.41 ±2.50 12.10 ±2.34 26.3 May 6 3.14 ±1.22 8.08 ±1.41 30.0 June 36 9.09 ±2.11 8.02 ±2.37 43. 9f *Mean values ± SE **Mean values obtained by pooling hemoymph from 3 or more individuals t76.7 jug/100 m£ if values of 150 and 200 jUg/100 m£ are included No free or nonesterified fatty acids could be detected in oyster hemolymph using standard analytical techniques or after lipid extraction followed by thin-layer chromatog- raphy. This is in agreement with results of other lipid analyses of oyster tissues (Watanabe and Ackman 1977, Bunde and Fried 1978, Ghassemieh 1978). Oysters held at constant temperature and in sea water of constant salinity tended to maintain their hemolymph glucose, LPS, and triacylglycerol concentrations over a 27-day period of starvation (Tables 2, 3. and 4); however, extremes in external conditions appear to affect the concen- trations of these metabolites. Groups of unfed oysters maintained in 24 ppt artificial sea water at temperatures of 4°C had significantly higher (p < 0.05) levels of hemolymph glucose and LPS when compared to values obtained from oysters kept at 20°C. Oysters held at 4°C had hemolymph glucose values of 19.3 ±3.5 mg/100 mC while those kept at 20°C had hemolymph glucose values of 8.41 ± 1.4 mg/ 100 mC. Similarly the mean LPS values were 17.56 ± 1.42 mg/mC and 9.76 ± 0.85 mg/mC for the animals at 4°C and 20°C, respectively. At a low ambient salinity of 1 2 ppt, oyster hemolymph glucose and LPS concentrations were significantly (p < 0.05) decreased when compared to the values found in oysters kept in water of 18 and 24 ppt (Tables 5 and 6). Study of Oyster Hemolymph 47 TABLE 2. Hemolymph glucose levels* (mg/100 m?) in starved oysters maintained in 24 ppt sea water at different temperatures. Temperature ( C) Number of Days 4 10 15 20 3 7 14 24 27 23.4 ±11.2 23.1 ± 10.6 13.2 ± 2.59a 19.6 ± 4.07(5) 13.3 ±10.0(3)3 19.3 ± 3.523 9.18 ±2.17 10.7 ±2.04(5) 11.3 ±1.74 7.72 ±1.83 6.63 ±1.75 11.7 ±3.81 10.2 ±2.81 5.33 ± 1.19 8.38 ±2.14 6.43 ±1.25 b 13.3 ±6.02(5) 5.92 ± 0.904b Group Mean 10.3 ±l.ll b 9.06 ± 1.33b 8.44±1.43 b *Mean value obtained from six individuals ± SE, unless otherwise indicated. Number in parenthesis shows number of oysters used. Means assigned the same or no superscript were not significantly different. Means assigned different superscripts were different at p ^0.05 level (compared across groups). TABLE 3. Hemolymph Lowry -positive substance levels* (mg/m?) in starved oysters maintained in 24 ppt sea water at different temperatures. Temperature ( C) Number of Days 4 10 15 20 3 7 14 24 27 26.0 ±4.18 a 15.1 ±1.52 14.1 ± 1.17 17.1 ±1.17(5)3 12.3 ±1.64(3) 17.5 ± 1.42 a 18.6 ± 3.18 14.5 ±2.86 15.1 ±1.68 19.4 ±4.18 (5) 16.8 ±1.32(5) 21.0 ±4.59 17.7 ±4.59 10.1 ±2.23(3)b 12.1 ±2.02(5) 6.7 ±1.23 11.1 ±1.84(5)b 8.71 ±1.93 Group Mean 16.1 ±2.57 18.7 ±3.56 9.76 ±0.85 b 'Mean value obtained from six individuals ± SE, unless otherwise indicated. Numbers in parenthesis show number of oysters used. Means assigned the same or no superscript were not significantly different. Means assigned different superscripts were different at p ^0.05 level (compared across groups). TABLE 4. Hemolymph triacylglycerol levels* (JLlg/100m6) in starved oysters maintained in 24 ppt sea water at different temperatures. Temperature ( C) Number of Days 4 10 15 20 3 11.7 25.0 6.25 16.9 7 13.4 23.8 6.25 55.0 14 8.33 6.25 6.25 47.5 24 6.25 — 25.0 113.0 27 — — — 27.3 *Pooled samples from 3 to 6 oysters. DISCUSSION Hemolymph glucose levels have been examined in other fasting molluscan species. In the terrestrial snail, Stropho- cheilus oblongiis (Miiller), hemolymph glucose values ranged from 2.5 mg/100 mx 1 to 16.88 mg/100 m2 (Marques and Falkmer 1976). Hemolymph glucose levels in the freshwater pulmonate snail, Lymnaea stagnalis jugidaris (Say), ranged from 1.86 to 5.68 mg/100 m2 (X = 3.0) and 1.9 to 4.0 mg/100 mx 1 (X = 2.9) in separate investigations (Friedl 1968, 1971 ). Hemolymph glucose concentrations in two freshwater bivalve molluscs, Anodonta cygnea (Linne) and Unio pictorum (Linne) averaged 9.4 ± 0.49 mg/100 m2 and 14.0 ± 1.6 mg/100 mS, respectively (Plisetskaya et al. 1978). The hemolymph glucose level in the Atlantic deep sea scallop, Placopecten magellanicus (Gmelin). was 2.6 ± 0.6 mg/100 mC (Thompson 1977); and the hemolymph glucose concentration in another marine bivalve, Mytilus edulis Linne, lies between 16.0 and 37.0 mg/100 m? (Bayne 1973). Inspection of these data leads to the conclusion that hemolymph glucose values during fasts in several molluscan species may vary over a wide range and are not directly related to terrestrial, freshwater or marine habitats. Thus, it may be inferred that these animals, including the oyster C. virginica, are more tolerant of larger variations of glucose concentrations in circulatory fluids than mammals. In this study, no deleterious effects of variations in hemolymph glucose levels could be detected. Groups of oysters with initial hemolymph glucose levels of 23.0 to 25.0 mg/100 mx 1 survived as long as those with initial hemolymph glucose values of 5.3 to 8.5 mg/100 mE. 48 SWIFT AND AHMED TABLES. The effect of ambient water salinity on hemolymph glucose levels* (mg/100 m£) of starved oysters. Temperature ( C) 20 15 Salinity (ppt) Number of Days 12 18 24 18 24 3 7 14 24 27 3.14 ±1.22(5) 2.46 ±0.47(5) 3.98 ±0.69 7.08 ±2.58(5) 3.74 ±0.92(5) 2.88 ±0.75(5) 3.72 ±0.93(5) 5.33 ±1.19 8.38 ±2.14 6.43 ±1.25 13.3 ±6.02(5) 5.92 ±0.90 6.72±0.92 b 10.2 ±1.09(5) 7.26 ±0.88(5) 2.16 ±0.51(5) 4.58 ±0.97(5) 7.87 ±1.82 6.63 ±1.76 11.7 ±3.82 10.2 ±2.86 Group Mean 3.24 ±0.59 4.57 ±1.00 6.53±1.00 b 9.02 ± 1.58 b *Mean values obtained from six individuals ± SE. Number in parenthesis shows number of oysters used. Means assigned the same or no superscript were not significantly different. Means assigned different superscripts were different at p ^0.05 level (compared across groups). TABLE 6. The effect of ambient water salinity on hemolymph Lowry-positive substance levels* (mg/m?) of starved oysters. Temperature ( C) 20 15 Salinity (ppt) Number of Days 12 18 24 18 24 3 7 14 24 27 8.08 ±1.41 5.98 ± 1.70(4) a 8.68±0.96(4) a 12.2 ±2.41(3) 6.51 ±0.73(4) a 9.99 ±1.99(4) 10.1 ±2.23(5) 12.1 ±1.23(5) b 6.73 ±1.23 11.1 ±1.84(5) 3.17 ±0.71(5) a 9.48 ± 1.74 7.68 ±1.94(5) 19.4 ±4.18(5) 16.8 ± 1.32 b 18.3 ±4.61 17.7 ±4.15 Group Mean 6.07 ±0.91 a 9.31 ± 1.15 b 9.64 ± 1.17 b 6.78 ± 1.15 b 17.5 ±2.10 b *Mean values obtained from six individuals ± SE. Number in parenthesis shows number of oysters used. Means assigned the same or no superscript were not significantly different. Means assigned different superscripts were different at p =Ss0.05 level (compared across groups). During the course of these studies hemolymph glucose levels were relatively stable within test groups. This indicates that some type of regulatory mechanism functions in the oyster. There is no direct evidence for the regulation of hemolymph glucose in other molluscs; however, indirect evidence concerning various aspects of this physiological mechanism has been published. Enzymatic activities which are necessary for the postulated regulation have been identified in several molluscs. For example, hexokinase and glycogen phosphorylase activities have been reported in Pecten maximus (Linne), O. eclulis, Ensis ensis flinne), Chlamys varius (Linne) (Zammit and Newsholme 1976), and C. gigas (Nakamuro et al. 1980). Glycogen synthase activity has been studied in M. edulis (Cook and Gabbott 1978, Gabbottet al. 1979). Of the hormones known to affect mammalian blood glu- cose levels, only insulinhasbeen investigatedin some molluscs. Hemolymph glucose levels in A. cygnea, U. pictorum (Plisetskaya et al. 1978), and S. oblongits (Marques and Falkmer 1976) are affected by insulin in ways analogous to those found in mammals. In addition, insulin-like proteins have been reported in several freshwater bivalves (Pliset- skaya et al. 1978), a terrestrial snail (Marques and Falkmer 1976). and in saltwater bivalves (Collip 1923, Fritsch and Sprang 1977), including O. edulis (DeMartinez et al. 1973). Hemolymph triacylglycerol levels in two other bivalves were at least 20 times those found in oysters in this study. Triacylglycerol concentration in the hemolymph of the hard clam, Mercenaha mercenaria (Linne), was 1 mg/100 m? (Hoskin and Hoskin 1977), and in the plasma of the deep- sea scallop,/ 1 , magellanicus, values ranged from 0.1 to 1 mg/ 100 m5 (Thompson 1977). The low levels of hemolymph triacylglycerols and free fatty acids in bivalve molluscs may be a consequence of their general metabolic strategy. As facultative anaerobes (Zandee et al. 1980) these animals would be more dependent upon carbohydrate for energy than lipid. Study of oyster Hemolymph 49 Few reports on the concentration of hemolymph proteins have appeared. Hand and Stickle (1977) examined ninhydrin- positive substances in whole hemolymph from C. virginica. Their values ranged from 193 to 702 mg/mC; however, those investigators were studying hemolymph which had not been subjected to centrifugation and. in addition, the ninhydrin method detects not only protein but also free amino acids. Thus, the large differences in data from the two laboratories may be explained. On the other hand, plasma from P. magellanicus contained LPS in the range of 1.55 to 2.17 mg/mB (Thompson 1977). The different levels of hemolymph glucose and LPS which were observed after the oysters were exposed to several temperature and salinity regimes may reflect adaptive metabolic mechanisms. These adaptive mechanisms would be necessary because oysters are sessile and, thus, subjected to the challenges of a changing euryhaline habitat. For example, successful acclimation to changing ambient salinity is apparently closely related to hemolymph amine concentration. Other investigators have found that hemo- lymph protein and amino acid levels not only in C. virginica (Hand and Stickle 1977), but also in Pyrazus ebeninus (BruguiJre) (Ivanovici et al. 1981) as well as the tissue free amino acid values (Lynch and Wood 1966), vary directly with ambient salinity. This phenomenon was readily observed with ambient salinity changes of > 6 ppt provided that the animals had been acclimated to the particular salinity for a period of at least two weeks. This is the first report that hemolymph glucose levels also vary with ambient salinity. Temperature also affects the metabolism of bivalve molluscs. Oysters that are held at elevated temperatures have increased metabolic rates as measured by increased oxygen utilization (Percy and Aldrich 1971, Percy et al. 1971, Shumway and Koehn 1981). As ambient tempera- ture increases, oyster hemolymph glucose levels decrease. Similarly short-term exposure (30 to 60 hours) of My tilus galloprovinciallus Lamarck to elevated temperature regimes caused a decrease in hemolymph glucose (Madar et al. 1980). The physiological importance of these findings remains to be explored. ACKNOWLEDGMENTS This work was supported in part by a Biomedical Research Support Grant No. 5S07. RR03561. from the General Research Support Branch, Division of Research Resources, National Institutes of Health. Bethesda, MD. REFERENCES CITED Bass, E. L. 1977. Influences of temperature and salinity on oxygen consumption of tissues in the American oyster (Crassostrea virginica). Comp. Biochem. Physiol. 58B : 125— 1 30. Bayne. B. L. 1973. Physiological changes in Mytilus edulis L. induced by temperature and nutritive stress. J. Mar. Biol. Assoc. U.K. 53:39-58. Bergmeyer, H. U. & E. Brent. 1974. Determination with glucose oxidase and peroxidase. Bergmeyer, H. U., ed.. Methods of Enzymatic Analysis. New York: Academic Press. Vol. 3 (2nd ed.): 1205-1212. Bunde, T. A. & M. Fried. 1978. The uptake of dissolved free fatty acids from seawater by a marine filter feeder, Crassostrea virginica. Comp. Biochem. Physiol. 60A:139-144. Collip, J. V. 1923. The demonstration of an insulin-like substance in the tissues of the clam (Mya arenaria). J. Biol. Chem. 55 :XXXIX. Cook, P. A. & P. A. Gabbott. 1978. Glycogen synthetase in the sea mussel, Mytilus edulis L. - I. Purification, interconversion and kinetic properties of the I and D forms. Comp. Biochem. Physiol. 60B:419-421. DeMartinez, N. R.. M. C. Garcia. M. Salas & J. L. R. Candela. 1973. Proteins with insulin-like activity isolated from oyster (Ostrea edulis L.) hepatopancreas. Gen. Comp. Endocrinol. 20:305-311. Fletcher. M. J. 1968. A colorimetric method for estimating serum triglyceride, din. Chim. Acta 22:393-397. Folch, J., M. Lees& G. H. Sloan-Stanley. 1957. A single method for the isolation and purification of total lipids from animal tissues. / Biol. Chem. 226:497-509. Friedl, F. E. 1968. Basal hemolymph glucose and dietary hypergly- cemia in the snail Lymnaea stagnalis. Am. Zool. 8:134 (Abstract) . 1971. Hemolymph glucose in the freshwater pulmonate snail Lymnaea stagnalis: Basal values and an effect of ingested carbohydrate. Comp. Biochem. Physiol. 39A:605-610. Fritsch. H. A. P. & R. Sprang. 1977. On the ultrastructure of poly- peptide hormone-producing cells in the gut of the ascidian. Ciona intestinalis L., and the bivalve Mytilus edulis L. Cell Tissue Res. 177:407-413. Gabbott. P. A.. P. A. Cook & M. A. Whittle. 1979. Seasonal changes in glycogen synthase activity in the mantle tissue of the mussel, Mytilus edulis L.: Regulation by tissue glucose. Biochem. Soc. Trans. 7:895-896. Gabbott. P. A. & A. J. M. Walker. 1971. Changes in the condition index and biochemical content of adult oysters {Ostrea edulis L.) maintained under hatchery conditions. /. Cons. Cons. Int. Explor.Mer. 34:99-106. Galtsoff, P. S. 1964. The American oyster Crassostrea virginica Gmelin. U.S. Fish Wildl. Sen: Fish. Bull. 64:381-396. Ghassemieh, M. B. 1978. The extraction and analyses of lipids in oyster Crassostrea virginica Gmelin. Washington, D.C.: Howard University. 58 p. Thesis Hand. S. C. & W. B. Stickle. 1977. Effects of tidal fluctuations of salinity on pericardial fluid composition of the American oysters Crassostrea virginica. Mar. Biol. (Berl.) 42:259-271. Hoskin, G. P & L. P. Hoskin. 1977. Partial characterization of the hemolymph lipids of Mercenaria mercenaria (Mollusca: Bivalvia) by thin-layer chromatography and analyses of serum fatty acids during starvation. Biol. Bull. (Woods Hole) 152: 373-381. Ivanovici, A. M., S. F. Raines & V. A. Wadley. 1981. Free amino acids in three species of mollusk: responses to factors associated with reduced salinity. Comp. Biochem. Physiol. 70A:17-22. Krishnamoorthy, R. V.. G J. Lakshmi. P. Biesiot& A.Venkataramiah. 1979. Variations in glycogen, total fat and caloric energies of the American oyster Crassostrea virginica (Gmelin) from natural reefs in the Mississippi Sound. Proc. Indian Acad. Sci. 88B: 397-409. Lowry, O. H.. N. J. Rosebrough, A. L. Fair & R. J. Randall. 1951. Protein measurement with the E'olin phenol reagent. J. Biol. Chem. 193:265-275. 50 SWIFT AND AHMED Lynch, M. P. & L. Wood. 1966. Effects of environmental salinity on free amino acids of Crassostrea virginica Gmelin. Comp. Biochem. Physiol. 19:783-790. Malins, D. C. & H. K. Mangold. 1960. Analysis of complex lipid mixtures of thin-layer chromatography and complementary methods. /. Am. Oil Chem. Soc. 37:576-578. Mann, R. 1979. Some biochemical and physiological aspects of growth and gametogenesis in Crassostrea gigas and Ostrea edulis at sustained elevated temperatures. /. Mar. Biol. Assoc. U.K. 59:95-110. Marques, M. & S. Falkmer. 1976. Effects of mammalian insulin on blood glucose level, glucose tolerance, and glycogen content of musculature and hepatopancreas in a gastropod mollusk, Stropho- cheilus oblongus. Gen. Comp. Endocrinol. 29:522-530. Madar, I., N. Sildan & E. A. Pora. 1980. Hyperthermic effects as a function of time on the hemolymph glucose content of Mytilus galloprovincialis. Stud. Cercet. Biol. Ser. Biol. Anim. 32:37-42. Nakamuro. K., K. Morimoto, M. Inui & S. Ishibashi. 1980. Basic biochemical studies on metabolism in the oyster. Properties of oyster hexokinase and its sensitivity to mercury compounds. EiseiKagaku 26:174-180. Percy, J. A. & F. A. Aldrich. 1971. Metabolic rate independent of temperature in mollusc tissue. Nature (Lond.) 231:393-394. & T. R. Marcus. 1971. Influence of environmental factors on respiration of excised tissues of American oysters, Crassos- trea virginica (Gmelin). Can. J. Zool. 49:353-360. Plisetskaya, E., V. K. Kazakov, L. Soltitskaya & L. G. Leibson. 1978. Insulin-producing cells in the gut of freshwater bivalve molluscs Anodonta cygnea and Unio pictonim and the role of insulin in the regulation of their carbohydrate metabolism. Gen. Comp. Endocrinol. 35:133-145. Riley, R. T. 1976. Changes in the total protein, lipid, carbohydrate, and extracellular fluid free amino acids of the Pacific oyster, Crassostrea gigas, during starvation. Proc. Natl. Shellfish. Assoc. 65:84-90. Ruddy, G. M., S. Y. Feng & G. S. Campbell. 1975. The effect of prolonged exposure to elevated temperatures on the biochemical constituents, gonadal development and shell deposition of the American oyster, Crassostrea virginica. Comp. Biochem. Physiol. 518:157-164. Shumway, S. E. & R. Koehn. 1981. Environmental and genetic influences on oxygen consumption in Crassostrea virginica. Paper presented to the Second International Conference on Aquacul- ture Nutrition, Rehoboth Beach, Delaware. Swift, M. L., K. Conger, J. Exler & S. Lakshmanan. 1975. Uptake of glucose and orthophosphate by the American oyster. Life Sci. 17:1679-1684. Swift, M. L., D. White & M. B. Ghassemieh. 1980. Distribution of neutral lipids in the tissues of the oyster Crassostrea virginica. Lipids 15:129-132. Thompson, R. F. 1977. Blood chemistry, biochemical composition, and the annual reproductive cycle in the giant scallop, Placopecten magellanicus, from southeast Newfoundland. J. Fish. Res. Board Can. 34:2104-2116. Watanabe, T. & R. G. Ackman. 1977. Effect of storage on lipids and fatty acids of oysters. Can. Inst. Food Sci. Technol. J. 10:40-42. Willis, S. A., W. K. Havens & R. M. Ingle. 1976. Quality improve- ment of oysters Crassostrea virginica (Gmelin) using artificial foods. Fla. Mar. Res. Publ. 20:6 p. Zammit, V. A. & E. A. Newsholme. 1976. The maximum activities of hexokinase, phosphorylase, phosphofructokinase, glycerol phosphate dehydrogenase, octapine dehydrogenase, phosphoenol- pyruvate carboxykinase, lactate dehydrogenase, nucleoside diphosphate kinase, glutamate-oxaloacetate transaminase and arginine kinase in relation to carbohydrate utilization in muscles from marine invertebrates. Biochem. J. 160:447-462. Zandee, D. I., D. A. Holwerda & A. deZwaan. 1980. Energy metabo- lism in bivalves and cephalopods. Giles, R. ed., Animals and Environmental Fitness. New York: Pergamon Press:185-206. Journal of Shellfish Research, Vol. 3, No. 1, 51-57. 1983. EFFECT OF RATION ON GROWTH AND GROWTH EFFICIENCY OF JUVENILES OF CRASSOSTREA VIRGINICA (GMELIN) EDWARD R. URBAN, JR., GARY D. PRUDER AND CHRISTOPHER J. LANGDON Center for Mariculture Research University of Delaware Lewes, Delaware 19958 ABSTRACT Juveniles of Crassostrea virginica were batch-fed on different rations of an algal diet of Tlialassiosira pseudonana and Isochrysis aff. galbana in experiments lasting three weeks and the resulting growth and growth efficiencies were determined. Maximum growth occurred when the oysters were fed on the highest daily ration tested which was equal, at the beginning of an experiment, to an algal dry weight of 4.6% of oyster live weight. Weight-specific rations decreased during each week of growth experiments because rations were only adjusted for oyster growth on a weekly basis. An initial daily ration of 4.6% was calculated to be equivalent to an effective daily ration of 2.8% of oyster live weight or 59.6% of oyster dry organic weight per week of an experiment. Highest growth efficiencies of 17.9 to 22.6% occurred with effective rations of 1.4 to 2.3% of oyster live weight. The experimental results indicated that weekly adjusted rations based upon previously reported formulae for the prediction of adequate rations for C. virginica may not be sufficient in meeting the requirements of juvenile oysters for maximum growth. KEY WORDS: ration, oyster, growth, algae, growth efficiency, Crassostrea virginica INTRODUCTION Successful rearing of bivalve molluscs for both research and commercial purposes depends upon the delivery of an adequate food ration. Despite many attempts to develop satisfactory nonalgal diets or supplements (e.g., Chanley and Normandin 1967, Winter 1974, Masson 1977, Epifanio 1979), algae remain indispensable as the principle food source for artificially reared bivalves. Growth studies have resulted in the determination of the relative food qualities of different algal species (for reviews see Epifanio [1983] and Webb and Chu [1983]); however, the relationship between ration size and bivalve growth rate has not been adequately studied for many bivalve species. The most complete studies on the relationship between ration size and growth of bivalves were conducted by Bayne and co-workers with Mytilus edulis L. (Bayne 1976, Widdows 1978a,b), and Navarro and Winter (1982) for Mytilus chilensis Hube. On the basis of measurements of the energy balance of Mytilus spp. fed on a range of algal rations under different conditions of algal cell density and animal body weight, numerical relationships were formu- lated that integrated these variables in a predictive model of "scope for growth." Scope for growth can be defined as the energy of the assimilated ration available for somatic and/or germinal tissue growth, once metabolic energy requirements have been met (Warren and Davies 1967). Bayne and Worall (1980) and Navarro and Winter (1982) found close agree- ment between growth of mussel populations in the field and growth predicted by such mathematical models. Less is known about the interrelationships among ration, metabo- lism, and growth for oysters, although assimilation and growth efficiences of Crassostrea virginica (Gmelin) have been reported by several workers (Tenore and Dunstan 1973, Langfoss and Maurer 1975, Romberger and Epifanio 1981, Valenti and Epifanio 1981). Predicting optimum algal rations for maximum oyster growth on the basis of caloric measurements and scope for growth determinations is of limited practical usefulness because algal diets vary in their nutritive value (Epifanio 1983, Webb and Chu 1983); thus, an algal ration may be calorifically satisfactory but biochemically deficient in some essential nutrient for growth. Because factors deter- mining algal food value are not fully understood, optimum rations for maximum oyster growth must be determined empirically. In this study, the effect of algal ration on the growth and gross growth efficiency of juveniles of C. virginica was determined. The tested rations were compared with the predicted rations for maximum oyster growth described by Epifanio and Ewart (1977), Pruder et al. (1977), and Epifanio (1979). MATERIALS AND METHODS Juveniles of C. virginica were fed different algal rations in a series of four experiments. In each experiment, groups of 20 oysters were randomly chosen from a population of similar sized oysters. Initial oyster live weight did not vary by more than one standard deviation of the population mean live weight. The identities of individual oysters were maintained during growth experiments by partitioning the oysters in 400 fim mesh trays, which were submerged in 4 2 of l-/im-filtered seawater at 30 ppt salinity and 25 C. The cultures were aerated to keep the algal cells in suspension and the seawater was changed daily. The animals were fed rations composed of a 50/50 mix- ture (based on dry weight [wt] )of Thalassiosira pseudonana 51 52 URBAN ET AL. Hasle and Heimdal (clone 3H) and Isochrysis aff. galbana Parke (clone T-ISO). This algal mixture supports excellent growth of juveniles of C. virginica (Ewart and Epifanio 1981). The algae were cultured in 250-8 containers at 19°C, illuminated with 550-600 /iW/cm 2 of light (cool white fluorescent lamps), and nutrient enriched with f/2 medium (Guillard 1975). Algal cell dry weights were assumed to be 1.32 X 10~ 8 mg/cell for T. pseudonana (Epifanio and Ewart 1977) and 2.01 X 10" 8 mg/cell for/, aff. galbana (S. Ali, Uni- versity of Khartoum, Port Sudan, Sudan, pers. comm.). Algal concentrations were determined using a hemocytometer. Initial algal rations that ranged in dry algal weight from 0.52 to 4.6% of oyster live weight were tested in growth experiments (Table 1). Algal concentrations ranged from 0.12 mg dry wt algae/C (10,000 cells/mC) to 2.60 mg dry wt algae/6 (217,000 cells/mx 1 ) (Table 1). By adding one-half the algal ration twice a day to the 4-2 culture vessels, it was possible to feed oysters algal cell concentrations which never exceeded 500,000 cells mC" 1 , and, therefore, were less than concentrations reported to cause pseudofecal produc- tion in C. virginica (Epifanio and Ewart 1977). Clearance of algal cells was greater than 95% per day in all treatments and, therefore, little loss of ration occurred. Oysters were weighed individually at the beginning of each experiment. Group live weights were used for weekly adjustments of rations to compensate for oyster growth during each week of the experiment. At the end of the experiments, oysters were reweighed individually, dried to constant weight at 60°C, weighed, and then ashed at 450°C for 24 to 48 hours and reweighed (Walne and Millican 1978). The difference between total dry weight and ash weight was assumed to be equal to total oyster organic weight. Individual live, dry, ash, and organic weights were similarly determined for an initial sample of 50 oysters at the beginning of each experiment. RESULTS The weight-specific daily rations decreased during each week of an experiment as a result of the growth of the animals and because the rations were only adjusted weekly (Figure 1). This decrease was greatest in treatments with rapidly growing oysters. To obtain a better estimate of the effective ration fed to the oysters, the geometric mean of the actual daily ration was determined for each week of an experiment. The overall effective ration for the 3-week experiment was calculated as the mean weekly effective ration (Table 1 ). Oyster growth rate increased with increasing effective ration over the range tested of 0.2 to 2.8% of oyster live weight (Table 1 and Figure 2). The highest effective algal ration of 2.8% of oyster live weight was equivalent to a ration of 59.6% of oyster dry organic weight, based on a mean dry organic contentof 4.7%foroystersfromtwoexperi- ments (Table 2). Regression analysis of log-transformed. TABLE 1. Initial, final, and effective percent rations and the resulting growth of juveniles of Crassostrea virginica after 3 weeks. Initial Ration Concentration (mg dry wt algae Z 1 ) Percent Rations* Initial Final Effective k Value 2.60 4.6 1.9 2.8 0.128 2.60 4.6 1.9 2.8 0.123 1.95 3.5 1.6 2.3 0.107 2.60 3.3 1.7 2.2 0.098 1.30 2.3 1.2 1.6 0.093 1.30 2.3 1.2 1.6 0.091 0.97 1.7 1.0 1.2 0.070 2.60 1.9 1.2 1.4 0.067 0.65 1.2 0.8 0.9 0.057 1.95 1.4 1.0 1.1 0.053 0.65 1.2 0.8 1.0 0.049 0.65 0.8 0.6 0.6 0.037 1.30 0.9 0.7 0.7 0.037 0.32 0.6 0.5 0.5 0.027 0.65 0.5 0.4 0.4 0.018 0.12 0.2 0.2 0.2 0.013 unfed 0.0 0.0 0.0 0.009 unfed 0.0 0.0 0.0 0.005 unfed 0.0 0.0 0.0 0.003 Percent Increase in Oyster Live Wt 1363 1226 847 687 604 585 338 305 231 203 183 120 107 78 45 31 22t 10* 6* *Percent ration = ([dry wt of algae per oyster live wt] X 100). Effective ration is the geometric mean ration for each week, averaged for the 3-week experiment (Figure 1). fk is the daily instantaneous relative growth rate (see RESULTS for formula). jLive weight increases of unfed oysters probably resulted from increases in inorganic shell weight because the organic content of unfed oysters decreased during the experiment (Table 2). Effect of Ration on Growth of Juvenile Oysters 53 weekly live-oyster weights plotted against time, indicated that growth occurred at a constant exponential rate for oysters fed on the 2.8% effective ration (r 2 = 0.997, F, 2615.3, p < 0.001). (1.6) o 3600 3200- o 2800 >J —I -o cr "5 O CD < £ o o rx o £ _J UJ 3 - Q _| >S v o^ UJ _l < < O 3 -I I- < > rr o 2400- 2000 1600 1200 800 H 400 Y=229t 0.l24t r^ 0.997 F = 26I5.3 p<0.00l (1,6) the effective % ration (geometric mean of actual % ration) 7 14 21 ELAPSED TIME -DAYS Figure 1. Change in percent daily ration for Crassostrea virginica fed an initial daily ration of 4.6%. The upper curve shows the growth of juveniles of C. virginica fed an initial daily ration of 4.6% over a 3- week period. The lower figure shows the change in the percent daily ration over the course of each week of the experiment. The vertical arrow indicates the effective ration for each week. The t-value in the exponential equation is in days. The daily instantaneous relative growth rate (k) was calculated for each ration (Table 1 ). where k= [(dWt/dt)/Wo] = (2.303/t) log (Wt/Wo) and Wo is the initial live weight (mg) and Wt is the final live weight (mg) after 21 days (t) of growth (Brody 1945). The k values and values for percentage increase in oyster live weight were both directly dependent on the weight- specific ration and were not greatly affected by the concen- tration of algae added to obtain the required ration (Table 1). — 14 O >< £ « i c « o e a Effective % Ration 1 2 3 elapsed time (weeks) Figure 2. Increase in live weight of juveniles of Crassostrea virginica fed different effective percent rations of an algal diet of Thalassiosira pseudonana and Isochrysis aff. galbana. Percent increase in live weight was calculated from initial live weight. Table 2 and Figure 3 give gross growth efficiencies for oysters fed different rations. Figure 3 shows that gross growth efficiency increased from -37.7% at an effective ration of 0.2% to a maximum of 22.6% with an effective ration of 1.4%. Gross growth efficiency declined slightly as rations were increased from 1.6 to 2.8%. From Figure 3, the maintenance ration for juvenile oysters cultured under the described conditions was 0.5% of oyster live weight. The organic content of both starved oysters and oysters fed a 0.2% effective ration decreased over the experimental period, compared with initial samples. Increases in total dry weights of starved oysters and oysters fed a 0.2% effective ration resulted, therefore, from increases in ash content, probably as a result of shell growth. DISCUSSION In bivalve growth experiments carried out by Langton and McKay (1975) and Gallager and Mann (1981), ration was not adjusted according to growth over the entire experimental period and the animals were fed a constant amount of food per individual. An important consequence of maintaining a constant ration with rapidly growing animals is that the weight-specific ration (expressed as a percentage of oyster live weight in this study) decreases as the animal grows (Figure 1 ). An example of large decreases in weight-specific ration is evident in Experiment 6 of Walne and Spencer (1974) in which a ration of Tetraselmis suecica (Kylin) Butch, fed to Ostrea edulis Linne decreased from 35 to 2% of oyster live weight over a 3-week period. This occurred even though the authors attempted to com- pensate for oyster growth by limited, but insufficient. 54 URBAN ET AL. TABLE 2. The relationship between the effective algal ration and the resulting growth and gross growth efficiency of juveniles of Crassostrea virginica. Increase in Increase in Dry Wt of Algae Percent Increase in Initial Oyster Oyster Oyster Dry Final Oyster fed per Oyster Live Wt/ Gross Growth Effective Ration* Live Wt Live Wt Organic Wt Organic Dry Experiment Dry Wt of Algae Efficiencyf (X 100) (mg) (mg) (mg) Wt/Live Wt (mg) fed per Experiment (GGE) 2.8 224.76 3,014.3 151.12 0.051 715.5 420 21.1 2.8 228.17 2,774.2 114.48 0.043 615.2 450 18.6 2.2 315.78 2,135.6 85.30 0.040 475.7 450 17.9 2.3 221.60 1,853.7 75.25 0.043 383.6 480 19.4 1.6 224.64 1,265.0 46.88 0.041 231.1 550 20.3 1.6 225.43 1,337.8 48.08 0.040 271.5 610 22.1 1.2 227.85 723.0 28.42 0.044 143.4 500 19.8 1.4 557.65 1,671.7 78.77 0.046 347.4 480 22.6 1.0 224.96 362.4 38.34 0.085$ 82.2 440 46.6 % 1.1 556.17 1,098.8 46.07 0.042 224.3 490 20.5 0.9 222.00 490.4 15.50 0.043 77.7 630 19.9 0.7 559.33 628.5 26.30 0.042 133.4 470 19.7 0.6 316.69 344.3 12.34 0.039 65.0 530 19.0 0.5 224.91 126.8 0.13 0.040 24.0 370 0.5 0.2 228.22 47.0 - 4.22 0.039 11.2 420 -37.7 0.4 556.11 219.7 6.35 0.038 57.2 380 11.1 unfed 219.17 48.2* 5.86 0.037 — unfed 220.73 22.7* - 6.43 0.039 — unfed 551.66 31.9* - 2.58 0.038 — *Effective percent ration = average weekly effective percent ration for a 3-week experiment (Figure 1), expressed as (mg dry wt algae per mg live wt oyster) X 100. fGross growth efficiency (GGE) = (increase in oyster dry organic weight/total dry weight of algae fed) X 100 for an experimental period of 3 weeks. tThese values are anomalous and may have resulted from analytical error. ♦Increases in the live weight of fed animals were adjusted by subtracting the mean increase in the live weight of starved animals. This was necessary to accurately determine gross growth efficiency (Winberg 1958). weekly increases in ration. Clearly, if a constant weight- specific ration is desired throughout a growth experiment, frequent adjustments of ration in proportion to bivalve growth are necessary. Such adjustments are especially important in growth experiments with juvenile animals in which weight-specific growth rates are high, and which result in significant changes in weight-specific rations over short periods of time, unless frequent ration adjustments are made. Changes in weight-specific rations will be less dramatic with large animals that have lower weight-specific growth rates. Under certain conditions the use of photo- electric devices to maintain constant algal concentrations may be useful (Winter 1973). Pruder et al. (1976, 1977), Epifanio and Ewart (1977), and Epifanio (1979) attempted to determine the maximum ration that could be ingested by bivalves under optimal growth conditions where excess food was available. Under those conditions, they assumed that the growth rate would be greatest when the animal was fed as much food as it could consume, i.e., a maximum ration (Epifanio and Ewart 1977). Because maximum ration is dependent on animal weight (Navarro and Winter 1982), several ration formulae, derived from measurements of the filtration rates of 40 IP (J c QJ O <^ UJ _c -t- o L o (/) U) o L 20 -20 -40 Effective % Ration Figure 3. Gross growth efficiencies of juveniles of C. virginica fed different effective percent rations for a period of 3 weeks. GGE = (increase in oyster organic dry wt/dry wt of algae fed) X 100. Effect of Ration on Growth of Juvenile Oysters 55 Crassostrea virginica, have been described in an attempt to predict the maximum ration on a weight-specific basis. Pruder et al. ( 1 976) reported an empirically derived equation relating oyster weight to a daily requirement of cells of a mixture of Thalassiosira pseudonana and hochrysis galbana. The equation Y = 5.3 W" 0,41 was derived on the basis of the maximum filtration rates of both laboratory-reared juvenile oysters and adult oysters from the field, where Y was the daily ration of algal cells of a 50/50 mixture (by cell number) of T. pseudonana and /. galbana X 10 8 per gram live weight of oyster and W was the individual oyster live weight in grams. Later, Pruder et al. (1977) repeated the work using only laboratory-reared oysters and the equation was modi- fied to Y = 8.2 W~°- 21 . The modification was required because laboratory-reared oysters had a higher content of organic material compared with wild oysters. Epifanio and Ewart (1977) determined the maximum dry weights of four algal species which could be filtered from suspension by laboratory-reared oysters (C. virginica) of 15 g live weight. They found that the maximum ration cleared varied from 4 mg/g/day (0.4% ration) for T. pseu- donana to 1 5 mg/g/day (1 .5% ration) for /. galbana. Using a maximum ration of 4 mg/g/day and a value for the exponent of -0.41 obtained from Pruder et al. (1976), Epifanio and Ewart (1977) derived the equation R/W = 0.01 W 0A \ where R was the daily ration of algae in mg dry weight, and W was the individual live weight of the animal in grams. In a later paper, Epifanio (1979) adjusted the value of the exponent to a theoretical value which was closer to the empirical value of Pruder et al. (1977) and the formula predicting ration size was given as R/W = 0.01 W -033 . The growth of C. virginica fed on rations derived from the formulae of Pruder, Epifanio, and co-workers has not been studied experimentally. In Figure 4, the predicted rations are compared with those of the present study. In the first week, the 4.6% initial ration was lower than the predicted ration of Epifanio and Ewart (1977), but higher than the rations of Pruder et al. (1977) and Epifanio (1979). As the animals grew, the predicted rations based on the weight-specific equations decreased and in the second and third week of the growth experiment, all were less than the 4.6% initial ration used in the present study. It was impossible to definitely determine which rations given in Figure 4 would support the greatest oyster growth. Juvenile oysters fed on the highest initial ration of 4.6% in this study grew at a constant exponential growth rate throughout the experimental period (Table 1, Figure 1), and were not adversely affected by the high algal concen- trations of the ration during the latter part of the experi- mental period. The optimal ration for maximum growth of juvenile oysters weighing 11 to 64 mg was, therefore, probably greater than that predicted by the weight-specific equations. Further study is necessary to test this hypothesis with juvenile oysters weighing less than 1 g, because the equations of Epifanio and Ewart (1977) and Pruder et al. (1977) were derived from experiments using larger oysters than those used in the present study. 7 LEGEND A - Ep.(amo& Ewart (1977) Mean oysier hve *l B " Epifanio (1979) 11 3 ma C ~ Thu paper 6 A D - Pruder eial 11977) c 5 Mean oyster hvewl Mean oysler live v o p _ i | 1 ?•<»! t , 627 4 B C A C C cc D £ 3 B A Id 2 — D B D c 1 — 0-1 1-2 2-3 Experimental period (weeks) Figure 4. A comparison of the initial percent rations used in the present paper and initial percent rations derived from reported equations for determining the maximum ration for Crassostrea virginica. The initial weekly mean individual live weights of oyster fed the 4.6% ration in the present study are indicated above each set of bars. These weights were used to calculate initial percent rations. The 8-part bar "C" indicates the eight rations used in the present study (Table 1). The relationship between ration size and gross growth efficiency (Figure 3) is similar to that reported for Mytilus edulis by Thompson and Bayne (1974) in that there was initially a dramatic increase in gross growth efficiency to a maximum value, followed by a slight decline with further increases in ration. At still higher rations, gross growth efficiency may decrease even more sharply, making it important for commercial oyster culturists to balance the cost savings of further improvements in growth rate with the increased costs of decreased utilization of expensive algal food. Comparisons between gross growth efficiencies of M. edulis and those reported in this paper for C. virginica are difficult because Thompson and Bayne (1974) used larger animals and also expressed gross growth efficiency in terms of tissue dry organic weight and not total organic weight (i.e., they did not include the contribution of the food to synthesis of the organic fraction of the shell). Price et al. (1976) reported that 39% of the total organic material of M. edulis (3.5 to 14.4 g live weight) was present in the shell and that 72% was present in the shell of adults of C. virginica (80.9 to 170 g live weight). For juveniles of C. virginica (10 to 30 mg live weight), the proportion of the total organic matter present in the shell is 33.8 ± 5.8% (C. Langdon, University of Delaware, Lewes, DE, unpub- lished data). Clearly, failure to take into account increases in 56 URBAN ET AL. the organic content of the shell may result in considerable underestimations of gross growth efficiencies (see Jorgensen 1976). Based on measurements of the total increase in the organic weight of juvenile oysters, Romberger and Epifanio (1981) reported a maximum gross growth efficiency of 36% for C. virginica fed a 50/50 mixture (by cell volume) of T. pseudonana and /. galbana at ration levels based on the predicted rations of Epifanio and Ewart (1977). Their maximum gross growth efficiency was, therefore, greater than the highest efficiency found in this study of 22.6% and may have resulted from differences in culture conditions. In conclusion, the use of high-algal rations and high concentrations of algae up to 500,000 cells mvT 1 need not be detrimental to oyster growth or growth efficiency when used in batch-feeding systems (Pruder and Greenhaugh 1978). The highest initial percentage ration tested in this study of 4.6% was greater than those recommended for oysters of the same size by the predictive equations discussed above. Constant adjustments of ration are required to compensate for increases in oyster weight during the course of growth experiments. An initial daily ration of 4.6%, which was equivalent to an effective daily ration of 2.8% per week, supported good growth of juveniles of C. virginica under the conditions of this study. Optimal rations for maximum oyster growth will vary according to culture conditions. Empirical growth studies, such as those described here, are useful because they integrate culture conditions with both the physiological and nutritional requirements of oysters for maximum growth. REFERENCES CITED Bayne, B. L. 1976. Marine Mussels: Their Ecology and Physiology. Cambridge and New York: Cambridge University Press. 506 p. & C. M. Worall. 1980. Growth and production of mussels, Mytilus edulis from two populations. Mar. Ecol. Prog. Ser. 3: 317-328. Brody, S. 1945. Bioenergetics and Growth. New York, NY: Reinhold Publishing Co. 1023 p. Chanley, P. & R. F. Normandin. 1967. Use of artificial foods for larvae of the hard clam, Mercenaria mercenaria. Proc. Natl. Shellfish. Assoc. 57:31-37. Epifanio, C. E. 1979. Comparison of yeast and algal diets for bivalve mollusks. Aquaculture 16:187-192. . 1983. Phytoplankton and yeast as foods for juvenile bivalves: A review of research at the University of Delaware. Pruder, G. D., C. J. Langdon & D. E. Conklin, eds. Proceedings of the Second International Conference on Aquaculture Nutrition: Biochemical and Physiological Approaches to Shellfish Nutrition. 1981 October 27-28. Rehoboth Beach, DE. World Maricult. Soc. Spec. Publ. 2:292-304. & J. Ewart. 1977. Maximum ration of four diets for the oyster Crassostrea virginica Gmelm. Aquaculture 1 1 : 1 3 — 29. Ewart, J. W. & C. E. Epifanio. 1981. A tropical flagellate food for larval and juvenile oysters, Crassostrea virginica (Gmelin). Aquaculture 22:297-300. Gallager, S. M. & R. Mann. 1981. The effect of varying carbon/ nitrogen ratio in the phytoplankton Thalassiosira pseudonana (3H) on its food value to the bivalve Tapes japonica. Aquaculture 26:95-105. Guillard, R. R. L. 1975. Culture of phytoplankton for feeding marine invertebrates. Smith, W. L. and M. H. Chanley, eds. Culture of Marine Invertebrate Animals. New York and London: Plenum Press, p. 109-133. Jorgensen, C. B. 1976. Growth efficiencies and factors controlling size in some mytilid bivalves, especially Mytilus edulis L.: review and interpretation. Ophelia 15:175-192. Langton, R. W. & G. U. McKay. 1976. Growth of Crassostrea gigas (Thunberg) spat under different feeding regimes in a hatchery. Aquaculture 7:225-233. Langfoss, C. M. & D. Maurer. 1975. Energy partitioning in the American oyster. Crassostrea virginica (Gmelin). Proc. Natl. Shellfish. Assoc. 65:20-25. Masson. M. 1977. Observations sur la nutrition des larves de Mytilus galloprovincialis avec des aliments inertes. Mar. Biol. (Berl.) 40:157-164. Navarro, J. M. & J. E. Winter. 1982. Ingestion rate, assimilation efficiency and energy balance in Mytilus chilensis in relation to body size and different algal concentrations. Mar. Biol. (Berl.) 67:255-266. Price, T. J., G. W. Thayer, M. W. LaCroix & G. P. Montgomery. 1976. The organic content of shells and soft tissues of selected estuarine gastropods and pelecypods. Proc. Natl. Shellfish. Assoc. 65:26-31. Pruder, G. D., E. T. Bolton, E. E. Greenhaugh & R. E. Baggaley. 1976. Engineering aspects of bivalve molluscan mariculture. Progress at Delaware, 1975. Proc. World Mariculture Soc. 7: 607-622. Pruder, G. D., E. T. Bolton & C. E. Epifanio. 1977. Hatchery techniques for a controlled environment molluscan mariculture system. Third Meeting of the International Council for the Exploration of the Sea Working Group on Mariculture. 1977 May 10-13. Brest, France. Actes Colloq. Cent. Natl. TExploit. Oceans 4:347-351. Pruder, G. D. & E. E. Greenhaugh, inventors. 1978. University of Delaware: assignee. Bivalve mollusc rearing process. U.S. patent 4,080,930. 1978 March 28. 4 p. Int. A01K 61/00. Romberger, H. P. & C. E. Epifanio. 1981. Comparative effects of diets consisting of one or two algal species upon assimilation efficiencies and growth of juvenile oysters, Crassostrea virginica (Gmelin). Aquaculture 25:77-87. Tenore. K. R. & W. M. Dunstan. 1973. Comparison of feeding and biodeposition of three bivalves at different food levels. Mar. Biol. (Berl.) 21:190-195. Thompson, R. J. & B. L. Bayne. 1974. Some relationships between growth, metabolism and food in the mussel, Mytilus edulis. Mar. Biol. (Berl.) 27:317-326. Valenti, C. C. & C. E. Epifanio. 1981. The use of a biodeposition collector for estimation of assimilation efficiency in oysters. Aquaculture 25:89-94. Walne, P. R. & P. F. Millican. 1978. The condition index and organic content of small oyster spat. /. Cons. Cons. Int. Explor. Mer. 38:230-233. Walne, P. R. & B. E. Spencer. 1974. Experiments on the growth and food conversion efficiency of the spat of Ostrea edulis L. in a recirculation system./ Cons. Cons. Int. Explor. Mer. 35:303-318. Warren, C. E. & G. E. Davies. 1967. Laboratory studies on the feeding, bioenergetics, and growth of fish. Gerking. S. D., ed. Tlie Biological Basis of Freshwater Fish Production. Oxford, England: Blackwell Scientific Publications, p. 175-214. Webb, K. L. & F. E. Chu. 1983. Phytoplankton as a food source for bivalve larvae. Pruder, G. D., C. J. Langon & D. E. Conklin, eds. Effect of Ration on Growth of Juvenile Oysters 57 Proceedings of the Second International Conference on Aqua- Winberg, G. G. 1958. Rate of Metabolism and Food Requirements culture Nutrition: Biochemical and Physiological Approaches to of Fishes. Nanchor. Trudy belojussk. gos. Univ. V. I. Lenina. Shellfish Nutrition. 1981 October 27-29. Rehoboth Beach, DE. (Translated from Russian by Fish. Res. Board Can. Trans!. Ser. World Maricult. Soc. Spec. Publ. 2:272-291. No. 194, 1960.) Widdows, J. 1978a. Physiological indices of stress in Mytilus edulis. Winter, J. E. 1973. The filtration rate of Mytilus edulis and its /. Mar. Biol. Assoc. U.K. 58:125-142. dependence on algal concentration, measured by a continuous . 1978b. Combined effects of body size, food concentration, automatic recording apparatus. Mar. Biol. (Berl.) 22:317-328. and season on the physiology of Mytilus edulis. J. Mar. Biol. . 1974. Growth of Mytilus edulis using different types of Assoc. U.K. 58:109-124. food. Ber. Dtsch. Wiss. Komm. Meeresforsch 23:360-375. Journal of Shell fish Research, Vol. 3, No. 1, 59-64, 1983. EFFECT OF DEPURATION SYSTEMS ON THE REDUCTION OF BACTERIOLOGICAL INDICATORS IN CULTURED MUSSELS (MYTILUS EDULIS LINNAEUS) AURORA LEDO, ENRIQUE GONZALEZ, JUAN L. BARJA AND ALICIA E. TORANZO Departamento de Microbiologia Facultad de Biologia Universidad de Santiago de Compostela Spain ABSTRACT Five bacteriological parameters (total coliforms, fecal coliforms, fecal streptococci, Escherichia coli, and total viable count) were used to examine depuration of cultured mussels (Mytilus edulis Linnaeus) by two different systems, one using chlorine as a disinfection agent for the water, and the other using untreated seawater. The most significant difference in post-depuration levels between chlorinated and untreated seawater systems was obtained for fecal coliforms (63.4 and 90.1% reduction, respectively), whereas reduction of the other bacteriological parameters were quite similar for both depuration methods. Although there was a large decrease in the fecal streptococci (> 74%), high residual numbers could be detected after depuration. From the identification of bacteria isolated from mussels, we found that the pathogens Salmonella and Yersinia were not recovered in the depurated samples, even though the genera Citrobacter, Enterobacter, and Escherichia coli were detected either before or after depuration. The drug-resistance patterns of the most representative members of the enterobacteria isolated from mussels were also determined. KEY WORDS: mussels, Mytilus edulis, shellfish depuration, pollution indicators, drug-resistance INTRODUCTION Since Dogson (1928) found that depuration was an effective method for reducing the microbial flora of contam- inated shellfish, this method has been adopted as the best technique for reducing the potential risk of public health hazards associated with the consumption of shellfish which might have accumulated high levels of bacterial or viral pathogens. In Galician "rias" (Atlantic coast of northwestern Spain), the production of cultured mussels (Mytilus edulis Linnaeus ) on rafts is a very important economic activity, reaching 200,000 metric tonnes in 1981. Approximately 50% of this production is destined for daily consumption and export, following depuration which is required by Spanish regulations. The depuration process is based on holding shellfish in tanks containing seawater that has been sterilized by physical or chemical means. The technology of depuration has been well studied (Huntley and Hammerstrom 1971, Neilson et al. 1978, Souness et al. 1979), and reviewed (Furfari 1976, Fleet 1978). Most countries have chosen to clean their shellfish in depuration plants rather than by relaying in natural waterways. Ultraviolet irradiation, ozonation, and chlorination are widely used to sterilize seawater for depuration (Kelly 1961, Wood 1961, Anon. 1972); however, Reynolds (1956) showed that the process could be simplified if depuration plants were located in areas with light or no contamination. In the former cases, the water sterilization step could be suppressed. Because of the special geography of Galician n'as, it is possible to find within 30 km (18 miles) depuration plants located in areas without microbial contamination, as well as others, nearer populated areas (on the middle upper part), that must use disinfection agents for water treatment. Our objective was to compare the reduction of bacterio- logical indicators of pollution in cultured mussels which were subjected to depuration systems that used either chlorinated seawater or untreated seawater. MATERIAL AND METHODS The sampling area selected for this study is located in northwestern Spain (Figure 1). Mussel samples were collected from January to June 1982, from rafts located in several shellfish-growing areas, and were treated in three different depuration plants; two plants used chlorinated seawater and the other used untreated seawater. During the sampling period, the water salinity ranged from 31.7 to 34.3 ppt and the temperature oscillated between 13 and 19°C. Total coliform levels of the water in the chlorine-treated systems ranged from 230 to 830 per 100 m2. The standard dose of chlorine for water treatment was 3 ppm. Treated water was dechlorinated by an appro- priate aeration period before the mussels were placed into the shellfish tanks. In the untreated system, the detected level of total coliforms was never higher than 9/100 mC. In both the treated and untreated systems the depuration time period was 48 hours. Samples were taken twice a month before and after depuration, transported to the laboratory in isotherm con- tainers, and immediately processed. Each sample was divided into two subsamples which were analyzed simul- taneously. Mussels were shucked aseptically according to 59 60 Ledo et al. Depuration of mussels by Two Different Systems 61 procedures recommended for shellfish by the American Public Health Association (APHA 1970). One hundred grams ( 1 00 g) of shellfish meat without mantle fluid (corresponding to six mussels) were weighed aseptically. After the addition of 1% of peptone water, the mixture (1 :9 w/v) was homoge- nized for 60 seconds in a sterile Waring blender. Each homogenate was transferred into a sterile flask and used as inoculum. Ten-fold serial dilutions of the homogenate were inoculated in triplicate on plate-count agar (Difco) and incubated at 37°C for 24 hours. After incubation, plates were counted and the results were expressed as colony- forming units (CFU) per gram. Total coliforms were estimated by the standard most probable number (MPN) method using three dilutions in three tube replication of lactose broth (LB) (Difco). Tubes were incubated at 35°C for 48 hours after which they were examined for growth and gas production (APHA 1970). Lactose broth tubes were reinoculated simultaneously into brilliant-green lactose bile broth (BGLB) (Difco) and into 1% triptone water, then incubated in a water bath at 44.5 ± 0.2°C for the indol test. Tubes showing growth and gas in BGLB were confirmed as fecal coliforms (FC). The MPN of Escherichia coli was determined from positive tubes for both tests, growth with gas at 44.5 ± 0.2°C and indol production. Fecal streptococci were determined by the MPN method in azide dextrose broth (Difco) at 35°C. Positive tubes of presumptive test were inoculated in ethylviolet-azide broth (Difco) at 35 C. Tubes showing violet sediment were con- sidered positives and the presence of fecal streptococci was confirmed by streaking on KF-streptococcus agar (Difco). Positive tubes from LB and BGLB of the MPN test were streaked on Levine-eosin methylene blue agar (Difco) and incubated at 37°C for 24 hours to isolate enterobacteria. Colonies were picked randomly from the plates, subcultured repeatedly to obtain pure cultures, and stored on agar slopes under mineral oil at room temperature. The isolates were subjected to taxonomic analysis using morphological, physiological and biochemical tests according to the pro- cedures of Edwards and Ewing (1972) and Bergey's Manual (Buchanan and Gibbons 1974). The drug-resistance patterns of the isolates were deter- mined by the diffusion disk assay method of Bauer et al. (1966) on Mueller-Hinton agar (Difco). The following anti- biotics and concentrations were used: ampicillin (10 jug), chloramphenicol (30 jug), erythromycin (15 /Jg), gentamicin (10 jug), polymyxin B (300 units), nalidixic acid (30 jug), kanamycin (30 jug), tetracycline (30 jug), and streptomycin (10 jug). RESULTS AND DISCUSSION The results obtained in this study of depuration levels of total coliforms (TC), fecal coliforms (FC), fecal strepto- cocci (FS). Escherichia coli, and total viable count (TVC) with the two systems used are shown in Table 1 and Figure 2. Total viable counts* 61.5 Total coliformsf 30.2 Fecal coliformsf 63.4 Escherichia co//f 91.5 Fecal streptococcif 74.0 In general, only small differences were observed between the two depuration systems. For the total viable count, similar values were obtained. The TVC decreased by 10-fold over the depuration time, but rarely went below values of 10 3 to 10 4 CFU/g of mussel. Similar results were found by Lee and Pfeifer (1974) who worked with oysters depurated by ultraviolet irradiated seawater and, as they indicated, that reduction in bacterial count in shellfish could have been due to the persistance of a stable population of micro- organisms in the mussels. In addition, Thi Son and Fleet (1980) obtained even lower reduction levels than ours in a laboratory depuration system with artificially contaminated oysters. TABLE 1. Comparison between the reduction levels of bacterial pollution indicators in Mytilus edulis obtained in two different depuration systems. Percent Reduction in Systems Using Bacterial Indicators Chlorinated Sea Water Untreated Sea Water 65.5 38.6 90.1 89.0 87.0 *Determined on plate-count agar medium at 37 C and expressed as bacterial numbers per gram. tDetermined by the most probable number (MPN) method and expressed as MPN/ 100 g. The most important different in the observed depuration in chlorinated and untreated seawater systems was obtained for FC, although in both methods most (about 90%) of this bacterial flora was represented by E. coli. The high depuration levels found for this organism agreed with the the results obtained by Thi Son and Fleet (1980) who attained depuration reductions greater than 97%. Considering only the reduction rates for E. coli, we found residual counts to be within the values allowed by Spanish regulation (500 E. coli/9.) in both depuration systems. If, however, we consider other regulations that use the number of FC as the indicator for bacteriological control, then the untreated seawater system appeared to be the most efficient method (Table 1). The FC levels in this system after depuration were below the recommended wholesale level of < 230/100 g (Slalyj 1980) suggested by the U.S. National Shellfish Sanitation Program for naturally harvested shellfish. Examination of bacteria isolated from mussels showed that the genera Citwbacter, Enterobacter and Escherichia coli were detected before and after depuration whereas other pathogens or potential pathogens such as Salmonella and Yersinia were not isolated from depurated samples of mussels (Figure 3). The elimination of organisms such as 62 LEDO ET AL. 4_. 9 O Z a. 3-. o 2_. 3 LL O o TVC TC VA FC FS E.coli WA *-/A VA Chl Oc Chi Oc Chi Oc Chi Oc Oil Oc Figure 2. Comparison between the reduction rates of bacteriological indicators obtained by the two different methods employed. 127 mm carapace width) and 51.6% were male. During 7 preoperational years (1968-74), crabs/pot aver- aged 4.06 at Kenwood Beach (33.3%), 3.94 at Plant Site (32.3%), and 4.18 at Rocky Point (34.3%). During 7 opera- tional years (1975-81), crabs/pot averaged 6.24 at Kenwood Beach (33.3%), 6.37 at Plant Site (34.0%), and 6.13 at Rocky Point (32.7%). Increased catch during the operational period was due to extreme abundance in 1981 when pots averaged nearly 17 crabs. Data analyses revealed no signifi- cant station differences other than a higher percentage of males at Kenwood Beach than at Rocky Point (p=0.005). There has also been a significant decrease in percent males since 1968 (p < 0.001) which has occurred equally at all stations. No effect of the CCNPP on crab populations was evident from these studies. REPRODUCTION AND DEVELOPMENT IN THE LUCINID CLAM CODAKIA ORBICULARIS LINNE PHILIP ALATALO 1 , CARL J. BERG 1 AND CHARLES N. D'ASARO 2 Marine Biological Laboratory Woods Hole, Massachusetts 02543 Department of Biology University of West Florida Pensacola, Florida 32504 The tiger lucine Codakia orbicularis is a large edible clam currently being investigated as a mariculture candidate in the Bahamas Islands. Gonad development and spawning seasons were assessed by monthly sampling of C. orbicularis from Grand Bahama Island and Key Biscayne, Florida. Histological examination of clams exceeding 20 mm in shell length showed most of the populations sampled ripe between April and November. Natural spawning probably occurs from May to October. Clams seldom respond to standard spawning techniques, including physical and chemical stimuli. Artificial fertili- zation by carefully stripping the gonads produced 15 to 20% viable embryos. Eggs are 108 to 112 nm in diameter and are encased in a thick capsular membrane. Following fertilization, the gastrula, trochophore, and early veliger stages develop within the capsular membrane. Upon hatching, the planktonic veliger ranges from 150 to 174 fxm in shell length and develops to the pediveliger stage in approximately 12 days. Metamorphosis occurs approxi- mately 16 days after fertilization. Larval growth and developmental features peculiar to C. orbicularis are discussed. EFFECTS OF INORGANIC PARTICLES ON THE GROWTH OF THE EASTERN OYSTER CRASSOSTREA VIRGINICA (GMELIN) SAYED M. ALI AND G. D. PRUDER College of Marine Studies University of Delaware Lewes, Delaware 19958 The effect of seven concentrations of inorganic particles (oxidized silt from the Broadkill River) on the growth of oysters (Crassostrea virginica) was studied at each of three algal ration levels. In the absence of silt (zero concentration) oyster growth was not significantly different between the selected algal ration levels. At the lowest algal ration, the addition of silt did not significantly affect oyster growth rate; however, at the medium and high algal ration levels oyster growth did increase with increasing silt concentra- tion up to 25 mg/£. Above 25 mg/2, up to 150 mg/8. the increased growth rate level was maintained showing neither further enhancement nor any adverse effect on oyster growth. The silt effect is discussed in terms of improved delivery of food, growth factors, toxic metabolites, increased digestability, resuspension of pseudofaeces, and increased filtration and ingestion rates. Implications of the findings for bivalve molluscan mariculture are suggested. The increased growth rate could not be explained by any single mechanism. 82 Abstracts, 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland APPLICATIONS OF FLOW CYTOMETRY TO CYTOGENETIC STUDIES IN BIVALVE MOLLUSCS: MEASURING CHANGES IN DNA CONTENT STANDISH K. ALLEN, JR. Marine Cooperative Fisheries Research Unit University of Maine Orono. Maine 04469 Flow cytometry is a relatively new approach to cyto- genetic studies in the biomedical field. This technique is of considerable utility in other fields, especially in measuring quantum shifts in DNA content. Diploid and triploid oysters and clams were subjected to tissue disaggregation and nuclei isolation techniques in an attempt to derive a suspended cell population for analysis. Tissue disaggregation was shown to be most effective and the principles of this method are des- cribed. Nonlethal analysis of DNA content in individual bivalves was also accomplished by sampling cells from hemo- lymph sinuses. An apparent quantum duplication of DNA between the sea scallop and bay scallop was demonstrated. Recommendations for continued investigations using flow cytometry are presented. OBSERVATIONS ON THE GROWTH AND SURVIVAL OF LABORATORY-REARED JUVENILE CONCHS, STROMBUS GIGAS AND S. COSTATUS R. S. APPELDOORN, D. L. BALLAN- TINE AND P. CHANLEY Department of Marine Sciences University of Puerto Rico Mayaguez, Puerto Rico 00708 A study of the culture and life history of the queen conch Strombus gigas Linne in Puerto Rico has been underway since 1981. Its objective is to develop suitable methods for the large-scale culture of larvae of S. gigas and subsequent release of juveniles to rebuild depleted natural stocks. Although efforts have concentrated on S. gigas, larvae of the closely related milk conch S. costalus Gmelin have also been raised. Larvae were raised from eggs collected from the field. The larval period was variable with settlement commencing from 12 to 19 (x = 15.6) days after hatching. Length at metamorphosis varied from 1.2 to 1.8 mm with a mode between 1.4 and 1.5 mm. Sets of over 1,000 juveniles were achieved with survival ranging from 4 to 7% from hatching to a postmetamorphosis size of 3 to 5 mm. After metamor- phosis growth increased noticeably. Initial postmetamorpho- sis growth was 0.2 mm/day, but the rate of growth continued to increase reaching a mean of 4 mm/day through the first 200 days. Feeding experiments of juveniles indicated that the macroalga SpyriJia filamentosa (Wulfen) was preferred. Pilot experiments involving the release of small (25 to 50 mm) tagged juveniles permitted the testing of suitable mark and recapture methods and the collection of prelimin- ary observations of juvenile behavior. These observations indicated that mortality was initially high but dropped over time. Dispersal has been slow and random. Observed growth was slow, probably caused by the large amount of time spent buried and hence inactive. A SIMULATION STUDY OF A STOCHASTIC MODEL FOR MY A ARENARIA JENNY A. BAGLIVO 1 , GEORGE E. LANG 2 AND DIANE J. BROUSSEAU 3 Department of Mathematics Fairfield University Fairfield, Connecticut 06430 and Department of Biostatistics Sloan-Kettering Institute New York, New York 10021 2 Department of Mathematics Fairfield University Fairfield, Connecticut 06430 Department of Biology Fairfield University Fairfield, Connecticut 06430 Field data presented by Brousseau (1978, 1979) provided estimates of age-specific fecundity and survival for the soft- shell clam Mya arenaria. We have used these values in a Leslie population model (1945, 1948) to estimate an equili- brium settlement rate for clams in the first age class (Brous- seau et al., in press). Settlement rates are highly variable in nature, however, and the modelling efforts incorporate this phenomenon!. An optimal harvesting strategy based upon the Leslie model was published by Rorres and Fair (1975). We have designed simulation studies which adapt their pro- cedure as well as other similar procedures to a stochastic environment and applied these strategies using the Mya model. Preliminary results show that these methods do not over exploit the population; however, they may be too conservative. AN EXPERIMENTAL HABITAT POT FOR PREMOLT CRAB CAPTURE JAMES M. BISHOP AND V.G. BURRELL,JR. Marine Resources Research Institute South Carolina Wildlife and Marine Resources Department P.O. Box 12559 Charleston, South Carolina 29412 Three years of testing premolt (peeler) crab capturing devices showed unbaited habitat pots to be a potential har- vest gear in South Carolina estuaries. Two and one-half- centimeter mesh wire was used for pot construction, and pot design was similar to that for baited hard crab pots. Tests were conducted 4 consecutive days/week in the Ashley National Shellfisheries Association, Baltimore, Maryland Abstracts, 1982 Annual Meeting, June 14-17, 1982 83 River from mid-April through mid-November, 1979, and daily in the Wando River from April through June, 1980 and 1981. Primary objectives were to increase pot efficacy and reduce pot construction cost and labor. Results showed that plastic flagging tape interwoven among the wire mesh did not increase catch rates: pots with and without tape averaged 0.7 peeler/gear-day (one pot with a soak time of 24 h). Two large entrance pots (61 X 61 X 45 cm) outfished 4 small entrance pots (61 X 61 X 30 cm) by 1.6 vs. 1.3 peelers/gear-day, respectively. Pots fished in shallow subtidal mudflats captured a mean of 1 .7 peelers/gear-day whereas those in deep water ( > 3 m) cap- tured only 0.7 peeler/gear-day. Highest capture rates were obtained in June during each year. A maximum of 3.5 peelers/gear-day was obtained when large habitat pots were fished on shallow water mudflats in June. Male peelers accounted for 63% of 1,832 peelers caught in habitat pots during 1981. Habitat pots require no bait and offer crabbers a method of harvesting peelers in relatively consistent num- bers throughout the shedding season. REFUGES FROM BLUE CRAB (CALLINECTES SAPIDUS RATHBUN) PREDATION FOR INFAUNAL BIVALVES IN THE CHESAPEAKE BAY JAY A. BLUNDON 1 AND VICTOR S. KENNEDY 2 Department of Zoology University of Maryland College Park, Maryland 20742 Horn Point Environmental Laboratories University of Maryland Cambridge, Maryland 21613 Direct measurements of valve strength of various sizes of Mya arenaria Linne, Macoma balthica (LinneV, Macoma mitchelli Dall, and Mulinia lateralis (Say) compared to measurements of blue crab chelae grip strength suggest that the shells of these infaunal bivalves confer no resistance to crushing by blue crabs. Also, blue crabs readily crushed these species in the laboratory. Possible refuges from predation afforded to theseinfaunal bivalves were investigated. Bivalve size, depth of burrowing, and density were measured in the field throughout spring and summer 1981 . This survey, in conjunction with labora- tory feeding experiments that offered M. arenaria burrowed at various sediment depths to blue crabs, suggested that M. arenaria and M. balthica obtain refuge from blue crab preda- tion at deeper sediment depths. Bivalves burrowed beneath an artificial submerged aquatic vegetation structure also gained additional protection. These refuges, however, were not absolute, but only relative to infauna burrowed less deeply or in bare sand (mud) environments. Yearly sampling of bivalve infauna in the Choptank River, Chesapeake Bay, suggested thatM mitchelli and M. lateralis are able to persist despite predation due to their high reproductive output. A RANDOM SAMPLE SURVEY TO ESTIMATE BLUE CRAB CATCH IN MARYLAND CHRISTOPHER F. BONZEK AND MICHAEL M. BURCH Maryland Department of Natural Resources, Tidewater Administration C-2 Tawes State Office Building Annapolis, Maryland 21401 In June 1981 the Maryland Department of Natural Resources (MDNR) began operating a new system to esti- mate the catch of blue crabs (Callinectes sapidus Rathbun) in Maryland waters. The basis of the system is a stratified, ran- dom sampling design developed by the Martin Marietta Corporation, which allows MDNR to reliably estimate total crab catch in Maryland by asking only a small fraction of all crabbers to report their catch each month. This method produced a total annual harvest estimate in 1981 of 29.5 X 10 6 kg (65 X 10 6 lb) live weight, nearly twice the highest estimate produced under past systems. The estimate is based on standard statistical techniques, and takes into account the previously ignored factors of non-reporting by some crabbers and the non-commercial catch. Estimates of fisher- man effort are produced concurrently. WHAT DETERMINES THE VULNERABILITY OF BIVALVE PREY TO HORSESHOE CRAB PREDATION? MARK L. BOTTON Department of Zoology Rutgers University P.O. Box 1059 Piscataway, New Jersey 08854 Adult horseshoe crabs, Limulus polyphemus (L.), were offered combinations of different size and species of bivalve prey in a large aquarium. Gemma gemma (Totten) (Veneridae), a small, thick shelled species, was avoided when larger, thinner shelled clams such as Mulinia lateralis (Say) (Mactridae) or Mya arenaria Linne (Myidae) were available. Crabs did not differentiate between M. lateralis and M. arenaria of comparable size; however, there was a preference for M. lateralis over hard-shell clams, Mercenaria mercenaria (Linne) (Veneridae), of equal size. Large individ- uals of M. lateralis, > 10-mm shell length, were preferred over smaller individuals of M. lateralis. Thus, both shell length and shell thickness appear to influence the preference of horseshoe crabs for bivalve prey. The largest available prey species offered to L. polyphemus was Spisula solidissima 84 Abstracts, 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland (Dillwyn) (Mactridae); clams up to 45-mm shell length were successfully opened. The method of consuming these bivalves differed from the manner in which smaller prey were handled, and is illustrated. CLAM PREDATION BY SCOTER DUCKS IN THE STRAIT OF GEORGIA, BRITISH COLUMBIA NEIL BOURNE Fisheries and Oceans, Pacific Biological Station, Nanaimo, B.C.. Canada V9R 5K6 Collections of three species of wintering scoter ducks, the white-winged scoter, Melanitta deglandi (Bonaparte), the surf scoter, M. perspicillata (Linnaeus), and the black scoter, Oidemia nigra (Linnaeus), were made at two clam beaches in southern British Columbia. Analyses of the crop and gizzard contents showed that these ducks were feeding primarily in the intertidal beach area. Molluscs, particularly bivalves, were the most important food items in the diet. The commercially important littleneck and Manila clams, Protothaca staminea (Conrad) and Tapes philippinarum (Adams and Reeve), respectively.comprised about two thirds of the gut contents of the scoters. Scoters are important clam predators in southern British Columbia; it was esti- mated that a wintering flock of 200 scoters could remove 5 to 14.5 metric tons of littleneck and/or Manila clams from these two beaches in a 6-mo period. DETERMINATION OF SETTLEMENT RATES IN SHELLFISH POPULATIONS USING MY A ARENARIA LINNE' AS A MODEL DIANE J. BROUSSEAU 1 , JENNY A. BAGLIVO 2 AND GEORGE E. LANG 3 Department of Biology Fairfield University Fairfield, Connecticut 06430 Department of Mathematics Fairfield University Fairfield, Connecticut 06430 and Sloan-Kettering Institute New York, New York 10021 Department of Mathematics Fairfield University Fairfield, Connecticut 06430 Egg loss, larval recruitment, and early post-larval mortal- ity are often limiting factors in the establishment and main- tenance of shellfish stocks; therefore, it is of interest to ecologists to be able to make estimates of settlement rates in such populations. This paper describes an indirect method for estimating mortality rates during settlement in shellfish populations for which demographic parameters (age-specific fecundity and survivorship) are available. The equilibrium settlement rate for a population olMya arenaria from Glou- cester, MA, was calculated using the Leslie matrix. Empiri- cally derived demographic parameters indicate that the theroretical settlement rate required to maintain a steady state population is 0.001462% or one egg out of approxi- mately 68,400 surviving to a size of 2 mm. HEAVY METAL BINDING TO PROTEINS OF THE BLUE CRAB CALLINECTES SAPIDUS RATHBUN M. BROUWER, D. ENGEL AND J. BONAVENTURA Marine Biomedical Center Duke Univeristy Marine Laboratory and NMFS Southeast Fisheries Center Laboratory Beaufort, North Carolina 28516 Hemocyanin is the large, extracellular oxygen transport- ing protein found in the hemolymph of the blue crab. The oxygen-binding site consists of a binuclear copper center. In addition to copper, blue crab hemocyanin invariably con- tains a small amount of tightly bound zinc (approximately 0.2 atom of zinc per oxygen-binding site). This observation, together with the fact that hemocyanins act at the interface between the organism and its environment, prompted us to investigate a possible role of these respiratory proteins in trace metal transport or toxicity in the blue crab. In vitro studies revealed that blue crab hemocyanin can indeed bind a variety of heavy metal ions, all with very high affinities (18 mercury, 14 cadmium, and 24 zinc ions per oxygen- binding site). The interaction of cadmium and zinc ions with blue crab hemocyanin increases its oxygen affinity ; mercuric ions have an opposite effect. All three heavy metal ions reduce the degree of cooperativity in oxygen binding. Cad- mium and zinc ions were found to substitute for calcium, which is a natural modulator of blue crab hemocyanin function. In vivo exposure of blue crabs to cadmium dissolved in a flowing seawater system at 0.1 ppm or to cadmium-ladened oysters did not result in measurable elevated levels of cadmium in the hemolymph. The sites of cadmium accumula- tion varied depending on the method of exposure. Seawater- exposed crabs accumulated most of the cadmium in the gills; the ions were bound to a low molecular-weight protein (MW~ 10,000). This protein was purified by gel-permeation chromatography and ion-exchange chromatography. Cad- mium was the only metal associated with the purified protein. Crabs exposed to cadmium-ladened oysters accumu- lated most of the cadmium in the hepatopancreas, where it was associated with a low molecular-weight cadmium/zinc- binding protein. Ion-exchange chromatography showed the gill and hepatopancreas proteins to be different, suggesting that these proteins, which are presumably involved in trace metal detoxification, are tissue specific. National Shellfisheries Association, Baltimore, Maryland Abstracts. 1982 Annual Meeting, June 14-17, 1982 85 THE ROLE OF CARBON FILTRATION IN CULTURING THE AMERICAN OYSTER CRASSOSTREA VIRGINICA CAROLYN BROWN National Marine Fisheries Senice, Northeast Fisheries Center, Milford Laboratory, Milford, Connecticut 06460 Embryos and larvae of the American oyster Crassostrea virginica (Gmelin) were reared in two types of "disinfected" seawater. One type was filtered through two 10-/im orlon filters and UV-irradiated; the second type was subjected to the same treatments, except that an additional filtration process through a carbon cartridge was inserted prior to the UV irradiation step. The study compared embryonic devel- opment of the 2-day-old larval stage, as well as survival and growth of larvae to metamorphosis in the two types of treated seawater. Data indicated that the percentage of live- normal development was significantly greater in seawater subjected to carbon filtration than in seawater without this added treatment. Other data suggested success in rearing oyster larvae to metamorphosis using carbon filtration only when the larval cultures were changed daily. Seawater treat- ment is but one aspect of the prevention regimen to be fol- lowed. Sound sanitary practices also are described to reduce the frequency of disease outbreaks in hatcheries. MOVING OUT THE LEARNING CURVE: AN ANALYSIS OF NURSERY OPERATIONS FOR THE HARD CLAM MERCENARIA MERCENARIA (LINNE') IN SOUTH CAROLINA JOHN W. BROWN 1 , JOHN J. MANZI 2 , HARRY Q. M. CLAWSON 3 AND FRED S. STEVENS 4 1 South Carolina Sea Grant Consortium, Charleston, South Carolina 29412 Marine Resources Research Institute Charleston, South Carolina 29412 3 'Trident Sea farms Co., 18 Broad St. Charleston, South Carolina 29401 Marine Resources Research Institute Charleston, South Carolina 29412 Trident Seafarms (a private corporation) and the State of South Carolina (SC Wildlife and Marine Resources Depart- ment) entered into a cooperative research agreement for the commercial production of hard clams in 1980. The SC Sea Grant Consortium provided partial funding for the scientific research and some staff time for the economic analysis of the first 15 months of nursery operation. Detailed cost and production analysis are provided, along with a description of the evolution of the nursery production protocols and of the nursery design. During the period from September 1980 to December 1981, 19,733,000 seed clams were imported into the nursery; of these 13,008,000 remained in the nursery at the end of the year, 3,402,000 were planted in the field with 14,700 returned to the nursery. The apparent mortality was 3,337,700 clams during the 15 months. This 16.9% mortality is misleading because of the rapidly increasing number of clams in the nursery over the period of the analysis. Beginning with the correction for mortality, a detailed budget analysis is given and linear programming is employed to determine optimal importation strategies. A SURVEY OF ALLOZYME VARIATION AND ESTIMATES OF GENETIC SIMILARITY AMONG THREE OSTREA SPECIES NORMAN E. BUROKER Bureau of Biological Research, Rutgers, The State University of New Jersey Piscataway, New Jersey 08854 Three nonsibling Ostrea species (i.e., O. edulis Linne, O. lurida Carpenter. andO. pennollis Sowerby) were studied by horizontal protein electrophoresis with relation to levels of genetic variation and similarity. The percentages of poly- morphic loci per species were estimated as 27.6, 37.0, and 52.0 for O. edulis, O. lurida, and O. permollis, respectively, based on an examination of 25 to 29 structural loci. The mean observed heterozygosities per individual were esti- mated as 9, 16, and 1 5% for O. edulis, O. lurida, and O. per- mollis, respectively. A pairwise comparison of loci was made between species which indicated that approximately 17% of the loci studies were genetically identical while 55% had no genetic similarity. The mean genetic identity across all loci among the three species was estimated as 24.5%. Finally, there seemed to be a correlation between the dispersal time of the planktonic larvae and the levels of genetic variation found within these nonsibling Ostrea species. THE SOUTHERN OYSTER DRILL: A PREDATOR OF TRAPPED BLUE CRABS EDWIN W. CAKE, JR. 1 AND VINCENT J. SMITH 2 Oyster Biology Section, Gulf Coast Research Laboratory Ocean Springs, Mississippi 39564 2 Route 3, Box F-52 Ocean Springs, Mississippi 39564 Southern oyster drills (Thais haemastoma floridana [Conrad] ) are reported for the first time to attack and kill mature blue crabs (Callinectes sapidus Rathbun) in commer- cial crab pots. Trapped blue crabs were attacked by as many as 54 drills of up to 80 mm in shell height. All affected crabs were either ovigerous or recently spent females, and all were simultaneously infested with the symbiotic acorn 86 Abstracts, 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland barnacle Chelonibia panda (Ranzani). Entry portals for the proboscis of feeding drills included: (1) open skeletal wounds caused by other trapped crabs, (2) internal skeletal openings between the branchial chamber and the infrabranchial sinuses at the bases of the gills, (3) stumps of autotomized pereio- pods, and (4) holes rasped in the exoskeleton by the snails' radulae. The attacks were attributed to at least two factors: the presence of large numbers of drills in the crab harvest area in the vicinity of Mississippi's offshore barrier islands, and the opportunistic feeding behavior of the drills, especially when confined with trapped crabs. Moribund and/or dead crabs also attracted another carnivorous snail, the cancellate cantharus, Cantharus cancellarius (Conrad). FACTORS AFFECTING DOCKSIDE PRICES FOR HARD BLUE CRABS IN CHESAPEAKE BAY ORAL CAPPS, JR. Department of Agricultural Economics Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061 The nature and the magnitude of selected factors hypo- thesized to influence the ex-vessel price of hard blue crabs in Chesapeake Bay were investigated. The data base used consisted of monthly observations for the period January 1973 to June 1980. Seasonality, landings of hard blue crabs in Chesapeake Bay, and the wholesale price of hard blue crabs had significant impacts on the ex-vessel price. Landings of hard blue crabs in the south Atlantic and the Gulf were not statistically significant in influencing the ex-vessel price of hard blue crabs in Chesapeake Bay. On the basis of the estimated flexibility coefficients, total revenue to harvesters could be incremented by increasing landings in Chesapeake Bay throughout each season of the year. MOLLUSCAN SHELL DISSOLUTION BY PENETRATING EUMETAZOAN INVERTEBRATES: AN HYPOTHESIS ON THE CHEMICAL MECHANISM BASED ON ULTRASTRUCTURE MELBOURNE R. CARRIKER College of Marine Studies University of Delaware Lewes, Delaware 1 9958 Of the 27 eumetazoan invertebrate phyla generally recog- nized, at least 8 widely separated ones are known to contain shell penetrating species (burrowers or borers): Platyhel- minthes, Bryozoa, Sipunculoidea, Phoronida, Annelida, Arthropoda, Brachiopoda, and Mollusca. The pattern of molluscan shell dissolution is similar at the ultrastructural level in species of four phyla that have been studied: polychaete Polydora websteri Hartman (Zottoli and Carriker 1974), barnacle Trypetesa lampas (Hancock) (Todd 1981). gastropod Urosalpinx cinerea (SayY) (Carriker 1978), and cephalopod Octopus vulgaris Cuvier (Nixon et al. 1980). A secretion weakens the shell surface by initially solubilizing the nonmineralized intercrystalline organic matrix between individual mineral cores of shell units, then dissolves exposed mineral cores; dissolution of organic matrix and mineral cores then proceeds at more or less equal rates, solubiliza- tion of the organic matrix ahead of mineral cores, the latter frequently irregular and pitted. The secretion of the accessory boring organ of U. cinerea, hypothesized to contain a com- bination possibly of HC1, chelating agent, and enzyme (Carriker 1981) could produce the differential dissolution observed ultrastructurally. Similarity of the pattern of etching produced in shell penetration of P. websteri, T. lampas, U. cinerea, and O. vulgaris suggests the existence of a generically similar chemical mechanism in the shell- penetrating Eumatozoa. COMPOSTING OF BLUE CRAB SCRAP: PROBLEMS AND SOLUTIONS THOMAS P. CATHCART, FRED W. WHEATON AND RUSSELL B. BRINSFIELD Department of Agricultural Engineering University of Maryland College Park. Maryland 20 742 Disposal of solid waste from blue crab processing plants became a major problem in Maryland with the closing of dehydrating plants. The dehydrated crab waste (scrap) was ground and sold for chicken feed. Presently, the scrap is disposed of in landfills; however, risk of ground water pollution and operational problems of placing crab scrap in landfills limits landfilling to a temporary solution. Com- posting of the crab scrap is a possible method of stablizing the waste and producing a useful soil additive for farmers, gardeners, the potted-plant industry, and others. Composting of crab scrap requires special provisions to eliminate noxious odors and prevent nuisance problems from developing. Studies to date have shown that the crab scrap pH must be maintained below 7.5 during composting, aeration must be supplied during part of the composting cycle, and a source of additional carbon must be added to the scrap. Solutions to these problems and methods of composting have been developed which produce high quality compost without noxious odor production. National Shellfisheries Association, Baltimore, Maryland Abstracts, 1982 Annual Meeting, June 14-17, 1982 87 OPTIMUM SALINITY REGIME FOR OYSTER PRODUCTION ON LOUISIANA'S STATE SEED GROUNDS MARK CHATRY AND R. J. DUGAS Lyle S. St. Amant Marine Laboratory Grand Terre Island. Louisiana 70358 Increased salinities have drastically reduced the produc- tive portion of Louisiana's public oyster seed grounds. Controlled freshwater diversions from the Mississippi River have been utilized or are now being planned in an attempt to reduce salinities and thereby reestablish formerly productive reefs. These diversions offer an unprecedented opportunity to manipulate salinities over a vast estuarine area for maxi- mizing seed oyster production. The purpose of this study was to determine the optimum annual salinity regime, using historical data, for the production of seed oysters on Louisiana's seed grounds. Salinity, spatfall, and seed oyster production data from three stations on Louisiana's productive seed grounds, 1971 — 1981, are presented. Salinity in the setting year was the prime factor determining production of seed oysters. Both high and low salinity extremes resulted in poor seed production. Insufficient setting was blamed for poor pro- duction at the low salinities and it was speculated that numerous organisms associated with the high salinities caused heavy mortalities in recently set oysters. The optimum annual salinity regime was derived from all of the year/ station salinity regimes which were followed in the ensuing year by good seed oyster production. This optimum regime accounts for the salinity dependent factors which limit seed production. GENE STRUCTURES OF ATLANTIC COAST POPULATIONS OF THE BLUE CRAB CALLINECTES SAPIDUS RATHBUN TIMOTHY J. COLE University of Maryland Center for Environmental and Estuarine Studies Horn Point Environmental Laboratories Box 775. Cambridge. Maryland 21613 Recent research has indicated that larvae of blue crabs are probably flushed from their parent estuary. Develop- ment continues in offshore waters, after which late-stage larvae or post-larvae return to the estuaries. A genetic study of blue crab populations was undertaken to determine if there is sufficient gene exchange among estuaries to prevent differentiation. Horizontal starch-gel techniques were used. Statistical analyses of frequencies of polymorphic loci indi- cate that blue crab populations south of Cape Hatteras are more genetically similar to each other than to those north of that cape. NATICID SNAIL PREDATION IN NEW ENGLAND: THE EFFECTS OF LUNATIA HEROS ON THE POPULATION DYNAMICS OF MY A ARENARIA AND MACOMA BALTHICA JOHN A. COMMITO Department of Biology Hood College Frederick, Maryland 21 701 The naticid snail predator Lunatia heros (Say) and two of its bivalve prey species, Mya arenaria Linne and Macoma balthica (Linne), were studied at an intertidal site in eastern Maine. The M. arenaria population was comprised largely of newly recruited individuals. Survivorship was low (3.5%/y) until the sixth year and increased thereafter. Lunatia heros preyed upon only those individuals of M. arenaria < 30 mm long. At that length the bivalve reached a size or depth refuge from predation. It delayed reproduction until it was 4 years old (20 mm long) and allocated its resources to rapid early growth instead (4.9 mm/y for the first 5 y). The dynamics of the population of M. balthica were different. There was a larger proportion of older individuals of M. balthica, and survivorship was higher (76.3%/y for the first 5 y). Macoma balthica grew to a length of 25 mm and never reached a size refuge. All sizes were susceptible to attack by L. heros, but the deeper burrow of M. balthica relative to individuals of M. arenaria of the same size may have afforded it some protection from predation. Macoma balthica grew slowly (2.7 mm/y for the first 5 y) and diverted its resources into reproduction at a younger age (3 y) and smaller size (10 mm). These different life-history patterns and the possible relationship between bivalve resource allocation and refuges from predation are discussed. THE EFFECTS OF POLLUTANTS ON LARVAL DEVELOPMENT OF THE BLUE CRAB CALLINECTES SAPIDUS RATHBUN J. D. COSTLOW AND C. G. BOOKHOUT Duke University Marine Laboratory Beaufort, North Carolina 28516 Since our initial rearing of all larval stages of the blue crab Callinectes sapidus from hatching to the juvenile crab, we have investigated the way in which a variety of pollutants may affect the survival, duration, and frequency of abnor- mality of larvae of this important commercial species. Having established the optimum temperatures and salinities required for total development, we have investigated the way in which a number of commonly used pesticides and heavy metals affect development, either singly or in combina- tion with those temperatures and salinities which are known to impose a stress on the developing larvae. Included among Abstracts, 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland the pesticides have been studies on Malathion, Methoxychlor. Mirex, Kepone, and Dimilin. Studies on the effects of heavy metals have included cadmium and mercury. Summary data involving these studies are presented and discussed. In all cases, small amounts of each of the chemicals tested reduced survival of the larvae. Even at "sublethal" levels, abnormalities in development were observed. In general, the larval stages were far more sensitive to pollutants than were the juvenile or adult crabs and any consideration of "water quality" should take into consideration this essential portion of the life cycle of the blue crab and the sensitivity of the various larval stages to extremely minute amounts of pollutants. ANALYSIS OF LOCAL POPULATIONS OF THE BLUE CRAB CALLINECTES SAPIDUS RATHBUN L. EUGENE CRONIN Chesapeake Research Consortium 4800 Atwell Road Shady Side, Maryland 20764 The catch of blue crabs and composition of that catch fluctuate rapidly and widely over time. Useful estimation of local availability, size structure, and sex composition is, however, essential for understanding and for management of the species. A procedure of obtaining such information is described and discussed. It involves detailed catch infor- mation from the best of samplers (selected professional crabbers) accompanied by appropriate quantitative observa- tion at frequent intervals on the composition of the catch. These can provide useful estimates of the number of each class of crab available per man day throughout the crabbing season. The advantage and limitations are considered. CHEMORECEPTION AND LIFE HISTORY OF STYLOCHUS ELLIPTICUS (GIRARD) PETER DANIEL 1 , TIMOTHY COLE 1 , AND DANIEL RITTSCHOF 2 1 Horn Point Environmental Labs University of Maryland Cambridge, Maryland 21613, and "College of Marine Studies University of Delaware Lewes, Delaware 1 9958 Stylochus ellipticus, a flatworm indigenous to the Atlantic coast of the United States, preys on oyster spat and barnacles. Adults have almost inflexible prey preferences. Little is known about early life stages. A prey chemolocation hypothesis was tested to explain ability of S. ellipticus to locate and discriminate prey species. Also, these studies initiated examination of life history and distribution of 5. ellipticus in Chesapeake Bay. Three apparatuses (chemossayer, Y-maze, and choice- chambers) were used to test adults for chemoreception. Effects of various environmental and biotic factors onchemo- reception were tested. The Atlantic oyster drill Urosalpinx cinerea (Say), an ecological analogue with an extensively studied chemobiology, was used to verify apparatus effec- tiveness and stimulus and control water attractiveness. Survivorship of larvae in nutrition and substrate preference settlement studies was determined. Distribution of S. ellipticus in Chesapeake Bay was determined from oyster bar survey reports (1980—81), occurrence in oyster hatch- eries (1980—81), and prior fouling plate studies (1963-65) (Shaw 1967). Studies of U. cinerea verified effectiveness of apparatuses and of stimulus and control water. Chemoreceptive behavior was indicated only in choice-chamber studies as long response time of adults rendered other apparatuses ineffec- tive. Light and starvation modified prey search. Stylochus ellipticus has a Gotte's larva which appears to be non- feeding and metamorphoses only on prey substrates. Though flatworm and prey densities often correlate, there were several instances of uninfested prey populations. Adults of S. ellipticus appear to prioritize behavior: (1) reproduction vs. prey search, and (2) prey search vs. escape. Barriers to larval dispersal probably allow some prey popula- tions to escape infestation. Earlier, nonreproductive life stages may influence prey preference establishment. EFFECT OF CRAB POT WIRE TREATMENT ON CRAB POT FOULING IN CHESAPEAKE BAY, MARYLAND RAY C. DINTAMAN AND J.F. CASEY Tidewater Administration, Maryland Department of Natural Resources Annapolis, Maryland 21401 It has been presumed that fouling on crab pots reduces the catch rate and contributes to a shortened fishing life or premature loss of the pot. Groups of standard anode pots, standard anode pots painted with an anti-fouling paint, and vinyl pots were compared for rate of fouling and catch. Crab pots treated with the anti-fouling paint fouled the least. Their fouling rate was 83% less than vinyl pots and 75% less than standard anode pots. Pots treated with anti-fouling paint accounted for 42% of the total crabs caught during the study. This study suggests that treatment of standard anode pots with anti-fouling paint could increase not only catch, but also pot life. AN OYSTER CULTCH COMPARISON : CLAMSHELL VS. LIMESTONE CHARLES N. DUGAS AND M. CHATRY Lyle S. St. Amant Marine Laboratory Grand Terre Island, Louisiana 70358 On 15 April 1981 four 70- X 70-cm trays containing equal National Shellfisheries Association, Baltimore, Maryland Abstracts. 1982 Annual Meeting, June 14-17, 1982 89 volumes of clamshell and graded crushed limestone were placed on the bottom at each of 10 stations in the Barataria Bay system of southeast Louisiana. At the end of 3 months two trays and their contents from each station were retrieved and replaced with two trays containing fresh material. After the following three months all trays were retrieved. Thus, the cultch materials were exposed to spat set for two suc- cessive 3-month periods and for one 6-month period. Spat set (spat/liter of cultch) was determined by counting live and dead spat on each piece of cultch material. The overall mean spat set/liter was 57.9 for limestone and 25.1 for clamshell. This ratio of approximately 2:1 also held true when the data were analyzed for each time period. Relative survival was slightly higher on clamshell; however, because of the greater set on limestone, there was still approximately twice the number of live spat on limestone as on clamshell. At current prices crushed limestone is approximately 60% higher than clamshell; however, since spat set on limestone was greater, the cost, using average prices, was about $0 .50/ 1 ,000 spat on limestone and $0 .70/ 1 ,000 onclamshell. INCIDENCE OF PATHOGENIC BACTERIA IN THE BLUE CRAB CALLINECTES SAPIDUS RATHBUN AND THE AMERICAN OYSTER CRASSOSTREA VIRGINICA (GMELIN) ELISA L. ELLIOT AND RITA R. COLWELL Department of Microbiology University of Maryland College Park, Maryland 20742 Blue crabs (Callinectes sapidus) and American oysters (Crassostrea virginica) were analyzed for the presence of human pathogenic bacteria. Live and cooked crabs, freshly picked crabmeat, and live, shucked, and washed oysters were obtained from a Maryland processing plant in the winter and spring of 1 981 -82. Cans of pasteurized crabmeat, purchased in Washington, DC, area stores, were also included in the study. All samples were subjected to standard plate- count determination and enrichment for the detection of specific pathogens. Sample analyses revealed low numbers of Staphylococcus aureus Rosenbach, Vibrio parahaemoly- ticus (Fujino et al.), other halophilic Vibrio spp., Aeromonas hydrophila (Chester), fecal coliforms, and presumptive Clostridium perfringens (Veillon and Zuber) spores; Vibrio cholerae Pacini and Salmonella spp. were not detected. Excluding S. aureus, all of the pathogens were present in highest numbers in the live crabs and oysters, suggesting that processing is effective in controlling the numbers of pathogens present in these foods. PREDATION ON SPAT OF THE AMERICAN OYSTER CRASSOSTREA VIRGINICA (GMELIN) BY THE AMERICAN LOBSTER HOMARUS AMERICANUS MILNE -EDWARDS. THE ROCK CRAB CANCER IRRORATUS (SAY), AND THE MUD CRAB NEOPANOPE SA YI (SMITH) R. W. ELNER 1 AND R. E. LAVOIE 2 Department of Fisheries and Oceans Biological Station St. Andrews, New Brunswick Canada E0G 2X0, and Department of Fisheries and Oceans Fisheries Research Branch Halifax, Nova Scotia, Canada B3J 2S7 Predation by lobsters, rock crabs, and mud crabs on oyster spat was compared in the laboratory at 13°C. Rock crabs (32- to 107-mm carapace width, CW) preyed on oysters up to 30 mm length, although they preferred smaller oysters. Preferred prey size increased with rock crab size. Lobsters (55- to 98-mm carapace length) demonstrated a broad preference for oysters of 1 0- to 25-mm length. Oysters up to 35-mm length were vulnerable to the lobsters. Preda- tion rate was highly variable but generally increased with predator size. Maximum mean lobster and rock crab preda- tion rates were 4.5 and 28.0 oysters/predator/day, respec- tively. Mud crabs (14- to 23-mm CW) and rock crabs (32- to 58-mm CW) feeding on oysters (2- to 9-mm length) attached to spat collectors ate approximately 0.5 oyster/predator/day. Lobsters used their mouthparts or chelae to open oysters by indiscriminate crushing. Rock crabs generally crushed the umbo, chipped away the shell margin, or punctured the prey shell. Mud crabs and rock crabs opened oysters still attached to the spat collector. Oyster fragments were found in the stomachs of 88 (44%) of 201 rock crabs collected around oyster beds in Caraquet Bay, New Brunswick. SEASONAL OCCURRENCE OF THE LARVAE OF CALLINECTES SAPIDUS RATHBUN IN DELAWARE BAY CHARLES E. EPIFANIO. C. C. VALENTI AND A. E. PEMBROKE College of Marine Studies University of Delaware Lewes, Delaware 19958 Blue crab larvae were collected weekly at a station in the mouth of Delaware Bay over a 16-wk period beginning in late June 1979. Collections were made with a 0.3-m Clark- Bumpus Sampler; discrete samples were taken at the surface, at 12 m, and at the bottom (25 m). On each sampling date, larvae were collected at the three depths every 3 h over one 90 Abstracts, 1982 Annual Meeting, June 14-17. 1982 National Shellfisheries Association, Baltimore, Maryland tidal cycle. Only Stage I zoeae and megalopae were collected during the course of the investigation. Peak abundance of Stage I occurred during late July and early August while peak occurrence of megalopae was observed 5 wk later. Stage I larvae were most abundant in seaward-flowing sur- face water and megalopae were distributed throughout the water column. We concluded that blue crab larvae are exported from the Bay as Stage I zoeae, undergo subsequent zoeal development on the continental shelf, and return to the estuary as megalopae. Adult of Crassostrea virginica were collected from 51 sites in Chesapeake Bay and its tributaries. Samples were analyzed for heavy metal, polychlorinated biphenyl (PCB), and pesticide contamination. Ranges, medians, means, and standard deviations were determined for the Maryland por- tion of Chesapeake Bay and for some major river systems. Trends indicated by the 1980 data are discussed. Data are compared to previously collected data. CHARACTERISTICS OF FECAL RIBBONS FROM JUVENILES OF CRASSOSTREA VIRGINICA (GMELIN) FED PHAEODACTYLUM TRICORNUTUM BOHLIN WITH AND WITHOUT THE ADDITION OF SILT: PRELIMINARY OBSERVATIONS JOHN W. EWART AND MELBOURNE R. CARRIKER College of Marine Studies University of Delaware Lewes, Delaware 19958 Two size classes of Crassostrea virginica were (edPhaeo- dactylum tricornutum at two cell concentrations with and without the addition of silt. The experimental treatments included 3-g and 21-g oysters, algal concentrations of 1.0 X 10 4 cells/ml and 1 .0 X 10 s cells/ml, and either natural or oxidized Broadkill River silt at a concentration of 50 mg/C. Each treatment was tested in replicate feeding trials lasting 24 h. Microscopic examination of fecal ribbon contents from oysters fed at the low algal concentration showed that the addition of silt resulted in a marked reduction in the number of whole cells of P. tricornutum. At the higher algal concentration the addition of silt had no effect on reducing the number of whole cells in the fecal ribbons. No differ- ences in the effect were found between oyster size classes. SEM examination of all fecal material indicated that silt- treated samples were different in appearance and composi- tion from those fed algae alone. The implications of silt additions in improving the nutritive value off. tricornutum are discussed. HEAVY METAL, POLYCHLORINATED BIPHENYL, AND PESTICIDE LEVELS IN CRASSOSTREA VIRGINICA (GMELIN) FROM CHESAPEAKE BAY MARY JO GARREIS AND F. A. PITTMAN Office of Environmental Programs Department of Health and Mental Hygiene, 201 W. Preston Street Baltimore, Maryland 21201 REDUCTION OF DISSOLVED ORGANICS IN BLUE CRAB PROCESSING PLANT EFFLUENT EUGENE L. GEIGER, RUSSELL B. BRINSFIELD AND FRED W. WHEATON Department of Agricultural Engineering University of Maryland College Park. Maryland 20742 Blue crab processing plants have difficulty meeting dis- charge guidelines for federal and Maryland state liquid effluents. Conventional treatment systems (e.g., foam flota- tion or aerated lagoons) do not represent viable options because of severe land and cost constraints. Research was initiated to develop: 1 ) a cost effective effluent treatment system and 2) a system producing effluent of sufficient quality to meet discharge guidelines. An attempt was made to utilize ultraviolet light as a substitute for chlorination. Crab cooking retort water, diluted to a 5% strength, was used as a consistent feed solution containing a high level of dissolved organics. Chemical floculation (with aluminum sulfate, ferric chloride, or ferrous sulfate), foam fractiona- tion, and aerobic biological treatment were examined in the laboratory using this solution to determine the most promis- ing treatment method. Because of the high dissolved organics concentration in the effluent, aerobic biological treatment proved to be the most effective treatment method. Various retention times in a sequential biological reactor were studied. A significant reduction in dissolved organic concen- trations was achieved, but substantial concentrations of col- loidal particulates were produced. Filtration with a fine sand filter greatly reduced the particulate concentrations. Final polishing by activated carbon absorption produced effluent transmission values in the range necessary for effective dis- infection by ultraviolet light. Water quality parameters were monitored between each treatment step. The quality of the water leaving the scale model system met federal and Mary- land state discharge limitations. National Shellfisheries Association, Baltimore, Maryland Abstracts. 1982 Annual Meeting, June 14-17, 1982 91 MORPHOMETRY PATTERNS IN INTERT1DAL BIVALVES REGINALD B. GILLMOR AND HERBERT HIDU Ira C. Darling Center Walpole, Maine 045 73 For several families of intertidal gastropods Vermeij (1973) has demonstrated low-to-high shore gradients in shell morphology which he interpreted in terms of adaptive responses to the dominant physical stresses of the shore environment. Evidence from a variety of studies suggests that similar responses may occur in bivalves. The present study examined this question further. Juveniles of six bivalve species (Argopecten irradians [Lamarck] , Modiolus modiolus [Linne] , Ostrea edulis Linne, Mytilus edulis Linne, Crassos- trea virginica [Gmelin] , and Geukensia demissa [Dillwyn] ) were grown at various tidal levels on a natural shore and in a laboratory tidal simulator. At the end of the treatment period, the bivalves were sacrificed and each specimen was measured for maximum shell dimension (MSD: length in the mussels, height in the other species) and width; dry meat and dry shell weights were also determined. Three morpho- metric ratios were calculated and compared among species and treatment groups: shell weight/(MSD X width) as an index of relative shell thickness; MSD/width as an index of relative shell globosity; and meat weight/shell weight. Bivalves that were grown intertidally tended to have thicker and more globose shells. These tendencies did not necessarily correlate with naturally occurring or experimental intertidal levels. Intertidal meat/shell ratios, however, corresponded closely to natural shore position; the lower-shore species had the lowest ratios and the higher-shore species had the highest. We concluded that inter-specific and, in some cases, intra-specific low-to-high shore gradients in morphometric relationships are present in bivalves. NONPLANKTOTROPHIC LARVAL DEVELOPMENT OF TWO SPECIES OF CONTINENTAL SHELF BIVALVES M. CASTAGNA' AND J. KRAEUTER" x Dept. of Oyster Culture, NJAES Cook College, Rutgers University New Brunswick, New Jersey 08903 VIMS. Wachapreague, Virginia 23480 Larvae of Periploma leanum (Conrad) and Astarte cas- tanea (Say) were reared under laboratory conditions. The larval stages of both species are lecithotrophic and have low dispersal capabilities. Spawning was induced in P. leanum with thermal stimulation and the addition of a gamete sus- pension following a period of intensive feeding. Individual eggs (dia. = 130 jim) were released inside of two-layered capsules. The outer gelatinous layer rapidly expanded and, within 24 hours, completely dissipated. After 4 to 6 days, straight-hinge larvae emerged from an opening at the restricted end of the inner capsule. After a planktic stage of < 24 h, the larvae (length = 170 p.m) assumed an inactive benthic existence; a functional foot was not observed until 15 to 18 days after fertilization. At no time during larval or early postlarval development were byssal threads observed. Astarte castanea was induced to spawn with thermal stimu- lation and the addition of a gamete suspension. Individual eggs (dia. = 170 yum) were released inside of double-walled, adhesive capsules. Prodissoconch I formation was extremely slow. The first sign of valve formation was observed after 6 to 10 days while the larvae rotated within the capsules. Movement within the capsule ceased between 8 and 15 days after fertilization when the valves first completely enclosed the soft tissues and closed against one another along their free margins. Between 22 and 26 days, young of A. castanea broke out of their capsules by pushing forcefully with their foot against the inner wall of the capsule. They emerged as benthic juveniles (In. = 240 (im). As a result of the adhesive nature of the encapsulated stages, the larval dispersal capa- bility of this species is estimated to be on the order of a few centimeters. THE ROLE OF THE VENTRAL PEDAL GLAND IN FORMATION OF AN EGG CAPSULE BY THE MURICID GASTROPOD EUPLEURA CAUDATA ETTERAE B. B. BAKER 195 1: AN INTEGRATED BEHAVIORAL, MORPHOLOGICAL, AND HISTOCHEMICAL STUDY GREGORY L. GRUBER College of Marine Studies University of Delaware Lewes, Delaware 19958 Several researchers described formation of egg capsules by females of a few neogastropods, but this process is still not well understood. Spawning behavior of females defined discrete times to sample egg capsules and spawning females before ventral pedal gland activity (VPGA), after peristaltic molding during VPGA, and after VPGA. Structure of these egg capsules and ventral pedal glands of females was examined with dissections, histology, polarizing microscopy, and histochemistry. Egg capsules before VPGA were ovoid, soft, and flexible. After peristaltic molding during VPGA, egg capsules were roughly shaped, loosely attached to a hard substratum, and still soft and flexible. Egg capsules after VPGA were completely shaped, firmly attached to a hard substratum, but now hardened and resilient. The apical plug, embryo chamber, and multilyatered fibrous wall of egg capsules before, during and after VPGA had similar 92 Abstracts, 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland morphologies. Histochemical composition of the wall of egg capsules before VPGA and after peristaltic molding during VPGA differed from that of the wall of the egg capsules after VPGA. The wall of whole egg capsules that were sam- pled before VPGA and exposed to filtered seawater for 5 days were soft, flexible, and showed no histochemical changes. These observations suggested that the ventral pedal gland molded an egg capsule into its final species-specific shape, firmly attached it to a hard substratum, chemically hardened the wall of the egg capsule, but did not secrete any layers of its wall. The ventral pedal gland has a columnar epithelium, two types of epithelial goblet cells, clusters of subepithelial gland cells, and a thin layer of circular and longitudinal muscle fibers between the epithelium and these gland cells. Each goblet cell type secreted different sulfated, acid mucosubstances that may act as lubricants during mold- ing of egg capsules. Subepithelial gland cells may secrete a noncarbohydrate, nonprotein substance that hardens the wall of the egg capsule. SOME RELATIONSHIPS AFFECTING GROWTH OF SEED OF THE HARD CLAM MERCENARIA MERCENAR1A IN RACEWAYS NANCY H. HADLEY 1 AND JOHN J. MANZI 2 Grice Marine Biological Laboratory 216 Ft. Johnson Rd. Charleston, South Carolina 29412 Marine Resources Research Institute Charleston, South Carolina 29412 Seed clams (y size = 3.9 mm) were maintained in race- ways for 6 months at densities corresponding to 740, 2220, 6660, and 19980 clams/m 2 . Each density was replicated eight times in the raceways and the highest and lowest densi- ties were replicated four times in subtidal field controls. Raceway clam populations were stocked in four different positions relative to water flow and in 19 different positions relative to total raceway biomass. Although nominal flow rate was constant, effective flow rate (water volume/clam volume/minute) was different for each replicate and decreased as clam biomass increased. Temperature and salinity were measured daily and inflow and outflow chlorophyll-a were monitored monthly from February to August 1981 to deter- mine growth and survival. Single classification ANOVA fol- lowed by SNK tests between means showed that growth was significantly reduced at the highest density in both the raceway and the field. The lowest density exhibited greater growth in the raceway than in the field, while the highest density showed no difference in growth between the two locations. In the raceway, growth rate was inversely propor- tional to distance from water inflow and to effective density (# clams/unit water). Although clams at the highest density consistently removed the greatest amount of chlorophyll-a, less chlorophyll was removed per clam as density increased. Growth was highly correlated with stripping rate (/ig chlorophyll-a/clam/day) and with effective water flow rate. These relationships are discussed and some implications for management of raceways in mariculture systems are made. MLXED-FUNCTION-OXYGENASE ENZYME SYSTEMS: PURPOSE AND POSSIBLE DELETERIOUS INTER- ACTIONS WITH ORGANIC POLLUTANTS IN THE BLUE CRAB ROBERT C. HALE Virginia Institute of Marine Science The College of William and Mary Gloucester Point, Virginia 23062 Mixed-function-oxygenases (MFO) are enzyme systems which have evolved in organisms to enable them to eliminate foreign compounds taken in from their environment. Often these compounds are toxic and lipophilic, possessing high accumulative potential (e.g., polynuclear aromatics, poly- chlorinated biphenyls, and chlorinated organic pesticides); therefore, they must be metabolized to biologically inactive, excretable forms. Occasionally, however, the resulting metabolites formed by the MFO system are more harmful than the parent compounds; some are potent carcinogens. Recent work has shown that the activity of the MFO sys- tem is greatest in mammals and decreases in fish, crustaceans, and mollusks, in that order. The enzyme system is also respon- sible for the synthesis and breakdown of certain steroid hormones. The molting hormone in crustaceans is believed to be a steroid compound. The activity of MFO in female blue crabs has been shown by others to be inversely related to the levels of crustecdysone, when examined over the course of a molt cycle. Elevated levels of aromatic hydro- carbons, caused by greater utilization of coal reserves and increased industrialization, are of concern to scientists. These and other pollutants have been found by workers to induce higher levels of MFO activity, and also to inhibit molting and limb regeneration in crabs. Levels of toxic organic compounds in the blue crab population of lower Chesapeake Bay are being determined using glass capillary gas chromatography and mass spectrometry. Differential abilities to metabolize aromatic compounds that may exist between molt and sex groups will be examined. ESTIMATES OF JUVENILE BLUE CRAB ABUNDANCE IN TEXAS BAYS PAUL C. HAMMERSCHMIDT Texas Parks and Wildlife Department Rt. 1, Box 368, Seadrift, Texas 77983 National Shellfisheries Association, Baltimore, Maryland Abstracts, 1982 Annual Meeting, June 14-17, 1982 93 Blue crab populations were monitored November 1 977— December 1981 by Texas Parks and Wildlife Department personnel using 18-m bag seines in the Galveston, Matagorda, San Antonio, Aransas, Corpus Christi, upper and lower Laguna Madre Bay systems. Seine samples and hydrological data were taken monthly at randomly selected stations in each of the sampled bay systems. Catch-per-unit-of-effort (CPUE), calculated as number of crabs/ha, as well as water temperature and salinity values are presented. These data were examined utilizing a 2-way ANOVA. Similarities in CPUE, water temperature, and salinity were examined between years and seasons within bay systems. THE SURF CLAM ALONG THE NEW JERSEY COAST: POPULATION SIZE. RECRUITMENT, GROWTH RATES HAROLD H. HASKIN, ERIC S. WAGNER AND MITCHELL L. TARNOWSKI Department of Oyster Culture N.J. Agricultural Experiment Station Rutgers the State University New Brunswick, New Jersey 08903 Over the last 10 years there has been regular and general settling of surf clam larvae along the New Jersey coast but. as indicated in earlier reports, mortality rates in early juve- niles are high and survival beyond the first summer is com- paratively rare. Exceptions to this will be discussed with emphasis on the 1976 year class which approximately doubled the standing stock in New Jersey waters. Since major portions of this year-class survived in areas where earlier year-classes were wiped out by anoxic waters in 1976, we have a unique opportunity to determine the effects of a variety of environmental conditions on growth rate. Results of some of these determinations will be presented, as will the most recent stock assessment. GROWTH PERFORMANCE OF CYTOCHALAZIN-INDUCED TRIPLOIDS OF AMERICAN OYSTERS AND SOFT-SHELL CLAMS HERBERT HIDU, STANDISH ALLEN AND JON STANLEY Department of Zoology University of Maine at Orono Orono, Maine 04469 We conducted extensive laboratory and field performance experiments in 1982 with 3-yr-old triploids of the American oyster Crassostrea virginica (Gmelin) and yearlings of the soft-shell clam Mya arenaria Linne. The Crassostrea triploids, which were created at meiosis I, grew significantly faster than the diploid controls, whereas those created later in the meiotic cycle exhibited no growth advantage over the dip- loids. The Mya triploids exhibited no growth advantage over diploid controls. Triploidy did not block gametogenesis in either species. Optimal methods are discussed for determin- ing the consequences of polyploidy in marine bivalves. PREDATION BY BLUE CRABS AND SPOT ON INFAUNAL COMMUNITIES IN CENTRAL CHESAPEAKE BAY ANSON H. HINES AND KATHRYN L. COM TO IS Chesapeake Bay Center. Smithsonian Institution. P.O. Box 28 Edgewater, Maryland 21037 The impacts of predation by blue crabs (Callinectes sapidus Rathbun)and spot(Leiostomusxanrhuri(sLacephde) on infaunal communities were compared for mud and sand sediments in the Rhode River, a typical subestuary of central Chesapeake Bay. The two species are the dominant benthic predators in the system, and their foraging activities from June to October correlated with the sharp seasonal decline in infaunal density and standing crop. Analysis of stomach contents showed that crabs preyed primarily on whole clams, whereas spot fed mainly on clam siphons and several species of polychaetes. Turnover rates of infaunal prey were estimated based on the density of predators taken in otter trawls, the weight of their stomach contents, and the weight of the standing crop of infauna. For total infauna, turnover rates were low (1— 7%/month) early in the season, when the standing crop was high; but turnover was high (30-60%/mo) in the top 5 cm of sediment late in the season, when the standing crop was low. For small clams, polychaetes, and amphipods in the top 5 cm of sediment, predation pressure by crabs and spot accounted for extremely high turnover rates (more than 100%/mo), whereas larger, deep-burrowing clams had turnover rates < 3%/mo. Experiments using pre- dator exclusion cages resulted in significantly higher densities of total infauna, clams, and some species of polychaetes within the cages than outside the cages. Survival of out- planted clams (Macoma balthica [Linne] ) was significantly higher in buckets with predator exclusion cages than in buckets without predator exclusion cages. Predation by blue crabs appears to have a major impact on small, surface- dwelling clams, whereas spot predation has a more general impact on clam siphons and a variety of invertebrates living in the surface sediment. Turnover of infauna in the surface sediment is very rapid. 94 Abstracts. 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland OCEANOGRAPHY OF THE SOUTHEASTERN BERING SEA AND RECRUITMENT PROCESSES IN TWO SPECIES OF TANNER CRAB LEWIS S. INCZE School of Fisheries WH-10 College of Ocean and Fishery Sciences University of Washington, Seattle, Washington 98195 Potential factors affecting the distribution and survival of the pelagic larvae of two species of tanner crabs, Chiono- ecetes bairdi Rathbun and C. opilio (Fabricius), that inhabit the wide continental shelf of the eastern Bering Sea were investigated as part of a large multi-institutional oceano- graphic program. The objective was to evaluate the relative importance of pelagic events in determining spatial patterns of recruitment to the benthos. The study emphasized the description of cause-and-effect relationships between physi- cal processes (mixing and transport) and biological (plank- tonic) conditions which affect feeding success and the ultimate survival and distribution of the larvae. Information on the timing of hatch-out, rates of growth and development, feeding physiology, and inter-annual differences in patterns of spatial distribution and relative abundance of the larvae are provided. How these data relate to regional oceanographic processes and their potential impact on population distribu- tion and age structure are stressed. SPECIES-SPECIFIC DIFFERENCES IN THE MEGALOPAL DISTRIBUTIONS RELATED TO WATER DENSITY PARAMETERS DAVID F. JOHNSON Department of Oceanography Old Dominion University Norfolk. Virginia 23508 The megalopae of 10 brachyuran crabs were sampled from July through September 1980 in the lower Chesapeake Bay and adjacent coastal waters. The megalopae are assigned to three apparent groups: retained estuarine. expelled estu- arine, and retained coastal recruitment types. The megalopae of estuarine species such as Hexapanopeiis angastifrons ( Benedict and Rathbun), Neopanope sayi (Smith ). Panopeus herbstii H. Milne-Edwards, and Pinnotheres ostreum Say are retained in estuarine epibenthic waters. The larvae of some estuarine species such as Callinectes sapidus Rathbun, Uca spp.. and Pinnixa spp. are expelled from the estuary, resul- ting in maximum megalopal concentrations on the shelf. Of the retained coastal species. Portunus spp. and Cancer irrora- tus Say are not abundant in the neuston of shelf waters, while Libinia spp. are most abundant in the epibenthos of near-shelf waters. The megalopae of 4 species show signifi- cantly different vertical distributions between stratified and homogenous water columns. Megalopae were not found to aggregate within pycnoclines. MECHANISM OF SHELL PENETRATION BY THE BURROWING BARNACLE TRYPETESA LAMP AS (HANCOCK), (CIRRIPEDIA: ACROTHORACICA): AN ULTRASTRUCTURAL STUDY TODD C. KAMENS College of Marine Studies University of Delaware Lewes. Delaware 19958 Trypetesa lampas is a soft -bodied, free-living cirriped that burrows in empty shells of gastropods inhabited by hermit crabs. Portions of this burrow are commonly lined with a limy, white material. Individuals of T. lampas were obtained from shells of Lunatia heros (Say) and Polinices duplicatus (Say) collected in the vicinity of Woods Hole, Massachusetts. Specimens of the mantle surface and burrow wall were examined with scanning electron microscopy to determine the mechanisms of shell removal and lining formation within the burrow by T. lampas and to correlate these activities with the microanatomy of the external mantle surface of the barnacle. Results confirm earlier hypotheses that bur- rowing by T. lampas is achieved through a combination of chemical and physical processes. Ultrastructural examination of fractures through the burrow reveal a gradual, shell- weakening process in which prismatic material within the surrounding gastropod shell is softened by preferential dis- solution of inter- and intra-crystalline matrix and subsequent solubilization of the bare calcareous prisms. Examination of thin sections through the mantle cuticle disclosed minute pore canals through which shell-dissolving secretions of the barnacle could be released. Dissolution of shell by T. lampas appears to be linked to the molt cycle, with most extensive stages of dissolution being observed in burrows of specimens that have just molted. Soft material remaining on the wall of the burrow after molting is removed with sharp spines covering the external surface of the barnacle's mantle. This material is subsequently used by T. lampas to thicken exist- ing parts of the lining and add new linings in areas that no longer fit snugly. TRACE METALS IN SHELLFISH AND GROWING AREA DESIGNATION JEFFREY KASSNER Department of Environmental Protection Town of Brookhaven Patchogiie, New York 11772 The level of coliform bacteria, as set forth by the National Shellfish Sanitation Program (NSSP), is the water quality National Shellfisheries Association, Baltimore, Maryland Abstracts, 1982 Annual Meeting, June 14-17, 1982 95 standard used to classify shellfish growing areas. It is the standard by which shellfish harvesting is regulated. Port Jefferson Harbor, NY, a moderately industrialized embay- ment of Long Island Sound, and Setauket Harbor, a more urbanized tributary basin of Port Jefferson Harbor, both have areas classified as certified (shellfishing permitted) and as uncertified (shellfish prohibited). Sediment analyses of the two harbors suggest that noncoliform pollutants, particu- larly trace metals, are present. Because of public health concerns, the hard clam Mercenaria mercenaria (Linne)was sampled for trace metals to determine how trace metal con- centrations in the shellfish tissues compared with the level of bacteriological pollution in the growing water and the NSSP classification. Hard clams were sampled from 5 loca- tions in each harbor and analyzed for copper, lead, zinc, and cadmium. From the metal and conform concentrations and their distributions in the two harbors, the following relation- ships were observed: in both harbors, hard clams from the station with the fewest coliform bacteria did not have the lowest metal concentrations; in Setauket, the variability in metal concentrations among the sampling locations was much less than in Port Jefferson; and in Port Jefferson, over- all metal concentrations were higher than in Setauket. The concentration of metals in the shellfish does not appear to be reliably related to the coliform level. BLUE CRAB PREDATION ON INFAUNAL BIVALVES: RELATION TO OPTIMAL FORAGING HYPOTHESES VICTOR S. KENNEDY, C. KING AND J. BLUNDON Horn Point Environmental Laboratories University of Maryland. Box 775 Cambridge, Maryland 21 613 Adult blue crabs {Callinectes sapidus Rathbun) were allowed to forage on equal numbers of 3 size classes of buried soft-shell clams (Mya arenaria Linnd); percentage of clams ingested increased with increasing clam size. This was also true in the case of juvenile blue crabs foraging on equal numbers of 5 size classes of buried specimens of Macoma balthica (Linne). When the largest size class of M. balthica was not available and equal numbers of the four remaining size classes could be preyed upon by juvenile crabs, the percentage of clams ingested increased with increasing clam size. This seems to indicate a pattern of optimal foraging by the crabs. Equal biomass of (a) two size classes of buried speimens of M. arenaria or (b) three size classes of buried specimens of M. balthica was then made availabe to adult or juvenile blue crabs, respectively. At the end of these experiments there was no statistically significant difference among size classes in percentage of clams ingested. This suggests that buried clams are preyed upon opportunis- tically by blue crabs. The results of the experiments using equal numbers of clams per class may have been influenced by the possibility that larger clams have a greater chance than smaller clams of being encountered by a sediment-probing crab because of their larger size. DEPARTMENT OF NATURAL RESOURCES AND UNIVERSITY OF MARYLAND FORM NEW COOPERATIVE SHELLFISH RESEARCH UNIT AT CRISFIELD GEORGE E. KRANTZ University of Maryland Center for Environmental and Estuarine Studies Box 775, Cambridge, Maryland 21613 The University of Maryland's Marine Products Laboratory located at Crisfield has become the site of a joint University/ Department of Natural Resources (DNR) program in shell- fish management effective 1 January 1982. The new joint research and management program will offer many advan- tages to the state's seafood industry by combining research and management functions in one unit as well as providing for the transfer of new hatchery technology through demon- strations of shellfish culture methods to watermen, seafood processors, and other interested groups. THREE INNOVATIVE TECHNIQUES THAT MADE MARYLAND OYSTER HATCHERIES COST-EFFECTIVE GEORGE E. KRANTZ, G. J. BAPTIST AND D. W. MERITT University of Maryland Center for Environmental and Estuarine Studies Cambridge, Maryland 21613 The combined use of 3 innovative techniques reduced the size of the physical plant of a Maryland oyster hatchery by 65% and reduced the labor by 55%. Tahitian Isochrysis, an unidentified algal strain that has an optimal growth tem- perature between 24 and 30°C, eliminated the need for a temperature controlled algae culture room in the hatchery. Algae cultures were grown at ambient room temperature and stored in a '"concentrated paste" after dewatering in a mechanical centrifuge. This technique permitted year round operation of a small algae culture laboratory rather than an intensive period of activity during the time of oyster larval culture (June through August). Oyster spat were collected directly from larval culture cones on a concrete-coated, wire device which also served as a growing substrate until the spat reached 2.5 to 3.5 cm. This growing device was transferred directly from the larval cone into the natural environment thereby eliminating the need for continuous flow of water in the hatchery and the labor involved with cleaning vast expanses of spat culture trays. Field trials of spat grown by 96 Abstracts, 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland these techniques will yield marketable oysters in the fall of 1983. EFFECT OF PROCESSING ON STEROL AND FATTY ACID COMPOSITION OF CRABMEAT JUDITH KRZYNOWEK National Marine Fisheries Service Northeast Fisheries Center Gloucester Laboratory Gloucester, Massachusetts 01 930 The use of water or brine or mechanical stress for crab- meat extraction and the freezing or further heating of crabmeat for canning purposes are processing techniques employed by the crabmeat industry. The impact of physical and chemical processing is discussed relative to the effect on the lipid portion of the meat (primarily on the sterol and fatty acid composition). Specific processing techniques to be discussed include: freezing, multiple freeze/thaw cycles, canning (both sterilized and pasteurized and the inclusion of bacteria in the product after canning), and three methods for meat extraction. ESTIMATION OF STANDING CROP OF MERCENARIA MERCENARIA (LINNE) IN THE JAMES RIVER, VIRGINIA, USING COMMERCIAL RECORDS ANDRE C. KVATERNIK AND WILLIAM D. DUPAUL Sea Grant Marine Advisory Services Virginia Institute of Marine Science College of William and Mary Gloucester Point, Virginia 23062 Commercial catch and effort records for boats using patent tongs to harvest hard clams from the James River were obtained for the years 1978-1981. Using Dickie's (1955) version of the Leslie method, catch-per-unit-effort of the sample fleet was regressed against accumulated catch to give estimates of the initial abundance. Estimates for 1978, 1979, 1980, and 1981 were 10,101 m 3 (280,605 bu), 14,625 m 3 (406,250 bu), 20,065 m 3 (557,250 bu), 1 2,397 m 3 (344.364 bu), and 14,297 m 3 (397,142 bu), respectively. The mean for the period 1978-1981, 14,297 m 3 (397,142 bu), was 30% below that estimated by Haven et al. (1981). Com- mercial catch records can be used in this application but limitations in the data must be understood. Abundance esti- mates from this method should be supplemented with addi- tional designed sampling strategies to give better accuracy. EFFECTS OF LIGHT AND GRAVITY UPON THE MOTILE BEHAVIOR OF TROCHOPHORE LARVAE OF MERCENARIA MERCENARIA (LINNE) MARK D. LESLIE AND ROBERT S. WILSON Department of Biology Southeastern Massachusetts University North Dartmouth, Massachusetts 02747 Adults of Mercenaria mercenaria were spawned in the laboratory and the fertilized eggs were reared to the trocho- phore stage. Responses of the larvae to light and gravity were observed. Distributions were determined under 5 experi- mental conditions: horizontal chamber in darkness, horizon- tal chamber with two different light intensities (2.5 and 15 W/M 2 ) shining from one end, vertical chamber in darkness, vertical chamber with light incident from above (2.5 W/M 2 ) and a vertical chamber with light incident from below (2.5 W/M 2 ). The results revealed a random distribution of the larvae in horizontal dark and horizontal light experiments, a substantial surface aggregation in the vertical dark chamber, and a decrease in surface accumulation with the light source shining from above and below the vertical chamber. Indivi- dual swimming paths of the larvae were analyzed using a computer-video system (viz., the Bug-system). The larvae were viewed in both the presence and absence of light in a vertical plane. Illumination from below caused a significant drop in vertical velocity and swimming speed and a small decline in the rate of change of direction. Phototaxis was not observed. Photostimulation caused the trochophores to exhibit a negative orthokinesis with a weakening in their negative geotactic behavior. GROWTH OF JUVENILES OF ARCTICA ISLANDICA (LINNE) IN EXPERIMENTAL CONTAINERS R. A. LUTZ 1 , J. G. GOODSELL 1 , M.CASTAGNA 2 AND A.P.STICKNEY 3 Dept. of Oyster Culture, New Jersey Agricultural Experiment Station, Cook College. Rutgers University New Brunswick, New Jersey 08903 Virginia Institute of Marine Science Wachapreague, Virginia 23480 Dept. of Marine Resources West Boothbay Harbor, Maine 04575 Laboratory -reared ocean quahogs {Arctica islandica) (n = 119) ranging in shell length (maximum antero-posterior dimension) from 1.8 to 4.3 mm (x = 2.5 ± 0.4 mm, SD) were placed during June in experimental mesh containers suspended from fixed and floating structures in marine waters off Boothbay Harbor, Maine. Shell length measure- ments were recorded at monthly intervals until the follow- ing March. Water temperatures at the locations of the con- tainers ranged from a high of 15.5°C during August to a low of 1 .0°C during February. Mean growth rages recorded during the warmer months from June through September National Shellfisheries Association, Baltimore, Maryland Abstracts, 1982 Annual Meeting, June 14-17, 1982 97 ranged from 2.0 to 2.4 mm/month. Reduced, yet measur- able, amounts of shell (x = 0.3 - 0.5 mm/month) were deposited during even the coldest winter months (January and February). Mortality during the study period was < 1%. By early March, the shell lengths of specimens (n = 117) ranged from 3.9 to 21.3 mm (x = 14.0 ± 2.8 mm, SD). Recorded growth rates were considerably faster than those heretofore reported for Arctica islandica and suggest that juveniles of this species have a potential for relatively rapid growth in certain environments. SIZE AND VOLUME RELATIONSHIPS IN JUVENILES OF MERCENARIA MERCENARIA (LINNE): A REVISION OF BELDINGS TABLES JOHN J. MANZI ' , F. S. STEVENS 1 , Y. M. BOBO 1 , V. G. BURRELL, JR. 1 AND NANCY H. HADLEY 2 Marine Resources Research Institute Charleston, South Carolina 29412 College of Charleston, Charleston, South Carolina 29402 Size and volume relationships in juveniles of the hard clam Mercenaria mercenaria were determined in commer- cial nursery populations over a 1-y period. Morphometric determinations included size (longest anterior-posterior dimension), displacement volume, and packed volume (wet). These data were used to establish empirical relationships between seed size and volume (displacement and wet packed) which are reported here as a revision of Belding's Tables. The empirical relationships, thus established, were iteratively employed in the construction of a model to pre- dict seed clam volume. The model assumed that the volume of a hard clam is proportional to the cube of a linear dimen- sion. The iterations allowed model refinements which pro- duced positive correlations between predicted and observed data. We summarize collected data on size/volume relation- ships in seed clams and present a model, based on truncated spheres, which attempts to relate size and volume character- istics in seed clams within the size range of 1 .0 to 1 5 .0 mm. A DESCRIPTIVE MODEL FOR THE CONSERVATION OF BLUE CRAB LARVAE IN THE VICINITY OF CHESAPEAKE BAY J. R. McCONAUGHA, D. R. JOHNSON AND A. J. PROVENZANO Department of Oceanography Old Dominion University Norfolk, Virginia 23508 An extensive series of plankton samples taken from the waters around Chesapeake Bay indicates that all larval stages of the blue crab Callineetes sapidus Rathbun are concentra- ted in the upper layers of the water column with maximum numbers in the upper 1 m. This distribution insures that stage I larvae hatched near the bay mouth are entrained in the outwardly flowing surface water. The general longshore current in the Mid-Atlantic Bight is southward which would tend to transport larvae towards Cape Hatteras. This would result in their being lost to the system. Recent evidence sug- gests that during the summer months, when peak spawning occurs, there is a wind generated counter-current on the inner shelf. The width and speed of this corridor is related to wind direction and velocity. Larvae entrained in this counter-current are returned to the vicinity of Chesapeake Bay and contribute to recruitment. The horizontal distri- bution of blue crab larvae from field samples is consistent with this hypothesis. A TEST OF A DART TAG FOR JUVENILE BLUE CRABS, CALLINECTES SAPIDUS RATHBUN R. E. MILLER University of Maryland Horn Point Environmental Laboratories Cambridge, Maryland 21613 A small dart tag was applied to the posterior junction between the ventral and dorsal parts of the cephalothorax of 80 juvenile blue crabs to test for success of molting and tag retention during the molting process. Sixty-one percent of tagged crabs which began ecdysis were successful in molting and retained the tag; however, overall mortality rate for tagged crabs was twice that of the untagged control group. METHODS FOR FIELD EXPERIMENTS WITH BAITED TRAPS ROBERT J. MILLER Fisheries Research Branch, Canada Department of Fisheries and Oceans Halifax, Nova Scotia, Canada, B3J2S7 The number of uncontrolled variables and the number of potentially testable variables in the field environment can be distracting and intimidating to the field technicians. This environmental complexity requires greater mental discipline to conduct good experiments in the field than is required in the tidier laboratory environment. Problems frequently encountered in conducting experiments on design and fishing strategy of baited traps are as follows. Testing of hypotheses using fishermen's logbook data commonly gives biased results and has poor resolution because fishing variables are neither controlled nor random and data are often incorrect. Because most fishermen lack appreciation for correct experi- mental procedures, even dictating an experimental design will not assure a properly executed experiment. Preliminary 98 Abstracts, 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland trapping should be carried out to locate an experimental area with uniform catch rates, to determine the optimum sample size, and to solve logistical problems in conducting the experiment. Experimental treatments should be randomized in space and time to avoid bias. An investigator rarely knows enough about the uncontrolled variables in the field to jus- tify a systematic allocation of treatments in space and time. Variance is controlled by careful attention to details of bait quantity and quality, by keeping traps in good repair, by standardizing soak time, and by standardizing time of day of setting traps. A FIRST ESTIMATE OF INDIRECT FISHING MORTALITY IN THE ICELAND SCALLOP CHLAMYS ISLANDICA (MU LLER) K. S. NAIDU Research and Resource Services Department of Fisheries and Oceans P.O. Box 5667. St. John's. Newfoundland, Canada A1C 5X1 Natural mortality in Iceland scallops (Chlamys islandica), computed from the ratio of cluckers to live animals, as might be expected, increased with age. Higher than average rates were found for the fully recruited ages (> 8 y) on heavily exploited grounds than in scallop beds subject to light or initial exploitation. The difference in mortality rates between near-virgin and fully exploited areas is ascribed to indirect fishing mortality associated with repetitive towing on productive grounds. THE ANNUAL GLYCOGEN CYCLE IN THE SOFT-SHELL CLAM MYA ARENARIA LINNE FROM MAINE CARTER R. NEWELL Program in Oceanography . University of Maine at Orono, Ira C. Darling Center Walpole, Maine 045 73 A field population of adults of Myaarenaria was sampled at approximately semi-monthly intervals for one year to determine glycogen levels in the meats. Highest levels occurred in late spring and early summer. Post-spawning late summer and fall levels were intermediate, and lowest levels occurred in the winter. Glycogen levels in juveniles and adults of M. arenaria were compared and the relationships between glycogen levels and gametogenesis, food availability, and temperature are discussed. THE EFFECTS OF SEDIMENT TYPE ON GROWTH RATE AND SHELL ALLOMETRY IN THE SOFT-SHELL CLAM MYA ARENARIA LINNE CARTER R. NEWELL Program in Oceanography, University of Maine at Orono, Ira C. Darling Center Walpole. Maine 04573 Hatchery-reared juveniles of Mya arenaria were grown for 1 1 weeks in replicated gravel, sand, mud, and pearl net treatments under flow-through seawater conditions in Maine. Analyses of variance showed significant differences between sediment treatments for final shell length, dry meat weight, chondrophore growth increment, and percent shell weight. Growth of juveniles of M. arenaria was more rapid in fine sediments than in coarse sediments or nets. The slopes of shell length vs. shell height and shell length vs. shell depth regressions also varied significantly between sediment treat- ments. Slower growing clams from nets and gravel were more globose than clams from sand or mud treatments. Clams reared in sand were longer and narrower than those reared in mud. Differences in growth rates and shell form were attributed primarily to the physical properties of the sub- stratum. PREFERENTIAL INGESTION OF ORGANIC MATERIAL FROM THE CONSUMED RATION BY THE OYSTER CRASSOSTREA VIRGINICA (GMELIN) ROGER I. E. NEWELL AND STEPHEN JORDAN Horn Point Environmental Laboratories University of Maryland. P.O. Box 775 Cambridge. Maryland 21613 Considerable debate exists in the literature as to whether suspension-feeding bivalve molluscs can preferentially ingest the organic component of the seston. Most of those discus- sions were based on circumstantial evidence rather than reli- able, quantitative measurements of the chemical composition of the oyster's food or biodeposits. This paper gives details of steady state measurements of the carbon, nitrogen, and energy content of the seston being fed to the oyster Cras- sostrea virginica and of the faeces and pseudofaeces being voided. The results indicate that, over the tested range of food concentrations (from 4—20 mg/1), the amount of energy (expressed as Joules/mg of dry weight of material) voided in the pseudofaeces by C. virginica can be reduced by 60% compared to the concentration in the food. Similar results were obtained from the carbon and nitrogen analysis. These data strongly indicate that C. virginica has the capa- bility of selecting certain particles from the total seston filtered from suspension, with the result that more food particles are rejected in the pseudofaeces. FACTORS LIMITING ABUNDANCE OF CALLINECTES SPP. ELLIOTT A. NORSE 1 AND VIRGINIA FOX-NORSE 2 Center for Environmental Education 624 9th Street NW, Washington. D.C. 20001 National Shellfisheries Association, Baltimore, Maryland Abstracts, 1982 Annual Meeting, June 14-17, 1982 99 United States Environmental Protection Agency, Office of Federal Activities, A- 104 Washington, DC 20460 The abundance of organisms varies in space and time be- cause the factors that limit abundance vary spatially and temporally. Understanding limiting factors and the ways organisms respond to them can lead to improved blue crab catches. Blue crab populations can be limited directly by (1) insufficient recruitment from the plankton; (2) inade- quate water quality, due either to natural or man-made causes; (3) insufficient resources, including food and cover;(4) inter- ference competition, especially from other crabs; and (5) removal by parasites, natural predators, and crabbers. Each of these classes of limiting factors can be tested experimentally. The results of these studies can suggest more effective ways to improve catches by managing not only the populations of blue crabs, but also the ecosystems to which they belong. TOTAL WIDTH -WEIGHT RELATIONSHIPS OF THE BLUE CRAB CALLINECTES SAPIDUS RATHBUN FROM THE ASHLEY RIVER, SOUTH CAROLINA EUGENE J. OLMI, III AND JAMES M. BISHOP Marine Resources Research Institute South Carolina Wildlife and Marine Resources Department Charleston, South Carolina 29412 Equations expressing total width-weight relationships of blue crabs (Callinectes sapidus) were calculated in relation to sex, sex by maturity, sex by molt sign, and sex by cara- pace form. All calculations were restricted to intermolt (Stage C) crabs except when molt sign was considered, and comparisons were restricted to crabs of similar size. Sex, maturity, molt sign, and carapace form significantly affected width-weight relationships. Overall, males were heavier than females of equal width. Mature males exhibited a greater mean weight than immature males, but mature females weighed less than immature females of similar size. Crabs with short lateral spines were heavier than those of the same sex with long spines. Intermolt and premolt (Stage D) males and females were heavier than recently molted (Stages A and B) males and females, respectively. Premolt females were heavier than intermolt females; a similar difference was not observed for males. Ashley River crabs were generally heavier than crabs from Florida, Texas, and Virginia. These differences may not be real, however, because many variables affect width-weight relationships of blue crabs and only sex differences were reported. Geographical variation is known to exist in crab populations, but only well defined compari- sons between populations should be considered. SIGNIFICANCE OF THE NEUSTON LAYER IN THE DISPERSAL OF LARVAE OF THE BLUE CRAB CALLINECTES SAPIDUS RATHBUN A.J.PROVENZANO,J.M. McCONAUGHA, AND D.F. JOHNSON Department of Oceanography Old Dominion University Norfolk, Virginia 23508 The distribution of larval blue crabs in the water column affects their transport out of Chesapeake Bay and during the larval period. The patterns of vertical distribution are not similar to those of other crab species in the region. First stage larvae are found predominantly in the neuston layer during the hatching season in the mouth of Chesapeake Bay and are carried seaward by the ebb tides. Later develop- mental stages, including the megalopae, are also found pre- dominantly in the neuston or upper 1 m, with very few being caught in intermediate layers or near bottom. Up to 99% of stage I larvae in the bay mouth and more than 70% of all Callinectes larvae of all stages even offshore were found above lm. No evidence of vertical migration of any stage was obtained. The effect of this distribution is to make larval blue crabs very susceptable to surface effects and wind driven currents during larval development and immediately after metamorphosis to the megalops. Studies which do not include the neuston layer may overlook a major fraction of the total population of blue crab larvae. Most previous studies of larval blue crab occurrence and distribution did not include sampling of the neuston and consequently some conclusions based on those studies were erroneous. GROWTH ENHANCEMENT OF MY A ARENARIA LINNE AND MERCENARIA MERCENARIA (LINNE) BY MARINE MACROALGAE HAUKE K. RASK Ira C. Darling Center, University of Maine Walpole, Maine 04573 Juveniles of My a and Mercenaria were Alizarin-stained and cultured for 12 weeks in flow-through tanks containing one of three different species of macroalgae. Clams grown with Ascophyllum nodosum Linnaeus and Laminaria longi- cruris De la Pylaic were significantly larger with respect to shell dimensions than controls and those grown with Ulva lactuca Linnaeus. Maximum enhancement was observed with Ascophyllum in all czses;Mya grown with Ascophyllum grew 4.54 times more than controls, while Laminaria treated My a showed 2.14 times more growth. A similar but less pronounced trend was seen for Mercenaria. Treatments with Ascophyllum and Laminaria were 12.6% and 9.6% larger than controls, respectively. Growth with Ulva was less 100 Abstracts, 1982 Annual Meeting, June 14-17. 1982 National Shellfisheries Association, Baltimore, Maryland than control treatments but differences were not significant. The mechanisms of growth enhancement from different macroalgae and their importance in aquaculture are discussed. ECONOMIC CONSIDERATIONS IN MANAGEMENT OF THE COMMERCIAL BLUE CRAB FISHERY RAYMOND J. RHODES Division of Marine Resources South Carolina Wildlife and Marine Resources Department Charleston, South Carolina 29412 From an economic prospective, the major consideration of common wealth fishery management is to maximize net benefits derived from the resource. In the case of commer- cial fisheries, net benefits accruing to society should include harvest revenues minus private costs (e.g., public adminis- tration and enforcement). In order to accomplish manage- ment objectives, private costs and public transactions costs need to be minimized. A simple review of various blue crab regulations germane to these economic concepts was performed. CHEMICAL ECOLOGY OF OYSTER DRILLS M. CARRIKER 1 , L. WILLIAMS\ AND L. WOOD 3 University of Delaware, College of Marine Studies, 700 Pilot town Road Lewes, Delaware 1 9958 Department of Botany. University of Washington, Seattle, Washington 98195 3 101 Whitcomb Circle Lafayette, Louisiana 70503 Oyster Drills are predatory snails that eat a wide spectrum of shelled prey such as oysters, mussels, and barnacles. Drills have a well documented ability to locate intact prey from a distance by following chemical trails. We have looked in detail at the molecular basis of prey location by drills. Newly hatched drills can locate only barnacles from a distance. This ability is apparently genetic as maternal diet and prey odor environment do not enable the young to locate other prey such as oysters or mussels. Once a newly hatched drill has fed for some time on oysters, however, it develops the ability to locate oysters. The molecules used by drills to locate either barnacles or oysters are similar peptides. Animals that can locate only barnacles, however, cannot use even high concentrations of oyster attractant to locate oysters. Drills cannot locate mussels from a distance even if they have fed upon mussels. In fact, mussels produce a molecule that suppresses the ability of drills to locate prey from a distance. This molecule is much different than the attractant molecules. It has a molecular weight less than 500 Daltons and does not appear to be a peptide. As a result of the differences between attractants and suppressants and the responses of inexperienced versus experienced drills we can measure levels of attractants and suppressants in natural waters. We hope that an understanding of the molecules and mechanisms involved in prey location can provide a means of drill control in the near future. DOCUMENTATION OF ANNUAL GROWTH LINES IN THE OCEAN QUAHOG.4/?77C4 ISLANDICA LINNE J. VV. ROPES 1 , D. S. JONES 2 , S.A. MURAWSKl',F.M.SERCHUK 1 AND A. JEARLD, JR. 1 U.S. Dept. Commerce. NMFS Woods Hole, Maine 02543 Dept. Geoi, Univ. Florida Gainesville, Florida 32611 About 42,000 ocean quahogs (Artica islandica) were marked for release at a deep (53-m) oceanic site off Long Island, NY, in 1978. Shells of live specimens recovered 1 and 2 years later have been radially sectioned, polished, and etched for preparation of acetate peels and examination by optical microscopy or microprojection; selected specimens were similarly prepared for examination by scanning elec- tron microscopy. Specific growth-line and growth-incre- ment microstructures are described and photo-illustrated. An annual periodicity of microstructure is documented. The observations form a basis for resource assessment ageing studies of the commercially important species. THE CHESAPEAKE BAY BLUE CRAB FISHERY: HISTORICAL TRENDS AND EMERGING ISSUES LEONARD A. SHABMAN AND TAMARA VANCE Department of Agricultural Economics Virginia Polytechnic Institute Blacksburg, Virginia 24061 Twenty-year trends in the Chesapeake Bay (Virginia and Maryland) blue crab fishery were measured with National Marine Fisheries Service data. Despite a recent downward trend in landings, Virginia continues to have the largest annual harvest of blue crabs in the U.S. While the total number of crabbers in Virginia has been stable, there have been decreasing numbers of users of trotlines and dredges and increases in users of pots. The mean harvest per crabber has fluctuated with a perceptible downward trend; but con- sistently rising ex-vessel prices have maintained rising gross income in the fishery. Maryland landings, like Virginia's, are a significant portion of U.S. harvest and have shown a slight downward trend. The number of Maryland crabbers has more than tripled over the observed period— predominantly from additions to the recorded number of part-time laborers. National Shellfisheries Association, Baltimore, Maryland Abstracts, 1982 Annual Meeting, June 14-17, 1982 101 There has not been a decline in the use of trotlines in Mary- land, as in Virginia, because of restrictions on the use of pots in certain Maryland waters. In Maryland, the mean harvest per crabber has fallen over the period. Consistently rising ex-vessel prices have resulted in an upward trend in mean labor income for Maryland pot crabbers, but there has been a drop in mean labor income for Maryland trotline crabbers. Based upon this review, three factors affecting the future growth of the industry are discussed: (1) state laws to protect brood stocks differ and confuse stock management efforts; (2) current public management programs(primarily licensing) may not be promoting maximum economic yield from the fishery ;and (3) economic uncertainties restrain development of processing facilities and, in turn, discourage harvest. MANAGEMENT OF THE BLUE CRAB FISHERIES IN NORTH CAROLINA: A CASE HISTORY TERRY M. SHOLAR North Carolina Division of Marine Fisheries, Washington, North Carolina 27889 Blue crabs support one of North Carolina's most impor- tant fisheries. The recent expansion of the crab fisheries has resulted in numerous management problems concerning resource allocations and gear conflicts. Regulatory authority for management in North Carolina has been delegated by the General Assembly to a 15-member commission which enacts regulations based on staff recommendations and input from the industry and general public. A key manage- ment tool is the proclamation authority which has been delegated by the Commission to the Secretary of the De- partment of Natural Resources and Community Develop- ment to respond rapidly to management needs. Proclama- tions can be issued to invoke a management action with a minimum of 48 h of public notice. This is generally done to open or close areas to a particular fishing method or to set seasons. This ability allows effective response to rapidly changing situations within the fisheries and the stocks. An example of North Carolina's management system involving the blue crab fisheries concerns resource allocation in certain tributaries of Pamlico Sound. Potting and trawling are in- compatible gears competing for space and resource. Each is controlled by proclamation. The decision to allow a certain fishery to occur is based on biological, economic, and social implications, with multiple-use resource management and protection being major factors in the decision. Tagging studies are being used to evaluate management strategies and their effect on maximizing crab harvest, and to deter- mine short-term migratory habits. Numerous other manage- ment issues affecting blue crabs and their fisheries such as minimum size limit, mandatory cull rings in pots, spawning sanctuaries, and nursery area protection are addressed. THE TEXAS OYSTER STUDY. I. RELATIONSHIPS BETWEEN AVAILABLE FOOD, OYSTER COMPOSITION, CONDITION, AND REPRODUCTIVE STATE THOMAS M. SONIAT 1 AND SAMMY M. RAY 2 Department of Biological Sciences University of New Orleans, Lakefront New Orleans, Louisiana 70148 Department of Marine Biology Texas A &M University at Galveston P.O. Box 1675, Galveston, Texas 77553 We examined the relationships between what is available for the oyster to eat, the oyster's proximate composition. its condition, and its reproductive state. Changes in the proximate composition of oysters were associated with changes in the annual cycle of fattening, storage, and repro- duction. The fattening phase was characterized by high dry- weight condition indices and elevated carbohydrate (glyco- gen) concentrations. A "storage cycle," the transition from stored glycogen to the lipid reserves in developing eggs, was evident in Crassostrea virginica (Gmelin). The gonadal index and percent lipid composition of the oyster were positively correlated. Spawned oysters had low lipid and carbohydrate concentrations, low condition and gonadal indices as well as high concentrations of water and protein. Available food for the oyster was measured as a food index. The food index was defined as the percentage food (food = lipid + carbohydrate + protein) in the total seston. The food index was higher in the spring and summer and was correlated with the gonadal index of oysters. Apparently, the amount of food was greatest at the time of greatest energy demand; that is, during gametogenesis. THE TEXAS OYSTER STUDY. II. MODELS OF OYSTER NUTRITION IN THE NATURAL ENFIRONMENT THOMAS M. SONIAT 1 . SAMMY M. RAY 2 AND REZENAT M. DARNELL 3 Department of Biological Sciences University of New Orleans. Lakefront New Orleans, Louisiana 70148 Department of Marine Biology Texas A &M University at Galveston P.O. Box 1675, Galveston, Texas 77553 Department of Oceanography Texas A&M University College Station, Texas 77843 Two FORTRAN models were developed to integrate in- formation about measured food levels (i.e., the food index) with the presumed needs of the oyster. One model assumed no selective ingestion on the part of the oyster. Another model assumed that the oyster could selectively ingest 102 Abstracts. 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland organic material. Although the results of the models are in fair agreement with published literature, this agreement could simply be fortuitous. The correspondence between the models we developed and other works, however, suggests the possibility that the food index is a useful measure of available food, that the simplifications made in the models are reasonable ones, and that enough particulate food was present to sustain oysters in the area studied. A CYTOGENETIC METHOD AS A TOOL FOR ASSESSING THE CONDITION OF SHELLFISH LARVAE S. STILES AND J. CHOROMANSKI National Marine Fisheries Service Northeast Fisheries Center Milford Laboratory Milford. Connecticut 06460-6499 As a means of assessing their condition at the cellular level, cultured oyster larvae were examined cytologically by employing a relatively simple squash technique. Chromo- some groups, and normal and abnormal cells and nuclei were evident. Bacteria also were discernible with this method. These observations were an indication of general health of the larvae in culture and provided some information regard- ing subsequent development and survival. In addition to being able to observe pathological states of the cells and bacterial infections, one could use the procedure to deter- mine the numbers of cells in mitosis as an indicator of growth rate. Larvae, potentially, could be pre-treated with colchicine to arrest cells in mitosis for counting the chromo- somes to obtain karyotypes as an aid in plankton identifica- tion. Cytological analyses of the larvae could have many uses in toxicological studies, including bioassays, as well as in hatchery rearing and breeding. ISOLATION AND PARTIAL CHARACTERIZATION OF A MALATE DEHYDROGENASE FROM CRASSOSTREA VIRGINICA (GMELIN) MARY L. SWIFT AND S. LAKSHMANAN Chemistry Department, University of Maryland, College Park, Maryland 20742 Final details of glucose metabolism in marine bivalve molluscs are yet to be elucidated. Malate dehydrogenase (E.C.I .1 .1.37) has been implicated in the catabolic path- way leading to the formation of succinate, a major end pro- duct of anaerobic metabolism in bivalve molluscs. Further clarification of this metabolic scheme may be gained by an examination of the properties of malate dehydrogenase (MDH). Homogenates of tissues of Crassostrea virginica contain at least 3 MDH isoenzymes. One of these was iso- lated from acetone powders of mantle and gill tissue by ammonium sulfate fractionation, gel permeation, and ion- exchange chromatography. Some properties of this prepara- tion were determined. The Michaelis-Menten constants were : K m(OAA) = 1.18xlO- 4 M; K m(NADH ) = 4.86xl(r 5 M; Km(mal) = 1.35xlO*M; K m(NAD) = 1.30xlO" 4 M. The following were not substrates: NADP + , a-ketobutyrate, a- ketovalerate, a -ketoglutarate.D-malate, pyruvate, succinate, oxomalonate. Tartronate, D-, L-, and mesotartrate were not substrates and were found to be competitive inhibitors of malate oxidation. The pH optima were: 7.6 for NADH oxi- dation and 9.5 for NAD + reduction. MDH was inhibited by p-chloromercuribenzoate and N-ethylmaleimide. Listed in decreasing order of effectiveness, Cd ++ , Zn ++ , Cu ++ , Co ++ and Ni ++ inhibited NADH oxidation by MDH. COMPARISON OF THE GROWTH OF CRASSOSTREA VIRGINICA (GMELIN) AT FIVE ALGAL RATION LEVELS WITH SPECIFIC REFERENCE TO PREDICTIVE FEEDING EQUATIONS EDWARD R. URBAN AND G. D. PRUDER College of Marine Studies University of Delaware Lewes, Delaware 19958 Mixtures of the algae Thalassiosira pseudonana Hasle et Heimdal (clone 3H) and Isochrysis aff. galbana Parke (T- ISO) were fed at each of 5 levels to juveniles of Crassostrea virginica. The oysters were grown for 3 wk at 25 °C and a salinity of 30 ppt. The relationship between algal ration level and oyster growth is presented. The results are discussed with specific reference to several feeding equations either published or in use or both. Recommended algal ration levels are compared for their relative effectiveness. We show that neither cell number, nor volume, nor weight constitute an acceptable parameter for comparing algal species or bivalve species. We recommend that feeding studies be carried out for any new combinations of algae until the nutritive value of the algal species can be correlated with physical characteristics and environmental conditions. The prudent use of predictive algal ration equations as management tools is discussed. A BLUE CRAB MANAGEMENT PLAN: OBJECTIVES AND RESPONSIBILITIES W1LLARD A. VAN ENGEL Virginia Institute of Marine Science The College of William and Mary Gloucester Point, Virginia 23062 The blue crab, Callinectes sapidus Rathbun. of the Atlantic and Gulf coasts supports one of the major marine fisheries of the United States. Regulatory authority National Shellfisheries Association, Baltimore, Maryland Abstracts. 1982 Annual Meeting, June 14-17, 1982 103 concerning licensing, size and sex limits, quotas, seasons, gear restrictions, and other controls over harvesting within its territorial waters rests with each state, retained by the respective state legislatures, but may be delegated to a com- mission. Regulations should be based on the best biological, economic, sociological, and environmental knowledge and provide for optimum yield from the resource. The blue crab industry's problems are not limited to regulation of the har- vest. They also include the need for federal and state assis- tance in processing, marketing, and research; conservation of the blue crab habitat; and an adequate data base. A com- prehensive blue crab management program should protect the resource, encourage and assist fishing with a minimum of regulations, and promote utilization of the product. THE BEHAVIORAL BASIS OF LARVAL DISPERSAL AND RECRUITMENT IN THE BLUE CRAB CALLINECTES SAPIDUS RATHBUN W. F. VAN HEUKELEM AND S. D. SULKIN Horn Point Environmental Laboratories University of Maryland Cambridge. Maryland 21613 Laboratory experiments have demonstrated that Stage I blue crab zoeae exhibit a number of behavioral traits which should result in distribution high in the water column. These traits include: negative geotaxis which is unaffected by salin- ity changes of 5 ppt; high barokiness at hydrostatic pressures exceeding 1 atmosphere; increased swimming rate with increased salinity; positive phototaxis at light intensities of > 1CT 3 W/m 2 ; maintenance of swimming speed with decreasing temperature ;and the ability to traverse haloclines of 10 ppt as well as sharp thermoclines. Because it is known that female blue crabs migrate to the mouth of Chesapeake Bay to spawn, these behavioral traits should result in massive export of virtually all Stage I zoeae in surface waters. Field evidence by other workers supports this contention. Mega- lopae possess behavioral traits that differ from late zoeal stages, chiefly, a highly sensitive pressure response, faster swimming speeds, negative geotaxis, and possibly locomotor rhythms that may enhance their transport back into estu- aries. Since larval development occurs on the continental shelf, recruitment success of megalopae back into estuaries is likely to be highly dependent on offshore climatological events that determine coastal circulation patterns during the summer and fall. REPRODUCTIVE PERIODICITY OF BUSYCON CARICA (GMELIN) IN WATERS OFF SOUTH CAROLINA DEBRA A. WEINHEIMER Department of Biology College of Charleston Charleston, South Carolina 29424 A total of 1237 knobbed whelks (Busy con carica) were collected over a 13-month period near Charleston Harbor, SC. Gonad maturation stages were determined by gonad color and histological sectioning. Monthly fluctuations in gonad weight, penis or nidamental gland weight, gonadal index, and reproductive index were also examined, of the six reproductive characteristics used in this study, gonadal index values were considered to be the best indicators of periodicity. The highest gonadal index values for males occurred in September, October, November 1979, and in March 1980. The highest values for females occurred from September 1979 through May 1980. Sex ratios fluctuated monthly. The number of females was significantly higher than the number of males from July 1979 through January 1980. This situation was reversed in April 1980 when the number of males was significantly higher than the number of females. Sex ratios also fluctuated when examined using shell-length classes. The smallest individuals in the monthly samples were females (60-64 mm). All individuals with shell- length values > 159 mm were female. Sex ratio relationships to reproductive periodicity are discussed. DISTRIBUTION, SIZE, AND SEX COMPOSITION OF THREE SPECIES OF CALLINECTES IN THE COASTAL HABITAT OF THE SOUTH ATLANTIC BIGHT ELIZABETH L. WENNER AND CHARLES A. WENNER Marine Resources Research Institute Charleston, South Carolina 29412 Collections by shrimp trawl during summer of 1980 at depth of 4.5 — 18 m between Cape Fear, NC, and Cape Canaveral, FL, showed that biomass of Callinectes sapidus Rathbun was greater than that of the other 72 decapod species collected. Callinectes similis Williams ranked fourth in abundance among the other decapod species collected, but C sapidus and C. ornatus Ordway were not as numerous. Catches of Callinectes spp. were greatest in the nearshore depth zone of 4.5— 8.5 m. Density and biomass totaled for all strata were greatest for C similis and C. ornatus off Flor- ida, and for C. sapidus off South Carolina. Few mature or ovigerous females of C. similis andC. ornatus were collected, whereas most females of C. sapidus were either mature or ovigerous. Significantly more females than males of C. sapi- dus were collected. The ratio of M:F for other Callinectes spp. varied with location. Sizes of crabs were not correlated with depth or distance from shore. 104 Abstracts, 1982 Annual Meeting, June 14-17, 1982 National Shellfisheries Association, Baltimore, Maryland NURSERY CULTURE OF THE BAY SCALLOP ARGOPECTEN IRRADIANS IRRADIANS (LAMARCK) IN SUSPENDED MESH ENCLOSURES JAMES C. WIDMAN, EDWIN W. RHODES AND P. A. BOYD Milford Laboratory. Northeast Fisheries Center, National Marine Fisheries Service 212 Rogers Avenue, Milford, Connecticut 06460-6499 Suspended mesh enclosures with bottom areas of 0.1 m 2 were used to grow hatchery-reared bay scallops in Long Island Sound in 1980 and 1981. The enclosures were con- structed of 3- or 7-mm polyethylene mesh and were anchored at a depth of 8 m and buoyed with styrofoam floats. Scallops as small as 4.6 mm were successfully grown to a size > 20 mm in the units. Acclimated scallops deployed in the spring of 1981 at temperatures as low as 5°C survived and subsequently grew normally as water temperatures increased. Scallop densities between 250 and 1 5,000/m 2 were tested in the enclosures, and although final shell height was inversely related to density, substantial growth occurred at all densities. Biovolumes of up to 3.9 C/enclosure were obtained. Some comparisons between culture of small scallops in mesh enclosures in Long Island Sound and in raceways were made and both systems were useful for nursery culture of this species. Journal of Shellfish Research, Vol. 3, No. 1, 105-115, 1983. ABSTRACTS OF TECHNICAL PAPERS Presented at 1982 Annual Meeting WEST COAST SECTION NATIONAL SHELLFISHERIES ASSOCIATION Olympia, Washington September 10-11, 1982 Olympia, Washington, September 10-12, 1982 Abstracts, 1982 NSA West Coast Section Meeting 107 CONTENTS Richard Albright Population Structure and Production of the Amphipod Corophium salmonis Stimpson in Grays Harbor, Washington 109 /. H. Beanie and J. Perdue Progress in the Development of Resistance Against Summer Mortality through Selective Breeding of Pacific Oysters 109 Clarke G. Beaudry Survival and Growth of the Larvae of Haliotis kamtschatkana Jonas at Different Temperatures 109 Richard Bumgarner Recent Developments in the Spot Prawn Fishery in Hood Canal, Washington 110 Ken Cooper Potential for Application of the Chemical DOPA to Commercial Bivalve Setting Systems 110 Flinn Curren Japanese Oyster Drill Studies Ill Catherine Falmagne Problems Associated with the Rearing and Setting of Larvae of the California Mussel Mytilus californianus Conrad in a Hatchery 112 Jill E. Follett A Histological Study of the Gastrointestinal Tract of the Tanner Crab Chionoectes bairdi Rathbun (Decapoda, Reptantia) 112 Thomas C Kline The Effect of Population Density on the Growth of the Butter Clam Saxidomus gigantus 112 Nancy Musgrove The Feeding Behavior of the Terebellid Polychaete Thelepus crispus Johnson in Response to Currents 113 Louisa Nishitani and Kenneth Chew Vertical Migration of Gonyaulax catenella: Potential Implications for Management of Paralytic Shellfish Poisoning (PSP) Problems 113 Scharleen Olsen Abalone and Scallop Culture in Puget Sound 113 Timothy Sample PSP: Its History, Processes and Impacts as Applicable to Puget Sound 114 A. Kimbrough Siewers Commercial Mariculture of a Bay Scallop Argopecten circularis (Sowerby) in the Ensenada of La Paz, Baja California Sur, Mexico 114 John J. Sullivan and Wayne T. Iwaoka PSP Research: Recent Advances in Analytical and Biochemical Studies 114 Louis Wachsmuth Disaster Ahead for the Yaquina Bay Oyster Industry? 115 Olympia. Washington. September 10-12, 1982 Abstracts, 1982 NSA West Coast Section Meeting 109 POPULATION STRUCTURE AND PRODUCTION OF THE AMPHIPOD COROPHIUM SALMONIS STIMPSON IN GRAYS HARBOR, WASHINGTON RICHARD ALBRIGHT Division of Aquaculture and Invertebrate Fisheries. School of Fisheries, University of Washington Seattle, Washington 98195 The tube-dwelling amphipod Corphium salmonis is a dominant benthic organism and important food resource in the estuarine mudflats of Grays Harbor, WA. Intertidal core samples were collected at two sites during the spring and summer of 1980 to determine the population structure, biomass, rate of growth, and production of C. salmonis. The abundance of C. salmonis ranged from 200 to 50.000 individuals per m 2 . Peak abundances occurred during July and August. Abundances at the 1 .8-m stations were higher than at the 0.6-m stations. Females of C. salmonis attained sexual maturity at alength of 4.0— 4.5 mm. Brooding of eggs began in April and continued through the end of sampling (30 September). Male-female ratios were lower for sexually mature individuals of C salmonis than for immature individ- uals, apparently as a result of predation on sexually mature males which wander over the tideflats in search of females. Male-female ratios decreased in the lower intertidal zone, apparently as a result of increasing predation pressure. Ratios also decreased over time at all stations, suggesting that predation pressure may also increase through the spring and summer. An inverse relationship between male-female ratios for mature and immature amphipods suggests a pos- sible genetic response to disparate sex ratios among mature individuals. Data from both natural populations and from cohorts which were artificially isolated inside in situ cages were used to obtain size-specific growth rate curves and production estimates for C. salmonis. Total Corophium pro- duction for each station between 1 April and 30 September varied from 3.6 to 10.7g/m 2 dry wt. Corophium production was higher at the upper intertidal stations. Turnover rates (the ratio of production to mean biomass) ranged rom 7.2 to 8.6. The production and turnover rates of Corophium salmonis are high relative to other invertebrate species. Thus, this amphipod is an important contributor to secondary production in Pacific Northwest estuaries, providing an important food resource for its predators, many of which are commercially or recreationally valuable. This production must be taken into consideration when making decisions relating to activities such as dredging and filling which have potentially adverse impacts on intertidal areas. PROGRESS IN THE DEVELOPMENT OF RESISTANCE AGAINST SUMMER MORTALITY THROUGH SELECTIVE BREEDING OF PACIFIC OYSTERS J.H. BEATTIE AND J. PERDUE Division of Aquaculture and Invertebrate Fisheries, School of Fisheries, University of Washington Seattle, Washington 98195 Since 1974 the University of Washington^ School of Fisheries has been conducting research in the genetics of the giant Pacific oyster Crassostrea gigas (Thurnberg). The main emphasis of this work has been the development, through selective breeding, of oyster stocks with high sur- vival potential during summer mortality. Summer mortality is a phenomenon that routinely accounts for losses of from 10 to 60% of harvestable 2-year-old oysters in bays of the states of Washington and California, and Japan. The breed- ing program began as a selection of individuals from wild populations. The selection process was based upon survival during elevated temperature (21 °C) challenges. The breed- ing of these individuals (one male mated with one female) produced families of oysters which could be tested and compared on growing grounds experiencing annual mortal- ities. On the basis of high survival during actual summer mortality, families were selected as the brood lines for future generations. Of 103 families tested since 1977, up to 78 have had higher survival than non-selected controls. The pri- mary goal of the breeding program is to provide brood stock to commercial hatcheries for production of oyster seed resistant to summer mortality. However, for the past three years, the families have also been monitored for growth, gonadal development, and glycogen storage. Since reduced gonadal development and high glycogen content are desirable commercial characteristics, these parameters have also been used in our overall breeding plan. Brood stocks which appear to show promise have been made available to commercial hatcheries since 1978. Data are now being processed and evaluated from the experimental families which will provide valuable information concerning heritability of glycogen levels, and experiments are being conducted on the effects of inbreeding. With every step, an understanding of oyster genetics is clearer and the goal of commercial production of superior oysters is closer. SURVIVAL AND GROWTH OF THE LARVAE OF HALIOTIS KAMTSCHATKANA JONAS AT DIFFERENT TEMPERATURES CLARKE G. BEAUDRY Division of Aquaculture and Invertebrate Fisheries. School of Fisheries, University of Washington Seattle, Washington 98195 110 Abstracts, 1982 NSA West Coast Section Meeting Olympia, Washington. September 10-12, 1982 Larvae of the pinto (or threaded) abalone 7/a//or/s kamt- schatkana were reared at four temperatures, 14, 16, 18.5, and 21°C in 2-C glass beakers. Survival at the end of the experimental period was best at 18.5° and worst at 21°. More rapid settlement observed at higher temperatures may have improved survival at those temperatures by shortening the vulnerable planktonic stage during which most mortalities occurred. Abalone at the highest temperature (21°) showed signs of thermal stress and experienced total mortality. During early embryonic development, from fertilized egg through the trochophore, the lowest temperature (14°) pro- duced the most normal larvae and highest survival. At higher temperatures progressively more mortalities and abnormal- ities occurred. Larvae reared at 18.5° were consistently of greatest size at settlement; however, abalone reared at 16° grew more rapidly and obtained the greatest length at the end of a 2-month period. RECENT DEVELOPMENTS IN THE SPOT PRAWN FISHERY IN HOOD CANAL, WASHINGTON RICHARD BUMGARNER Washington Department of Fisheries Point Whitney Shellfish Laboratory Brinnon, Washington 98320 Hood Canal, a major arm of Puget Sound, is located in northwestern Washington about 48 km (30 mi) west of Seattle. This is the only area in Washington that has consis- tently produced commercial quantities of the spot prawn Pandalus platyceros Brandt. Harvest for both commercial and personal use (recreation) has been restricted to shell- fish pot gear since the early 1950's. Increased commercial fishing pressure and poor recruitment between 1972 and 1974 resulted in a decline in spot prawn abundance and serious conflict between commercial and recreational fisher- men. This necessitated emergency season closures in 1974, 1975, and 1976. The year 1977 marked the beginning of a new management approach for the Hood Canal spot prawn stocks and associated fisheries. Season lengths and opening dates were set according to the results of a preseason stock assessment and anticipated fishing effort. To ensure an equi- table share of the available surplus for recreational fishermen the season was opened first to sport fishing and later to commercial harvest. By 1979, all fishermen were restricted to the use of shellfish pot gear having a mesh size of > 2.2 cm (7/8 in). This was initiated to protect juvenile prawns and to increase total yield. Changes in management appear to be working well. Since 1977, stock abundance has increased from a pre-season index of 1 .1 3 kg (2.5 lb) to 3.06 kg (6.75 lb) per pot in 1982. Harvest is also at an all time high. Nearly 95 metric tons were taken in both 1981 and 1982. Improved fishing success has also, in part, led to a tremendous increase in fishing pressure. The rate of increase has averaged nearly 50% per year since 1977. Better methods of effort-control are now needed to deal with the rapid expansion of this fishery. POTENTIAL FOR APPLICATION OF THE CHEMICAL DOPA TO COMMERCIAL BIVALVE SETTING SYSTEMS KEN COOPER Department of Biology Humboldt State University Areata, California 95521 Simple chemical compounds have been shown to trigger attachment and metamorphosis of the larvae of several species of marine invertebrates. The simplest molecules in which settlement inducing activity has been demonstrated are L-3. 4-dihydroxyphenylalanine (DOPA), gamma- aminobutyric acid (GABA), and choline. These molecules occur in the marine environment as covalently bounded compounds associated with adhesives, lubricants, exoskeletal proteins, and pigments. A review of numerous studies clearly implicated these chemical cues in successful habitat selec- tion by invertebrate at the termination of the planktonic stage of the life cycle. The similarity between these mole- cules and neurotransmitters suggests that the chemoreceptors are modified either ontogenetically or phylogenetically from receptors specific to the neurotransmitters dopamine. GABA. and acetylcholine. Selectivity in response by larvae to a given chemical appears to depend on the neurotransmitter- like portion of the compound, whereas specificity appears to depend on the protein, carbohydrate, or lipid constituents. Pediveligers of the blue mussel Mytilus edulis Linne and the giant Pacific oyster Crassostrea gigas (Thurnberg) settle in response to the amino acid DOPA. Implementing the use of chemicals to commercial setting systems depends on being able to either modify the chemoreceptors so that they respond to an inexpensive and easily available chemical and/or manipulating settlement behaviors. The initial objectives of my study were to determine the response of oyster larvae to DOPA, to examine the potential for applica- tion to existing commercial setting systems, and to determine the effect of several environmental factors on the degree of response. Aliquots of hatchery -reared pedivel- igers of C. gigas were tested for attachment in culture dishes to both aged oyster shells and the smooth glass surface of culture dishes. The pediveligers were reared at 34 ppt and at 25 °C. Within individual tests, the settlement response by the pediveligers was examined following exposure to DOPA at 0.00001 M while varying the salinity (25 to 35 ppt) and Olympia, Washington, September 10-12, 1982 Abstracts, 1982 NSA West Coast Section Meeting 1 1 1 temperature (20 to 30°C). Controls were run without the addition of DOPA. The results presented are preliminary findings and only indicate observed trends. In tests which offered only a smooth glass surface for settlement, attach- ment of the larvae to the glass occurred after 24 hr with but not without the addition of DOPA to the seawater. In tests to which DOPA was added the highest percentage of attach- ment occurred at a salinity /temperature combination of 35 ppt/30°C. The pediveligers also attached to the glass surface at the following salinity/temperature combinations listed in order of decreasing percent response: 35 ppt/25°C, 35 ppt/ 20°C. and 30 ppt/30°C. After 18 hr. a relatively high num- ber of pediveligers attached to the glass surface in the runs without DOPA at a salinity/temperature combination of 35 ppt/30°C. Also at 35 ppt/30°C in the runs with DOPA a smaller, but significant, percentage of the pediveligers meta- morphosed (indicated by new shell growth) without attach- ing to the glass surface. This did not occur in any of the other runs. The oyster pediveligers were next tested for attach- ment to aged oyster shells in response to the addition of DOPA. Preliminary results indicate that there was a slightly greater set after 24 hr onto the shells in the tests with DOPA. However, exposure of the larvae to DOPA also promoted attachment to the glass surfaces of the culture dishes. The consequence was that after 48 hr, the set onto the shell was greater in the runs without DOPA, although the total percentage of larvae which undergo metamorphosis appeared to be the same. In the runs with DOPA a signifi- cant percentage of the larvae either attached to the glass surface or metamorphosed without attaching to any sub- strate. These findings suggest that DOPA will not increase the percentage of set onto oyster shells when the setting is allowed to occur over several days. Rather, these findings clearly suggest that the use of DOPA promotes extraneous setting onto otherwise unfavorable substrates. However, these findings do not discount the possibility that chemicals can be used to obtain a more rapid set. The use of chemical cues appeared more applicable to setting systems in which no preferred setting substrate is used, such as in the setting of clams and clutchless oysters. JAPANESE OYSTER DRJLL STUDIES FLINN CURREN Division of Aquaculture and Invertebrate Fisheries, School of Fisheries University of Washington Seattle, Washington 98195 The Japanese oyster drill Ocenebra inomata (Recluz) is an economically important predator of oysters in areas along the west coast, as well as in its native Japan. Since its acci- dental introduction into Puget Sound with shipments of Pacific oyster seed, attempts to control this snail have included expensive hand picking and mercuric chloride. These animals aggregate during certain times of the year, and it is suspected that this behavior is cued by water-borne pheromones (chemical substances which enable communica- tion between animals). Pheromones are currently being used in the control of several insects (e.g., gypsy moth) and might have potential as a control technique for the Japanese oyster drill. It was necessary, therefore, to develop an appropriate bioassay to test different water extracts for pheromones. Bioassays consist of subjects (in this case snails), stimuli (water with suspected chemical agents), and responses (which should be easy to identify . associated with the stimuli, reproducible, and rapid). Bioassays should also minimize the water used for stimulus and control to decrease efforts involved in chemical extraction and concentration. Large numbers of snails must be assayed to give statistical credi- bility to sometimes subjective behavioral data. Several bio- assays have been based on the snail's rheotactic response (in a current of water, the snail moves upstream). The Pratt choice chamber was rejected because large volumes of water were needed with only one snail per run. Riffle flumes were rejected because turbulent flows were encountered. Cephalic antennal elongation (after pipetting a small amount of water in front of the snail) was also rejected because of ( 1 ) the highly subjective nature of the response (i.e., when are the antennas elongated?), and (2) the large time requirement of (10 min/subject) with the undivided attention of the research. The inadequacies of these bioassays led to work currently being done on a trough bioassay. A test chamber 1 X 1.5 m (39 X 50 in.) was constructed with stimuli and controls (aged sea water) entering the flume through over- flowing 1-2 beakers. Several hundred snails were placed 1 m from the beakers and the numbers of snails climbing up the beakers during a 6-hr period are noted. Current research using this apparatus includes: (1 ) dye studies to determine the water depth necessary for good mixing; (2) determina- tion of the threshold flow rate to induce rheotaxis in oyster drills; (3) testing of flow rates with a known stimulus (oyster effluent); and (4) testing of stimuli from whole ground snail extracts and field-filtered effluents from aggre- gations. Stimuli found to be effective in these bioassays may eventually be used to bait traps or disrupt snail behavior to control Japanese oyster drills on oyster beds. 112 Abstracts, 1982 NSA West Coast Section Meeting Olympia, Washington, September 10-12, 1982 PROBLEMS ASSOCIATED WITH THE REARING AND SETTING OF LARVAE OF THE CALIFORNIA MUSSEL MYTILUS CALIFORNIANUS CONRAD IN A HATCHERY CATHERINE FALMAGNE Division of Aquaculture and Invertebrate Fisheries, School of Fisheries University of Washington Seattle, Washington 98195 Mytilus californianus was successfully spawned and its larvae were reared through metamorphosis in the University of Washington hatchery at Manchester, WA. Although suc- cess in spawning and rearing may vary with the hatchery location and methods, data indicated the unreliability of induced spawning at any given time. Some effects resulting from different experimental combinations of temperature and salinity have been observed. Survival of larvae to the pediveliger stage at 18°C and 32 ppt was 31%. The larvae all settled at the lower part of the suspended seed ropes because they have a tendency to sink to the bottom of the tank throughout metamorphosis. Further, higher numbers of the larvae settled when the water was "conditioned" with adult mussels. A HISTOLOGICAL STUDY OF THE GASTROINTESTINAL TRACT OF THE TANNER CRAB CHIONOECTES BAIRDI RATHBUN (DECAPODA, REPTANTIA) JILL E. FOLLETT Alaska Dept. of Fish and Game 333 Raspberry Rd. Anchorage, Alaska 99502 The tanner crab Chionoecetes bairdi is a commercially important species in Alaska about which little is known of its histology. In this study of the tanner crab, the morphol- ogy and histology of the gastrointestinal tract is examined and compared to that of the blue crab Callinectes sapidus Rathbun. Three histological stains were used: hematoxylin and eosin, periodic acid-Schiff (PAS), and the Feulgen reaction with picro-methyl blue. The foregut, midgut, and hindgut were examined. The fore- and hindguts are both of ectodermal origin, and exhibit similar cuticular layers, epi- thelial cells, and tegmental glands. The endodermally derived midgut and caeca differ significantly from the fore- and hindgut both in their lack of cuticle, and in the vacuolation of the epithelial cell nuclei. One morphological difference that was noted between the tanner and blue crabs was the absence of aborizations in the posterior midgut caecum of the tanner crab. The function of this caecum may be for osmoregulation. Prolonged osmoregulation in brackish and fresh water occurs to a significant extent in the blue crab but not in the tanner crab because it remains in a marine environment. This difference in habitats may explain the variation in caecum structure. In most other aspects, the histology and morphology of C. sapidus closely resembled those of C. bairdi. THE EFFECT OF POPULATION DENSITY ON THE GROWTH OF THE BUTTER CLAM SAXWOMUS GIGANTUS THOMAS C. KLINE Division of Aquaculture and Invertebrate Fisheries, School of Fisheries University of Washington Seattle. Washington 98195 Butter (or smooth Washington) clams, Sax idomus gigan- teus (Deshayes), were grown for 2 yr at 4 population densi- ties (96, 48, 24, and 12 clams/0.25 m 2 plots) in a Latin Squares arrangement at the — 0.5-m tide level (MLLW) on a privately owned beach approximately 1 km west of Port Gamble on Hood Canal in Washington State. The clams, dug up from within 10 m of the experimental site, and were individually numbered and measured in length, width, and thickness to the nearest 1 mm and placed into three groups, each containing one third of the naturally occurring popula- tion, depending on the clam length. The medium sized group ranged from 76 to 80 mm, with the small and large groups taking the remainder. The plots were filled by randomly selecting from the three groups, with one third of each plot represented by each of the three size groups. The clams were planted in 1978 during the spring tidal series closest to the summer soltice. They were removed, remeasured and replanted at a similar tide in 1979. In 1980, the clams were removed for the last time, during the soltice tidal series. In order to compare the growth differences in the 4 population densities, Walford plots of length at one time versus length at another were made. Walford plots were also made for width and for the product of length and width. The result- ing plots showed that there was an appreciable difference in growth between the 48 and 24 clams/plot. The 96 clams/ plot had the same growth slope as the 48/plot. The differ- ence between the 12 and 24 clams/plot was also negligible. The data indicated that the maximum density for best growth is 24 clams/0.25 m 2 (96/m 2 ). The experiment also demonstrated the usefulness of Walford plots to optimize population in a grow-out situation as used in shellfish aquaculture. Olympia. Washington, September 10-12, 1982 Abstracts. 1982 NSA West Coast Section Meeting 1 1 3 THE FEEDING BEHAVIOR OF THE TEREBELLID POLYCHAETE THELEPUS CRISPUS JOHNSON IN RESPONSE TO CURRENTS NANCY MUSGROVE Division of Aquaculture and Invertebrate Fisheries, School of Fisheries, University of Washington Seattle, Washington 98195 The role of currents in determining the feeding behavior of Thelepus crispus was investigated as part of a large-scale research project on the response of bottom-dwelling com- munities to organic enrichment and pollution. Live worms were collected from the intertidal beach at Garrison Bay on San Juan Island, WA. They were placed in natural sediments, in specially designed flow tanks at the Seattle Aquarium and at the University of Washington Friday Harbor Labs. After the worms reconstructed their tubes, the feeding behaviors were observed under three different current velocities ranging from 1 to 8 cm/sec. Particle settlement experiments were also conducted at the three velocities to determine if flow affected the settlement of food around the feeding worms. To clarify any morphological limitations which might affect the choice of food or feeding method in Thelepus, the tentacles of preserved specimens were examined under a scanning electron microscope. To corroborate findings in laboratory experiments, field observations and How measure- ments were made using SCUBA gear at Garrison Bay, WA. When Thelepus is exposed to different current velocities it orients its feeding tentacles in response to the direction of flow and the areas of maximum particle settlements. At speeds < 2 cm/sec, particle settlement is relatively even around the worm mounds and Thelepus spreads it tentacles in all directions on the sediment as well as in the water column. It is under this type of flow condition that Thelepus is abundant in the field. Suspension feeding may play an important role in food gathering for Thelepus. At higher current speeds (4 to 8 cm/sec) particle settlement becomes differentiated between upstream and downstream areas around the worm. The upstream face of the mound has relatively few particles settling out. The downstream face and area immediately behind the worm mound has greater amounts of particles settling out. The placement of tentacles mirrors the settlement patterns of particles. The strength of the current is an important consideration as to how Thelepus feeds and where it gathers its food. VERTICAL MIGRATION OF GONYAULAX CATENELLA: POTENTIAL IMPLICATIONS FOR MANAGEMENT OF PARALYTIC SHELLFISH POISONING (PSP) PROBLEMS LOUISA NISHITANI AND KENNETH CHEW School of Fisheries University of Washington Seattle. Washington 98195 The diel vertical migration pattern of the dinoflagellate Gonyaulax catenella Whedon et Kofoid which produces paralytic shellfish poisons, may have important implications for management decisions by industry, public health agen- cies, and research groups. This migration pattern influences the length of time shellfish at different tide heights or depths below rafts are exposed to G. catenella. The exposure should be considered by health agencies, along with tide height or depth, when planning routine sampling and by the shellfish industry when selecting bivalve species to plant or dredging depths. Because the vertical migration pattern is greatly affected by the degree of stratification of the water a predictive model which involves field studies of the effects of changes in density gradients on density of G. catenella should be developed. The vertical migration pattern appears to be extremely important in the development of large populations of G. catenella in certain sheltered bays, from which significant numbers of G. catenella may then be exported to waters outside the bay. An understanding of the functioning of such bays may be useful in determining timing and sites for monitoring and in selection of sites for controlling G. catenella with the parasite, Amoebophrya (if laboratory tests indicate such control would be safe, desirable, and feasible). ABALONE AND SCALLOP CULTURE IN PUGET SOUND SCHARLEEN OLSEN Washington Department of Fisheries, Point Whitney Shellfish Laboratory, Brinnon, Washington 98320 Three new species were cultured at Point Whitney Shell- fish Laboratory during 1979-82; the native pinto (or threaded) abalone Haliotis kamtschatkana Jonas, the red abalone Haliotis rufescens Swainson, and the purple hinge (or giant) rock scallop Hinnites multirugosus (Gale). A pilot hatchery system was developed and various culture condi- tions, methods, and temperatures were investigated. Growth of the pinto abalone was followed over a period of 3 yr in the hatchery. Comparisons of growth and survival rates between juvenile pinto and red abalone were investigated over a one-year period. The pinto growth rate was affected by the type of culture container used and by the presence or absence of light. At one year of age, pinto abalone shells averaged 20 mm. At two years, mean shell length was 37 mm, and the oldest year-class averaged 59 mm at three years of age. Various scallop culture methods, feeding densities and container configurations affected the scallop growth rates. Salinity tolerance was studied and salinities < 23 ppt were detrimental to normal growth and survival. Field plantings at Lopez Island, Port Blakley, Willapa Bay, Manchester. Belling- ham Bay, and Point Whitney were studied for growth and survival of juvenile rock scallops. Growth rates of 4.2 mm/mo were achieved at some locations. 1 14 Abstracts, 1 982 NSA West Coast Section Meeting Olympia, Washington, September 10-12, 1982 PSP: ITS HISTORY, PROCESSES AND IMPACTS AS APPLICABLE TO PUGET SOUND TIMOTHY SAMPLE METRO, Water Quality Division Seattle, Washington 98104 This report provides a synopsis of available information concerning the history, processes, and impacts associated with paralytic shellfish poisoning (PSP) in Puget Sound. Paralytic shellfish poisoning is a form of food poisoning in which extremely lethal toxins, produced by certain dino- flagellates, are accumulated in shellfish and passed on to humans. Outbreaks of PSP appear to be spreading to previ- ously unaffected areas. They are increasing in intensity worldwide as well as within the Puget Sound basin. This report includes a review of these trends and of the current toxicity monitoring program established in the state of Washington to protect the public from PSP. Attention is also given to what causes toxic dinoflagellate blooms, partic- ularly dinoflagellate cysts, and contributing environmental factors (i.e.. temperature, precipitation, and nutrients). Apparently, numerous environmental factors may influence development of a bloom from newly emergent germlings. In addition, the introduction of certain organic compounds, called chelators, to coastal waters may create an environ- ment favoring growth of the dinoflagellate population by controlling the availability of certain growth-regulating trace metals. A discussion of the nature of dinoflagellate toxins and their possible effects on man and other organisms is included. The recent discovery that dinoflagellate toxins may be lethal to organisms other than man has serious impli- cations: for example, consumption of toxic shellfish may prove fatal to certain species of birds. Additionally, recent investigations indicate that lethal levels of dinoflagellate toxins can be accumulated, retained, and passed up the food chain by herbivorous zooplankton that feed on toxic dinoflagellates. COMMERCIAL MARICULTURE OF A BAY SCALLOP ARGOPECTEN CIRCULARIS (SOWERBY) IN THE ENSENADA OF LA PAZ, BAJA CALIFORNIA SUR, MEXICO A. KIMBROUGH SIEWERS Cultivos Marinos de Baja California S. A. de C. V. RioNazas 163-401 Mexico 5, D. F. (and) Pigeon Point Aquaculture Center 921 Pigeon Point Road Pescadero, California 94060 Mexico's first private shellfish aquaculture company was formed in La Paz, BCS. A local bay scallop, the Pacific calico scallop Argopecten circularis, is grown in lantern nets suspended from long lines. Scallop spat are collected by putting sticks of plastic mesh in nylon "onion bags" which are tied five to a weighted line and hung from long lines. Collectors are set out in the spring and the seed scallops are removed 2 to 5 mo later. Significant numbers of scallop spat also regularly set on the lantern nets. Seed scallops are grown in pearl nets during the nursery phase of culture, then grown to market size in lantern nets. Fouling is removed from the nets by a saltwater spray from a gasoline-powered water pump. Scallops are stocked at a density of 25/0.1 m 2 (50 per level) for the final growth stage. Market size (5 to 6 cm) is reached in 5 to 7 mo. Four metric tons of scallops were marketed in Mexico City in the first year of production. A pufferfish, Spheroides annulatus (Jenyns), preyed on cultured scallops by chewing open the bottom compartments of some lantern nets. This was alleviated by shortening the lantern nets by 3 levels. A hatchery was constructed, and in the first experiment scallops were conditioned, spawned, and the larvae reared to juvenile stage. Improvements in the grow out system should include using 5-level lantern nets in 2 mesh sizes (12 and 21 mm), and submerging the long lines by 0.5 m. An annual production of 10 tons appears necessary for profitability, with 20 to 30 tons possibly optimum. PSP RESEARCH: RECENT ADVANCES IN ANALYTICAL AND BIOCHEMICAL STUDIES JOHN J. SULLIVAN AND WAYNE T. IWAOKA Institute for Food Science and Technology School of Fisheries University of Washington Seattle, Washington 98195 Paralytic shellfish poison (PSP), or "Red Tide," is a persistent problem in the northern coastal areas of the United States and monitoring of shellfish is accomplished via mouse bioassays. We have developed an alternate analyti- cal technique for measuring the toxins using high pressure liquid chromatography. Comparison of both techniques showed high correlation when toxin content in shellfish samples contained about 60 /ig toxin per 100 g meat. The mean variation was 25% when higher amounts of toxin were present. Variation in the mouse bioassay is ± 20%. Preliminary and proposed work will be reported on the biochemical aspects of uptake, storage, and release of the PSP toxins in shellfish. Olympia, Washington, September 10-12, 1982 Abstracts, 1982 NSA West Coast Section Meeting 115 DISASTER AHEAD FOR THE YAQUINA BAY OYSTER INDUSTRY? LOUIS WACHSMUTH Oregon Oyster Company 208 SW Ankeny Street Portland, Oregon 9 7204 After 115 years of fishing and farming, the future of Yaquina Bay is as uncertain and bleak as ever, with one exception. The history of this bay, located in Newport, OR, parallels histories of other west coast growing areas. The oyster schooners from San Francisco, the old-time oyster tongers, the replacement of the native Pacific oyster Ostrea lurida Carpenter by the giant Pacific oyster Crassostrea gigas (Thurnberg), the wood products pollution, the local town's sewage, the infamous tidal wave, and the massive siltation problem are all elements and events of the past 1 15 years. The current crisis seems to be of major proportions and threatens the future of oyster farming. Generally speaking, oysters are no longer growing to full potential. Kumamoto oysters (variants of C. gigas), which were grown on the bottom 15 years ago, now grow only from rafts. Giant Pacific oysters, as of 8 years ago, became stunted after the second year of growth, only putting on thick layers of blistered shells that were filled with a foul-smelling exudate. They seldom reached "medium" size even after 6 years. Perhaps related to this is the fact that several other forms of sea life have almost disappeared from our area over the past 30 years. The source of this problem is unknown, but could be related to the destruction of the ocean food chain over the years. The stunting problem also has been observed in other locations on the west coast. The only ray of hope for this company is to repeat the great switch of the 1920's. That is, change species of oysters once again. The Japanese oyster, Crassostrea ariakensis (Wakiya) (= Ostrea/ Crassostrea rivularis), seems to be the answer. After experimenting for five years, we discovered these advantages: (1) 50% faster growth than C. gigas, thereby shortening the growth cycle by one year; (2) good flavor; (3 ) absence of the stunting and blistering problem ; (4 ) larger maximum size than C. gigas; (5) higher spawning tempera- tures resulting in a firm and tasty meat during August and September; and (6) uniform shell shape and attractive interior shell surface. INFORMATION FOR CONTRIBUTORS TO THE JOURNAL OF SHELLFISH RESEARCH Original papers dealing with all aspects of shellfish research will be considered for publication. Manuscripts will be judged by the editors or other competent reviewers, or both, on the basis of originality, content, merit, clarity of presentation, and interpretations. Each paper should be carefully prepared in the style followed in Volume 3, Number 1, of the Journal of Shellfish Research (1983) before submission to the Editor. Papers published or to be published in other journals are not acceptable. Title. Short Title, Key Words, and Abstract: The title of the paper should be kept as short as possible. Please include a "short running title" of not more than 48 char- acters including space between words, and approximately seven (7) keys words or less. Each manuscript must be accompanied by a concise, informative abstract, giving the main results of the research reported. The abstract will be published at the beginning of the paper. No separate summary should be included. Text: Manuscripts must be typed double-spaced throughout one side of the paper, leaving ample margins, with the pages numbered consecutively. Scientific names of species should be underlined and, when first mentioned in the text, should be followed by the authority. Abbreviations, Style, Numbers: Authors should follow the style recommended by the fourth edition ( 1978) of the Council of Biology Editors [CBE] Style Manual distrib- uted by the American Institute of Biological Sciences. All linear measurements, weights, and volumes should be given in metric units. Tables: Tables, numbered in Arabic, should be on separate pages with a concise title at the top. Illustrations: Line drawings should be in black ink and planned so that important details will be clear after reduction to page size or less. No drawing should be so large that it must be reduced to less than one third of its original size. Photographs and line drawings preferably should be prepared so they can be reduced to a size no greater than 17.3 cm X 22.7 cm, and should be planned either to occupy the full width of 17.3 cm or the width of one column, 8.4 cm. Photographs should be glossy with good contrast and should be prepared so they can be repro- duced without reduction. Originals of graphic materials (i.e., line drawings) are preferred and will be returned to the author. Each illustration should have the author's name, short paper title, and figure number on the back. Figure legends should be typed on separate sheets and numbered in Arabic. No color illustrations will be accepted unless the author is prepared to cover the cost of associated reproduction and printing. References Cited: References should be listed alpha- betically at the end of the paper. Abbreviations in this section should be those recommended in the American Standard for Periodical Title Abbreviations, available through the American National Standards Institute, 1430 Broadway, New York, NY 10018. For appropriate citation format, see examples at the end of papers in Volume 3, Number 1. of the Journal of Shellfish Research or refer to Chapter 3, pages 51-60 of the CBE Style Manual. Page Charges: Authors or their institutions will be charged S25.00 per printed page. If illustrations and/or tables make up more than one third of the total number of pages, there will be a charge of $30.00 for each page of this material (calculated on the actual amount of page space taken up), regardless of the total length of the article. All page charges are subject to change without notice. Proofs: Page proofs are sent to the corresponding author and must be corrected and returned within seven days. Alterations other than corrections of printer's errors may be charged to the author(s). Reprints: Reprints of published papers are available at cost to the authors. Information regarding ordering reprints will be available from the National Shellfisheries Association at the time of printing. Cover Photographs: Particularly appropriate photo- graphs may be submitted for consideration for use on the cover of the Journal of Shellfish Research. Black and white photographs, if utilized, are printed at no cost. Color illustrations may be submitted but all costs associated with reproduction and printing of such illustrations must be covered by the submitter. Correspondence: An original and two copies of each manuscript submitted for publication consideration should be sent to the Editor, Dr. Roger Mann, Woods Hole Oceano- graphic Institution, Woods Hole, Massachusetts 02543. JOURNAL OF SHELLFISH RESEARCH Vol. 3, No. 1 June 1983 CONTENTS Brian F. Beat Predation of Juveniles of the Hard Clam Mercenaria mercenaira (Linne) by the Snapping Shrimp A Ipheus heterochaelis Say and A Iphens normanni Kingsley 1 Rodney Dal ton and Winston Menzel Seasonal Gonadal Development of Young Laboratory-Spawned Southern (Mercenaria campeehiemis) and Northern (Mercenaria mercenaria) Quahogs and their Reciprocal Hybrids in Northwest Florida . : 11 Paul J. Flagg and Robert E. Malouf Experimental Plantings of Juveniles of the Hard Clam Mercenaria mercenaria (Linne) in the Waters of Long Island, New York 19 / D. Andrews Transport of Bivalve Larvae in James River, Virginia 29 Catherine Enright, Donna Krailo, Larry Staples, Maria Smith, Carl I aughan, Debra Ward, Pamela Gaul, and Elisabeth Borgese Biological Control of Fouling Algae in Oyster Aquaculture 41 Mary L. Swift and Mohammed Ahmed A Study of Glucose, Lowry-Positive Substances, and Triacylglycerol Levels in the Hemolymph of Crassostrea virginica (Gmelin) 45 Edward R. Urban, Jr., Gary D. Pruder and Christopher J. Langdon Effect of Ration on Growth and Growth Efficiency of Juveniles of Crassostrea virginica (Gmelin) 51 Aurora Ledo, Enrique Gonzalez, Juan L. Barja and Alicia E. Toranzo Effect of Depuration Systems on the Reduction of Bacteriological Indicators in Cultured Mussels (Mytilus editlis Linnaeus) 59 C. B. Calloway and R. D. Turner Documentation and Implications of Rapid Successive Gametogenic Cycles and Broods in the Shipworm Lyrodus floridanus (Bartsch) (Bivalvia. Teredinidae) 65 RESEARCH NOTE C. F. Phleger and S. C Cary Settlement of Spat of the Purple-Hinge Rock Scallop Hinnites multirugosus (Gale) on Artificial Collectors 71 Abstracts of Technical Papers Presented at the 1982 Annual Meeting National Shellfisheries Association, Baltimore, Maryland - June 14-17, 1982 75 Abstracts of Technical Papers Presented at the 1982 Annual Meeting National Shellfisheries Association, West Coast Section, Olympia, Washington - September 10- 1 2, 1982 105 COVER PHOTOMICROGRAPH: Female specimen of Alpheus heterochaelis Say (Decapoda; Alpheidae) collected 26 June 1982 from an oyster reef near Beaufort, North Carolina, USA (scale bar = 5 mm). Photo- graphed with 4 X 5-inch Graphic (Graflex) camera and Xenar lens (# 1 : 4.7/1.35) using Kodak Tech Pan 2415 film and processed in HC 110, F-dilution. (Exposure = 5 sec at f 45.) [Photograph provided by Henry E. Page, University of North Carolina, Institute of Marine Science, Morehead City, North Carolina./ JOURNAL OF SHELLFISH RESEARCH VOLUME 3, NUMBER 2 DECEMBER 1983 moratory LIBRARY AUG 26 1985 5 Hoie, Mass. The JOURNAL OF SHELLFISH RESEARCH (formerly Proceedings of the National Shellfisheries Association) is the official publication of the National Shellfisheries Association Editor-in-Chief Dr. Roger Mann College of William and Mary Virginia Institute of Marine Science Gloucester Point, Virginia 23062 Managing Editor Dr. Edwin W. Cake, Jr. Gulf Coast Research Laboratory Ocean Springs, Mississippi 39564 National Shellfisheries Association Publications Committee Prof. Melbourne R. Carriker College of Marine Studies University of Delaware Lewes, Delaware 19958 Dr. Rober E. Hillman Battelle New England Marine Research Laboratory Duxbury, Massachusetts 02332 Mr. Michael Castagna College of William and Mary Virginia Institute of Marine Science Eastern Shore Laboratory Wachapreague, Virginia 23480 Dr. Richard A. Lutz Department of Oyster Culture New Jersey Agricultural Experimental Station Cook College, Rutgers University New Brunswick, New Jersey 08903 Journal of Shellfish Research Volume 3, Number 2 ISSN: 00775711 December 1983 Journal of Shellfish Research, Vol. 3, No. 2, 117-128, 1983. SYMBIOTIC ASSOCIATIONS INVOLVING THE SOUTHERN OYSTER DRILL THAIS HAEMASTOMA FLORIDANA (CONRAD) AND '" ^ f,> - LIBRARY MACROCRUSTACEANS IN MISSISSIPPI WATERS, EDWIN W. CAKE, JR. Oyster Biology Section Gulf Coast Research Laboratory Ocean Springs, Mississippi 39564 AUG 26 1985 .Woods Ho/* ka»* ABSTRACT The symbiotic relationships between the southern oyster drill Thais haemastoma floridana (Conrad) and two species of crabs, the blue crab Callineetes sapidus Rathbun and the striped hermit crab Clibanarius vittatus (Bosc), were investigated in Mississippi. The crabs provided passive transport and food (attached fouling organisms) for the attached drills; 99 blue crabs_ carried 203 drills (X= 2.0 ± 2.1 drills crab , range = l-17;mode = 1, N = 55 crabs); 233 hermit crabs carried 299 drills (X= 1.3 ± 0.8 drills crab" 1 , range =1-6; mode = 1, N = 194 crabs). Drills attached to blue crabs were twice the mean height and six times the mean weight of those attached to hermit crabs (36.8 mm and 8.9 g versus 18.5 mm and 1.4 g. respectively). During one survey period 30 of 423 blue crabs (7.19"c) and 97 of 1,360 hermit crabs (7.17r) carried drills. The oyster drill/blue crab symbiosis persisted while spawning female crabs congregated around Mississippi's offshore barrier islands during the early fall of 1980 and ceased when the crabs died or migrated to deeper water during late fall. The oyster drill/hermit crab symbiosis was continuous. Drills attached while the crabs were buried at the seawater/ substrate interface, resting under peat outcroppings, or while scavenging among grass roots and jetsam. Once mounted, the drills were not readily dislodged by movement of the crabs. In the laboratory drills more readily mounted hermit crabs with attached drills and/or acorn barnacles than hermit crabs without these organisms. A typical mounting took only seconds to complete; drills readily attached to moving hermit crabs. Drills dismounted from hermit crab shells when in the immediate vicinity of live oysters. The drills preyed on acorn barnacles (Chelonihia patula [Ranzani] , Balanus spp.), oysters (Crasso- strea virginica [Gmelin] , Ostrea equestris Say), and slipper shells (Crepidula spp.) that fouled the blue crab carapaces and hermit crab shells. Two other gastropods (Cantharus cancellarius [Conrad] and Odostomia impressa [Say] ) were occasion- ally attached to blue and hermit crabs that carried oyster drills. KEY WORDS Blue crab, commensalism. macrocrustaceans. oyster drill, phoresis, striped hermit crab, symbiosis INTRODUCTION The southern oyster drill Thais haemastoma floridana (Conrad) (Gastropoda; Muricidae) is the most destructive predator of the American oyster Crassostrea virginica (Gmelin) in coastal Gulf of Mexico waters from Florida to Texas (Burkenroad 1931; St. Amant 1938; Butler 1953, 1954;Gunter 1953, 1979; McConnell 1953;Chapman 1955, 1958;Menzel and Hopkins 1955; Menzel et al. 1957, 1966; Hofstetter 1959; May and Bland 1970; May 1971 ; Pollard 1973; Breithaupt and Dugas 1979). Mortalities of oysters from drills can be as high as 50% a year (St. Amant 1938). Drills prefer water salinities that usually exceed 18 to 20 ppt (St. Amant 1938, Gunter 1979), and thus oyster reefs located near open Gulf waters are subjected to drill preda- tion during periods of drought or reduced freshwater inputs (from extended closures of water impoundments). Offshore, high salinity areas serve as reservoirs for drills; when inshore salinities increase, the drills invade nearshore reefs as planktonic veliger larvae. The larvae spend about a month in the plankton and are widely dispersed (Butler 1953). After metamorphosis, juvenile drills grow rapidly and can grow an average of 28 mm a year (range = 20 to 42 mm) (Butler 1953). During the fall of 1980 I observed southern oyster drills attached to many blue crabs (Callineetes sapidus Rathbun) and gastropod shells occupied by striped hermit crabs (Clibanarius vittatus [Bosc] ) in shallow waters around Mississippi's barrier islands. The drills were being passively transported by the crabs. St. Amant (1938) and Fothering- ham (1976) reported this symbiotic relationship, but not to the extent that 1 observed along Horn and Ship islands. St. Amant found four and five drills attached to two blue crabs in the vicinity of Grand Island, LA. He also noted that drills occurred on flotsam . Fotheringham found juvenile drills on 1 .7% of all gastropod shells (> 20 g) that were occupied by C. vittatus along the Texas coast. Mark Chatry (Louisiana Dept. Wildl. Fish., St. Amant Marine Laboratory, Grand Isle, LA. pers. comm.) found drills on blue and striped hermit crabs in lower Barataria Bay, LA, in the vicinity of Grand Isle during 1980 and 1981. The late Capt. L. J. Gorenflo of Biloxi. MS, photographed two blue crabs with two and four drills attached, respectively, that were trawled from Biloxi Bay channel in Mississippi Sound in 1953 (photograph provided by W. J. Demoran. Gulf Coast Research Laboratory, Ocean Springs. MS). Capt. Gorenflo noted on the photograph that most of the crabs were alive, but some were weak and dead. His photograph is the only evidence that this drill/crab symbiosis occurred previously in Mississippi waters. Other muricid oyster drills participate in similar drill/ crab symbioses along the Atlantic coast (Table 1). Federighi (1931) reported that the Atlantic oyster drill Urosalpinx 117 118 Cake cinerea Say utilized hermit crabs as a means of transport in lower Chesapeake Bay. Harold Haskin reported (in Carriker 1955) that on two occasions in Delaware Bay he found five previously marked drills (U. cinerea) on shells of the Atlantic moon snail Polinices duplicatus Say that were inhabited by the flat-clawed hermit crab Pagimts pollicaris Say. Some of the drills were attached to shells of hermit crabs that were no larger than their own shells. One marked drill attached to and was transported 3.5 in by a large hermit crab within 15 minutes of release. Haskin suggested that hermit crabs may play an important role in the distribu- tion of oyster drills. MacKenzie (1962) reported that large numbers of the thick-lipped oyster drill Eupleura caudata (Say) and lesser numbers of U. cinerea were transported on the carapaces of most horseshoe crabs (Limulus polyphemus [Linnaeus] ) that he dredged from Long Island Sound. One crab carried 761 thick-lipped and 4 Atlantic oyster drills. He described their symbiotic association as phoresy. (Cheng [1973] defined phorsey as a nonparasitic association in which the smaller species, the phoront, is mechanically carried on or in the larger species, the host, and no metabolic interaction or dependency occurs.) Richards Nelson (in Carriker 1955) reported as many as 140 Atlantic oyster drills per horseshoe crab in New Haven Harbor, CT. Fred Sieling (Maryland Dept. Nat. Resour., Annapolis. MD, pers.comm.) and Michael Castagna (Virginia Inst. Mar. Sci., Wachapreague. VA, pers.comm .(reported that they hadoccasionallyobserved blue crabs transporting one or two thick-lipped drills in lower Chincoteague Bay, VA, in the mid-1 950's. Federighi (1931) suggested that oyster drills obtained food from fouling organisms attached to the hermit crabs. Others (St. Amant 1938, MacKenzie 1962) found no evidence of drilling on the transport crab. Although several of these authors alluded to a symbiotic association of macrocrusta- ceans and muricid oyster drills, none attempted to docu- ment or quantify the extent of those associations. This paper describes the nature and extent of the drill/ crab symbioses that existed along Mississippi's barrier islands during the fall of 1980. It examines the factors that initiate and control these symbioses which appear to have character- istics of commensalism and phoresis. Hereinafter, 1 shall refer to the crabs as hosts and to the drills as symbionts. Occassionally, I shall utilize the terms "infestation" and "drill-infested" when presenting and discussing occurrence data and when describing the existence of drills on the shells of hosts. The use of these terms is not intended to infer any parasitic relationship; they are simply utilized in the absence of more appropriate terms. The crustacean and molluscan taxonomies utilized herein follow those of Williams (1965) and Abbott (1974), respectively. TABLE 1. A synoptic review of oyster drill/crab symbioses along the Atlantic and Gulf coasts of the United States. Oyster Drill Species Tliais haemastoma haysae T. haemastoma T. haemastoma T. haemastoma T. haemastoma floridana Eupleura caudata E. caudata Urosalpinx cinerea U. cinerea U. cinerea U. cinerea Calotrophon ostrearum Crab Species Callinectes sapidus C. sapidus C. sapidus Clibanarius vittatus C. vittatus C. sapidus C. vittatus 2 Limulus polyphemus C. sapidus L. polyphemus L. polyphemus L. polyphemus "Hermit crabs" Pagurus pollicaris Pagurus impressus Locality Grand Isle. LA Mississippi Sound. Ocean Springs, MS Lower Barataiia Bay. LA Texas coast Horn and East Ship islands, MS Lower Chincoteague Bay, VA (1956) Long Island Sound, NY Long Island Sound, NY New Haven Harbor, CT Lower Chesapeake Bay, VA Delaware Bay, DE Reference St. Joseph Bay, FL St. Amant 1938 (L. J. Gorenflo, 1953 photograph) Mark Chatry, LA DW&F, pers. comm. 1981 Fotheringham 1976 (Present study) Fred Sieling (MD DNR), Mike Castagna (VIMS), pers. comm. 1981 MacKenzie 1962 MacKenzie 1962 Richards Nelson (in Carriker 1955) Federighi 1931 Harold Haskin (in Carriker 1955) (E. W. Cake 1981, field observation) Identified as Tliais floridana haysae. "Occupying the following gastropod shells: Busy con contrarium alatus. and Tliais haemastoma floridana. Occupying the shells of P. duplicatus. Occupying the shells of 5. alatus. B. spiratum plagosum. Murex fulvescens. Polinices duplicatus, Stromhus Southern Oyster Drills Infest Macrocrustaceans 119 materials and methods Blue crabs and striped hermit crabs with attached oyster drills were collected at four stations on Horn and Ship islands; those islands form the southern boundary of Mississippi Sound (Figure 1). The crabs were collected in shallow water (< 1 m) with dip nets or crab tongs. On several occasions, all crabs encountered were collected to determine the incidence of infestation. Field observations were made on the behavior of the crabs and drills in their shared habitats. The drills and potential prey items on the crabs' shells (e.g.. acorn barnacles, oysters, and slipper shells) were examined for evidence of predation. Infested crabs and their attached drills were placed in individual plastic- bags and transported alive in coolers to the Gulf Coast Research Laboratory. Ocean Springs. MS. where they were measured and weighed, and the numbers of drills, barnacles, oysters, and slipper shells per crab (shell) were determined. In the laboratory, studies of the oyster drill/hermit crab symbiosis were conducted in 70- to 95-C all glass aquaria using sand, seawater, and animals from Horn Island. The experimental crabs occupied the shells of the lightning whelk Busy con contrarium Conrad, the pear whelk B. spiratum plagosum (Conrad), the southern oyster drill T. h. floridana, and the Atlantic moon snail Polinices duplicates (Say), and each was initially infested with acorn barnacles (Balanus spp.) Each trial utilized 5 to 8 crabs. 40 to 50 drills (height range. 15 to 75 mm), and lasted 2.0 to 3.5 hours. Various combinations of crabs (with and without attached barnacles), substrates (sand and oyster shells), oysters (live and empty shells), and in-tank locations of same were utilized during the experiments. Observations were made on the behavior of the drills in relation to the crabs and oysters. RESULTS Description of Habitat Independent drills and crabs and infested crabs shared habitats in the inlets to Horn Island lagoons and in adjacent shallow waters of Mississippi Sound (Figure 1). Those habitats consisted of ( 1 ) exposed roots of salt-marsh grasses; (2) submerged grassbeds and root-debris mats; (3) shallow depressions in sandflats and under solid jetsam (e.g., boards and timbers); (4) crevices in and ledges under peat out- croppings; (5) small clumps of oysters; and (6) large groups of oysters attached to submerged structures (e.g., wrecked vessel debris, tree stumps, etc.). The drills and crabs fre- quently made contact in those habitats, especially when the drills crawled across partially buried blue crabs or quiescent hermit crabs. Drills, crabs, and infested crabs were also trawled-up together from Dog Keys Pass at the west end of Horn Island ( Figure 1 ). MISSISSIPPI SOUND V &&M ^ 5 km i I i I i_) 89°45' GULF OF MEXICO Figure 1. Location of stations where drill-infested blue crabs and striped hermit crabs were collected. 120 Cake Mean water salinities and temperatures in the study area were 28.4 ppt (24.0 to 30.5 ppt) and 26.TC (20.0 to 28.0°C), respectively, when drill-infested blue crabs were collected (October 1980), and 21.5 ppt (20.0 to 30.5 ppt) and 19.4°C (18.5 to 27.0°C), respectively, when drill-infested hermit crabs were collected (October and November 1980). Because of visibility and collecting device limitations, all collections were made at depths of 1 m or less. The pre- dominant substrate was well sorted and rounded quartz beach sand, except in the lagoon inlets and along parts of island shorelines where relict peat outcroppings existed. Oyster Drill/Blue Crab Symbiosis Ninety-nine infested blue crabs (98 9, 1 6) were collected from four stations on Horn and East Ship islands on 2, 7, 9, and 14 October 1980. All crabs were adults, while the majority of the drills were juveniles. Mean sizes and weights of the crabs and drills are given in Table 2. The crabs carried a total of 203 drills (X = 2.0 ± 2.1 drills crab" 1 , range = 1-17) (Figure 2); the drills were attached to the carapace (200), the chelae (2), and the abdomen (1). The drill infestation mode was 1 drill crab -1 (N = 55 crabs; 55.5% of total); 22 crabs (22.2%) carried 2 drills apiece; 14 (14.1%) carried 3 drills apiece; 2 crabs each (2.2%) carried 5, 6, and 7 drills apiece, respectively; and 1 crab each (1.1%) carried 9 and 17 drills apiece, respectively (Figure 2). No drill-infested blue crabs were observed during three surveys in November (1.2, and 3 November 1980) and none was seen during numerous surveys during the summer and fall of 1981. On four occasions at two stations on Horn Island (Stn. 1.2 and 1.3, Figure 1) all blue crabs encountered were collected. Seven percent (30 of 432) of the crabs carried a total of 44 drills (Table 3). Results of regression analyses of the drill infestation and drill/crab meristic data are presented in Table 4. Only a weak correlation existed between the number of attached drills and the three crab meristics tested (carapace width, weight, and the cross product of the width and weight). In general, however, the larger the crab the larger the number of attached drills. Other Symbionts on Drill-Infested Blue Crabs The most abundant epizoan on the drill-infested blue crabs was the symbiotic acorn barnacle Chelonibia patula (Ranzani) (see Overstreet 1978, 1982). Each crab carried a mean of 81.8 ± 33.8 (12 to 287) live barnacles on its entire exoskeleton and 35.2 ± 23.7 (2 to 122) live barnacles on its carapace. The numbers of live barnacles on the entire crab and also on the carapace alone were negatively corre- lated with the number of attached drills (Table 4). Thus, the larger the number of barnacles, the smaller the number of attached drills (i.e.. barnacles reduce the space available for attaching drills). Crabs with light barnacle infestations carried 1.4 and 1.7 times as many drills as those with moderate and heavy infestations, respectively; and crabs with moderate barnacle infestations carried 1 .2 times as many drills as those with heavy infestations (Table 5). Thirteen (13.1%) of 99 drill-infested blue crabs had recently dead (empty) barnacles (C. patula) on the carapace or abdomen (X = 30 barnacles crab" 1 , range = 1 to 8, N = 39 barnacles). Two oyster drills were observed feeding on barnacles attached to two crabs during the study, but the barnacles did not appear to be an important food source for the drills in general; only 39 of 8,099 (0.5%) barnacles on the 99 drill-infested crabs were dead (empty). Two (2.2%) of the 99 drill-infested crabs also carried one specimen each of the buccinid gastropod Cantharus cancellarius (Conrad), a common omnivore of mud/sand bottoms in high salinity areas of Mississippi Sound. (Five TABLE 2. Summary of data from crabs that were infested with oyster drills at four stations on Horn and East Ship islands, Mississippi. Striped Category Blue Crabs Hermit Crabs Number drill-infested crabs 99 233 Total number drills 203 299 Mean number drills crab" 2.0+2.1 1.3 ±0.8 (Range) (1-17) (1-6) Infestation mode (drill crab ) 1 (N = 55) 1 (N= 194) Percent infested 7.10% 7.13% Mean size of crab 152 ±13 mm 82 ±32 mm (Range) (117- 183mm) (23- 159 mm) Mean weight of crab 152±38g 49.2 ±27.5g (Range) (71- 269 g) (3.2- 120 g) Mean height of drill 36.8 ±11.5 mm 18.5 ±7.6 mm (Range) (3.0- 73.8 mm) (4.2-47.3 mm) Mean weight of drill 8.9 ±9.1 g 1.4 ± 1.9 g (Range) (0.1- 53.6 g) (0.1- 14.1 g) Number barnacle- infested crabs 99 103 Total number live barnacles 8.099 896 Mean number barnacles crab 81.8 ±33.8 8.7 ±18.8 (Range) (12-287) (0- 120) Number Crepidula- infested crabs 14 106 Total number Crepidula 14 603 Mean number Crepidula crab 1.0 5.7 ±5.7 (Range) (1) (0-26) ^ata from crab subpopulations (see Table 3). 2 Blue crab (carapace width); hermit crab (gastropod shell height). 3 Blue crab plus fouling organisms; hermit crab plus gastropod shell plus fouling organisms. 4 Chelonibia patula (on blue crabs); Balanus spp. (on hermit crabs). Southern Oyster Drills Infest Macrocrustaceans 121 Figure 2. Female blue crab (Callinectes sapidus) with 16 southern oyster drills (Thais haemastoma floridana) on carapace and 1 (not visible) on chela. Crab width (carapace) and weight: 150mm and 15 1 g, respectively. Mean drill height and weight, (ranges): 31.8 mm (12.8 - 40.3) and 4.8 g (0.3 - 8.4). respectively. Total weight of all drills: 81.6 g. Infested crab was captured at the west end of Horn Island, MS, (Stn. 1.1) on 2 October 1980. TABLE 3. Incidence of oyster-drill infestation on crabs collected at four stations on Horn and East Ship islands, Mississippi. Category Blue Crabs 1 Striped Hermit Crabs 2 Total number crabs 423 1,360 Number infested crabs 30 97 Percent infested 7.10% 7.13% Number drills 44 119 Mean number drills on infested crabs 1.47 1.23 Mean number drills on all crabs 0.10 0.09 'Combined data: Chimney Lagoon, Stn. 1.2 (7 & 14 October 1980) and Ranger Lagoon, Stn. 1.3 (9 & 14 October 1980). Combined data: Chimney Lagoon. Stn. 1.2 (3 November 1980) and Ranger Lagoon. Stn. 1.3 (2 & 3 November 1980). other blue crabs in addition to the 99 drill-infested crabs were infested with specimens of C. cancelhrius only.) Fourteen (1 4.1%) and three (3.3%) of the 99 drill-infestedcrabs were also infested with single slipper shells (Crepidula spp.) and pyram shells (OJostomia impressa [Say] ), respectively. Oyster Drill/Hermit Crab Symbiosis Two hundred thirty-three drill-infested striped hermit crabs were collected at four stations on Horn and East Ship islands on 2, 7. 9, and 14 October and 1, 2, and 3 November 1980 (Table 2. Figure 3). The hermit crabs occupied the shells of 100 oyster drills (T. h. jloridana) (42.9%), 70 lightning whelks (B. contrariwri) (30.0%), 42 moon snails (P. duphcatiis) (18.0%), 17 pear whelks (B. s. plagosum) (7.3%), 2 giant eastern murexes (Murex fulvescens Sowerby) (0.9%), and 2 Florida fighting conchs (Strombus alatus Gmelin) (0.9%). The hermit crabs carried a total of 299 drills (X = 1 .3 + 0.8 drill shell" 1 , range = 1-6). The infesta- tion mode was 1 drill crab" 1 (N = 194 crabs, 833% of total); 22 crabs (9.4%) carried 2 drills apiece; 12 crabs (5.2%) carried 3 drills apiece; 3 crabs (1 .3%) carried 5 drills apiece; and 1 crab each (0.4%) carried 4 and 6 drills apiece, respec- tively (Figure 3). Mean sizes and weights of the crabs (including the shell and attached epifauna but excluding the drills) and the drills are given in Table 2. In general. the larger the size of the hermit crab shell, the greater the number of attached drills and the larger the size of the attached drills. On three occasions at two Horn Island locations (Sta. 1.2 and 1.3, Figure 1) all of the striped hermit crabs encountered were collected. Seven percent (97 of 1,360) of 122 Cake TABLE 4. Results of regression analyses on data from drill-infested crabs collected at four stations on Horn and East Ship islands, Mississippi Host Crab Species Correlations (versus number drills)* r-Value F-Value Regression Equation Callinectes sapidus Clibanarius vittatus Carapace width Total crab weight Cross product (width X weight) Number live barnacles crab Number live barnacles carapace Maximum crab (shell) dimension Weight of crab (+shell) versus weight of individual drills Cross product (size X weight) Number live barnacles shell Number live Crepidula shell - 0.0346 0.166 Y = 2.8935 - O.0O55X + 0.0326 0.102 Y = 1.7755 +0.0018X + 0.0176 0.030 Y = 1.9339 +0.0050X -0.1699 2.883** Y = 2.6023 - 0.0067X - 0.2508 6.508** Y = 2.8419 - 0.0225X + 0.1836 8.062** Y = 0.9212 +0.0044X + 0.2353 17.409** Y = 0.5851 +0.0156X + 0.2297 12.871** Y = 1.0820 +0.0426X + 0.0160 0.059 Y = 1.2797 +0.0009X + 0.1373 4.438** Y = 1.2268 +0.0218X *(Unless otherwise indicated.) **(F-Value significant at the 50% covered. the crabs carried a total of 1 19 drills (Table 3). The 97 drill- infested crabs occupied 44 shells of the oyster drill T. h. floridana (45.4%), 26 shells of the lightning whelk B. contrarium (26.8%), 20 shells of the moon snail P. duplicates (20.6%), 6 shells of the pear whelk B. s. plagosum (6.2%), and 1 shell of the fighting conch S. alatus (1 .0%). Regression analyses were performed on three host categories versus the number and/or weight of attached drills (Table 4). All three correlations were weak but positive. In general, the larger the occupied hermit crab shell, the larger the number and size of the attached drills. Several noteworthy differences existed between the two drill/crab symbioses (Table 2). Drills that were attached to blue crabs were twice the mean height as those on hermit crabs (36.8 versus 18.5 mm) and consequently, six times the mean weight (8.9 versus 1.4 g). Infested blue crabs carried more drills than hermit crabs (X= 2.0 ± 2.1 versus 1.3 ± 0.8 drills crab" 1 , respectively). The maximum number of drills carried by a blue crab (17) was 2.8 times the maximum number carried by a hermit crab (6). Drill- infested blue crabs carried approximately 9.4 times as many live acorn barnacles as drill-infested hermit crabs (81.8 versus 8.7 barnacles crab" 1 , respectively); however, the number of drills on blue crabs was inversely related to the number of barnacles, and the number of drills on hermit crabs was directly related to the number of barnacles (Table 4). Although no additional collections were made, the drill/hermit crab symbiosis continued into the fall of 1981, whereas the drill/blue crab symbiosis was not observed when spawning ceased and the onset of colder water temperatures caused blue crabs to migrate into deeper water (late fall, 1980). Other Symbionts on Drill- Infested Hermit Crabs Acorn barnacles (896 of Balanus spp.) and slipper shells (603 of Crepidula spp.) were the most abundant epifaunal organisms on drill-infested striped hermit crabs (Table 6). The mean numbers (and ranges) of barnacles and slipper shells per hermit crab shell were 8.7 ± 18.8 (1 - 120) and 5.7 ± 5.7 (1 — 26), respectively. Weak but positive correla- tions existed between the numbers of live barnacles and slipper shells on the hermit crab shells and the number of Southern Oyster Drills Infest Macrocrustaceans 123 Figure 3. Shell of lightning whelk (Busycon contrarium) inhabited by striped hermit crab (Clibanarius vittatus) and infested with Five southern oyster drills (Thais haemastoma floridana) and three spotted slipper shells (Crepidula maculosa). Height and weight of whelk shell (including attached fouling organisms, except drills): 132 mm and HOg, respectively. Mean drill height and weight, (ranges): 18.1 mm (11.0 - 31.0) and 1.1 g (0.2 - 3.7), respectively. Infested crab was captured in the inlet of Ranger Lagoon (Stn. 1.3), Horn Island, MS, on 2 November 1980. TABLE 6. Epifauna of gastropod shells occupied by drill-infested striped hermit crabs at four stations on Horn and East Ship islands, Mississippi Mean and Mean and Mean and Shell Species Occupied (N) Crepidula spp. Range Balanus spp. Range Ostrea equestris Range Busycon contrarium (70) 385 5.5, 1-26 204 2.9, 1- 55 4 0.1,(1) B. spiratum plagosum (17) 108 6.4, 2-20 29 1.7. 1- 12 8 0.5, 1-7 Murex fulvescens (2) 1 0.5, 1 1 0.5, 1 3 1.5,(3) Polinices duplicatus (42) 56 1.3,1- 7 75 1.8, 1- 20 Strombus alatus (2) 12 6.0,(6) Tfwis haemastoma floridana (100) (233) 41 603 0.4, 1- 5 2.6, 1-26 587 896 5.9, 1-120 18 33* 0.2, 1-3 Grand totals, means, ranges 3.8, 1-120 0.1. 1-7 *9 live; 4 dead with right valve drilled; 20 dead with only left valve remaining. attached drills (Table 4). In general, the greater the number of slipper shells on a drill-infested hermit crab shell, the larger the number of drills (Table 4). Thus, the presence of attached prey species is directly related to the attractiveness of the crab's shell to foraging drills. The positive correlation in the case of barnacles on hermit crab shells (as opposed to the negative correlation in the case of barnacles on blue crab carapaces) is a function of the total shell space available for foraging drills to attach. (Blue crabs heavily infested with C. patula have limited space on the carapace for drills to attach.) Several oyster drills were observed feeding on epifauna attached to hermit crab shells. One 34-mm drill had rasped a hole and was feeding on a 32-mm oyster spat (C. virginica) which was attached to the outside of a 107-mm lightning whelk shell when the host hermit crab was collected. Another 36-mm drill was rasping a hole along the margin of a 29-mm slipper shell {Crepidula plana Say) which was attached inside the aperature of a 122-mm lightning whelk shell when the host hermit crab was collected. Only 9 (27.3%) of 33 crested oysters {Ostrea equestris Say) found on drill-infested shells occupied by hermit crabs were alive; 4 shells were empty and drilled by a muricid gastropod (probably T. h. floridana); and 20 were represented only by their attached left valves. 124 Cake Additional Drill/Crab Symbioses During the study, several additional examples of oyster drill/crab symbioses were observed in the vicinity of Horn Island, MS. Several large horseshoe crabs (L. polyphemus) along the island's beach had one or two moderate-to-large oyster drills attached to their abdomens. Two additional drill-infested crab species were collected in commercial blue crab traps in deeper water (> 3 m) off the island's north beach. One stone crab {Menippe merceiiaria [Say] ; 95 mm, 184 g) carried five drills (66 to 71 m) and five live barnacles (C. patula) on its carapace. One spider crab (Libinia dubia H. Milne Edwards; 73 mm. 148 g) carried one drill (62 mm, 34 g) and 85 live and 10 dead barnacles (C. patula) on its carapace. Those symbiotic associations may have been artificially produced, however, because the crabs were confined in a trap that attracted and permitted the entry of large numbers of oyster drills. An additional oyster drill/hermit crab symbiosis was observed during the summer of 1981 in St. Joseph Bay, FL. Four red hermit crabs (Pagurus impressus [Benedict] ) occupying shells of the Florida fighting conch {S. alatus) carried six mauve-mouth oyster drills (Calotrophon ostrearum [Conrad] ) (Figure 4). The mean height of the conch shells was 86 mm (78 — 94 mm) and the mean weight of the shell plus crab was 76 g (52 - 93 g). The mean height and weight of the drills were 21.0 mm (17.5 - 23.6 mm) and 1.0 g (0.5 - 1.6 g), respectively. The conch shells were also occupied by five crested oysters (O. equestris), one of which was incompletely drilled, and numerous slipper shells (Crepidula maculosa Conrad and C. plana) of various sizes. Behavior of Oyster Drills and Hermit Crabs When given the opportunity to interact with hermit crabs in laboratory aquaria, the oyster drills behaved as follows: 1. The drills more frequently mounted hermit crab shells that had live barnacles attached, and also those that had other drills attached. When live barnacles were present. 179 (91.8%) of 195 drills mounted hermit crab shells; 124 drills (63.6%) attached if other drills were already attached to the hermit crab shells; and 55 drills (28.2%) attached when no other drills were present. When no live barnacles were present on the hermit crab shells, 1 1 drills (5.6%) attached in the presence of other drills and 5 drills (2.6%) attached in the absence of other drills. Figure 4. Shell of fighting conch (Strombus alatus) inhabited by red hermit crab (Pagurus impressus) and infested with two mauve-mouth oyster drills (Calotrophon ostrearum) and one spotted slipper shell (Crepidula maculosa). Height and weight of conch shell (including fouling organisms, except drills): 94 mm and 93 g, respectively. Drill heights and weights: 21.4 and 23.6 mm, 1.2 and 1.6 g, respectively. Slipper shell length and weight: 27.0 mm and 2.0 g, respectively. Infested crab was captured in the vicinity of Presnell's Fish Camp, St. Joseph Bay, Port St. Joe, FL, on 15 June 1981. Southern Oyster Drills Infest Macrocrustaceans 125 2. The usual drill-to-crab mounting occurred in the following manner: When the hermit crab shell was encountered, the drill raised its tentacles and siphon, extended them forward, and examined the shell; the drill then raised the forward portion of its foot, attached to the shell, and when most of the foot was connected, it pulled its body and shell onto the host's shell. Once mounted, the drill usually moved around the shell for a few minutes before becoming quiescent. The drill-to-shell mounts were relatively fast and were completed approximately 5 seconds after initial contact. 3. Most drills mounted the part of the hermit crab's shell that was initially encountered, regardless of the position and activity of the host crab's tentacles, eyes, and chelipeds. Drills were able to mount hermit crab shells that were moving when encountered. 4. Drills mounted hermit crab shells from sand and solid substrates with relative ease; 61% of the mountings were from sand and 39% were from aquarium sides, other crab shells, dead oyster shells, and pieces of brick. Drills also attached to passing crab shells while upside-down (shell aperature up) in the sand. 5. On three occasions 15 drills mounted one hermit crab shell (6 drills per hermit crab shell was the greatest infestation observed in the field). Fifteen drills mounted one crab within 50 minutes (0.3 drill min" 1 ). The greatest attach- ment rate on one hermit crab shell was 1 1 drills within 6 minutes ( 1 .8 drills min" 1 ). 6. Apparently, oyster drills were attracted to barnacles on the hermit crab shells and remained on the shell until more preferred prey such as oysters were encountered or until dislodged for other reasons. The drills dismounted from hermit crab shells onto or immediately adjacent to live oysters, but rarely remounted the crab shells once on live oysters. Twenty-four (57.1%) of 42 drills in three experiments were transported to live oysters by hermit crabs. DISCUSSION Factors Controlling Oyster Drill /Crab Symbioses Southern oyster drills were attracted to and mounted blue crabs and striped hermit crabs for at least one of the following reasons: 1 . Foraging and the presence of potential food. The presence or probable presence of acceptable prey species of the southern oyster drill appeared to be the most important controlling factorin the drill/crab symbioses. St. Amant (1938) and Butler (1953, 1954) reported that drills, especially young ones, will consume barnacles, and that drills of all sizes will prey on oysters and mussels. During this study I observed direct and indirect evidence of drill predation on epifauna of blue crab and hermit crab shells. Direct evidence included actual feeding of drills on barnacles (on blue crabs) and indirect evidence included Thais drill holes in dead oysters and in-progress drilling of a slipper shell (on hermit crab). This is the first known evidence of slipper shell predation by the southern oyster drill. All drill-infested blue crabs had live barnacles attached to their exoskeletons, but if the crab's carapace was heavily infested (> 50% coverage) with barnacles, space availability appeared to limit the number of attached drills. The numbers of barnacles and slipper shells on drill-infested hermit crabs were, however, positively correlated with the number of attached drills. Foraging drills are negatively geotactic; they will move upward when placed under water, unless they encounter acceptable food in which case they remain with the food species "indefinitely" (Butler 1979). The act of crawling up onto any solid substrate including crab shells or aquarium walls is a normal foraging behavior of oyster drills. Butler (1979) reported that the South Australian drill Thais orbita (Gmelin) moved up the walls of a container in the absence of barnacles, but remained with and fed on barnacles (Balanus glandula Darwin) when present. Whether the drill's negative geotaxis was automatic or in response to the release of metabolites by potential prey species was not determined and, in the case of relatively small substrates like crab shells, the two behaviors may be inseparable. In the case of these drill/crab symbioses, most initial attach- ments probably resulted from foraging, but were enhanced if acceptable prey species were present. 2. The presence of other attached drills (gregarious factor). Southern oyster drills are normally gregarious, especially during feeding and spawning when food by- products and pheromones, respectively, are released (St. Amant 1938, Gunter 1979). The presence of 16 drills on the carapace of one blue crab is an example of gregarious- ness (Figure 2). The 16 drills were clumped together; however, only five live barnacles were present and no feeding or spawning activities were in progress. Thus, some other factor attracted and held the drills on the crab's carapace. During the laboratory experiments, only 11 (5.6%) of 195 drills attached to crabs which had other attached drills but no live prey (barnacles). Thus, this appears to be a minor factor. When the initial field collections were made, the drill spawning season had ended and no reproductive activities were observed among the young drills used in the laboratory behavior trials. 3 . The availability of solid, stable substrates for pro- tection or shelter. Oyster drills, especially recently settled juveniles, are normally associated with and attached to firm substrates such as oyster shells, rocks, and submerged objects (timbers, stumps, etc.) for food (epifauna), protection (from predators), and shelter (from currents, waves, and potential 126 Cake loss of attachment and subsequent abrasion, burial, or predation). Because of the dearth of such substrates in the vicinity of the barrier islands, the attachment of young drills to the crab shells may have been a defensive as well as a foraging behavior. Small drills which were attached to crab shells had a lower probability of being consumed by fish and crab predators than unattached drills. Although striped hermit crabs will kill oyster drills (Gunter 1979), they are unlikely to leave the protection of their gastropod shell to attack attached drills; however, small drills within the aperature of the hermit crab shell may be subjected to such predation. Blue crabs will remove attached drills if they are within reach of the chelae and the crabs can dis- lodge attached drills by "rubbing" them against aquarium walls. In either case, protection is lost, and the drills may be subject to predation. 4. The presence of eggs on ovigerous blue crabs. Eggs or their by-products which are released from ovigerous crabs may biochemically attract foraging drills. Sixty-four (65.3%) of the 98 drill-infested females were ovigerous, 26 (26.5%) were "spent" (the zoeae had recently hatched), and the remaining 8 (8.2%) had not yet spawned. The probability of drill infestation is greater when the females are ovigerous than when they are not. Of 55 females infested with a single drill. 31 (56.4%) were ovigerous; 16 (72.7%) of 22 females with two drills were ovigerous; 11 (78.6%) of 14 females with three drills were ovigerous; and 6 (75.0%) of 8 females with five or more drills were ovigerous. In a related study of drill damage to blue crabs in commercial traps north of Horn Island. I observed several drills feeding on the "sponge" of ovigerous females. The highly protrusile proboscis of oyster drills permits them to rasp and feed on crab eggs while attached to the carapace and abdomen of ovigerous females. 5. The presence of biochemical stimulants or by- products from wounded or moribund blue crabs. Wounded, moribund, or dead crabs, especially blue crabs, may represent a potential food source for the other- wise carnivorous drills. On several occasions in November 1980, when large numbers of spawned-out females were dead or dying, a few were stranded on the beach at low tide with drills still attached to their carapaces. Were the drills waiting for passive transportation to continue or were they waiting for a meal? During a related study of drill damage to commercially trapped blue crabs north of Horn Island in the spring of 1981. I observed that drills pene- trated the crabs' exoskeletons via wound holes, autotomized appendage stumps, thin appendage joints, and occasionally via holes drilled in the carapace. The drills also used their protrusile proboscis to penetrate the thin membranes at the bases of the gills within the branchial chambers to gam access to thoracic muscle tissues. No such crab predation was observed during the present study of 203 drills that were attached to 99 live blue crabs. 6. Increased random attachment to available substrates by an exploding drill population. Environmental conditions near the barrier islands may have promoted the drill/crab symbioses. Extended drought conditions during 1979-1981 increased salinities in Mississippi Sound and permitted the settlement of relatively large numbers of young drills in normally marginal habitats. Those drills became abundant in habitats con- taining barnacles, mussels, and oysters around the barrier islands. The sheer abundance of the drills and their random attachment to all firm objects may account for their presence on crabs. In those instances when infestation prevalence was determined, blue crabs and hermit crabs exhibited the same prevalence (7.17c). Although I made no attempt to document the presence of drills on flotsam and jetsam around the barrier islands, Federighi (1931) and L. A. Stauber (in Carriker 1955) reported that young oyster drills (U. cinerea) were distributed by attaching to floating algae as well as to other flotsam and jetsam. I routinely observed drills on submerged planks and other discarded items in barrier island lagoons during this study. 7 . In response to a programmed symbiotic phenomenon. If the drill/crab symbioses are as well established as shown by this and other studies (Table 1). then muricid drills may be programmed to seek crabs for their trans- portation potential. The availability of transportation to unpopulated areas, especially those with abundant food supplies, may foster the symbiotic relationship. Probable Role of Macrocrustaceans in the Migration of Southern Oyster Drills Along the Gulf of Mexico coast, blue crabs and striped hermit crabs are common inhabitants of estuaries and oyster reefs (McDonald 1940, Galtsoff 1964, McClellan 1965, Fotheringham 1976, Bahr and Lanier 1981) where they tolerate a wide range of water salinities and tempera- tures (Christmas and Langley 1973). Blue crabs move about extensively (Darnell 1959) and can travel as much as 1.6 to 2.0 km day" 1 (H. Perry, Gulf Coast Research Labora- tory, Ocean Springs, MS, and M. Oesterling, Virginia Institute of Marine Science, Gloucester Point, VA, unpub- lished data). Thus, they could carry oyster drills from barrier island habitats to inshore oyster reefs within a week. In contrast, striped hermit crabs travel much less and usually remain within the littoral and shallow, sublittoral zones (Fotheringham 1975). They travel as much as 156 m day" 1 (Hazlett 1981) and. thus, could carry oyster drills (to nearby oyster reef) but not as far as blue crabs. On the other hand, oyster drills do not migrate (Butler 1953); unless carried by crabs or other means, the drills probably remain within the vicinity where they originally settled. Southern Oyster Drills Infest Macrocrustaceans 127 In Mississippi Sound, at least three species of crabs (blue, striped hermit, and horseshoe) were observed trans- porting drills during this study. Thus, the drill/crab symbioses may be important in distributing juvenile and young adult drills. The quantity of drills transported by this means is small when compared with the number of larval drills that are distributed in the plankton to high salinity areas following reproduction. Nevertheless, the crabs might carry drills to areas where currents do not carry larval drills, and they can transport drills throughout the year. Along the Atlantic and Gulf coasts of the United States, at least four species of muricid oyster drills [Calotrophon ostreanim, Eupleura caudata, Thais haemastoma floridana, and Urosalpinx cinerea) and five species of arthropods ( Callinectes sapidus, Clibanarius vittatus, Pagiims impressus, Pagitrus pollicaris, and Limulus polyphemus) (Table 1) participate in drill/crab symbioses. Although relatively few reports about these symbioses appear in the literature. I suspect that they are common and have an important role in extending the distributions of oyster drills. MacKenzie (1962) concluded that horseshoe crabs (L. polyphemus) were important distributors of Atlantic coast oyster drills (E. caudata and U. cinerea) throughout Long Island Sound and perhaps beyond. Harold Haskin (in Carriker 1955) concluded that hermit crabs (P. pollicaris) played an important role in the distribution and migration of Atlantic oyster drills (U. cinerea) in Delaware Bay. The distributory effects of these drill/crab symbioses may be somewhat negated, however, because blue crabs and striped hermit crabs prey on small oyster drills. Blue crabs in Horn Island lagoons (pers. observ.) and in nearby Lake Pontchartrain, LA (Darnell 1958), readily consume small gastropods which they ingest whole. Gunter (1979) reported that striped hermit crabs killed southern oyster drills in Apalachicola Bay, FL, by pinching their tentacles until they bled to death; thereafter, the crabs pulled the drills from their shells, consumed the flesh and occupied the newly emptied shell. Of 1,360 striped hermit crabs collected during November 1980, from the Horn Island lagoons (Stn. 1.2 and 1.3), 825 (60.7%) occupied shells of the southern oyster drill. (The next most frequently occupied shell was that of the moon snail P. duplicatus [23.0%].) Rudloe (1971) documented the attack of a striped hermit crab on a live pear whelk Busycon spiratum (Lamarck) in which the crab killed the whelk with its chelae, extracted and consumed the flesh, and occupied the new shell briefly before returning to its "old" shell. Drill/Crab Symbioses: Commensalism or Phoresis? Cheng (1967) discussed the importance of commensalism and phoresis in the marine environment and pointed out that the two symbioses differed primarily with regard to nutritional aspects. He defined "commensalism" as a more or less intimate relationship in which the commensal (in this case the drill) generally derives physical shelter from the host (the crab), is nourished on food organisms that are associated with but not a part of the host (barnacles, oysters, slipper shells), and is not metabolically dependent on the host. Literally, commensalism means "eating at the same table." It is a loose type of nonobligatory relationship (Cheng 1967). He defined "phoresis" as a loose, nonobliga- tory relationship in which one species, the host (crab), merely provides shelter, support, or transport for the other species, the phoront (drill). Metabolic dependency is not involved. In a more restrictive definition, Cheng (1973) considered phoresis as an association in which the smaller of the two species, the phoront. is mechanically carried in or on the larger species, the host, and no metabolic inter- action or dependency occurs. It does not involve a sharing of food as does commensalism. According to Cheng's definitions of phoresis, those animals, commonly referred to as being epizootic or epizoic. are engaged in phoretic associations with their hosts. The symbiotic relationships between southern oyster drills and crabs in Mississippi Sound share components of commensalism and phoresis. The two symbioses can overlap according to Cheng (1967). and this is apparently the case with the drill/crab associations described herein. In a limited sense, the drills derive passive transport (cf., phoresis), shelter (cf., phoresis and commensalism), albeit negligible, and support (cf., phoresis) from the crab hosts. The drills derive nutritional benefit in a nonobligatory fashion (cf., commensalism) from the epifauna on the crab hosts, but the drills do not "share" those prey species in the traditional sense (cf.. commensalism) such as do hermit crabs and attached sea anemones. On the other hand, if drills consume eggs from ovigerous blue crab females or attack and kill free-living blue crabs, then the relationship can be considered predatory. If food availability and utilization are the primary controlling factors in the drill/crab symbioses, the relation- ships should be categorized as modified forms of commens- alism. On the other hand, if, as MacKenzie (1962) observed, the drills primarily derive passive transport from the crabs, the relationships should be categorized as modified forms of phoresis. Cheng (1967) noted a considerable overlapping between commensalism and phoresis, yet he provided no examples of symbioses that shared characteristics of both. He suggested that one type of symbiosis may evolve into another. In that case, neither the commensalitic nor phoretic behavior of the two symbionts appears to be dominating. I suggest, therefore, that the commensalistic components probably evolved first and the phoretic components occurred secondarily. The drill/crab symbioses in Mississippi waters appear to be primarily commensalistic and secondarily phoretic, and perhaps should be defined as phoretic commensalism. Of the seven controlling factors discussed at the beginning of this section, foraging and the presence of attached prey species (on blue and hermit crab shells) 128 Cake and egg masses (on ovigerous blue crabs) probably initiated the relationships', food availability, gregariousness, and sub- strate stability (protection and/or shelter) probably pro- longed them; and the foraging for new food sources or dislodgment probably terminated the relationships. The drill's "predatory" behavior toward wounded or moribund blue crabs appeared to be an expression of the drill's normal opportunistic feeding, especially when it occurred in commercial crab traps. The possibilities of random attachment to solid substrates and "programmed" trans- portation attempts appeared to be the least plausible controlling factors. ACKNOWLEDGMENTS I gratefully acknowledge the financial and logistical support of the Gulf Coast Research Laboratory and the cooperation and assistance provided by officials of the Gulf Islands National Seashore (National Park Service). Roger Jennings and Rick Sherrard assisted with field collections and laboratory measurements; Gary Licht conducted prelim- inary studies of oyster drills and hermit crabs: and Vincent Smith provided occasional boat transportation and infested specimens from his commercial crab traps. Valarie Hebert provided statistical and computer assistance and Lucia O'Tooleand Cindy Dickens typed the manuscript drafts. REFERENCES CITED Abbott, R. T. 1974. American Seashells. Tlic Marine Mollusca of the Atlantic and Pacific Coasts of North America. 2nd ed. New York, NY: Van Nostrand Reinhold Co. 663 p. Bahr, L. M. & W. P. Lanier. 1981. The ecology of intertidal oyster reefs of the south Atlantic coast: A community profile. U.S. Fish Wildl. Sen: Biol. Sen: Program FWS/OBS-81/15 : 105 p. Breithaupt, R. L. & R. J. Dugas. 1979. A study of the southern oyster drill [Tliais haemastoma) distribution and density on the oyster seed grounds. La. Wildl. Fish. Comm. Tech. Bull. 30:1-20. Burkenroad, M. D. 1931. Notes on the Louisiana conch, Tliais haemastoma Linn., in its relation to the oyster. Ostrea virginica. Ecology 12<4):656-664. Butler, A. J. 1979. Relationships between height on the shore and size distributions of Tliais spp. (Gastropoda; Muricidae). /. Exp. Mar. Biol. Ecol. 41:163-194. Butler, P. A. 1953. The southern oyster drill. Natl. Shellfish. Assoc, Conven.Pap. 1953:67-75. . 1954. Summary of our knowledge of the oyster in the Gulf of Mexico. U.S. Fish Wildl. Serv. Fish. Bull. 89:479-489. Carriker, M. R. 1955. Critical review of biology and control of oyster drills Urosalpinx and Eupleura. U.S. Fish Wildl. Serv. Spec. Sci. Rep. Fish. 148:1-150. Chapman, C. R. 1955. Feeding habits of the southern oyster drill, Tliais haemastoma. Proc. Natl. Shellfish. Assoc. 46:169-176. . 1958. Oyster drill [Tliais haemastoma) predation in Mississippi Sound. Proc. Natl. Shellfish. Assoc. 49:87-97. Cheng, T. C. 1967. Marine molluscs as hosts for symbioses. Adv. Mar. Biol. 5:1-424. . 1973. General Parasitology. New York, NY: Academic Press, Inc. 965 p. Christmas, J. Y. & W. Langley. 1973. Estuarine invertebrates, Mississippi. Christmas, J. Y., ed.. Cooperative Gulf of Mexico Estuarine Inventory and Study. Mississippi. Gulf Coast Research Laboratory, Ocean Springs, MS. Section 4:255-319. Darnell, R. M. 1958. Food habits of fishes and larger invertebrates of Lake Pontchartrain. Louisiana, an estuarine community. Publ. Inst. Mar. Sci. Univ. Tex. 5:353-416. . 1959. Studies of the life history of the blue crab iCallinectes sapidus Rathbun) in Louisiana waters. Trans. Amer. Fish.Soc. 88(41:294-304. Federighi, Henry. 1931. Studies of the oyster drill [Urosalpinx cinerea Say). U.S. Bur. Comm. Fish. Bull. 47:85-115. Fotheringham, Nick. 1975. Structure of seasonal migrations of the littoral hermit crab Clihanarius vittatus (Bosc). J. Exp. Mar. Biol. Ecol. 18:47-53. . 1976. Population consequences of shell utilization by herit crabs. Ecology 57(3):570-578. Galtsoff, P. S. 1964. The American oyster Crassostrea virginica (Gmelin). U.S. Fish Wildl. Serv. Fish. Bull. 64:1-480. Gunter. G. 1953. The relationship of the Bonnet Carre spillway to oyster beds in Mississippi Sound and the "Louisiana Marsh." with a report on the 1950 opening. Publ. Inst. Mar. Sci. Univ. Tex. 3(1):17-71. . 1979. Studies of the southern oyster borer, Thais haemastoma. Gulf Res. Rep. 6(31:249-260. Hazlett, B. A. 1981. Daily movements of the hermit crab Clihanarius vittatus. Bull. Mar. Sci. 3 1 ( 1 ) : 1 77— 1 83. Hofstetter, R. P. 1959. The Texas oyster fishery. Tex. Parks Wildl. Dep. Bull. 40:1-39. MacKenzie, C. L., Jr. 1962. Transportation of oyster drills by horseshoe "crabs." Science 137(35231:36—37. May, E. B. 1971. A survey of the oyster and oyster shell resources of Alabama. Ala. Mar. Resour. Bull. 4:1-53. & D. G. Bland. 1970. Survival of young oysters in areas of different salinity in Mobile Bay. Proc Southeast. Assoc. Game Fish Comm. 23:519-521. Menzel. R. W. & S. H. Hopkins. 1955. Crabs as predators of oysters in Louisiana. Proc. Natl. Shellfish. Assoc. 47:177- 184. Menzel, R. W., N. C. Hulings & R. R. Hathaway. 1957. Causes of oyster depletion in St. Vincent Bar. Apalachicola Bay, Florida. Proc. Natl. Shellfish. Assoc. 48:66-71. . 1966. Oyster abundance in Apalachicola Bay. Florida, in relation to biotic associations influenced by salinity and other factors. Gulf Res. Rep. 2(2):73-96. McClellan, H. A. 1965. A survey of the ectoparasites and commensals of the American oyster, Crassostrea virginica. in Mobile Bay, Alabama. Tuscaloosa, AL: Univ. of Alabama. 39 p. Thesis. McConnell, J. N. 1953. The Gulf coast conch, Thais haemastoma. Natl. Shellfish. Assoc. Conven.Pap. 1953:76-77. McDonald, H. J. 1940. Animal associates of oysters at Beaufort, North Carolina. Durham, NC: Duke Univ. 49 p. Thesis. Overstreet, R. M. 1978. Marine Maladies? Worms, Germs, and Other Symbionts from the Northern Gulf of Mexico. Ocean Springs, MS: Mississippi-Alabama Sea Grant Consortium. MASGP-78- 021:140 p. . 1982. Metazoan symbionts of the blue crab. Perry, H. M. and W. A. Van Engel, eds.. Proceedings of the Blue Crab Collo- quium. 1979 October 16-19; Biloxi, MS: Gulf States Mar. Fish. Comm. Ann. Meet. Available from: GSMFC, Ocean Springs, MS. 7:81-87. Pollard, J. F. 1973. Experiments to re-establish historic oyster seed grounds and to control the southern oyster drilL La. Wildl. Fish. Comm. Tech. Bull. 6:1-82. Rudloe, J. 1971. The Erotic Ocean. New York, NY: World Publishing Co. 448 p. St. Amant, L. S. 1938. Studies on the distribution of the Louisiana oyster drill, Tliais floridana haysae Clench. Baton Rouge, LA: Louisiana State Univ. 108 p. Thesis. Williams, A. B. 1965. Marine decapod crustaceans of the Carolinas. U.S. Fish Wildl. Serv. Fish. Bull. 65(11:1-298. Journal of Shellfish Research, Vol. 3, No. 2, 129-134, 1983. PREDATION ON AMERICAN OYSTERS (CRASSOSTREA VIRGINICA [GMELIN] ) BY AMERICAN LOBSTERS (HOMARUS AMERICAN VS MILNE-EDWARDS), ROCK CRABS (CANCER IRRORATUS SAY), AND MUD CRABS (NEOPANOPE SAYI [SMITH] ) ROBERT W. ELNER 1 AND RENE E. LAVOIE 2 1 Department of Fisheries and Oceans Invertebrates and Marine Plants Division Biological Station St. Andrews, New Brunswick, Canada EOG 2X0 2 Department of Fisheries and Oceans Fisheries Research Branch Halifax, Nova Scotia. Canada B3J 2S 7 ABSTRACT Predation on the American oyster Crassostrea virginica (Gmelin) by the American lobster Homarus americanus Milne-Edwards, the rock crab Cancer irroratus Say, and the mud crab Neopanope sayi (Smith) was studied in the laboratory. When provided with a range of oysters from 10 to 35 mm shell length (SL), lobsters (55-98 mm carapace length [CL)) and rock crabs (32-107 mm carapace width [CW]) could all open oysters of 25 to 30 mm SL, but they usually selected smaller oysters. Oysters of 30 to 35 mm SL appeared to be a critical size as they were rarely preyed upon by either lobsters or rock crabs. Although all lobsters had a similar broad preference for oysters of 10 to 25 mm SL, larger rock crabs preferred larger oysters than smaller rock crabs. Predation rates were variable among individuals but were generally faster in larger than in smaller lobsters and rock crabs. Maximum mean lobster and rock crab predation rates were 28.0 and 4.5 oysters • predator ■ day , respectively. Groups of rock crabs (32-58 mm CW) and mud crabs ( 14-23 mm CW) that fed on oysters of 2 to 9 mm SL which were attached to spat collectors, averaged 0.59 and 0.44 oyster • crab -1 ■ day , respectively. Larger rock crabs (50-76 mm CW). foraging on spat collectors, consumed 1.26 oysters * crab day . No discernible sexual differences existed in either oyster size selection or predation rates for the lobsters and crabs. Patterns of oyster destruction were predator-specific. Lobsters opened oysters by indiscriminate crushing, whereas rock crabs exploited weak spots around the shell margin and thin areas in the cup valve. Isolating oyster < 30 to 35 mm SL from decapod predators and barring rock crabs and mud crabs from spat collectors would reduce oyster mortality. KEY WORDS: Oysters, Crassostrea virginica, lobsters, crabs, predation. selectivity, mariculture INTRODUCTION did not consider opening techniques or size-specific predation rates. The American oyster Crassostrea virginica (Gmelin) is In Caraquet Bay, NB, culturists collect oyster spat on cultivated extensively on the eastern coast of North America "chinese-hat" collectors, so-called because of the conical and has economic importance in many communities. Studies collector plates, the 0.33-m diameter plates are stacked in of oysters have identified cancrid. portunid, and xanthid bundles of 12 with a gap of 30 mm between adjacent crabs (Menzel and Hopkins 1955, McDermott 1960. Krantz P^tes. The bundles are suspended, from rafts or fences, and Chamberlin 1978, MacKenzie 1981). oyster drills 0.3 m above the sea bottom. Culturists detach oyster spat (Galtsoff 1964), and starfish (Galtsoff 1964) as the principle from the collectors at a shell length of about 23-25 mm predators. (See MacKenzte [1981] for a comprehensive and use various P rotective reari "g techniques to grow the oysters to seed. The oysters are exposed to lobsters and review of oyster mortality factors.) We found in our laboratory study that American lobsters rock crabs when relaid as seed onto growing areas at shell lengths from 25 to 60 mm. (Homarus americanus Mhne-Edwards) are also predators of We conducted a laboratory investigation on the shell- oysters. Previously it was shown that lobsters prey on op ening behavior and feeding rates of lobsters and rock several types of molluscs and other invertebrates (Ennis crabs on unattached oysters from Caraquet Bay. Predation 1973, Elner and Jamieson 1979, Elner 1980, Scarratt 1980). by rock crabs and mud crabs {Neopanope sayi [Smith] ) on The rock crab Cancer irroratus Say preys on oysters in oysters attached to chinese-hat collectors was also Caraquet Bay, New Brunswick, Canada (R. W. Elner, considered. Both crabs are abundant on the collectors unpublished data), and on other molluscs and invertebrates suspended in the bay and cause oyster mortality. We were in other areas (Scarratt and Lowe 1972, Elner and Jamieson particularly interested in obtaining information about the 1979, Elner 1980). The only previous quantitative investi- predators that could be used to improve culture strategy gation of oyster predation by rock crabs (MacKenzie 1981 ) and ultimately increase oyster yields. 129 130 Elner and Lavoie MATERIALS AND METHODS Lobsters and large rock crabs were captured by otter trawl in Passamaquoddy Bay, New Brunswick, Canada, near the mouth of the Bay of Fundy . Small rock crabs, mud crabs, and oysters were collected from commercial oyster beds in Caraquet Bay, NB. The oysters had been grown on the cement substrate which coats the chinese-hat collectors, and thus had cement cultch bonded in their left (cupped) shell valves. During experiments, the predators and oysters were kept in 0.35- X 0.5-m glass aquaria filled to a depth of 0.25 m with running seawater. The seawater temperature was 13 ± 1°C and the salinity range was 29 to 32 ppt throughout our investigations. Lobster size was measured as carapace length (CL) from the posterior edge of an eye socket to the posterior edge of the carapace, parallel to the longitudinal axis. The sizes of rock crabs and mud crabs were determined by measuring carapace width (CW) between the tips of the distal marginal teeth. Maximum shell dimension ("length") (SL) was measured to express oyster size. All measurements are accurate to the nearest 0.1 mm. Predators were held without food for 2 days before feeding experiments to standardize hunger levels. Only uninjured, apparently healthy predators and oysters were used. Predation techniques used by lobsters from 55 to 98 mm CL and rock crabs from 32 to 107 mm CW on oysters of 5 to 35 mm SL were observed. We also observed the tech- niques used by small roc