PROCEEDINGS NATIONAL SHELLFISHERIES ASSOCIATION 0 Volume 67 EDITORIAL BOARD EDITOR Dr. Robert E. Hillman Battelle-Columbus Laboratories William F. Clapp Laboratories, Inc. P.O. Drawer AH Duxbury, Massachusetts 02332 ASSOCIATE EDITORS Dr. Jay D. Andrews Virginia Institute of Marine Science Gloucester Point, Virginia 23062 Dr. Anthony Calabrese National Oceanic and Atmospheric Administration National Marine Fisheries Service Biological Laboratory Milford, Connecticut 06460 Dr. Kenneth Chew University of Washington College of Fisheries Seattle, Washington 98105 Dr. Thomas W. Duke U.S. Environmental Protection Agency Gulf Breeze Laboratory Sabine Island Gulf Breeze, Florida 32561 Dr. Paul A. Haefner, Jr. Department of Biology Rochester Institute of Technology 1 Lomb Memorial Drive Rochester, New York 14623 Dr. Herbert Hidu The Ira C. Darling Center for Research, Teaching and Service University of Maine The Marine Laboratory Walpole, Maine 04573 Dr. E. S. Iverson University of Miami School of Marine and Atmospheric Science Miami, Florida 33149 Dr. Stanley C. Katkansky Portland General Electric Co. Electric Building 621 S.W.Adler Street Portland, Oregon 97205 . Dr. Thomas L. Linton Office of Marine Affairs 410 Oberlin Road Raleigh, North Carolina 27605 Dr. J. C. Medcof P.O. Box 83 St. Andrews, New Brunswick Canada EOG 2XO Dr. Gilbert Pauley Washington Cooperative Fishery Unit College of Fisheries University of Washington Seattle, Washington 98195 Dr. Daniel B. Quayle Fisheries Research Board of Canada Nanaimo, B. C, Canada Dr. Aaron Rosenfield National Oceanic and Atmospheric Administration National Marine Fisheries Service Biological Laboratory Oxford, Maryland 21654 Dr. Saul B. Saila University of Rhode Island Narragansett Marine Laboratory Kingston, Rhode Island 02881 Dr. Roland L. Wigley National Oceanic and Atmospheric Administration National Marine Fisheries Service Biological Laboratory Woods Hole, Massachusetts 02543 PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION OFFICIAL PUBLICATION OF THE NATIONAL SHELLFISHERIES ASSOCIATION; AN ANNUAL JOURNAL DEVOTED TO SHELLFISHERY BIOLOGY VOLUME 67 Published for the National Sliellfisheries Association, Inc. by The Memorial Press Group, Plymouth, Massachusetts JUNE 1977 PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION CONTENTS Volume 67 — June 1977 List of Abstracts by Author of Technical Papers Presented at 1976 NSA Annual Meeting, Miami, Florida v Michael Castagna and John N. Kraeuter Mercenaria Culture Using Stone Aggregate for Predator Selection l -\ Robert E. Malouf and Wilbur P. Breese Food Consumption and Growth of Larvae of the Pacific Oyster, Crassostrea gigas (Thunberg), in a Constant Flow Rearing System 7-4- Kwang H. Im and Don Langmo Economic Analysis of Producing Pacific Oyster Seed in Hatcheries 17 -v Joseph G. Loesch and John W. Ropes Assessment of Surf Clam Stocks in Nearshore Waters along the Delmarva Peninsula and in the Virginia Fishery South of Cape Henry 29 Victor G.Burrell, Jr. Mortalities of Oysters and Hard Clams Associated with Heavy Runoff in the Santee River System, South Carolina in the Spring of 1975 35 Raymond J. Rhodes, Willis J. Keith, Peter J. Eldridge and Victor G. Burrell, Jr. An Empirical Evaluation of the Leslie-DeLury Method Applied to Estimating Hard Clam, Mercenaria mercenaria, Abundance in the Santee River, South Carolina 44 George E. Krantz and Donald W. Meritt An Analysis of Trends in Oyster Spat Set in the Maryland Portion of the Chesapeake Bay 53+ Gary H. Cole, Ronald L. Copp, and David C. Cooper Estimation of Lobster Population Size at Millstone Point, Connecticut, By Mark-Recapture Techniques, 1975-1976 60 in G.P. Ennis Determination of Shell Condition in Lobsters (Homarus americanus) by Means of External Macroscopic Examination 67 + T.O Thatcher An Effect of Chlorination on the Hatching of Coon Stripe Shrimp Eggs: So What? 71 -^ Herbert Hidu, Mark S. Richmond, and Allison H. Price, II Morphological variability in Sea Scallops, Placopecten magellanicus (Gmelin) Related to Meat Yields 75 John E. Huguenin The Reluctance of the Oyster Drill (Urosalpinx cinerea) to Cross Metallic Copper 80-f- Richard B. Nickerson A Study of the Littleneck Clam (Protothaca staminea Conrad) and the Butter Clam (Saxidomus giganteus Deshayes) in a Habitat Permitting Coexistence, Prince William Sound, Alaska 85 Richard A. Macintosh and A.J. Paul The Relation of Shell Length to Total Weight, Tissue Weight, Edible-Meat-Weight, and Reproductive Organ Weight of the Gastropods Neptunea heros, N. lyrata, N. pribiloffensis, and U. ventricosa of the Eastern Bering Sea 103 + Clyde L. MacKenzie, Jr. Sea Anemone Predation of Larval Oysters in Chesapeake Bay (Maryland) 113 1 Abstracts: NSA Annual Meeting NSA Pacific Coast Section 118 IV LIST OF ABSTRACTS BY AUTHOR OF TECHNICAL PAPERS PRESENTED AT THE 1976 NSA ANNUAL MEETING MIAMI BEACH, FLORIDA Aven M. Anderson and Michael D. Bilger Growth of Marked Rangia Clams in the Potomac River 118 Kenneth K. Chew Manila Clam Reseeding Prospects in Washington State 118 Dexter S. Haven Growth and Survival of Cultchless Spat Planted in Nomini and Lower Machodoc Creeks in 1973 118 Jonathon P. Houghton Age and Growth of Protothaca staminea (Conrad) and Saxidomus giganteus (Deshayes) at Kiket Island, Washington 119 Rene E. Lavoie The Systematic Identification of Commercially Useable Sources of Natural Oyster Spat in Eastern Canada 119 Richard A. Lutz Annual Structural Changes in the Inner Shell Layer of Geukensia ( = Modiolus) demissa 120 (Hyde L. MacKenzie, Jr. and Arthur S. Merrill Observations of Sea Scallop Stocks on Georges Bank and Middle Atlantic Shelf in 1975 120 John J. Manzi and Victor G. Burrell, Jr. A Comparison of Growth and Survival of Subtidal Crassostrea virginica (Gmelin) in Four South Carolina Salt Marsh Impoundments 120 Sara V. Otto and George E. Krantz An epizootic of "Dermo" Disease in Oysters in the Maryland Portion of the Chesapeake Bay 121 Patrick R. Parrish, James M. Patrick, and Jerald Forester Effects of Three Toxicants in Oysters (Crassostrea virginica) Exposed Continuously for Two Years 121 Anthony J. Provenzano, Jr. and Joseph W. Goy Evaluation of a Sulphate Lake Strain of Artemia as a Food for the Grass Shrimp, Palaemonetes pugio 122 v Raymond J. Rhodes, Willis J. Keith, and V.G. Burrell, Jr. South Carolina's Hydraulic Escalator Harvester Fishery 122 Oswald A. Roels, Thomas E. Dorsey, Kenneth Rodde, Scott Lawrence, Richard Lyon, and Paul W. McDonald The Efficiency of "Nitrogen" Transfer in Artificial Upwelling Mariculture. I. The Conversion of Deep-Sea Water Dissolved Nitrate to Phytoplankton Protein to Tapes semidecussata Meat-Protein in a Fully Managed System 123 Paul A. Sandifer and Theodore I.J. Smith Preliminary Observations on a Short-Claw Growth Form of the Malaysian Prawn, Macrobrachium rosenbergii (DeMan) 123 Virginia R. Van Sickle, Barney B. Barrett and Ted B. Ford Barataria Basin: Salinity Changes and Oyster Distribution 124 N.T. Windsor and J.L. Dupuy Some Spatial and Nutritional Effects on the Culturing of the Larvae of Crassostrea virginica, the American Oyster 124 ABSTRACTS OF NSA PACIFIC COAST SECTION John (Hal) Beattie Development of University of Washington's Experimental Oyster Hatchery 125 W.P. Breese, P.L. Donaghay, R.E. Malouf, and L.F. Small Algal Chemostats and Oyster Larvae 125 Linda Chaves-Michael Mussel Studies in Seabeck Bay and Clam Bay 125 Kenneth K. Chew Preliminary Findings on a Recent Summer Kill Of Pacific Oysters 126 Thomas F. Gaumer Recent Clam Studies in Oregon's Estuaries 126 W. S. Grant, L. Bartlett and F.M. Utter Biochemical Genetic Identification of Species and Hybrids of the Bering Sea Tanner Crab, Chionoecetes bairdi and C. opilio Richard S. Johnston and Larry O. Rogers The Economic Feasibility of Brine Shrimp Culture Under Semi-controlled Conditions 127 Richard S. Johnston and Nelson A. Swartz The Demand for Pacific Oysters: A Preliminary Report 127 Chris Jones Sea Urchins — Washington's Newest Fishery Presents Some Prickly Problems 128 Mark Miller, Charles D. Magoon, Lynn Goodwin and Chris Jones Manila Clam Reseeding Studies in Puget Sound 128 Herb Tegelberg and Ron Arthur Dungeness Crab Mortality from Channel Maintenance Dredging in Gray's Harbor, Washington 128 Ronald E. Westley Current Status of Shellfish Harvest Problems in Washington State 129 vn Proceedings of the National Shellfisheries Association Volume 67 — 1977 MERCENARIA CULTURE USING STONE AGGREGATE FOR PREDATOR PROTECTION1 Michael Castagna and Joint N. Kraeuter VIRGINIA INSTITUTE OF MARINE SCIENCE WACHAPREAGUE, VIRGINIA 23480 ABSTRACT A low technology method utilizing hatchery-raised seed clams and field grow-out techniques is presented. This technique appears to be economically feasible and can be carried out by non-technical personnel with a minimum of training. The hatchery uses the Wells-Clancy (centrifuged, incubated seawater) method for raising food for the larval clams. The larvae set in 8 - 10 days and the seed are supplied with flowing seawater until they grow to 2 mm. The 2 mm seed were placed in nursery plots and protected from predation by a layer of gravel or crushed stone aggregate. Movement of the small clams was prevented by a system of baffles which enclosed and dissected the nursery areas. Field survival of a 1975 test group of 600,000 clams ap- proached 75% . Costs of raising the clams for the first year are included. INTRODUCTION The hard clam or quahog, Mercenaria mercenaria (Linne, 1758), is a commercially im- portant bivalve species along the Atlantic coast of the United States. Larval culture of Mercenaria has been carried out in laboratories (Calabrese and Davis, 1966; Chanley, 1961; Chanley and An- drews, 1971; Davis, 1958; Davis and Calabrese, 1964; Loosanoff, 1937, 1954; Loosanoff and Davis, 1950, 1963; Loosanoff, et al, 1951; and Wells, 1924, 1926) and a series of growth rate data for hybrid and natural populations in a number of geographical areas have been summarized by Ansell (1968). Most other work of commercial in- terest centers on enumeration of local stocks or the examination of habitat variables (Kerswill, 1941; Loesch and Haven, 1973; Pratt, 1953; Pratt and Campbell, 1956). In spite of this extensive knowledge, no economic culture system for quahogs has been 1 Contribution No. 707 from Virginia Institute of Marir Science. developed. In order to be economically feasible and competitive with wild harvest, mariculture of M. mercenaria must be based on simple inexpen- sive hatchery and culture techniques. In order to grow clams inexpensively, it appears that they must be grown in natural waters for at least part of their lives and harvested when they reach the most desirable size. Control of the seed population during field growth is critical. Predation and loss of small clams that are washed out of the substrate by currents and/or wave action are the most serious problems in field maintenance of clams. The method described here eliminated or con- trolled many of these problems. Seed clams as small as 2 mm were successfully reared in prepared beds and predation was controlled to ac- ceptable levels. This simplified method appears to be adaptable to culture of other infaunal species, if appropriate alterations are incorporated for local conditions. We have chosen to explain in detail the equipment and methods utilized because there are no literature sources we are aware of that pro- vide this information. The methods adopted and described provided a i-4 M. CASTAGNA AND J. N. KRAEUTER compromise between cost-effective methods and available technology. Description of Area The hatchery and grow-out facilities were located on the eastern side of the Delmarva Penin- sula in Wachapreague, Virginia. The nursery area was located in Bradford's Bay about \ 4 mile from the hatchery. This small shallow bay was part of a logoon system separated from the Atlantic by a series of barrier islands on the east and the penin- sula to the west (Newman and Munsart, 1968!. The bay has a muddy substrate fringed with Spai- tina alterniflora-dominated salt marshes and a mean tidal amplitude of 1.2 m. The salinity and temperature of this area ranged from 17-32 %o and 2-28 C respectively. Due to the winds, tide and the shallowness of the bay, it was usually extreme- ly turbid. Ice cover formed over the entire bay for short periods in cold years, and fringing ice was common in January and February. The nursery area was subtidal except for spring tides when approximately half to two-thirds of the bottom was exposed. At mean low tide it was covered by about 20 cm of water. EQUIPMENT AND METHODS Larval rearing facilities The larval food and larvae culture was housed in a 7.3 X 19.2 X 3.2 m wood-framed solarium covered by corrugated fiberglass panels. The 8 tanks for culturing unicellular algal food were 1.2 X 2.4 X 1.2 m and held about 3000 L. The tanks for growing larvae were 1.2 X 1.2 X 1.2 m and held approximately °50 L. Both types of tanks were constructed of plywood coated with fiberglass and had exterior wood and metal braces. In addition, larvae were also grown in cylindrical tanks referred to as conicals. The con- icals held 1000 L and were 1.2 m in diameter and 70 cm tall with a 30 cm deep cone shape bottom. These containers were formed fiberglass and were used interchangeably with the wooden larval con- tainers. Screens Assorted screens were used for separating eggs and larvae from seawater, and for sorting larvae and post-set plantigrades into different size groups. These screens were constructed of 25 or 30 cm diameter plexiglass tubing cut to the proper length to act as frames tor the various screens. Woven nylon mesh cloth (Nitex) was then glued to the tubing using 1, 2-dichloroethane. Heat exchanger A 7.6 m coil of polyethylene tubing 1.2 cm in diameter immersed in a fresh water bath was used to raise water temperature in the spawning trough. The water bath was a 75 liter polyethylene container (trash can) with a 40 amp electric calrod immersed in the water. The temperature ot the water in the spawning trough was controlled by diluting the flow of warmed seawater (flowing through the tubing in the water bath) with am- bient seawater. Seawater system Seawater was pumped from a tidal creek in front of the laboratory. Water temperature ranged from 12.0 C to 28 C during the period of opera- tion, and salinity was similar to that reported for Bradford's Bay. The pumps were 5 cm cast iron centrifugal single volute pumps driven by a 3 hp electric motor. The intake and all saltwater lines and valves were plastic. Seawater entering the grow- out facility was used without modification. The seawater pumped to the solarium passed through an industrial model (Sharpies AS-14 clarifier) con- tinuous flow centrifuge which spun the water at 15,000 RPM in a 15 cm stainless steel tube, exer- ting a centrifugal force of 13,200 x G. Centrifuging seawater removed most of the silt and clay particles, larger diatoms and all zooplankton and eggs from 1700 L'hr and only particles and algae with a density about equal to seawater remained. The centrifuged water was piped to 3000 L algal growing tanks, where the water was gently aerated to prevent the algae from settling. This water remained in these tanks while the algae bloomed. At temperatures over 22 °C blooms occurred within 24 hours, but at lower temperatures (14 C) 48 hours or longer were necessary. A typical mid-summer bloom would contain Heteromastix. Clmetoceros, Nitzschia, Chorella and others. This culture of mixed wild algal species varied in quantity and composition MERCENARIA CULTURE from season to season, but under most conditions there was more than sufficient food for the developing larvae. The incubated water was used undiluted as the growing media for the eggs through early post larval seed. Ultraviolet light To control bacterial infections of larvae, the in- cubated seawater was flowed through an ultraviolet radiation unit similar to the Kelly- Pur- dy unit (described by Kelly, 1961) before it was pumped into the larval tanks. The UV unit is 2.8 m long, 90 cm wide, 10 cm deep, and water being sterilized flowed over a series of staggered baffles 1.5 cm high. A reservoir and dam at the intake end of the unit and an overflow reservoir at the discharge end controlled the depth of the water be- ing treated to 5 mm. The unit had 12 40-watt sterilamps 92 cm long spaced equidistant across the unit. Tests indicated the Kelly-Purdy unit reduced the bacterial content of seawater to accep- table densities at a flow rate of 150 L/min (Presnell and Cummins, 1972). Since the maximum flow in our system was 126 L/min, it was assumed that bacterial densities were reduced to acceptable levels. This ultraviolet unit was used only when high mortality rates, high densities of bacterial- feeding protozoans or obvious bacteria swarms were observed in cultures. Cleaning All containers used for larvae were washed after each use with mild biodegradable detergent and fresh water, thoroughly rinsed with hot water and allowed to drain dry. Immediately before use they were rinsed with clarified seawater, drained and filled with clarified incubated water. Larval culture technique Spawning stocks. Adult clams were collected primarily from wild populations. By utilizing clams from the southern coastal states of Georgia, South Carolina and North Carolina, culture of larvae was started in early March without condi- tioning. Spawning was accomplished through Oc- tober by selecting clams from different regions, moving north as the wild clam stocks become ripe. Since clams were easily shipped from place to place, this system eliminated the need, equipment and cost of conditioning. In addition, the faster growing clams from previously grown groups were also used as spawners when they became ripe. Spawning. Spawning was accomplished in a fiberglass trough 3 m long x 30 cm wide x 13 cm deep. Incubated seawater at about 22-24 °C was streamed over 50 to 100 clams for about 30 minutes or until most of the clams had siphons ex- tended. The temperature of the seawater was then increased to as high as 32 °C and dropped back to 24 "C by draining and adding cooler water at about 30-minute intervals to induce spawning. If these temperatures and depth fluctuations did not induce spawning, a clam was sacrificed, and the gonadal material stripped and added to the trough. This usually induced spawning in a few in- dividuals, and since clams are gregarious spawners, a mass spawning followed. The water containing the sperm and eggs was drained through a 25 fi nylon screen. The sperm water passing through the screen was collected in a con- tainer and, if necessary, returned to the spawning trough to further stimulate the clams. The eggs were trapped on the screen. As the screen clogged with eggs, cultured water was used to rinse them into a calibrated 20 L container. When several million eggs were in the container, they were thoroughly mixed by stirring with a plastic plunger and subsampled. The 1 ml subsample was withdrawn by pipette and placed on a 1 ml Sedgwick-Rafter counting cell and the eggs were counted under a microscope. While the eggs were being counted, the larval growing containers were filled with the clarified incubated seawater. The eggs were distributed in- to the filled containers at a density of approx- imately 15-20 eggs/ml. About 40 hours after fertilization the larval tanks were drained through 35 y. mesh screen which caught the veliger larvae. These were con- centrated in 10-15 liters of clarified water and poured through a series of screens ranging from 80 \x to 35 yi mesh. The larvae collected by each screen were placed in separate 20 L calibrated con- tainers, the containers were filled to 10 or 15 liters and subsampled and counted using the same technique described above. The larvae were measured and observed microscopically. If large numbers of abnormal or poorly developing larvae M. CASTAGNA AND J. N. KRAEUTER were present in a given screen size, they were usually discarded. After counting, larvae were redistributed in clean larval growth tanks tilled with new clarified and incubated water. This pro- cedure was followed on Monday, Wednesday and Friday until the larvae metamorphosed and set. Setting. Metamorphosis and setting occurred after 8-12 days under normal operating temperatures. During this process the velum degenerates and the plantigrades creep about with a well-developed foot or fasten to the slide with a byssus (Carriker, 1961). Setting was apparent when the larval tanks were drained. The set clams were attached by a byssus to the tank sides and bottom and often required a jet of water to dislodge them. The larvae did not all set at the same time, but the set clams were easily separated from the veliger larvae by pouring the water and swimming veligers from the containers in which they were concentrated into another container. The set clams remained attached to the bottom by their byssus and were then taken to the grow-out facility. Grow-out facility Equipment. The grow-out wet tables were 1.2 x 2.4 m and t> cm deep constructed of wood and coated with fiberglass resin. There was a dam 8 cm high at the head end and a 6 cm dike at the outlet end. The tables were supplied with a continuous flow of unaltered seawater. The salt water system in the grow-out facility was similar to that in the solarium. It had duplicate intakes, pumps and pipes, which allow- ed one set to be in use for one week while the other system was allowed to stand without draining. The stagnant water in the pipes becomes anoxic, causing the death of fouling organisms which may have attached in the pipes. After a week this line was flushed out and put into use while the other line was allowed to stagnate. Grow-out techniques Grow out of seed. The newly settled clams were moved from the larval facility and washed onto the grow-out tables. The flow rate was regulated to about 1 L min when clams were first placed on the table, and later increased to about 10 L/min. Fouling organisms, especially sea squirts, Molgula manhattensis, were a problem on the grow out tables. The Molgula larvae entered with the water, set on the tables, and smothered the clams. The Molgula were controlled by draining the tables and allowing the clams on the tables to air dry for about 3 hours per day, 5 consecutive days of each week. Any remaining squirts were removed during the two to three week screening when accumulated sediments were removed and the clams sorted by size. The clams were kept in this system about 6 weeks or until they could be collected on a 2 mm mesh sieve. They were then planted in the field nursery plots. Field nursery techniques Equipment. Current baffles were constructed of 1 cm diameter steel rod and 7X7 mm mesh plastic screen. The steel rod was made into a rec- tangular frame 0.6 m high and 1.5 m long with a 0.9 m leg extending down on each side. The plastic screen was fastened to the 0.6 X 1.5 m frame with 3 mm polypropylene line. To install current baf- fles the legs were pushed into the bottom until the plastic touched the substrate. A 13 mm mesh plastic net 2 m tall surrounded the clam planting site. This net was supported by 10 X 10 cm poles pushed into the bottom approx- imately every 3 meters. The net bottom was weighted down by a 6 mm chain fastened to the bottom with 3 mm polypropylene line. This chain was embedded about 10 cm into the soft mud bot- tom. Predator protection. A major predator of small clams is the blue crab, Callinectes sapidus. Preliminary experiments indicated that crushed rock aggregate provided some protection against this predator. Approximately 75% more small clams survived in aggregate than in control plots. To further reduce crab predation, baited commer- cial crab traps were also placed in the clam plan- ting area and fished 3 to 5 times per week. Another group of major predators on juvenile and adult clams are rays of the families Dasyatidae, Myliobatidae and Rhinopteridae. Large schools of these rays may enter an area and destroy the clam populations. The 2 m high net protects the seed clams against these predators and prevents larger blue crabs from entering the plots. MERCENARIA CULTURE Wash out prevention. The second major pro- blem, when 2 mm seed clams were used, was that they pushed out or were washed out of the bottom by waves and were carried away by tidal currents. Laboratory experiments indicated that clams 3 mm in width could be moved by current velocities as low as 15 cm/sec (0.3 knot current). The battles prevented the current from moving small clams from the aggregate bed. Preparation and planting of nursery plot. The nursery area was prepared by first placing a series of current baffles in squares. To conserve baffles, they were placed next to each other to share a common panel between two squares (Fig. 1). A crushed stone aggregate of 1-3 cm chips was then broadcast into each square to a depth of ap- proximately 4 cm. The aggregate was then leveled with a rake and allowed to stand. After about one week the nursery pens were examined. The pens should contain a thin layer of silt over the gravel or aggregate. If this layer does not appear, the cur- rents across the bottom are too strong and more baffles should be implanted. If the silt becomes too heavy, some baffles should be removed or the small clams will work up into this layer above the protecting aggregate. Once the area had been stabilized, small clams were broadcast over the ag- gregate at an average density of approximately -a3£/sq. m. ■oo RESULTS Hatchery production utilizing this method has yielded sets of 120, 97 and 55 X 10" Mercenaria in 1975, 1974 and 1973 respectively. Additonal species have been produced concurrently. Estimated hatchery production to field size for Mercenaria (again as one of several species) is 15 - 20% of set for each year. These latter estimates could be substantially improved with greater care given to one species. At the growth rate exhibited by these clams, it was estimated that they would reach the desired little neck size (1" depth) in 22 to 28 months. This estimation later proved to be correct. The cost for 600,000 clams planted in the field averaged $0,015 per clam. This cost included estimated interest on a loan sufficient to begin a hatchery, labor, utilities and all supplies. The only additional charges would be maintenance of the field plots and harvesting. Maintenance costs for two years should not exceed $0,005 per clam and harvesting cost is estimated to be about $0,002. The curent market value for prime sized little neck clams in this area is about $0.05. Following the first winter's growth, five samples were taken in each of 41 squares of aggregate ( ! : of the test squares). Average survival was in ex- cess of 75%, and included samples of 5 squares of clams planted at less than 2 mm which were lost. FIG. 1. View of nursery area showing baffles in foreground and 2 m net in background. M. CASTAGNA AND J. N. KRAEUTER During the summer some severe predation was observed, but this was controlled by raising the height of the fence so that it was submerged only during spring tides. In addition the number of crab pots inside the fence was increased from 4 to 8. Preliminary calculations indicated that a commer- cial operation would be economical with 40% sur- vival of the planted clams. Experiments to reduce the cost of the clams planted in the field are being conducted this sum- mer. These experiments are designed to eliminate unnecessary components and thus reduce costs. LITERATURE CITED Ansell, A.D. 1968. The rate of growth of the hard clam Mercenaria mercenaria (L.) throughout the geographical range. J. Cons. Perm. Int. Ex- plor. Mer. 31:364-409. Calabrese, A. and H. C. Davis. 1066. The pH tolerance of embryos and larvae of Mercenaria mercenaria and Crassostrea virginica. Biol. Bull. 131:427-436. Carriker, M. R. 1961. Interrelation of functional morphology, behavior and autecology in early stages of the bivalve Mercenaria mercenaria. J. Elisha Mitchell Sci. Soc. 77:168-241. Chanley, P.E. 1961. Inheritance of shell markings and growth in the hard clam, Venus mercena- ria. Proc. Nat. Shellfish. Assn. 50:163-168. Chanley, P. and j.D. Andrews. 1971. Aids for identification of bivalve larvae of Virginia. Malacologia 11:45-110. Davis, H. C. 1958. Survival and growth of clam and oyster larvae at different salinities. Biol. Bull. 114:296-307. Davis, H. C. and A. Calabrese. 1964. Combined effects of temperature and salinity on develop- ment of eggs and growth of larvae of M. mercenaria and C. virginica. U.S. Fish Wildlife Serv., Fish. Bull. 63:643-655. Kelly, C. B. 1961. Disinfection of sea water by ultraviolet radiation. Amer. J. Publ. Hlth. 51:1670-1680. Kerswill, C. J. 1941. Some environmental factors limiting growth and distribution of the quahog, Venus mercenaria L. Ph.D. thesis, Univ. Toron- to, 104 pp. Loesch, J. G. and D. S. Haven. 1973. Estimated growth functions and size-age relationships of the hard clam, Mercenaria mercenaria in the York River, Virginia. Veliger, 16:7o-81. Loosanoff, V. L. 1937. Spawning of Venus mercenaria (L.). Ecology 18:506-515. Loosanoff, V. L. 1954. New advances in the study of bivalve larvae. Amer Scient. 42:607-624. Loosanoff, V. L. and H. C. Davis. 1950. Condi- tioning V. mercenaria for spawning in winter and breeding its larvae in the laboratory. Biol, bull. 98:60-05. Loosanoff, V. L. and H. C. Davis. 19o3. Rearing of bivalve mollusks. In Advances in Marine Biology, Vol. I. F. S. Russell, Ed. Academic Press Inc., London and New York, 1-136. Loosanoff. V. L., W. S. Miller, and P. B. Smith. 1951. Growth and setting of larvae of Venus mercenaria in relation to temperature. J. Mar. Res. 10:50-81. Newman, W. S. and C. A. Munsart. 1968. Holocene geology of the Wachapreague lagoon. Eastern Shore Peninsula, Virginia. Marine Geol. 6:81-105. Pratt, D. M. 1953. Abundance and growth of Venus mercenaria and Callocardia morrhuana in relation to the character of bottom sediments. I. Mar. Res. 12:60-74. Pratt, D. M. and D. A. Campbell. 105b. En- vironmental factors affecting growth in Venus mercenaria. Limnol. Oceanogr. 1:2-17. Presnell, M. W. and J. M. Cummins. 1972. Effec- tiveness of ultra-violet radiation units in the bactericidal treatment of seawater. Water Research, 6:1203-1212. Wells, W. F. 1024. First report of station for shellfish culture. New York State Conservation Commission Annual Report, 98-117. Wells, W. F. 1926. Report of experimental shellfish station. New York State Conservation Commission Annual Report, 112-130. Proceedings of the National Shellfisheries Association Volume 67 — 1977 FOOD CONSUMPTION AND GROWTH OF LARVAE OF THE PACIFIC OYSTER, CRASSOSTREA GIG AS (THUNBERG), IN A CONSTANT FLOW REARING SYSTEM1 Robert E. Malouf and Wilbur P. Breese OREGON STATE UNIVERSITY DEPARTMENT OF FISHERIES AND WILDLIFE NEWPORT, OREGON 97365 ABSTRACT This study of food consumption and growth of larvae of the Pacific oyster, Crassostrea gigas, Thunberg, is intended to provide some of the basic information necessary for the efficient operation of commercial oyster hatcheries. As a part of the study, the caloric content of larvae of various sizes was determined to provide a mean- ingful measure of mass. The effects of algal concentration (Monochrysis Lutheri) and larval density on the growth of larval oysters was measured in a flow-through feeding system. Algal concentration and larval food consumption can be optimized in a flow- through system to provide for maximum larval growth and efficient use of algal food. Removal of algae at the rate of 2,600 algal cells per larva per hour does not result in ap- preciably better growth than the removal of 1,300 algal cells per larva per hour. A modified flow-through feeding system, applied to a commercial oyster hatchery, could improve utilization of cultured algae, and yield increased larval growth. INTRODUCTION The principal commercial oyster grown in the Pacific Northwest, the Pacific oyster, Crassostrea gigas, was originally introduced from Japan. The industry has been dependent for years on seed oysters imported from Japan, supplemented by seed collected at a few locations in Washington and British Columbia. However, the price of im- ported seed is increasing, and there have been periodic failures of oyster spawning in Northwest waters. The lack of a dependable seed source has depressed the oyster industry. To provide a reliable source of oyster seed, a number of commercial oyster hatcheries have This work was supported by NOAA. Office of Sea Grant, Department of Commerce through the Oregon State Univer- sity Sea Grant College Program under grant No. 04-5-158-2. Technical Paper No. 4323, Oregon Agricultural Experiment Station. been built in the Pacific Northwest in recent years. The purpose of this study was to determine some of the basic relationships between algal food den- sity, food consumption, and growth in oyster lar- vae. A clear understanding of the interactions of these variables is needed for the operation of a successful oyster hatchery. Our secondary objec- tive was to evaluate a technique for continuously feeding oyster larvae as an alternative to batch feeding in hatchery operation. METHODS An apparatus was designed to maintain a cons- tant flow of algal cells at a known density through test chambers containing a known number of oyster larvae (Fig. 1). The test chambers consisted of 76 cm sections of 4" diameter PVC pipe which were fitted on one end with 116 micron nylon screen. The screen served to retain the larvae while allowing uneaten algae and feces, to pass R. E. MALOUF AND W. P. BREESE algae inflow / u • 4- PVC, JO' long . coupling 116 micron screen FIG. 1. Apparatus used for constant flow ex- periments. through. These chambers, filled to a depth of 61 cm, contained five liters of salt water. Forty-five L. carboys fitted with siphons were used to main- tain a flow of 14 ml per minute of algae suspension of the appropriate cell density to each test chamber. Only the algal densities and the number and size of the larvae varied. The total volume of water passing through the chambers was kept con- stant. All experiments were conducted at a temperature of 20 C and at a salinity of 25 ppt. The density of the algae entering and leaving the test chambers was determined at least once every 24 hours. The difference in algal density between the inflow and the outflow of the test chambers gave a measure of larval food consumption at dif- ferent algal and larval densities. Monochrysis lutheri, cultured in five gallon car- boys, was used as larval food. This naked flagellate is easily cultured and has been shown to be an excellent food organism for larval bivalves (Walne, 1970). Algal densities in the cultures and in the test chambers were determined using a Coulter Counter model B, following the pro- cedures outlined by Sheldon and Parsons (1967). The algal cultures generally reached a density of about two million cells per ml within seven days. Cultures having densities of two to five million cells per ml were used in all feeding experiments. In preliminary tests, the Coulter Counter model J plotter was used in conjunction with the counter to make size frequency determinations of cells from stock algal cultures and from test chambers containing larvae. No differences in particle size distribution were evident between the stock cultures and algal populations that had been graz- ed upon by oyster larvae for several hours in a flowing water system. The similarity of the histograms indicated that counts made on samples taken from the test chambers in which grazing was being measured did not contain a significant amount of cell fragments and detritus that were within the size range of the counter. The Pacific oyster larvae used in this study came from the pilot oyster hatchery at the Oregon State University Marine Science Center. The adults were spawned artificially (Loosanoff and Davis, 1963). The larvae used in any one experi- ment were drawn from a single hatchery tank after they had been thoroughly mixed. No attempt was made to control parentage. Larvae were caught on a stainless steel screen and concentrated in a beaker for counting and measuring at the start of each experiment. It was found that reproducible counts could be obtained by withdrawing 1 ml samples with an automatic pipette while the water in the beaker was being gently agitated by raising and lowering a per- forated plexiglass disc in the beaker. Preliminary 1 ml counts gave a rough approximation of the volume of water containing the desired number of larvae for each test chamber. Ten 1 ml subsamples were subsequently withdrawn from the aliquots prepared for each test chamber, and adjustments were made in the number of larvae in each aliquot before they were added to the test chambers. Counts were also made of the larvae while the experiments were in progress. The larvae were uniformly suspended in each chamber with a plex- FOOD CONSUMPTION OF PACIFIC OYSTER TABLE 1. The results of dichromate oxidations showing the relationship between larval shell length and the caloric content per larva of Pacific Oyster larvae. Length of larvae in microns" 96±2 100±2 133±5 189±4 249±7 274±5 306±5 Number of larvae used per sample Calories per larva (sample No. 1) Calories per larva (sample No. 2) Mean caloric content per larva 7,400 5,360 .00033 .00039 .00038 .00039 .00036 .00039 8,325 16,550 10,280 9,080 4,200 00075 .0019 .00372 .00581 .00655 00077 .0019 .00369 .00578 .00666 00076 .0019 .00371 .00579 .00661 "Means with °5 percent confidence interval estimate. iglass plunger before a piece of 11 mm diameter glass tubing was extended down into the chamber to collect larvae from the entire length of the water column. This procedure was repeated three times so that a total of about 200 ml was collected from each chamber and placed in a small beaker. Ten subsamples were then withdrawn from the beak- ers using the method described above. These lar- vae were killed with AFA, counted, and measured with the use of an ocular micrometer. The re- mainder of the 200 ml sample was returned to the test chamber. Biomass Determination. The growth of larvae is most easily observed from increases in the average shell length. These measurements were made on larvae which were sampled for periodic deter- mination of larval density in the test chambers. Shell length does not adequately reflect the animal's mass. It is difficult to compare the realtive biomasses of populations consisting of lar- vae of different sizes where shell length is the only criterion used. To overcome this problem, a rela- tionship was established between shell length and total biomass in terms of organic content. The method used is a modification of the dichromate chemical oxygen demand determina- tion described in the 10th edition of Standard Methods for the Examination of Water, Sewage, and Industrial Wastes (American Public Health Association, 1°55). A known excess of oxidant (potassium dichromate) was added to a sample of a known number of larvae in distilled water. After oxidation was completed, the amount of oxidant remaining was determined by titration. The amount of reacting oxygen was calculated from the difference between the initial and the final quantity of oxidant. These values were then multiplied by an oxycalorific coefficient of 3.42 caloreis per mg 02 to obtain an estimate of the caloric content of the sample (Maciolek, 1962). Determinations of the caloric content of seven batches of Pacific oyster larvae are given in Table 1 and are shown graphically in Figure 2. Larval length in microns when plotted against the caloric content per larva on a semi log scale (Fig. 2) yields /\ From experimental data for C gigos O Modified from Walne(l965) for 0 edulis 200 i Micrometers FIG. 2. The relationship between shell length and caloric content of Pacific and European oyster lar- vae. 10 R. E. MALOUF AND W. P. BREESE a straight line. This shows that there is an ex- ponential increase in the caloric content of Pacific oyster larvae with increases in shell length. Walne (1965) found a similar relationship between the shell length of European oysters and their total dry weight. He also found that the shell of an oyster larva comprises a constant 75 per cent of its total dry weight. Miller and Scott (1967) confirmed that the shell of European oyster larvae makes up about 75 per cent of the total dry weight. Walne's (1965) data for the total dry weight of European oyster larvae were multiplied by .25 to obtain an approximation of the organic matter per larva and then by 5,000 calories per gram to estimate the caloric content of the animal tissue (Davis and Warren, 1968; Maciolek, 1962). these values are compared with data obtained ex- perimentally in the present study in Figure 2. Three experiments with the constant flow ap- paratus provided information on the relationships between larval growth and food consumption and algal density in a flowing system. The experiments will be described individually. Growth rates were determined from periodic measurements of shell length for a sample of lar- vae. The mean length was converted to caloric content by use of the experimentally determined curve in Figure 2. Instantaneous growth coefficients, K, were calculated for each test group from the formula: K = log.C2+log.C, x 1000 t Where: C, = The initial caloric content per larva C2 = the final caloric content per larva t = the time in days The use of the instantaneous growth coefficient, K, facilitates comparison of larval growth obtain- ed in a number of experiments of differing dura- tion and using larvae of different initial and final size. This equation was applied to mussel larvae by Bayne (1965) and to oyster larvae by Walne (1963), based on changes in shell length. These experiments were intended to describe general relationships between larval density and algal inflow density as reflected by the growth of TABLE 2. Stocking rate, mortality, final mean length, and instantaneous growth coefficient of Pacific oyster larvae fed at four different algal densities (Experiment J). Initial larval length was 189 microns. Length of larvae (microns) Initial Mean algal Mean algal Instantaneou! Chamber number of Percent Final mean Standard inflow outflow growth number larvae/ml. mortality value error density density rate (K) 1 1 43 268 5.18 6,474 4,350 82.1 2 2 21 263 3.79 6,783 2,751 75.0 3 4 16 222 3.90 6,845 2,192 37.6 4 8 18 195 4.45 6,714 1,641 3.3 5 1 97 11,762 10,738 6 2 53 237 5.77 11,812 9,760 50.5 7 4 94 11,704 10,302 8 8 30 208 3.50 11,747 2,026 22.2 9 1 49 309 4.33 21,318 13,869 118.9 10 2 23 303 6.51 22,077 11,359 114.9 11 4 100 22,079 17,782 12 8 39 240 5.09 21,834 3,338 76.6 13 1 38 254 5.45 40,482 33,097 26.8 14 2 18 283 6.03 41,388 25,868 36.4 15 4 95 41,630 26,098 16 8 40 276 4.66 42,286 11,834 34.4 FOOD CONSUMPTION OF PACIFIC OYSTER 11 larvae. Because of inherent differences among broods of larvae and procedural differences among the experiments, growth data from the three experiments were analyzed separately. The procedures did not involve replication of in- dividual experiments, so a valid error term for testing differences in responses was not available. Experimental Design. In Experiment 1, four algal inflow densities (5,000, 10,000, 20,000 and 40,000 cells per ml) were given in all possible com- binations to larval cultures with one, two, four, and eight larvae per ml. In Experiment 2, four algal inflow densities (10,000, 20,000, 40,000, and 80,000 cells per ml) were given in all possible combinations to larval cultures with two and sixteen larvae per ml. In the final experiment, Experiment 3, algal in- flow densities were increased to 20,000, 40,000, 80,000, and 160,000 cells per ml and were fed in all possible combinations to larval cultures with two and sixteen larvae per ml. RESULTS Larval Growth Experiment 1. Mortality exceeded 50 per cent in five test chambers (Table 2). Because of un- predicted high mortality in the five chambers, an analysis of variance was not carried out. Instead, 95 per cent confidence intervals of the mean sizes of the larvae at the end of the experiment were calculated. From these data it is possible to make FIG. 3. General relationship between algal inflow density and larval growth rate at four larval den- sities (Experiment 1). some general conclusions concerning the effects of algal inflow density and larval density on larval growth in this system. Larval growth rate increased with increasing algal density up to 20,000 cells per ml, but declin- ed at an algal inflow density of 40,000 cells per ml (Fig. 3). Maximum growth rates in Experiment 1 were achieved at a mean algal outflow density of 11,000 to 14,000 cells per ml (Table 3). These algal outflow densities resulted when an inflow density of 20,000 cells per ml was passed through test chambers initially stocked with one or two larvae per ml (chambers 9 and 10). TABLE 3. Stocking rate, mortality, final mean length and instantaneous growth coefficient of Pacific oyster larvae fed at two different algal inflow densities (Experiment 2). Initial larval length was 193 microns. Length of larvae (microns) Initial Mean algal Mean algal Chamber number of Percent Final mean Standard inflow outflow Growth number larvae/ml. mortality value error density density rate (K) 1 2 0 242 3.04 10,927 3,389 83.7 2 16 9 199 2.23 10,943 1,237 9.4 3 2 0 272 3.15 20,923 6,397 131.7 4 16 12 202 2.03 20,770 2,143 18.2 5 2 100 6 16 100 7 2 6 296 2.34 79,477 49,588 153.4 8 16 17 247 2.72 77,997 3,500 88.2 12 R. E. MALOUF AND W. P. BREESE Two fold increases in larval density had a mark- ed effect on larval growth rates at an algal inflow density of 5,000 cells per ml (Table 3). At an algal inflow density of 40,000 cells per ml, larval densi- ty had no effect on larval growth rates. Hnousands of cell FIG. 4. Relationship between algal inflow density and larval growth rate for two larval densities (Ex- periment 2). Experiment 2. Larval mortality rates in chambers receiving 40,000 cells per ml (chambers 5 and 6) were high early in the experiment and ap- proached 100 per cent at the conclusion of the study. A flow of algae was maintained through both chambers 5 and 6 for the duration of the ex- periment despite the heavy mortality. This was done to measure the effects on algal density of a large population of an unidentified ciliate that developed in the chambers as the larval popula- tion declined. It was found that the densities of aJgae in the inflow and outflow of these chambers were not significantly different. As in Experiment 1, no cause was determined for the extreme mor- tality in the two chambers. It is apparent that the growth rate of the larvae was directly related to the density of algae in both the inflow and the outflow in Experiment 2 (Table 3). In all cases higher algal inflow densities resulted in higher growth rates in this experiment (Fig. 4). Maximum growth rate was achieved at an algal inflow density of 80,000 cells per ml through a test chamber containing two larvae per ml (chamber 7). The results of Experiment 2 differed in several respects from those of Experiment 1. There was no indication of reduced larval growth at algal den- sities above 20,000 cells per ml (Figure 4). Growth rates of larvae at a density of 2 per ml appeared to level off above an algal inflow density of 20,000 cells per ml. However, the growth rates of larvae kept at a density of 16 per ml continued to increase at algal densities in excess of 20,000 cells per ml. Experiment 3. Larval mortalities were high in only one test chamber during this experiment. Although the mortality in this chamber exceeded 50 per cent, most of this mortality occurred during the last two days of the experiment. Careful ex- amination of dead and dying larvae revealed what appeared to be fungus infection. Stained with neutral red, the fungus was very similar in ap- pearance to one described and tentatively iden- tified as Sirolpidium sp. by Davis and Loosanoff (1954) and Vishniac (1955). Although it appeared that the fungus attacked still living larvae, We cannot be certain that the fungus was the cause of the recorded mortalities. The data from Experiment 3 (Table 4 and Fig. 5) indicate that larval growth rate increased in algal inflow density for those chambers having a larval density of 16 per ml. However, increases in algal inflow density did not result in higher larval growth rates in those test chambers that had an in- itial larval density of two per ml. In fact, as Figure 5 shows, larval growth rate declined as algal in- flow density exceeded 40,000 cells per ml in those chambers containing two larvae per ml. HG. 5. Relationship between algal inflow density and larval growth rate for two larval densities (Ex- periment 3). FOOD CONSUMPTION OF PACIFIC OYSTER 13 TABLE 4. Stocking rate, mortality, final mean length, and instantaneous growth coefficient of Pacific oyster larvae fed at four different algal inflow densities (Experiment 3). Initial larval length was 199 microns. Length of larvae (microns) Mean algal Mean algal Chamber Initial Percent Final mean Standard error inflow density outflow density Growth number number of mortality value of the mean (cells/ml.) (cells/ml.) rate (K) larvae/ml. 1 2 29 267 4.99 20,749 8,193 107.5 2 16 11 215 2.85 21,008 1,514 25.8 3 2 25 285 3.63 40,916 21,453 113.4 4 16 8 223 3.82 41,472 2,458 39.2 5 2 24 265 4.05 80,402 58,906 104.4 0 16 58 233 5.01 80,716 14,951 53.6 7 2 29 233 4.39 164,353 138,897 56.6 8 16 14 249 4.60 166,582 28,051 82.3 Chambers 1 and 8 are of particular interest in this experiment. Chamber 1 contained 10,000 lar- vae and were fed at an algal inflow density of 20,000 cells per ml. Chamber 8 contained 80,000 larvae that were fed at 160,000 cells per ml. Although chambers 2 and 8 were fed the same amount of food per larva, the larvae in chamber 1 showed significantly better growth than those in chamber 8. The results from Experiment 3 were similar to those obtained in Experiment 2 at similar algal densities. However, it appears that the maximum larval growth rate at a density of 16 larvae per ml may occur at an algal density of 160,000 cells per ml or higher. DISCUSSION There are a number of possible explanations for the growth differences among the three ex- periments. The relatively low maximum growth rate and rapid decline in growth at higher algal densities during Experiment 1 may have been due, in part, to the procedures employed in that experi- ment. Larvae used in experiment 1 were subjected to slightly cooler temperatures than were those us- ed in Experiments 2 and 3. The profound effect of temperature on larval growth has been shown by a number of workers including Bayne (1965), Davis and Calabrese (1964), Loosanoff and Davis 1963, and Walne (1956, 1965). Results of these studies showed that larval growth rates increased rapidly with increasing temperature. In part, the growth differences could have been due to genetic differences among larvae produced by different parents. In studies of C. gigas, Lannan (1973) found a statistically significant genetic com- ponent of the variance of larval growth and sur- vival through metamorphosis. Our experience with larval growth in hatchery tanks suggests that progeny from different parental crosses produce widely different growth rates among batches of larvae that are treated similarly. Since the experiments were not conducted con- currently, differences in the quality of the algal cultures among the three experiments are possible and may have affected larval growth. Taub and Dollar (1965) reported that the chemical composi- tion of the alga Chlorella varied significantly as light intensity and culture medium were varied. It is possible that the composition and therefore the food value of M. lutheri varied in the present ex- periment. Despite the existence of variation in larval growth among the three experiments, there are certain features that are common to all three growth curves. Growth rates of the larvae increas- ed as the mean algal inflow density was increased to 20,000 cells per ml in all three experiments. Subsequent increases in algal density yielded less significant increases in larval growth rates, these 14 R. E. MALOUF AND W. P. BREESE results indicate that at larval densities of eight per ml or less there was little or no advantage to an algal inflow density of more than 20,000 cells per ml. The reduction of larval growth rates at high algal densities shown in Figures 4 and 6 is similar to results reported by other workers. Loosanoff et al. (1953) reported that the growth of clam larvae was retarded by over-feeding. They also found that larvae of the mussel, Mytilus edulis, grew best at intermediate algal densities, and that any increase beyond the optimum caused a reduction in growth rates for the larvae of the Eastern oyster, Crassostrea virginica. They concluded that at high algal densities the larvae were adversely af- fected by metabolic products of the algae. Walne (1966) reported similar results in his experiments with batch feeding of lsochrysis galbana to the lar- vae of the European oyster. He concluded that there was no advantage to batch feeding at a con- centration of 120,000 cells per ml compared to 30,000 cells per ml under the conditions of his ex- periments. Our observations indicate that another impor- tant cause of reduced growth at high algal den- sities is the excessive formation of pseudofeces. Yonge (1926) described the pseudofeces of oyster larvae as long strings of mucus with algal cells embedded singly and in clumps along the length of the mucus string. These strings trail behind and often entangle and trap a swimming larva. I O ■ «P1 n 28 . i «pl m 0 stf O A A o a. 96 O ,°0o * o 3 64 °/ _i 12 0 0/ d & «oo eoo 1200 1600 2000 2400 2800 FIG. 6. The relationship between the rate of clearance by oyster larvae of suspended algal cells and the resultant growth rate of the larvae. Data from three experiments plotted by inspection. A normally-feeding larva produces a con- tinuous flow of mucus in which food particles are trapped and are carried to the mouth to be in- gested (Yonge, 1926). Normally, the bulk of the mucus that is produced is reingested. A larva that is producing excessive pseudofeces is not only removing its food supply from suspension and making it unavailable for consumption, but is pro- ducing and losing large amounts of mucus. The consequences of this type of superflouous feeding is shown in Figure 6, which relates the ap- parent food consumption rate of larvae and their resultant growth rate for all three experiments. The data are from all test chambers in the three ex- periments which showed less than a 50% mortali- ty during the growth period, received between 5,000 and 80,000 cells per ml in the inflow water, and had initial larval densities of from one to six- teen larvae per ml. Food consumption data, ex- pressed as the number of cells cleared per larva per hour, are based upon mean values determined throughout the growth period, as previously described, the larvae were about 200 u long at the beginning of the growth period, and all three ex- periments were terminated as soon as any setting occurred in any chamber. Obviously, the ab- solute value of growth and food consumption ob- tained under different conditions and for different growth periods would vary, but the general rela- tionship would very likely resemble Figure 6. The removal, by larvae, of 2,600 algal cells per hour did not result in appreciably greater growth than did the removal of 1,300 algal cells per hour. These results suggest that many of the cells that were removed by the larvae at the higher feeding rates were not ingested and assimilated but were ingested and incompletely assimilated or weie simply cleared and rejected as pseudofeces. Millar (1955) described the mechanisms of food movement in the gut of larvae of the European oyster. He described a muscular pulsation of the digestive diverticulua, the site of absorption, that withdrew partially digested food materials from the stomach. He concluded that it was entirely a matter of chance whether food particles drawn off into the midgut and into the rectum had been in the stomach and diverticula for a long or short time, and therefore to what extent they had been digested and assimilated. It appears then that a FOOD CONSUMPTION OF PACIFIC OYSTER 15 larva exposed to a very high concentration of algae could ingest algal cells and pass them through its system without gaining any nutritional benefit from them. This would further reduce the rate of growth relative to the number of algal cells consumed. This study has shown that a constant flow system for feeding larvae can be regulated to maintain some algal density that is most favorable for larval growth. If such a system could be adapted to a hatchery operation it would provide maximum return in terms of larval growth from algae used in feeding. The commonly used technique of batch feeding larvae is subject to a number of disadvantages. In order to supply larvae in one feeding with enough food to last them 24 or 48 hours, it is necessary to raise the algal density to a high level. As has been shown the larvae feed actively at this initially high algal density, but an undetermined amount of their food supply is tied up as pseudofeces or otherwise utilized ineffeciently. Then, as the algal density declines with time, the larvae become less able to obtain food. As a consequence of batch feeding, larvae are overfed part of the time and essentially starved the rest of the time. Since the conditions of the constant flow system differed in a number of ways from those common- ly employed in hatchery operations it is difficult to compare growth rates with those typically record- ed in a hatchery. These experiments were con- ducted at 20 C to enhance survival of the algae in- the test chambers. Additionally, the experiments utilized only one species of algae for the sake of simplicity. Improved growth can be obtained by using a mixture of algal species (Davis and Guillard, 1958). In spite of these disadvantages, growth rates obtained from the experimental systems were found to be only slightly lower than those commonly experienced in the hatchery. ACKNOWLEDGEMENTS The authors wish to thank Dr. William McNeil for his assistance and encouragement during the course of this study. We would also like to thank Mr. Dean Satterlee for his assistance in construc- tion of the test apparatus and in preparation of the manuscript. The manuscript was critically review- ed by Dr. James Lannan and Dr. Gerald Davis. The study was supported by a grant from NOAA Office of Sea Grant, U.S. Department of Commerce, through the Oregon State University Sea Grant College Program. BIBLIOGRAPHY American Public Health Association, 1955. Stan- dard methods for the examination of water, sewage, and industrial wastes. 10th ed., New York, 522 p. Bayne, B.L. 1965. Growth and delay of metamor- phosis of larvae of Mytilus edulis L. Ophelia 2:1-47. Davis, G. E. and C. E. Warren. 1968. Estimation of food consumption rates. In Methods for assessment of fish production in fresh waters. W. E. Ricker, Ed. Oxford, Blackwell Scientific Publications, 204-225. Davis, H. C. and A. Calabrese. 1964. Combined effects of temperature and salinity on develop- ment of eggs and growth of larvae of M. mercenaria and C. virginica. U.S. Fish and Wildlife Serv. Fish. Bull. 63:643-655. Davis, H. C. and R. R. Guillard. 1958. Relative value of ten genera of micro-organisms as food for oyster and clam larvae. U.S. Fish and Wildlife Serv., Fish. Bull. 58:293-304. Davis, H. C. and V. L. Loosanoff. 1954. A fungus disease in bivalve larvae. Proc. Nat. Shellfish. Assoc. 45:151-156. Lannan, J. E. 1973. Genetics of the Pacific oyster: biological and economic implications. Ph.D. thesis, Corvallis, Oregon State University. 104 numb, leaves. Loosanoff, V. L. and H. C. Davis. 1963. Rearing of bivalve molluscs. In Advances in marine biology, F. S. Russel, ed. Vol. 1, London. Academic Press, 1-136. Loosanoff, V. L., H. C. Davis, and P. E. Chanley. 1953. Behavior of clam larvae in different con- centrations of food organisms. Anat. Rec. 117(3):586-587. Maciolek, J. A. 1962. Limnological organic analysis by quantitative dichromate oxidation. U.S. Fish and Wildlife Serv., Research Report 60. 61 p. Millar, R. H. 1955. Notes on the mechanism of food movement in the gut of the larval oyster, 16 R. E. MALOUF AND W. P. BREESE Ostrea edulis. Quart. J. Microscop. Sci. 96(4):539-544. Millar, R. H. and J. M. Scott. 1967. The larvae of the oyster Ostrea edulis during starvation. J. Mar. Biol. Assoc. U.K. 47:475-484. Sheldon, R. W. and T. R. Parsons. 1967. A prac- tical manual on the use of the Coulter Counter in marine research. Toronto, Coulter Elec- tronics Sales Company. 66 p. Taub, Frieda B. and Alexander M. Dollar. 1965. Control of protein level of algae, Chlorella. J. of Food Sci. 30(2):359-364. Vishniac, Helen S. 1955. The morphology and nutrition of a new species of Sirolpidium. Mycologia 47:633-645. Walne, P. R. 1956. Experimental rearing of the larvae of Ostrea edulis in the laboratory. United Kingdom Ministry of Agriculture, Fisheries, and Food, Fishery Investigations, ser. 11, 20(9):l-23. Walne, P. R. 1963. Observations of the food value of seven species of algae to the larvae of Ostrea edulis. ]. Mar. Biol. Assoc. U.K. 43:767-784. Walne, P. R. 1965. Observations on the influence of food supply and temperature on the feeding and growth of the larvae of Ostrea edulis. United Kingdom Ministry of Agriculture, Fisheries, and Food, Fishery Investigations, ser. 11, 24(l):l-45. Walne, P. R. 1966. Experiments in the large-scale culture of the larvae of Ostrea edulis L. United Kingdom Ministry of Agriculture, Fisheries, and Food, Fishery Investigations, ser. 11, 25(4):l-53. Walne, P. R. 1970. Present problems in the culture of the larvae of Ostrea edulis. Helgolander wiss. Meeres 20:514-515. Yonge, C. M. 1926. Structure and physiology of the organs of feeding and digestion in Ostrea edulis. ]. Mar. Biol. Assoc. U.K. 14:295-386. Proceedings of the National Shellfisheries Association Volume 67 — 1977 ECONOMIC ANALYSIS OF PRODUCING PACIFIC OYSTER SEED IN HATCHERIES Kwang H. Im and Don Langmo DEPARTMENT OF AGRICULTURAL AND RESOURCE ECONOMICS OREGON STATE UNIVERSITY CORVALLIS, OREGON ABSTRACT Total costs and average costs for the hatchery production of Pacific oyster (Crassostrea gigas) seed were determined for 5 levels of output, ranging from 6,000 to 14,000 cases per year. Costs, returns, and economies of size for each model plant were further established for 5 different options (methods) of cultch preparation. Within the limits of the study, it was found that hatchery production of Pacific oyster seed in the Pacific Northwest is economically feasible, and as the plant capacity increased, the net returns increased. Among the 5 options of cultch preparation, Option 1, which pumps salt water with city power for the cleaning operation, is the most favorable; Option 2, which pumps salt water with own generated power, is the next most favorable; and Option 5, which buys already-cleaned cultch from a local dealer, is the least favorable in terms of pro- duction cost. INTRODUCTION Commercial oyster hatcheries in the Pacific Northwest have operating conditions — econo- mic, location, technical, and biological — all of which, in one way or another, affect the cost of production. In the past, most of the Pacific oyster seed has been imported from Japan at high cost, and often with an extremely low survival rate. Oyster growers could not depend entirely on imported oyster seed, mainly because of high cost and uncertainty of seed supply. The importance of oysters can be seen readily if one considers that, in terms of ex-vessel value, oysters (all species taken together) rank seventh among seafoods landed in the United States. Table 1 shows the value of U.S. major seafood species in 1975. Even though the use of oysters is relatively important in the American diet, the supply of domestic hatchery seed for oyster propagation is not sufficient to meet the demand at current market prices. Figure 1 shows the historical trend of U.S. oyster supply for the last two and one-half decades. As shown in Figure 1, oyster imports (mostly canned) have increased significantly, while domestic landings have decreased substantially, primarily due to high domestic production costs, pollution of oyster beds, and foreign competition. The quantity of domestic landings has decreased by 23.7, 30.4, 31.8, and 32.4 percent in 1960, 1965, 1970, and 1975, respectively, compared with a base period of 1950-54. Conversely, for the same years the quantity of oyster imports has in- creased tremendously by 821, 1,032, 1,874, and 1,532 percent, respectively, compared with the same base period. Nevertheless, in these same years the total supply of oysters for U.S. con- sumption has decreased by 15.6, 20.3, 13.6, and 17.4 percent. Population growth and oyster supply, in terms of domestic landings plus imports, are expressed 17 18 K. H. IM AND D. LANGMO TABLE 1. Ex-Vessel Value of U.S. Seafood Species, 1975 Species Value million dollars Shrimp 226 Salmon 116 Tuna 108 Crab 84 Lobster 59 Menhaden 49 OYSTERS 43 Clams 41 All others 245 SOURCE: U.S. Department of Commerce, Fish- ery Statistics of the United States, 1975. 85 85 80 75 z 70 o 65 60 55 50 44 TOTAL SUPPLY 80 L 50 55 60 65 YEAR FIG. 1. U.S. Oyster supply. SOURCE: U.S. Department of Commerce, Fishery Statistics of the United States, various issues. as oyster consumption per capita in Figure 2. Oyster consumption per capita in 1950 was .51 pounds, but in 1975 it was only .31 pounds. Notably, oyster supply has not kept up with population growth. 220 -r 85 180 of .1 FIG. 2. U.S. population, oyster supply, and oyster consumption per capita. SOURCE: U.S. Department of Commerce, Statistical Abstract of the United States, and Fishery Statistics of the United States, various issues. METHODS This study was initiated to provide expected cost and return data for hatchery-produced Pacific oyster seed and to investigate short-run and long- run cost-volume relationships. Factors that affect the cost of oyster (Crassostrea gigas) seed produc- tion were established for hypothetical plants with different capacities of output and with different methods of cultch preparation. Five different capacities were selected on the basis of conditions considered to be practical to commercial hat- PACIFIC OYSTER SEED 19 cheries under current technological capabilities. Data analyzed were adapted from existing com- mercial facilities. Production techniques and operational methods differ by hatcheries. In general, commercial hat- cheries operate nine months per year, and during the winter months the operators are laid off. However, in this study it was assumed that all employees (one manager, one supervisor, two operators, and one half-time bookkeeper), except for those involved in cultch preparation, work on a full-time basis all year round. They produce 15 batches per year, one batch in February, one and one-half batches each in March and April, two batches in each month from May through Septem- ber, and only one batch during the winter period (October through January). During the winter months most labor is devoted to repairing and maintaining the facilities and equipment. Even though the business flow may not be sufficient at all times, especially during the winter months, to keep these operators working at capacity, it is necessary to employ them full-time in order to have these highly skilled operators available when they are needed. Because of the variable production by seasons of the year, the cost analysis has been developed both month-by-month and on an annual basis. This provides a cost-and-income relationship by month as well as by year. The basic model, Plant I, was synthesized to provide general information on building and equipment costs, labor inputs, and other costs in- curred in producing oyster seed. Based on produc- tion costs for Plant I, which has a production 1 The explanation of the economic-engineering approach which is given by Madden is that: In the economic-engineering or synthetic-firm approach, budgets are developed for hypothetical firms, using the best available estimates of the technical coefficients — resource requirements and expected yields — and charging market prices or opportunity costs for all resources. Hypothetical firms are developed in much the same way that an architect or engineer bidding for a construction contract designs a proposed factory or bridge, and estimates the performance and cost of the finished product. Economic-engineering or synthetic-firm analysis is an ap- propriate technique when either of two research questions is asked: (1) What is the average cost per unit of output or pro- fit that firms of various sizes could potentially achieve using modern or advanced technologies, or (2) what are the dif- ferences in average cost per unit of output attributable strict- ly to the differences in size of firm . capacity of 6,000 cases per year, costs for four other model plants, Plants II to V, were projected. Table 2 shows the capacities of the model plants, and Figure 3 shows the variation in projected oyster seed production by month and by plant. Construction costs of a new hatchery building, based on interviews with several contractors in Oregon and Washington, were estimated, at cur- rent prices, to be $25 per square foot, including wiring and piping. Total costs were developed by an economic-engineering approach,1 and analyzed for both short- and long-run conditions. In this study, "short-run" refers to the situation in which the plant's building and equipment are assumed to be invariant with respect to output, while the long-run situation permits changes in building and equipment levels with different output rates. Cultch is the material (usually oyster shell) to which oyster seed attach themselves shortly after hatching. This material, which usually is automatically cleaned in the natural environment, must be cleaned by other means for use in a hat- chery. Five different methods of cultch prepara- tion were used in estimating costs and returns. Selection of an option will be governed by the utility services available at the hatchery site. The descriptions of the options are: Option 1: Pump salt water with city power. Option 2: Pump salt water with own generated power. Option 3: Use city water and power. Option 4: Use city water with own generated power. Option 5: Buy already-cleaned cultch from a local dealer. TABLE 2. Capacities of Model Plants (15 Batches/ Year) Number of cases" Number of cases Plant per year per batch I 6,000 400 II 8,000 534 III 10,000 666 IV 12,000 800 V 14,000 934 "One case is equivalent to 2V4 bushels, and will contain ap- proximately 1,000 to 1,500 pieces of oyster shell, broken and unbroken with an average spat count of 20 spat per shell 20 K. H. IM AND D. LANGMO 1,900 PLANT V 1,500 - o 1,000 - 500 FIG. 3. Production cycle, by plant. Oyster seed production is dependent on six main stages of operation: (1) conditioning adult oysters for spawning; (2) spawning; (3) algal food production; (4) larval rearing; (5) larval setting; and (6) cultch preparation. The only situation in which an operating stage occurs outside of the hat- chery is the case in which precleaned cultch is pur- chased from a supplier. The current price of precleaned cultch is $4.50 per case F.O.B. Initial investment costs consist mainly of land, building, and equipment. Cost of initial invest- ment in land is highly variable in relation to loca- tion and site, and was omitted in this study. This is not a major item affecting the cost of produc- tion, and it is also non- depreciable. Initial invest- ment costs for the oyster hatchery building were estimated on the basis of area requirements at $25 per square foot. Initial investment costs for equip- ment were estimated on the basis of process re- quirements for each plant and option. Fixed costs are the costs that are not a function of the level of output, but are incurred regardless of output level. Costs considered in this study as being fixed include depreciation, interest on in- vestment, insurance and taxes, repair and main- tenance charges for building and equipment, and supervision, administrative, and full-time labor costs. Travel expenses for the manager and other personnel are considered to be fixed. In estimating fixed costs, the following pro- cedures and values were used. Depreciation was calculated using the straight- line method, assuming no salvage value, on the basis of 10 years life for equipment and 30 years for the building. Interest on investment was calculated at 12 percent of undepreciated balance on building and equipment, i.e., 6.6 percent on equipment and 6.2 percent on building, according to the following formula: Average interest = i f2^ where i = interest rate estimated as 12 percent, and n = number of useful years Insurance and taxes were equal to 1 percent and 1.6 percent, respectively, of the total initial invest- ment costs. Repair and maintenance charges were allocated as 1.5 percent of the total initial invest- ment costs for building and equipment. Four and one-half employees, allocated to supervision, ad- ministrative, and full-time labor were considered to be fixed labor for plants I through V. The total estimated wages and salaries, including fringe benefits, were $50,815. Variable costs used in this study include such items as wages of part-time labor, costs of utilities, materials, and supplies, and other expenses direct- ly related to oyster seed production. With the designed model and technology, 4 part-time workers are required to clean 200 cases of oyster shell per day. The labor requirements vary month by month with the plants and options chosen. Also, 2 additional part-time workers, 4 days per week for 5 months (May through September), are necessary for plants IV and V to support full-time workers. Wage rates for these workers, including fringe benefits, were estimated at $3.64 per hour. In addition to having fixed costs associated with repair and maintenance, some machinery requires maintenance which varies with length of usage. PACIFIC OYSTER SEED 21 Some machinery is used for 24 hours per day, regardless of output level, and some is used de- pending on weather conditions. Costs for variable repair and maintenance for machinery were estimated at 0.5 percent of the initial investment costs for that machinery per 100 hours of opera- tion. Some charge for items such as electrical de- mand, water and sewer, garbage, and telephone, are semi-fixed on a monthly basis, regardless of output level. Costs of packaging, advertising, and transportation were not included in the average and the total cost figures. RESULTS AND DISCUSSION Initial Investment Costs. The estimated initial in- vestment costs for building and equipment are presented in Table 3. In each plant the equipment costs of Option 5 are the lowest among all options because, under this option, no equipment is need- ed for cultch preparation. In comparing Plants I and V, there is about a 40 to 54 percent increase in equipment investment costs, varying with cultch preparation options. In turn, Plant V has 133 per- cent greater output capacity than Plant I. The respective total initial investment costs for building and equipment varied from $184,572 to $199,872 for Plant I, and from $286,170 to $301,470 for Plant V. Initial investment costs for building alone accounted for up to 72 to 78 per- cent of the total. Total Costs. Total costs and costs per case for Plants I through V are presented in Table 4. These costs, including both fixed and variable costs, are expressed as annual costs. Total costs vary with options, months of the year, and size of plant. In Options 1 to 4 for the five different plants, the proportion of fixed costs to total costs falls be- tween the range of 61 and 76 percent, but in Op- tion 5 for those plants, the proportion drops to the range of 52 to 67 percent. This is because, under Option 5, cultch preparation costs are more sub- ject to variation with output levels than is the case under the other options. Plant I's average costs vary between $18 and $20 per case, depending upon the option chosen, and Plant V operates between $11 and $13 per case. Since the current market price of hatchery seed is about $23 per case, the difference between market price and average costs in each option and plant would be the net returns per case for that particular option and plant. Monthly and cumulative seed production, and total receipts and costs for Plant I, appear in TABLE 3. Total Initial Investment Costs for Building and Equipment for Each Option and Plant Plant number Item I II III IV V Output capacity per year. 6,000 8,000 cases 10,000 12,000 14,000 Building Equipment: Option 1 Option 2 Option 3 Option 4 Option 5 Total building & equipment: Option 1 Option 2 Option 3 Option 4 Option 5 144,250 51,622 55,622 50,322 54,322 40,322 195,872 199,872 194,572 198,572 184,572 166,675 57,948 61,948 56,648 60,648 46,648 224,623 228,623 223,323 227,323 213,323 . dollars 184,425 62,806 66,806 61,506 65,506 51,506 247,231 251,231 245,931 249,931 235,931 203,700 67,755 71,755 66,455 70,455 56,455 271,455 275,455 270,155 274,155 260,155 223,875 73,595 77,595 72,295 76,295 62,295 297,470 301,470 296,170 300,170 286,170 22 K. H. IM AND D. LANGMO TABLE 4. Total Costs and Average Costs Per Case for Each Option and Plant Plant number Item III IV V Output capacity per year . . . cases 6,000 8,000 10,000 12,000 14,000 Option 1: Fixed costs . . . Variable costs . Total costs . . . Average costs . Option 2: Fixed costs . . . Variable costs . Total costs . . . Average costs . Option 3: Fixed costs . . . Variable costs . Total costs . . . Average costs . Option 4: Fixed costs . . . Variable costs Total costs . . , Average costs . Option 5: Fixed costs . . . Variable costs . Total costs . . . Average costs . 82,119 26,364 108,483 18.08 82,947 26,286 109,233 18.20 81,850 29,760 111,610 18.60 82,678 30,070 112,748 18.79 79,780 39,687 119,467 19.91 86,478 32,601 119,079 14.88 87,306 32,639 119,945 14.99 86,209 36,974 123,183 15.40 87,037 37,414 124,451 15.56 84,139 50,507 134,646 16.83 dollars 89,897 38,845 128,742 12.87 90,725 38,998 129,723 12.97 89,628 44,005 133,633 13.36 90,456 44,574 135,030 13.50 87,558 61,334 148,892 14.89 93,543 50,529 144,072 12.01 94,371 50,797 145,168 12.09 93,273 56,405 149,678 12.47 94,101 57,102 151,203 12.60 91,203 77,595 168,798 14.07 97,495 56,768 154,263 11.01 98,323 57,153 155,476 11.10 97,226 63,345 160,571 11.47 98,054 64,173 162,227 11.58 95,156 88,415 183,571 13.12 Figure 4. This figure reveals the distribution of total receipts, total costs, and total returns which would be generated through the year for Plant I. The vertical distance between total receipts and total costs represents cumulated net returns. Cumulative total costs of Option 5 are the highest, and those of Option 1 are the lowest. Cumulative total costs of all other options, Options 2 through 4, fall within this range. Average Costs Per Case. Average costs are estimated by taking total costs and dividing by cases produced. Average costs vary with options, size of plant, and months of the year. During the winter months (October through January), aver- age costs per case for Plant I and Plant V are around $80 and $43, respectively. During the sum- mer months (May through September), depending on the option, these costs for Plant I varied be- tween $12 and $14, and for Plant V they varied be- tween $8 and $10. These are the extreme cases — the highest and the lowest costs — through the year. But annual average costs for Plant I ranged from $18 to $20, and those costs for Plant V rang- ed from $11 to $13. Figure 5 shows these relation- ships. PACIFIC OYSTER SEED 23 a a CUMULATIVE PRODUCTION AND RECEIPTS 7 / CUMULATIVE TOTAL COSTS ~T FOR OPTION 5 CUMULATIVE TOTAL COSTS FOR OPTION 1 MONTHLY PRODUCTION AND RECEIPTS 138 115 92 69 46 23 JFMAMJJASOND MONTH FIG. 4. Monthly and cumulative oyster seed pro- duction, total receipts, and costs for Plant I. Net Returns. Net returns refer to the total receipts after deducting all costs incurred to the production of oyster seed. In general, winter months (October through January) are the only months which have negative net returns. Following this period, net returns increase and reach the peak during the summer months (May through September). Table 5 shows the efficiency between net returns and total costs for the various plant sizes and cultch preparation options considered. This table gives some idea how much average net returns would be created for each dollar of total costs for each option and plant. Table 6 shows the estimated average net returns per case for the various plant sizes and cultch preparation options considered. In this study the net returns per case increase with the capacity of the plant. Cosfs of Cultch Preparation — As stated earlier, Option 5 does not have a cultch preparation stage. Therefore, in Table 6 the comparison of average « ANNUAL AVERAGE COSTS PER CASE MARKET PRICE PER CASE Upper limit I I I I I I I I Winter costs {HI 1 1 1 1 ♦ 1 1 1 1 1 1 1 1 1 1 1 1 -H- I l I I I I I I I I I I I I I I I I I I I I I I TttT* Mil' I I I I MM1 l I l I « . I I rrr i i i i i i it* }i» Lower Unit - Summer costs Option 12345 12345 12345 12345 12345 FIG. 5. Range of average costs per case through the year for each option and plant. net returns per case between Options 1 through 4 and Option 5 indicates by how much costs can be reduced in each operation if cultch is prepared within the hatchery's own facilities. More specifically, depending on the option chosen, $1.12 to $1.83 and $1.54 to $2.11 per case are sav- ed in Plant I and Plant V, respectively, if cultch is prepared in the hatchery's own operations. This result is dependent, of course, upon the assumed cost conditions for cultch preparation and the assumed price of $4.50 per case of precleaned cultch. Table 7 shows the average costs per case associated with cultch preparation. Since there is no cultch preparation stage in Option 5, the cur- rent market price of $4.50 per case (as of April 1976) was assigned in Option 5. The average cultch preparation costs per case for Plant I ranged from $2.67 to $3.38, and for Plant V from $2.39 to $2.96. These costs decrease with increasing plant capacity, mainly because the charges of city water and power are lower per unit as use increases. Table 8 demonstrates that costs of cultch prepara- tion contribute a substantial percentage of total costs required per case of oyster seed. For Option 5, purchased cultch accounts for about 23 to 34 percent of the total costs. 24 K. H.IMANDD. LANGMO i TABLE 5. Efficiency: Average Net Returns Per Dollar of Total Costs Option Plant Cases number capacity dollars. I 6,000 .27 .26 .24 .22 .16 II 8,000 .55 .53 .49 .48 .37 III 10,000 .79 .77 .72 .70 .54 IV 12,000 .92 .90 .84 .83 .64 V 14,000 1.09 1.07 1.01 .98 .75 TABLE 6. Average Net Returns Per Case Option Plant Cases number capacity dollars. I 6,000 4.92 4.80 4.40 4.21 3.09 II 8,000 8.12 8.01 7.60 7.44 6.17 III 10,000 10.13 10.03 9.64 9.50 8.11 IV 12,000 10.99 10.91 10.53 10.40 8.93 V 14,000 11.99 11.90 11.53 11.42 9.88 TABLE 7. Average Costs Per Case Associated with Cultch Preparation Option Plant Cases number capacity 12 3 4 5 dollars I 6,000 2.67 2.79 3.19 3.38 4.50 II 8,000 2.55 2.66 3.07 3.23 4.50 III 10,000 2.48 2.58 2.97 3.11 4.50 IV 12,000 2.43 2.52 2.90 3.03 4.50 V 14,000 2.39 2.48 2.85 2.96 4.50 TABLE 8. Cosrs of Cultch Preparation as a Percentage of Total Costs Per Case Option Plant Cases number capacity 1 2 3 4 5 . . percent . . I 6,000 14.8 15.3 17.2 18.0 22.6 II 8,000 17.1 17.7 19.9 20.8 26.7 III 10,000 19.3 19.9 22.2 23.0 30.2 IV 12,000 20.2 20.8 23.3 24.0 32.0 V 14,000 21.7 22.3 24.8 25.6 34.3 PACIFIC OYSTER SEED 25 Major Components Affecting the Cost of Produc- tion. Labor costs are the major component affec- ting the cost of production, ranging from 39 to 50 percent of the total in Options 1 to 4 for most plants. But in Option 5 these proportions fall to a range of 30 to 43 percent, largely because of the higher costs devoted to purchased cultch. Cost of utilities, materials, and supplies are the next major component affecting the cost of pro- duction, ranging from 19 to 29 percent of the total in Options 1 to 4, and from 31 to 43 percent in Op- tion 5. Figure 6 shows these relationships. Costs of labor as a percentage of the total decrease as plant size increases, but the reverse is true for utilities, materials, and supplies. Costs other than these components, as a percentage of the total, are relatively stable for the several options and plants. Economies of Size. As the size of plant and the scale of operation become larger, certain economies are usually realized. That is, after ad- justing all inputs optimally, the unit cost of pro- UTILITIES, MATERIALS, & SUPPLIES LABOR FOR SUPERVISION, ADMINISTRATIVE, FULL-TIME AND PART-TIME Option 1234512345123451234512345 Plant I II III IV V FIG. 6. Cosfs categories as a percentage of total costs per case for each option and plant. duction can often be reduced by increasing the size of plant. Two broad forces — specialization of labor, and technological factors — enable pro- ducers to reduce unit cost by expanding the scale of operation. These forces give rise to the negatively-sloped portion of the long-run average cost curve. (Ferguson, 1969). Analysis of size economies is usually considered in terms of short- and long-run situations. Ac- cording to Madden (1967), short-run economies are viewed as resulting from fuller utilization of the fixed plant, and long-run economies as resulting from efficiencies obtained by changing plant size, presumably involving a longer time period. The treatment of any resources as "fixed" is usually based on the length of the planning horizon being examined, the longevity of the resources involved, and the costs of changing these resources. Which resources are treated as "fixed" in the short-run has no effect on the even- tual shape of the long-run average cost curve. The long-run average cost curve assumes all resources are variable, including those designated as fixed in the short-run. A curve that is drawn tangent to the short-run curves approximates the long-run economies-of-size curve for that range of output represented by the short-run curves. This curve in- dicates the average total cost of production that would be experienced by firms of different sizes under assumed price relationships and technolo- gies. The long-run production cost curve or function is a relationship between costs and output which shows the minimum average production costs for any level of output when all inputs are variable. Figures 7 and 8 show the relationship of short-run to long-run average production costs. These are "fixed" factors associated with each of these figures, however. In Figure 7 the Option 1, cultch preparation method, applies to all of the curves while, in Figure 8, the Option 5 is used. As can be seen in these figures, the solid lines are the long-run average production cost curves or the long-run planning cost curves, and the dotted lines are the short-run average production cost curves for the fixed plants, Plants I to V. The long- run average cost curve is downward sloping, and means that as the size of plant increases, the average costs per case decrease when plants are 26 K. H. IM AND D. LANGMO w in < u W Bu, en H c/} O u z o l-l H U Q o w o ss > < 26 24 22 20 16 14 12 10 9 0 I III V LONG-RUN COSTS — COSTS FOR FIXED PLANTS O FULL CAPACITY 90% CAPACITY V 80% CAPACITY 26 24 22 20 18 16 14 12 _]_ > 10 9 0 4,000 6,000 8,000 10,000 12,000 PRODUCTION PER YEAR, CASES FIG 7. Relation of short -run to long-run average production costs in Option 1 for Plants I to V 14,000 26 < u a w p- CO H to o O z o M H U Q 8 U o ss > < 24 - 22 20 18 16 14 12 10 I III V » * t LONG-RUN COSTS COSTS FOR FIXED PLANTS O FULL CAPACITY • 90% CAPACITY X 80% CAPACITY AC C- 26 J_ _L - 24 22 20 18 -\ 16 14 12 10 9 FIG 0 4,000 6,000 8,000 10,000 12,000 14,000 PRODUCTION PER YEAR, CASES 8. Relation of short-run to long-run average production costs in Option 5 for Plants I to V. PACIFIC OYSTER SEED 27 operating at near capacity. The short-run average production cost curves, all the dotted lines in figures 7 and 8, are moving toward the long-run average cost curve until they coincide, when the rate of output nears capacity. For any plant, operating at below capacity increases average costs significantly. There are definite economies of size with in- creased plant capacity. As the plant capacity in- creases from 6,000 to 14,000 cases per year, average costs per case decrease around 35 to 40 percent for all options. The downward-sloping long-run average production cost curve indicates that further economies of scale might exist for even larger plants. The slopes of the long-run average production cost curves are negative and, within the output range examined, do not become parallel to the horizontal axis, because each suc- cessive plant has a lower average cost per case when it operates at its planned capacity. Figure 9 shows the long-run average production cost curves for the different options. Option 1 is the most favorable, and Option 2 is the next favorable situation, compared with other options, and Option 5 is the least favorable. Finally, this study concludes that, under the conditions of current market prices, producing Pacific oyster seed in hatcheries is economically feasible in the Pacific Northwest. ACKNOWLEDGEMENTS This work is a result of research sponsored by the Oregon State University Sea Grant College Program, supported by NOAA Office of Sea Grant, Department of Commerce, under Grant #04-6-158-44004. The authors would like to express their ap- preciation to Professor Wilbur Breese, Depart- ment of Fisheries and Wildlife, Oregon State University, who was the principal investigator of the project of which this study is a part. In addi- tion, particular appreciation is due Dr. Vance Lipovsky, Hatchery Manager, Coast Oyster Com- pany, South Bend, Washington, who was the ma- w < o a. w c- in H to O o z o M H u Q O OS PL, w o > < 26 r- 24 - 22 20 J. 18 16 1A 12 10 9 o'-v^- FIG 0 4,000 6,000 8,000 10,000 12,000 PRODUCTION PER YEAR, CASES 9. Long-run average production costs in different options for Plants I to V. 14,000 28 K. H. IM AND D. LANGMO jor cooperator in providing data and contributing ideas for this study. Cooperation also was gratefully received from government agencies, industries, and commercial oyster farmers. Only the authors are, of course, responsible for any deficiencies that may be pre- sent in this study. LITERATURE CITED Ferguson, C.E., 1969. Microeconomic Theory, rev. ed. Richard D. Irwin, Illinois, 521 p. Madden, J. P., 1967. Economies of Size in Farm- ing: Theory, Analytical Procedures, and a Review of Selected Studies. U.S. Dept. Agr., Agr. Econ. Rep. No. 107, 83 p. READING LIST Im, K.H., R.S. Johnston, and R.D. Langmo, 1976. The Economics of Hatchery Production of Pacific Oyster Seed: a Research Progress Report. Proc. Nat. Shellfish. Assoc. 66: 81-94. Park, W.R., 1973. Cost Engineering Analysis: A Guide to the Economic Evaluation of Engineer- ing Projects. John Wiley & Sons, N.Y., 308 p. Peeler, R.J., Jr. and R.A. King, 1963. In-plant Costs of Grading and Packing Eggs. A.E. In- formation Series No. 106, Univ. of North Carolina, Raleigh, 36 p. Quayle, D.B., 1969. Pacific Oyster Culture in British Columbia. Bull. No. 169, Fisheries Research Board of Canada, Ottawa, 193 p. Reed, R.H., 1959. Economic Efficiency in Assembly and Processing Lima Beans for Freez- ing. Calif. Agr. Exp. Sta., Giannini Foundation, mimeo. Rep. No. 219, 106 p. Proceedings of the National Shellfisheries Association Volume 67 — 1977 ASSESSMENT OF SURF CLAM STOCKS IN NEARSHORE WATERS ALONG THE DELMARVA PENINSULA AND IN THE FISHERY SOUTH OF CAPE HENRY1 2 Joseph G. Loesch and John W. Ropes VIRGINIA INSTITUTE OF MARINE SCIENCE GLOUCESTER POINT, VIRGINIA and NATIONAL MARINE FISHERIES SERVICE MIDDLE ATLANTIC COASTAL FISHERIES CENTER OXFORD, MARYLAND ABSTRACT In 1974 the abundance of surf clams was sampled from Delmarva Peninsula, Delaware south to North Carolina. Surf clams were not found in commercial densities in the inshore waters along the Delmarva Peninsula. Off shore and south of Cape Henry, an area of intense surf clam fishing, the estimated standing crop was 10 million bushels. A length-age relationship was estimated and it implies that recruitment to the fishery occurs at approximately age 2, at an average annual rate of about 8%. It is con- cluded that because of the low recruitment rate relative to the heavy fishing pressure that Virginia surf clam stocks have been overharvested in recent years. INTRODUCTION The fishery for surf clams, Spisula solidissima, presently supplies meats for about 80% of all clam products in the United States. In the late 1940s and early 1950's the surf clam was a relatively unknown resource, but the fishery has since ex- perienced dramatic growth. In 1950, for instance, only 8 million lbs. of surf clam meats were landed; by 1974, however, the reported meat landings were 96 million lbs. (Current Fishery Statistics, 1974). Beds located off the New Jersey coast were the major source of surf clams until the late 1960's (Ropes, 1972). Since then effort has shifted to beds off the Delmarva Peninsula and Virginia. Virginia landings of 58 million lbs. of surf clam meats in 1974 were 60% of the United States total. Contribution No. 805. Virginia Institute of Marine Science, Gloucester Point, Virginia 23062. Research sponsored by NOAA, National Marine Fisheries Service, Contract No. 03-4-043-357. Declining surf clam densities in the overfished beds off New Jersey promoted consideration of management plans for the fishery. In June, 1973, representatives from industry, the National Marine Fisheries Service (NMFS) and the States of New York, New Jersey, Delaware, Maryland, and Virginia formed a Surf Clam Technical Committee and a Sub-Council. The functions of the Technical Committee are to direct investigations of the resource and identify management alternatives. The Sub-Council, guided by the findings of the committee, is to formulate management policy. These two bodies are part of a more comprehen- sive State-Federal Fisheries Management Program administered by the Northeast Marine Fisheries Council. This report is an account and analysis of the in- vestigation of the surf clam resource in October, 1974, in the inshore waters of the Delmarva Penin- sula, and in the area offshore of Cape Henry, Virginia and south to upper North Carolina. The 29 30 J. G. LOESCH AND J. W. ROPES inshore investigation along the Delmarva Penin- sula complemented an offshore investigation in this region by NMFS in August, 1974. The main objectives of the study were to estimate the distribution and abundance of adult and juvenile surf clams along the Delmarva Peninsula and in areas of intense harvesting off the Virginia coast. The project was a joint undertaking by NMFS and the Virginia Institute of Marine Science (VIMS). MATERIALS AND METHODS Surf clams were sampled by a hydraulic tow dredge operated from the VIMS research vessel RETRIEVER. The dredge, supplied by the NMFS, is similar to those employed in the surf clam fishery, but smaller. It has a 76.2 cm (30 inches) blade versus blades ranging up to 254 cm (100 in- ches) on industrial models. The dredge has a reten- tion bag constructed of 5.1 cm (2 inches) rings ver- sus 7.6 cm (3 inches) rings or cage bars generally used throughout the industry. The relationships of sample catch and its size composition to the total population is unknown since the catch-efficiency of the dredge with respect to surf clams less than 5.1 cm is not known. Vessel speed was estimated to be 0.5 knot while towing the dredge, thus it was assumed that a standard 5-minute tow provided a sampling unit of 58.8 m2 (632.9 ft2) for stock assessment. Arbitrarily, a surf clam catch ^ 45 clams was considered satisfactory in the sense that the im- mediate area would warrant future replicate sampling to determine a reliable average catch and the extent of the local distribution. This figure (45) was derived in consideration of the necessity to maintain a constant sampling unit, whereas an ex- perienced fisherman would make gear adjustments to enhance catch according to sea conditions and bottom type. Sampling stations along the coast of the Delmarva Peninsula were established along lines of latitude at intervals of 1.8, 3.7, and 5.6 km (1, 2 and 3 nautical miles) offshore of the 1 fathom line indicated on the National Ocean Survey chart no. 1109 (Figs. 1 and 2). These transects were spaced at intervals of 9.3 km (5 nautical miles) from just below Cape Henlopen (Rehoboth Beach area), Delaware, to Cape Charles, Virginia. An addi- tional transect of three stations in a north to south direction was sampled inshore near Cape Henry, a a HG. 1. Location of sampling stations in the near- shore waters of the upper Delmarva Peninsula. Numbers above the stations indicate the catch of surf clams. HG. 2. Location of sampling stations in the near- shore waters of the lower Delmarva Peninsula. Numbers above the stations indicate the catch of surf clams. ASSESSMENT OF SURF CLAM STOCKS 31 Virginia (Fig. 3). Offshore of Cape Henry and fur- ther south, sampling was conducted along a rec- tangular grid constructed of six stations on each of 12 transects, in which both stations and transects were at intervals of 4.6 km (2.5 nautical miles). The grid duplicated one sampled by NMFS in August 1974. Surf clam density was approximated from the product of average catch and area. Sampling did not follow a stratified random sampling procedure or systematic sampling as defined by Cochran (1963) since all station locations were selected. Because there was no underlying probability model, standard errors could not be validly calculated nor interval estimates of densities established. A constant of 12.6 lbs. of usable meats per bushel was used to estimate standing crop in terms of meat weight. This value, an overall average yield per bushel for 1974 and 1975, was reported by Mr. N. Doughty, owner and operator of C & D Seafood Inc., Oyster, Virginia (Loesch, 1977). The constant of 17 lbs. of meats per bushel used in the U.S. Current Fishery Statistics for converting 75°|i0 ¥ . T 4 • ? ? JSI 82 12 54 214 * • * 34 IS • ■'*.-* T . 58 33 )3 5 4 • i •: M s (■i FIG. 3. Location of sampling stations off the coast of lower Virginia and upper North Carolina. Numbers above stations indicates the catch of surf clams. bushels to meat weight includes the viscera which is not used by the surf clam industry. A station is referred to by the transect number followed by its offshore position, e.g., T4(3) is the third station, counting from inshore to offshore, on transect 4 (Fig. 1). Three stations, T14(1), T20(2), and T33(5) were not sampled. At each station, the catch of surf clams to the nearest 0.1 bushel of clams was measured for length (longest linear dimension). Two growth curves published by Yancey and Welch (1968) for surf clam stocks of Long Island, New York and off New Jersey were re-evaluated in this report. The age-length relationship for the Long Island clams was ascertained from the growth curve in the unpublished manuscript of Westman and Bidwell (1946); the New Jersey surf clam data were supplied by Welch (personal com- munication). The Walford analysis (Walford, 1946) was used to transform asymptotic growth functions to the linear form: Lr+1=Loo(l— k)+kL, where L, = length at time t; L,*, = length at the end of a constant time interval (one year in the present cases); Lood — k) = regression line in- tercept; k = the regression coefficient; and Loo is the asymptotic size, i.e., the average maximum size. The equation is independent of age, but the age-length relationship was estimated by using 0.24 mm (0.01 inch), the midpoint of the general size range of newly settled surf clam spat reported by Loosanoff, et al. (1966). At this time, when the larvae leave the planktonic environment and become members of the benthic community, they were established to be age zero. Substitution of the estimated average (0.24 mm) at age zero into the growth function produced an estimate of length at age 1. Growth curves were generated by continu- ing this process until arbitrarily terminated at age 20. Average annual recruitment since 1969, the year the area was last surveyed by NMFS, was estimated by assuming a maximum length for age 5 occurred at the mid-point between its average length and the succeeding age group's average length. The short-comings (size overlap by age groups) of this procedure are recognized by the authors, but methods for determining the in- dividual age of surf clams and, thus, stock age 32 J. G. LOESCH AND J. W. ROPES structure have not been developed. (Perhaps growth and age estimates from cross-sectioned shells as presently done with several bivalve species (e.g., Kennish and Olsson, 1975) may eventually be shown applicable. ] A Smith-Maclntyre benthic sampler was employed at each station to sample for juvenile clams. A single 0.1 m2 (1.08 ft2) grab sample was taken at each station and wet sieved on a 1 mm (0.04 inch) mesh screen. The portion retained was preserved in 5% formalin and returned to the laboratory for examination. RESULTS AND DISCUSSION Distribution and Abundance. A commercial density of surf clams was not found in the inshore waters along the Delmarva Peninsula (Figures 1 & 2). Surf clams were obtained at only six of 58 sta- tions sampled. The total catch was 271 and the average catch was 4.7 clams per standard tow. Commercial abundance was indicated at only one site, T4(2) where the catch, 233 clams, was about 87% of the total catch along this Peninsula. This concentration of surf clams was very limited in its distribution since no clams were taken at the adja- cent sites T4(l) and T4(3), nor along transect T3, and only two clams were taken along transect T5. No surf clams were taken at the three inshore stations (T21) off Cape Henry (Fig. 3). Offshore of Cape Henry and south to upper North Carolina, 71 stations were sampled (Fig. 3). A total of 2,474 surf clams were taken, averaging 34.8 clams per tow. Two areas of heavy surf clam density were apparent. One was along T23 and T24 where 8 of 12 catches ranged from satisfac- tory (>45 clams) to the highest recorded (394 clams). Another group of five spatially associated high catches occurred along T26 and T27. Only four other stations had catches ^ 45 (T24(6); T28(5); and T29(2&4)]. The catch distribution for the NMFS surf clam cruise in August, 1974, ex- hibited a similar trend (Ropes, 1974). Standing crop estimates, derived from the average catches, are presented in Table 1 for the entire area, transects T22 through T33 (343.75 miles2), and also for the area between T23 and T29 (187.5 miles2), the north-south boundaries of the highest observed densities for both the NMFS and VIMS cruises. Approximately 89% of the estimated standing crop of about 10 million bushels of surf clams occurred within the T23-T29 boundaries. Estimation of Growth. The surf clam growth curve presented by Westman and Bidwell (not shown) does not appear realistic for Virginia stocks. By the 17th year the curve still does not tend toward an asymptotic size (Loo) and the Walford analysis indicated that Loo would not be attained until about age 38. Thus, one would have to assume the surf clam lived for well over 40 years. Surf clam longevity is not known, but about 17 years has been suggested (Ropes, et al., 1969). The growth function ascertained from Welch's data by least squares analysis of length at suc- cessive check marks which he interpreted as an- nual marks is: L,+, = 47.05 + 0.6807 L, where length is expressed in mm. Substitution of age zero length, i.e., 0.24 mm, the average length of newly settled spat, and the subsequent substitu- tion of each estimated average length at 1 year in- tervals produced a curve which appears to be a reasonable approximation of surf clam growth in the Virginia fishery area (Fig. 4). This contention is supported by the reported size of juvenile surf clams of known age off Chincoteague, Virginia (Ropes, et al. . 1969). A more intense growth study TABLE 1. Standing crop estimates for surf clams in the Virginia fishery area south of Cape Henry. VIMS cruise, October, 1974. Area Number Samples Average Catch Bushels Per Acre Total Bushels (X 10") Meat Wts. (lbs.) (X106)* T22-T33 T23-T29 71 42 34.8 56.6 34.2 55.5 9.96 8.84 125.50 111.38 'Estimates based on 12.6 lbs of usable meat per bushel. ASSESSMENT OF SURF CLAM STOCKS 33 100 8 10 12 14 AGE (YEARS) FIG. 4. The length-age relationship for surf clams derived from the data of Welch. is needed if long term management of the fishery is considered, since Figure 4 was derived from the data of Welch, who made only 90 check-mark measurements. The average maximum length (Loo) was estimated to be 147 mm (5.8 inches) and theoretically reached at about age 14 (Table 2). Of more practical importance are the estimates that 95% of Loo occurs at age 8 and 50% by about age 2. The growth curve indicates that recruitment to the Virginia surf clam fishery occurs at age 2, since 76.2 mm (3-inch) rings or cage bars are used in the commercial dredges. Thus, there are not several 0.500 H o o 0400 z o 0300 0200 0.1 00-1 0000 "Shell cavity water North Santee Bay North Santee River Bottom water South Santee River 1.000 (-0 900 0.800 0.700 o t- < 0.600 °=- > 0500 > 0400 2 a 2 0 300 8 0 200 0 100 0 000 313 9IYI7H24IYI7 137 202 671 871 3m 3HI 9E 1 717 2417 I 7 1 37 205 621 912 I7H24I2I2 l32 205 621 2 HI 2 IE 1975 1976 1975 1975 1976 HARD CLAMS FIG. 2. Hard clam inner shell cavity conductivities and bottom water conductivities/ salinities in Santee River 1975-1976. 40 V. G. BURRELL, JR. on these dates shell liquor conductivity exceeded surrounding bottom water, following the pattern at the other two sampling sites. Subtidal oyster conductivity from North Santee Bay exceeded outside water conductivities on each sampling date (Fig. 3). The difference was much greater during periods of high runoff than at times of low runoff. A greater range of conductivity was observed in oysters than hard clams. In North Santee River, liquor conductivity of subtidal oysters was much higher than ambient during periods of high runoff. However, on two sampling dates, when outside salinities were above 28 o/OO, oyster liquor conductivity was lower than am- bient. Subtidal oysters from the South Santee followed the pattern of the other two systems ex- cept in June, when inner shell fluid samples and an extremely high conductivity reading. In both the North and South Santee River, intertidal oysters maintained higher liquor conductivities than the surrounding water during periods of high runoff, but generally had lower inner shell water con- ductivity ratios than did subtidal oysters (Fig. 4). When external salinities increased, liquor con- ductivities fell below that of surrounding water. Liquor conductivities of subtidal oysters and hard clams in North Santee Bay, though much higher than ambient in periods of high runoff, did reflect changes in external ionic concentration throughout the entire sampling period in 1975. During periods of low runoff in 1976, clam liquor conductivity closely reflected outside salinity (Fig. 2). The same was true for subtidal oysters in North Santee Bay and South Santee in 1976 (Fig. 3). Conductivities from intertidal oysters reflected their position above the lower water mark. The lower the position, the lower the conductivity of inner shell water (Fig. 4). Starfish, oyster drills, whelks, and boring sponges were not observed in any dredge samples taken from subtidal shellfish beds or in intertidal oyster beds in the Santee system in spring of 1975. Starfish (Asterias forbesi) and the common oyster drill (Urosalpinx cinerea) were present in dredge samples taken over Santee subtidal shellfish beds in 1976. < tr o ■=> o z o o 3 12 9El7E24J2l2 132 202621 62E 3fli 31 9E I7I224EI2 132 202 621 9B I7E 2«E l¥ 132 202 621 2JD 21 1975 1976 1975 1975 1976 SUBTIDAL OYSTERS FIG. 3. Subtidal oyster inner shell cavity conductivities and bottom water conductivities/ salinities in Santee River 1975-1976. MORTALITIES IN SANTEE RIVER 41 1.000 0 900 0800H 0700 O 2 0600 >- t- > 0 500- g 0400 o o 0300 0 200 0 100- 0000- * • Shell cavity water North Santee River • ■ Bottom woter . South Sontee River I 000 0900 0800 0700 o -0 600 c > t- -0500 > i- o 0400 g 0300 0200 0 100 0000 O -i 1 1 1 1 1 1 1 — 9EZ I7U24I3IY 132 202 621 912 I7r22412l2 132 202 621211 2IE 1975 1975 1976 INTERTIDAL OYSTERS FIG. 4. Intertidal oyster inner shell cavity conductivities and bottom water conductivities/ salinities in Santee River 1975-1976. DISCUSSION The distribution of hard clams indicates that they are more stenohaline than oysters (Wass, 1972). However, short-term tolerance of low salinity appears greater in clams than oysters (Haven et al., 1975). This may be due to the abili- ty of clams to cease or restrict pumping for longer periods than oysters, or that clam microhabitats are more conducive to less metabolic activity than oyster microhabitats. Dugal (1939) indicated that shell calcium buffered the acid products of anaerobic metabolism in both clams and oysters and that the efficiency of this mechanism increas- ed with decreased temperatures. Santee clams were buried and subject to lower ambient temperatures because of buffering effect in sediments to rising spring temperatures (Carriker, 1967). Subtidal oysters, which for the most part lie on a shell bottom or intertidal oysters which may be exposed to as much as 10 hours of highly variable air temperatures daily, are subject to higher ambient temperatures. Clam shell cavity conductivity showed an upward trend while am- bient salinities were low, indicating reduced or suspended pumping activities. Observed changes in ionic concentration of South Santee Bay hard clam and subtidal oyster liquor, remaining higher but paralleling external conductivities, indicate that some exchange, however, probably occurs during low salinity periods. Hard clam con- ductivity fluctuated less with changing salinities than did oysters. This might explain to some degree the higher mortalities found by Haven et al. (1975) in relaid clams. If these clams were not as well buried in the bottom, and thereby in- sulated from higher water temperatures, then ability to reduce pumping might well be affected. Andrews et al. (1959) found oysters conditioned to low salinity at low temperatures were able to withstand low salinities in a state of "narcosis" for long periods. This reaction to low salinity pro- bably did not occur in Santee oysters because low 42 V. G.BURRELL, JR. salinity came during a period when oysters were actively pumping (Fig. 3). The higher mortality of subtidal oysters in North Santee may be a result of higher freshwater discharge though this distribu- tary resulting in more pronounced lower salinities. Salinity also might be responsible for higher losses in intertidal than subtidal oysters in South Santee and the equally high losses in intertidal oysters in North Santee. The samples showed a strong sur- face to bottom salinity gradient was present at times during the high runoff in spring of 1975. This would subject intertidal oysters to lower salinities than subtidal oysters as they come in contact with surface water at some stage of nearly every tide. Low dissolved oxygen concentrations in the water column often accompany high freshwater run off (McHugh 1967). Such conditions, together with the hydrogen sulfide produced when lower strata of the water column become anaerobic, have been implicated in mortalities of shellfish (Carpenter and Cargo, 1957; Tamura, 1966). No evidence that this may have occurred in the Santee System during 1975 was observed either by low dissolved oxygen values or black shell coloration of dead or dying oysters. In addition, highest mor- talities occured in the intertidal zone where ox- ygenation of the water column would be highest. Predators often implicated in high shellfish mor- talities were not observed in any samples taken in Spring of 1975. Therefore, most if not all mor- talities observed during the spring of 1975 can be attributed to low salinity. Starfish and drills were abundant in all subtidal areas during 1976, and were probably responsible for many of the deaths recorded in these samples. SUMMARY 1. High mortalities occurred in subtidal and in- tertidal oyster populations in the Santee River during the spring of 1975, concurrent with high freshwater runoff. 2. Deaths among hard clam populations were substantially fewer than in oysters. 3. Liquor conductivity suggested that hard clams remain closed for longer periods than oysters. 4. The ability of an organism to remain closed might be related to microhabitat, with the sedi- ment providing insulation against rising tempera- ture. 5. Common shellfish predators were not present in 1975 during the low salinity period. 6. Low dissolved oxygen did not accompany low salinity in the Santee River in 1975. ACKNOWLEDGEMENTS I am indebted to my colleagues, John J. Manzi, Dale R. Calder and Paul A. Sandifer for their critical review of the manuscript, to Mrs. Lexa Ford for the typing and Mrs. Evelyn Myatt for drafting the figures. I particularly appreciate the dedicated manner in which W. Z. Carson carried out the field sampling program. This research was funded in part through the National Sea Grant Program under grant numbers 04-5-158-5 and 04-6-158-44009. LITERATURE CITED Andrews, J. D., D. Haven and D. B. Quayle. 1959. Freshwater kill of oysters (Crassostrea virginica) in James River, Virginia, 1958. Proc. Nat. Shellfish. Assoc. 49:29-49. Baughman, J. L. 1948. An annotated bibliography of oysters with pertinent material on mussels and other shellfish and an appendix on pollu- tion. Texas A & M Res. Found. College Station, Texas. 794 p. Carriker, M. R. 1967. Ecology of estuarine benthic invertebrates: A perspective. In Estuaries, G. Lauff (ed,) AAAS. Washington, D. C. Carpenter, J. H. and D. G. Cargo. 1957. Oxygen requirement and mortality of the Blue Crab in the Chesapeake Bay. Chesapeake Bay Institute. The Johns Hopkins University Tech. Rept. 13. 22. p. Castagna, M. and P. Chanley. 1973. Salinity tolerance of some marine bivalves from inshore and estuarine environments in Virginia waters on the western mid-Atlantic Coast. Malacologia 12(l):47-96. Dugal, L. P. 1939. The use of calcareous shell to buffer the product of anerobic glycolysis in Venus mercenaria. J. Cell. Comp. Physiol. 13:235-251. Eldridge, P. J., W. Waltz, R. C. Gracy and H. H. Hunt. 1976. Growth and mortality rates of hat- chery seed clams, Mercenaria mercenaria, in MORTALITIES IN SANTEE RIVER 43 protected trays in waters of South Carolina. Proc. Nat. Shellfish. Assoc. 66:13-20. Galtsoff, P. S. 1964. The American oyster, Crassostrea virginica Gmelin. U.S. Fish. Wildl. Serv. Fish. Bull. 64:1-480. Galtsoff, P. S. 1972. Bibliography of oysters and other marine organisms associated with oyster bottoms and estuarine ecology. G. K. Hall & Co. Boston, Mass. 857 p. Godcharles, M. F. 1971. A study of the effects of a commercial hydraulic clam dredge on benthic communities in estuarine areas. Fla. Dept. Nat. Resour. Mar. Res. Lab. Tech. Ser. No. 64. 51 p. Haven, D. S., W. J. Hargis, J. G. Loesch and J. P. Whitcomb. 1975. The effect of tropical storm Agnes on oysters, hard clams, soft clams and oyster drills. Chesapeake Research Consortium Publ. No. 34:D170-D208. Joyce, E. A., Jr. 1972. A partial bibliography of oysters, with annotations. Fla. Dept. Nat. Res. Sp. Sci.Rept.No.34.846p. Lunz, G. R. 1938. The effects of the flooding of the Santee River in April 1936 on oysters in the Cape Romain area of South Carolina. Mimeo Rept. Corps of Engineers, U. S. Army. 24 p. Manning, J. H. and E. A. Dunnington. 1955. The Maryland soft shell clam fishery: a preliminary investigational report. Proc. Nat. Shellfish Assoc. 46:100-110. McHugh, J. L. 1967. Estuarine nekton p. 581-620. In Estuaries, G. Lauff, (ed.) AAAS, Washing- ton, D. C. Stallings, J. S. 1967. South Carolina stream flow characteristics. Low-flow frequently and flow duration. U. S. Dept. Int. U.S.G.S. Open file Rept. 83 p. Tamura, T. 1966. Marine aquaculture. National Science Foundation. Trans, from Japanese. 1970. United States Geological Survey. 1976. Water resources data for South Carolina. Water Year 1975. U.S. G. S. Water Data Rept. S.C. 75-1. 210 p. Wass, M. L. 1972. A check list of the biota of lower Chesapeake Bay. Va. Inst. Mar. Sci. Spec. Sci. Rept. No. 65. 290 p. AN EMPIRICAL EVALUATION OF THE LESLIE-DeLURY METHOD APPLIED TO ESTIMATING HARD CLAM, MERCENARIA MERCENARIA, ABUNDANCE IN THE SANTEE RIVER ESTUARY, SOUTH CAROLINA1 Raymond]. Rhodes, Willis J. Keith, Peter J. Eldridge, and Victor G. Burrell, Jr. SOUTH CAROLINA WILDLIFE AND MARINE RESOURCES DEPARTMENT CHARLESTON, SOUTH CAROLINA ABSTRACT This paper estimates the abundance of hard clams, Mercenaria, mercenaria, in the Santee River estuary based upon catch and effort data generated by hydraulic escalator clam harvesters between 1974 and 1976. Using the Leslie method, catch per unit of standardized effort at each time interval was regressed on the cumulative catch. The resulting regression equations had regression coefficients (estimates of catchability) of .0006, .0014, and .0006 for the South Santee River, North Santee River and North Santee Bay, respectively. There were an estimated 6.4 million, 5.0 million and 10.7 million clams in the legal harvesting areas of the South Santee River, North Santee River, and North Santee Bay respectively. The density of clams in the preferred fishing areas varied between 18/ m2 and 24/ m2. In this analysis, the Leslie-DeLury method had two major limitations: first, the lack of effort estimates for specific locations, and second, significant gear competition. It is suggested this method should be considered only for supplementing designed, direct sampling. INTRODUCTION equipment. Since 1974, commercial hydraulic , .i c ,„„, .i . i . escalator harvesting in the Santee Delta has In the spring or 1974, the environmental impact ,,..,. , , . ■ c ., .-,.. c l j i- resulted in a significant increase in revenues from and commercial feasibility ot using hydraulic , . ... ,_ . . . ., , , . , i ■ . j ■ i the clam hshery ( 1 able 1) and has become a source escalator clam harvesters was investigated in the . . J r , , , , c n. c 0 ., r, ,. „ , ot seasonal income tor commercial hshermen liv- Santee Kiver estuary of South Carolina. Based on . . , . ~, „ ,, „ , „ , ,. |. j . , , . , ing in McClellanville, bouth Carolina, sampling results and interest by commercial ° , , , , . , , . , , r. i lr, , -„„,s ,f. , r I he Maryland hydraulic escalator harvester has hshermen (Cracy, et ah, 1976), the South Santee , ,:,,,, >. ,„^„ n. , c , .. r , been described by others (e.g., Manning, 1957; Kiver was opened for harvesting ot clams, ,,„,., „ /. , . , ^ ™ , ,„„ , MacPhail, 1961; Mathieson and DeRocher, 1974). The harvester (Figure 1) is basically a cluster of Mercenaria mercenaria, from September to December in 1974. In subsequent clam seasons, ., nx ., 0 . , ., , ~ n water lets in front of a scoop (escalator head) the North Santee river and North Santee Bay were u u . i j l also opened to mechanical harvesting by th is Water jets loosen the substrate; clams and other benthos are flushed onto the conveyor belt and carried to the surface for hand sorting. With a nor- 1 Contribution No. 68 from the South Carolina Marine mal amount of propeller thrust, the escalator head Resources Center. References to firms in this paper do not ., ,, . .„ . _ ,,, ... imply endorsement of commercial products by the State of can be forced through 40 to 45 cm of ... solid South Carolina. bottom . . ." (Manning, 1959). 44 HARD CLAM ABUNDANCE 45 TABLE 1. Reported commercial clam, Mercenaria mercenaria, landings (U.S. bushels) from 1971 thru 1975 shown by method of harvest. Harve. ;t Method Qam Othei Hydraulic Escalator Total Season Quantity Percent" Quantity Percent Total Exvessel Value 1971-72 5,296 100% 0 0 5,296 $ 17,370 1972-73 11,292 100% 0 0 11,292 44,273 1973-74 4,594 64% 2,582 36% 7,176 45,339 1974-75 11,302 27% 30,917 73% 42,220 213,382 "Percent of total clam harvest for the clam season. The Leslie (Leslie and Davis, 1939) and DeLury, (DeLury, 1947) methods for estimating population abundance have been employed for many fishery stocks (e.g., Omand 1951; Ketchen, 1953; and Dickie, 1955). Loesch and Haven (1973) employed the Leslie method for estimating clam abundance when using a hydraulic escalator as a molluscan sampling gear. For the South Atlantic states of North Carolina, South Carolina, Georgia and Florida, no published estimates of clam abundance derived from commercial catch and effort data ex- ist. Consequently, this analysis was performed to evaluate the feasibility of applying the Leslie method to commercial data for estimating clam abundance, and to document the commercial yield from the original populations for future resource management decisions. The second objective seems especially relevant because much of the Cooper River freshwater flow will be rediverted into the Santee River over the next four years. Kjerfve (1976) believes the rediversion will destroy the hard clam and seed oyster beds in the lower Santee River. VESSEL U/v UARY- SELF PROPELLED BARGE LOA 43' BEAM 17' DRAFT 4' NET TONNAGE 16 BUILT 1973, GEORGETOWN, S C ENGINE GU 6-17 DIESEL (2 I REDUCTION) SPEED 8 KNOTS FIGURE 1. A diagram of a hydraulic escalator shellfish harvester employed in the Santee River from 1974 to 1976. 46 R. J. RHODES, W. J. KEITH, D. J. ELDRIDGE, V. G. BURRELL, JR. METHODS AND MATERIALS Description of Harvesting Areas. The Santee River flows southeast through South Carolina, draining a river basin of approximately 41,000 km2 (Anonymous, 1973) before it empties into the Atlantic Ocean 75 km northeast of Charleston, South Carolina (Figure 2). With the completion of the Santee-Cooper Dam in 1942, most of the fresh water discharge in the Santee River was diverted into the Cooper River, thereby decreasing the flow rate of the Santee from 525 m3/s to 80 m3/s (Cum- mings, 1970). At present, approximately 85% of the fresh water reaching the lower Santee Rivery system flows through the North Santee River channels (Cummings, 1970) compared to 15% in the South Santee River channels. Marine processes are eroding the present Santee River delta front, and within the old distributary channels of the North Santee, coarse marine channel sands have accumulated while finer bar sands are being deposited in the shallower water (Stephens, et. al., 1976). The lower Santee River is a partially mixed estuary, although during flood conditions it ap- proaches the vertically homogeneous type (Kjerf- ve, 1976). In the North Santee River mouth, bot- tom salinity ranges from 35 o/oo at flood slack to 32 o/oo at ebb slack (Stephens, et. al., 1976). Bot- 79°25 HARVESTING AREAS SOUTH SANTEE RIVER NORTH SANTEE RIVER NORTH SANTEE BAY 79°20' 79°15 33: 05' TH SANTEE RIVER »\ SOUTH SANTEE RIVER 79*25 79°20 79°I5' 05' HGURE 2. A chart of the Santee River delta, South Carolina, and designated legal harvesting areas from 1974 to 1976. HARD CLAM ABUNDANCE 47 torn temperatures range from 8°C in January to 30°C in August (Shealy, 1976). Bottom dissolved oxygen apparently varies between 5.6 ml/1 to 4.2 ml/1 during the year (Burrell, 1976). Logbook Catch and Effort Data. Under authorization of Section 28-775, S. C. Code of Laws, permits were issued for hydraulic escalator harvesters during the 1974-75 and 1975-76 South Carolina clam seasons. The legally designated fishing areas for harvesting hard clams will be referred to as the "South Santee River", "North Santee River", and "North Santee Bay" (See Figure 2). Permit holders were required to maintain a logbook for each harvesting day. The following information was reported: start of fishing time; end of fishing time; an estimate of time spent repairing gear while in the harvesting area; quanti- ty and grades of clams harvested; quantity of oysters harvested; date of harvesting; vessel operator's name and vessel's permit number. In this study, fishing time is defined as an estimate of the hours spent searching for and harvesting clams. Most operators did not indicate time spent for meals or rest periods. The authors observed that during fishing times that exceeded six hours, rest periods and "lunch breaks" usually lasted one hour except when repairs were performed. An hour was subtracted from reported daily fishing time when the total exceeded seven hours. If repairs were indicated in the logbook form, then the one hour adjustment was not subtracted, but the repair time was subtracted because meals and rest periods were usually taken during this time. After the 1975-76 clam season harvester operators were interviewed to document the spatial distribution of harvesting effort within the legal fishing areas (Figure 2) during previous seasons. Standardization of Fishing Effort. Fishing effort was standardized by selecting a vessel with a "Maryland type" displacement, vessel No. 5, as a standard, and assigning it a relative fishing power of 1. As Beverton and Holt (1957) indicated, the selection of the vessel for a standard can be ar- bitrary, and it need not be an "average" vessel for the fishery. Number of fishing days was considered when selecting the standard vessel to allow the greatest number of direct comparisons. Fishing effort in the South Santee River (nine permits) during 1974 was not standardized due to the changing of vessel operators during the 1974 period and the lack of any previous experience by these new operators. The standardization method was adapted from Gulland (1956) for the ith vessel, fishing in the time period, t, Y„ = cP1-f,-,-D,-e,-, (1) where Y„ = catch P, = relative fishing power of the ith unit f,, = time spent fishing of ith boat in tth interval D, = clam density in tth interval c = proportionality factor e,, = random error term A catch per unit of effort (Time) logarithmical transformation of (1) as suggested by Gulland (1956) becomes, Y log -~ = log c + log P, + log D, — log e„ (2) Two-way analyses of variance tests were perform- ed by utilizing BMD05V, General Linear Hypothesis (Dixon, 1967), to determine if relative fishing powers of vessels and densities of clams varied significantly during the tenure of the two fisheries. Fishing powers and density of clams for both fisheries did vary significantly (See Table 2). Since catches of every vessel in the fishery were used to calculate relative fishing powers, it was TABLE 2. Results of the two-way analysis of variance tests for North Santee River and North Santee Bay fisheries. Source of Variation d.f. North Santee River Relative Fishing Power 8, 114 Density of Clams 20,114 9.14" 14.21** Source of Variation d.f. North Santee Bay Relative Fishing Power 6, 97 Density of Clams 18, 97 15.68** 11.64** The method for calculating F„ summarized by Dixon (1967). 48 R. J. RHODES, W. J. KEITH, D. J. ELDRIDGE, V. G. BURRELL, JR. not necessary to perform regression analysis relating vessel attributes to relative fishing powers. Calculated relative fishing powers obtain- ed by the use of the BMD05V program are shown in Table 3. As described previously, vessel No. 5 was assigned a relative fishing power of 1; hence, it was considered the standard vessel. The Leslie-DeLury Method. The Leslie-DeLury methods are based upon the decline in catch per unit of effort due to the removal of individuals in the exploited population. Since the catch per unit of effort is altered by factors independent of population density (e.g., weather), a series of tem- poral estimates is usually calculated. According to Ricker (1975), the predictive regression line in the DeLury method will underestimate catchability and consequently overestimate the original population abundance. The Leslie method was selected because it was considered preferable to underestimate resource abundance for the management decision process. In the original Leslie method, catch per unit ef- fort at each time interval was regressed on the cumulative catch at the start of the time interval. Using a modification by Braaten (1969) and ter- minology described by Ricker (1975), the equation for the linear regression line is: C, T; = qN„ K, (3) N0 = original population abundance K, = cumulative catch to start of interval t plus half of that taken during the interval (See Braaten, 1969) q = catchability, the fraction of the population taken by 1 unit of fishing effort C, f = catch per unit of effort during the interval t The regression equation is estimated by the least squares procedure, and the Y-axis intercept is the absolute value of the regression coefficient (qN„). RESULTS AND DISCUSSION Regression Equation and Correlation Coeffi- cients. The catch per unit effort, C,/f,, for each fishing area, was regressed on cumulative catch, k,, as previously described. The resulting regres- sion equations are shown in Table 4. The analysis 1 .81 2 .36 3 .62 4 .73 5 1.00 6 .74 7 .72 8 .88 9 .72 TABLE 3. Relative fishing power of hydraulic escalator clam harvesters in North Santee River and North Santee Bay, South Carolina. Vessel Relative Fishing Power Code North Santee River North Santee Bay 1.49 .93 1.21 1.21 1.00 1.19 .88 N" N "This vessel did not harvest in the North Santee Bay. TABLE 4. The regression equations of the Leslie method for the South Santee and North Santee Rivers and North Santee Bay." South Santee River (t = 7 days) c f_ = 15.4609 — 0.0006K, North Santee River (t = 2 days) C ft = 28.0393 — 0.0014K, North Santee Bay (t = 2 days) C f = 25.6450 — 0.0006K, "See Gracy, et ai. 1976, for legal limitation on harvesting days. of variance results (Table 5) indicate a negative regression coefficient significantly different from zero. The correlation coefficient (R) was .858, .969 and .965 for the South Santee River, North Santee River and North Santee Bay, respectively. As previously mentioned, South Santee River data variability was probably caused by the entrance of inexperienced harvester operators and the depar- ture of experienced operators during September and October, 1974. It was assumed by December, HARD CLAM ABUNDANCE 49 TABLE 5. Analysis of variance results with regression for the South Santee, North Santee Rivers and North Santee Bay. South Santee River Source of Variation d.f. Sum of Squares Mean Square F Regression 1 115.5955 115.5955 36.3607*** Residual 13 41.3287 3.1791 Total 14 156.9243 North Santee River Source of Variation d.f. Sum of Squares Mean Square F Regression 1 944.9495 944.9495 310.0777*** Residual 20 60.9492 3.0475 Total 21 1005.8987 North Santee Bay Source of Variation d.f. Sum of Squares Mean Square F Regression 1 359.0181 359.0181 235.2657*** Residual 17 25.9422 1.5260 Total 18 384.9603 TABLE 6. Catchability (regression coefficient), q, and regression statistics for the Leslie method applied to clam, Mercenaria mercenaria, harvesting data from South Carolina. South Santee River North Santee River North Santee Bay Number of observations 15 22 19 d.f. 13 20 17 Catchability 0.0006 0.0014 0.0006 Standard Error 0.0001 .0001 .0001 95% Confidence Limits P> .0004 .0012 .0005 P< .0009 .0016 .0007 1974, and January, 1975, that their skills had im- proved significantly. Catchability. The regression coefficients (estimates of catchability), q, were .0006, .0014 and .0006 for the South Santee River, North Santee and North Santee Bay respectively (Table 6). Confidence limits of catchability for the three harvesting areas are shown in Table 6. A com- parison between North Santee River's catchability and North Santee Bay (Table 7) indicates a signi- ficant difference. According to Braaten's (1969) analysis, the con- stancy of catchability within the fishing season should be a major concern of investigators at- tempting to estimate the size of a population employing the Leslie-DeLury method. For exam- ple, if actual catchability decreases during a fishing season, the Leslie-DeLury method results in an increased (higher) estimate of catchability, which in turn results in a lower initial population size estimate. Harvesting Areas. Because effort was much greater per area (Table 8) in the North Santee River, one of the major assumptions of the Leslie- DeLury method, i.e, units of fishing effort during the season not being competitive, may have been violated for the North Santee River. Moreover, due to the immobility of hard clams, their spatial 50 R. J. RHODES, W. J. KEITH, D. J. ELDRIDGE, V. G. BURRELL, JR. TABLE 7. The analysis of covariance for catchability (regression coefficients) of the North Santee River and North Santee Bay." Error (Residual) Location d.f. Sum of Squares Mean Square F North Santee River 20 60.9495 3.0475 1.9970 North Santee Bay 17 25.9422 1.5260 d.f. Error (Residual) Source of Variation Sum of Square F Within Locations 37 86.8917 2.3484 Pooled 39 770.1354 19.7471 Between Slopes 1 683.2437 683.2437 290.9401*** "Table format adapted from Snedecor and Cochran (1967). TABLE 8. Estimates of original clam (Mercenaria mercenaria) population size, N0, available to hydraulic escalator harvesters in the Santee Delta, South Carolina from 1974 thru 1976. LOCATIONS South Santee River North Santee River North Santee Bay Surface Area 104 ha. 53 ha. 159 ha. Effort" 1517 hr. 1111 hr. 1521 hr. Effort /Surface Area 14.58 hr. 20.96 hr. 9.57 hr. Y — Intercept 15.4609 28.0393 25.6450 No 6,442,042 clams 5,007,000 clams 10,685,417 clams (25,768 "bags"') (20,028 "bags") (42,742 "bags") 95% Confidence Limits,, d.f. 13 20 17 ?> 4,829,020 4,624,000 9,499,750 P< 7,251,039 5,354,250 14,149,000 Percent of Actual Harvest 59.4% 76.6% 60.7% "Effort not standardized for the South Santee River (See Text). 'SeeDeLury (1951) for formula emplyed in estimating the confidence limits of N„. 'There are approximately 250 ungraded clams per "bag". distribution in the North Santee River probably increased in heterogeneity (i.e., aggregation) com- pared to other harvesting sites. This could have in- creased the probability of a given unit of effort be- ing expended on a substrate nearly "swept clean" by previous effort and may have resulted in a declining catchability coefficient during the season. Estimate of Density. The y-axis intercept was divided by catchability to obtain an estimate of the original population (N0) in each harvesting area (Table 8). If these abundance estimates are divided by the total surface area in each legally designated fishing area there would be an overall average of approximately 6.5 clams/m2, 9.5 clams/m2, and 6.7 clams/m2 in the South Santee River, North Santee River and North Santee Bay, respectively. However, unpublished data revealed the presences of two distinct clam density strata in the Santee River estuary; the lowest strata having a density of less than one clam/m2. The higher density strata in the three areas varied between 22 and 27 clams/m2. Higher density strata was HARD CLAM ABUNDANCE 51 characterized as "sand-shell" substrate and the lower density strata as "sand". Interviews of harvester operators indicated that they had con- centrated their effort in the higher clam density strata. As Gullard (1969), has emphasized, fishermen generally expend their effort where they believe the highest stock densities occur. Based upon interview information and the estimate of N„ in each harvesting area, the density of clams in the preferred fishing areas varied between 18/m2 and 24/m2. These are in good agreement with un- published data, but due to the subjective nature of interviewing and the difficulty in defining strata precisely, density estimates obtained in conjunc- tion with the Leslie-DeLury method should be considered approximations. Evaluation. The accuracy of the Leslie-DeLury method when applied to simple commercial fishing catch and effort data has two major limita- tion compared to designed, direct sampling (e.g., Loesch and Haven, 1973): first, the lack of effort estimates for specific locations, and second, potential of commercial gear interaction. The first limitations might be removed by requiring that vessel operators keep detailed records of their harvesting location. This approach would ob- viously increase the time (cost) which the operator must devote to record keeping. In contrast, the simple logbook system used in this investigation was generally compatible with the record system of the owners and vessel operators, since most of them maintained daily records of their catches. In areas where harvestable quantities of clams are apparently clustered in a small site (e.g. North Santee River), the second limitation is not avoidable and consequently makes an abundance estimate questionable. Because of these difficulties, we suggest that this method should be considered only for supplemen- ting direct sampling activities. Such sampling would describe not only the spatial distribution, but also the possible existence of clam density strata so that the influence of distribution on the assumptions required in the Leslie-DeLury method can considered. ACKNOWLEDGEMENTS Our gratitude is earnestly extended to all who contributed to this study, especially to Messrs. Michael Bailey, Holland Mills, Dale Theiling and Glenn Ulrich for assistance in the field; Ms. Karen Swanson for preparation of figures; Ms. Nickie Jenkins for assistance in computer analysis; and Ms. Cheryl Oswald, Ms. Patricia Godsell and Catherine Roehrig for typing the manuscript. The fishing community of McClellanville provided valuable information to our field personnel. Cap- tain Eugene Morrison, James Leland, Thomas Duke, Jr., Robert Ashley, Jr., and James Scott provided pertinent information on clam harvesting sites. We also thank Mr. Charles Bearden and Dr. Edwin Joseph for their en- couragement and direction, and Drs. Paul San- difer, John Manzi and Dale Calder for their editorial assistance. LITERATURE CITED Anonymous. 1973. Santee River Basin, a review and summary of available information on physical, chemical, and biological characteris- tics and resources. Envir. Protect. Agency, Surveillance and Analysis Div., Athens, Ga. 122 pp. Beverton, R. J. H. and S. J. Holt. 1957. On the dynamics of exploited fish populations. U. K. Min. Agric. Fish., Invest. (Ser2) 19:533 pp. Braaten, D. O. 1969. Robustness of the DeLury population estimator. J. Fish. Res. Board Can. 26:339-355. Burrell, V. G. 1976. Mortalities of oysters and hard clams associated with heavy run-off in the Santee River system of South Carolina in the spring of 1975. Nat. Shellfish Assoc. Miami Beach, Fla., June 20-24, 1976, (Abstract). Cummings, T. R. 1970. A reconnaissance of the Santee River estuary, South Carolina. U.S. Geol. Survey, Water Res. Div., Columbia, S.C. 96 pp. DeLury, D. B. 1947. On the estimation of biological populations. Biometrics 3:145-167. DeLury, D. B. 1951. On the planning of ex- periments for the estimation of fish populations. J. Fish. Res. Board Can. 8:281-307. Dickie, L. M. 1955. Fluctuations in abundance of the giant scallop, Phacopecten magellanious (Ornelin), in the Digby area of the Bay of Fun- dy. J. Fish. Res. Board Can. 12:797-857. Dixon, W. J. 1967. (ed.). Biomedical Computer 52 R. J. RHODES, W. J. KEITH, D. J. ELDRIDGE, V. G. BURRELL, JR. Programs. U. of Calif. Press, Los Angeles, Calif. 600 pp. Gracy, R. C, W. J. Keith and R. J. Rhodes. 1976. Management and development of the shellfish industry in South Carolina. Final Rpt. for PL 88-309, Project 2-179-D, S. C. Wildlife and Marine Res. Dept., Charleston, S. C. 73 pp. (unpublished). Gulland, J. A. 1956. On the fishing effort in English demersal fisheries. U. K. Min. Agric. Fish., Fish. Invest. (Ser. 2) 20:41 pp. Gulland, J. A. 1969. Manual of methods for fish stock assessment. Part 1. Fish population analysis. FAO (Food Agric. Organ. U.N.) Man. Fish. Sci. 4:154 pp. Haven, D. S. and J. G. Loesch, 1973. An in- vestigation into commercial aspects of the hard clam fishery and development of commercial gear for the harvest of molluscs. Final Report for PL 88-309, Project 3-124-R, Virginia In- stitute of Marine Science, Gloucester Point, Virginia. 119 pp. Ketchen, K. S. 1953. The use of catch-effort and tagging data in estimating a flatfish population. J. Fish. Res. Board Can. 10:459-485. Kjerfve, B. 1976. The Santee-Cooper: A study of estuarine manipulations. Belle W. Baruch Inst. Mar. Biol. Coastal Res., Contri. 127, U. S. C. Columbia, S. C. 16 pp. Leslie, P. H. and D. H. S. Davis. 1939. An attempt to determine the absolute number of rats on a given area. J. Anim Ecol. 8:94-113. Loesch, J. G. and D. S. Haven. 1973. Estimates of hard clam abundance from hydraulic escalator samples by the Leslie method. Chesapeake Sci. 14:215-216. MacPhail, J. S. 1961. A hydraulic escalator shellfish harvester. Bull. Fish. Res. Board. Can. 28:24 pp. Manning, J. H. 1957. The Maryland soft-shell clam industry and its effects on tidewater resources. Md. Dept. Res. Educ. Resour. Study Rep. 11:25 pp. Manning, J. H. 1959. Commercial and biological uses of the Maryland soft clam dredge. Proc. Gulf Carib. Fish. Inst. 12th Ann. Sess.: 61-67. Mathieson, J. H. and P. DeRocher. 1974. Applica- tion of Maryland clam dredge on the Maine coast. Maine Dept. Mar. Res. Sea Grant Pub. Augusta, Me. 51pp. (unpublished). Omand, D. H. 1951. A study of populations of fish based on catch-effort statistics. J. Wildl. Manage. 15:88-98. Ricker, W. E. 1975. Computation and interpreta- tion of biological statistics of fish populations. Bull. Fish. Res. Board Can. 191. 383 pp. Shealy, M. H. 1976. An environmental base-line study of South Carolina estuaries with emphasis on the North and South Santee and Cooper Rivers. Coastal Plains Reg. Comm. Contract No. 10340031. S. C. Wildlife and Marine Res. Dept., Charleston, S. C. (unpublished data). Snedecor, G. W. and W. G. Cochran. 1967. Statistical methods. Iowa State U. Press, Ames, Iowa, 6th Edition. 593 pp. Stephens, P. G., D. S. Van Nieuwenhuise, P. Mul- lin, C. Lee, and W. H. Kanes. 1976. Destructive phase of deltaic development: North Santee River delta. J. Sed. Petrology 46:131-144. AN ANALYSIS OF TRENDS IN OYSTER SPAT SET IN THE MARYLAND PORTION OF THE CHESAPEAKE BAY George E. Krantz and Donald W. Meritt HORN POINT ENVIRONMENTAL LABORATORIES CAMBRIDGE, MARYLAND 21613 ABSTRACT Statistics on commercial harvest of oysters and on spat set on natural cultch that oc- curred from 1939 to 1975 in the Marx/land portion of the Chesapeake Bay, were analyz- ed to explain recent fluctuations in abundance of oysters. Data appear adequate to predict a decline in oyster production in the immediate future. A period of reduced spat set that began in 1968 is the lowest reproductive success recorded for Maryland waters. This condition has occurred in every river system and geographic subunit in the Bay. Spat set has declined significantly in some of Maryland's prime seed oyster production areas. Biological explanation of this phenomena is difficult and Hurricane AGNES which devastated some regions of the Bay cannot be the sole cause of the problem. Other environmental factors and oyster management programs are discussed in reference to the reduction in natural oyster recruitment. INTRODUCTION Natural history observations on aquatic animals such as oysters have long been a favorite pastime of biologists studying the Chesapeake Bay (Ferguson et al. 1880, Yates 1913). Beginning in the late 1930's biologists at the Chesapeake Biological Laboratory (Natural Resources Institute of Maryland) standardized oyster bar sampling techniques and developed a "field data sheet" where observations on various aspects of the oyster bar communities found in the Chesapeake Bay were systematically recorded. Over the years, oyster bars were inspected annually and the field data sheets have been prepared by a large number of individuals with diverse reasons (management, research, Health Department survey, etc.) for ex- amining the oyster bars. Personnel at the Chesapeake Biological Laboratory conducted the majority of the surveys until the late 1950's (Beaven 1955) when the field work was assumed by personnel of the State management agency — The Department of Natural Resources. During the past two years we have assembled, compiled and analyzed various types of data from these sheets. One very interesting phenomena which these field studies document is the change in pattern and density of oyster spat set on natural cultch on the various natural oyster bars in Maryland's portion of the Chesapeake Bay. These data may be found in a recently published docu- ment by Meritt (1976). This report is a synopsis of the data which were extracted from over 990 legal- ly defined oyster bars that occupy over 215,000 acres of Chesapeake Bay bottom. Statistics on the commercial harvest of oysters from the Chesapeake Bay have been collected with various degrees of accuracy since the late 1800s. Data collected by the Fisheries Statistics Branch of National Marine Fisheries Service (Anon, 1973) show a dramatic decline in oyster production from 1890 to 1935, (Figure 1) followed by a more gradual trend toward lower levels of annual harvest. Although Maryland waters have con- tributed from 55 to 80 percent of the marketable oysters removed from the Chesapeake Bay during 53 54 G. E. KRANTZ AND D. W. MERITT OYSTER PRODUCTION 45 50 55 60 65 70 75 i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — t — i — UNITED STATES FIG. 1. Production of oysters Crassostrea virginica in Chesapeake Bay and in the United States, 1880 to 1975. the past 30 years, Maryland oyster harvest sta- tistics (Figure 2) have shown several periods of fluctuation with a marked decline in harvest dur- ing the early 1960s. A combination of changes in the social structure of the Maryland population, major shifts in consumer tastes, the rising cost of seafood processing labor, changes in environ- mental conditions in the Bay, and a decline in the number of oysters available for harvest are col- lectively responsible for the general decline in landing statistics shown in Figure I. However, several biological factors have also been responsi- ble for the fluctuations in oyster harvest during the past 30 years. Some of these factors have been described through studies of oyster populations in the Chesapeake Bay that were conducted by bio- logists employed at the University of Maryland, Chesapeake Biological Laboratory; the Oxford Biological Laboratory (NMFS); and the Fisher- ies Administration of the Maryland Depart- ment of Natural Resources. Collectively these studies have produced records of changes in the quality and quantity of adult oysters, mortalities due to diseases and natural phenomena, and the periodicity of successful reproduction, or spat set1. OYSTER RECRUITMENT One Underlying theme found in these studies and in the "field data sheets" which appears to ex- HARVEST to UJ 'I X 5 The reader is referred to bibliographies available from these agencies which list published papers as well as unpublished internal reports. FIG. 2. Harvest of oysters from the Maryland por- tion of the Chesapeake Bay in millions of Bushels per year (1945-1974). plain the recent and past variations in oyster harvests is the periodicity and geographic distribu- tion of oyster spat-set on natural cultch in the Chesapeake bay. The annual spat set, or natural recruitment, and subsequent growth of the oysters sustains the population of harvestable oysters throughout the Bay. Bay-wide survey techniques that were developed in the late 1930's to determine the number of spat per bushel of shell and /or oysters taken from the natural oyster bars each year, can be used to measure annual variation in recruitment throughout the Bay (Figure 3). Even though recruitment is expressed as a Bay-wide average in this figure (as are harvest statistics found in Fig. 2) two periods of low recruitment are obvious. The first period, 1952 to 1960, is follow- ed by five years of high spat set. The second period of depressed recruitment, 1966 to 1975, is the period of current concern. OYSTER SPAT TRENDS 55 FIG. 3. Spat Production on natural cultch for Maryland portion of Chesapeake Bay 1939-1975. Each geographic area of the Bay (such as Potomac River, Choptank River, Eastern Bay, Tangier Sound, etc.) has experienced different levels of recruitment during the 35 year study period (Fig. 4). Reasons for variation in recruit- ment in natural populations of oysters are numerous. Environmental disasters during the natural spawning period have obvious effects on the delicate, swimming oyster larvae. Beaven (1946) and Engle (1946, 1955) have shown a rela- tionship between the periodicity of discharge from a given river and the subsequent spat set in that river system. Hurricane AGNES (which changed the temperature, salinity, and water quality throughout the whole Bay) in June 1972 is an ob- vious example of an environmental disaster (Anon, 1975). However, most environmental changes are so subtle that the lay public, fishermen and even shellfish biologists do not notice any problems in the oyster populations. One of these "subtle" disasters was the impact 1966 1973 spot per bushe 50-75 25-50 10-25 0-10 FIG. 4. Distribution of oyster spat set on natural cultch in the Maryland portion of the Chesapeake Bay 1939-1965. HG. 5. Distribution of oyster spat set on natural cultch in the Maryland portion of the Chesapeake Bay 1966-1975. 56 G. E. KRANTZ AND D. W. MERITT of the protozoan disease, Minchinia nelsoni, ("MSX") on oysters in the Chesapeake Bay. Effects of this disease are reflected in reduced harvests (Fig. 2) and poor recruitment on specific oyster bars in the Southeastern portion of the Bay (Meritt 1976) during the mid 1960's (Andrews 1960, Sindermann 1968). The impact of "MSX" disease was also severe enough to depress the Bay-wide average spat set for several years even though "disease-free" areas of the Bay had very high recruitment. Of immediate concern is the period of reduced spat set from 1966 to 1975. Spat set during this period has been lower than ever encountered dur- ing biological studies of the Chesapeake Bay and should have an impact on abundance of harvest- able oysters during the next 4 to 7 years. Figure 5, which presents geographic distribution of spat set during this period (1966-1975), shows that the problem occurred in all areas of the Bay. Areas with traditionally low levels of spat set (Figure 4) have had virtually no recruitment for 6 years; whereas areas of high annual spat set (Eastern Bay, Broad Creek, Harris Creek, Honga River, St. Mary's River) have experienced 50 to 85 percent reduction in spat recruitment. The reader is refer- red to Table 1 for actual data from which Figure 5 was prepared. This table indicates the percent decline in spat set during 1966 to 1975 period in specific areas of the Bay. It is noteworthy that a very high percent reduction of spat set occurred Bay-wide and prime areas for spat set in the Bay were the most severely damaged. Many of the highest spat set areas are also used to collect spat on the dredged shell transplantation program. TABLE 1. Oyster Spat Set on Natural Cultch in Geographical Regions of the Chesapeake Bay — 1966-1975. Decline is by Comparison to Mean of Data for 1939 to 1965. SPAT AREAS 1966-1975 SP 1. Kedges Straits 2. Wicomico River (Potomac) 3. Eastern Bay North 4. Dorchester Shore 5. Hooper Straits 6. Upper St. Marys River 7. Honga River bu. % Decline 71.9 77 63.4 >01 60.3 57 59.1 76 58.2 73 57.0 81 56.1 66 8. Broad Creek 9. Holland Straits 10. Smith Creek 11. Eastern Bay South 12. Lower St. Marys River 13. Harris Creek 14. Talbot Shore 15. Wye River 16. St. Georges Creek 17. St. Marys Shore 18. Lower Potomac River 19. Middle Tangier Sound 20. Little Choptank River 21. Miles River 22. Manokin River 23. Lower Choptank River 24. Poplar Is. Narrows 25. Fishing Bay 26. Lower Patuxent River 27. Middle Choptank River 28. Middle Patuxent River 29. Nanticoke— Wicomico Rivers 30. Lower Tangier Sound 31. Pocomoke Sound 32. Upper Calvert Shore 33. Upper Bay East 34. Lower Calvert Shore 35. Big Annemessex River 36. Upper Tangier Sound 37. South — Rhode Rivers 38. Upper Bay West 39. Tred Avon River 40. Upper Patuxent River 41. Upper Choptank River 42. Kent Shore 43. Upper Anne Arundel Shore 44. Little Annemessex River 45. Upper Chester River 46. Tar Bay 47. Severn River 48. Trippes Bay 49. Lower Chester River 50. Middle Potomac River 51. Lower Anne Arundel Shore 52. St. Clements-Breton Bays 53. Upper Potomac River 54. Lower Bay East 55. Magothy River "Denotes % increase 50.9 68 48.6 78 42.0 70 39.0 63 37.6 61 37.3 82 35.0 47 35.4 37 34.7 46 33.8 06 33.0 54 31.3 34 23.7 83 26.1 73 21.1 81 20.9 69 20.7 46 18.7 67 18.0 23 23.7 83 14.3 14 13.6 59 12.6 73 10.8 85 10.3 16 9.8 36 9.0 80 9.0 88 8.3 89 8.3 58 7.2 232* 7.0 83 6.5 64 6.0 78 5.6 89 4.8 02* 4.5 87 4.4 67 4.0 97 3.7 78 3.5 93 3.3 73 2.8 80 2.7 55 2.4 92 0.6 93 0.0 100 0.0 No Data OYSTER SPAT TRENDS 57 Lack of spat set on the planted shell has nullified the effectiveness of this once successful manage- ment technique. During the recent period of little or no recruit- ment, oyster harvest has continued at very high levels, and recent harvest statistics have shown a shift of oyster harvest from traditional areas throughout the Bay to the middle Eastern Shore. This increase in fishing pressure and resultant loss of oyster stocks in specific geographic regions of the Maryland portion of the Bay has been noted by local watermen, seafood packers, and manage- ment agencies. Published biological studies (Hidu, 1969; Hidu et ai, 1969) and numerous unpublished observa- tions have determined that three to four years are required for an oyster spat to grow to a desirable market size in the Chesapeake Bay. However, oysters usually found in the commercial harvest range from four to seven years old. Therefore, oyster spat set in a given year will begin entering 45 50 55 60 65 70 75 1940 45 50 55 60 65 FIG. 6. Comparison of spat set on natural cultch (Bottom scale) to harvest statistics (top scale) which have been adjusted 5 years in time. the commercial harvest about four years later and reduced recruitment will be expressed for an addi- tional two to three years. Figure 6 is a combina- tion of data on annual spat recruitment from Figure 3 and on oyster harvest from Figure 2. Figure 2 has been shifted to the left in time to demonstrate the lag of six to eight years in the ef- fect of recruitment on harvest following a period of good spat set (1964-1965) or a period of poor spat set (1950-1954). Correlation coefficients calculated for various lengths of time between spat set and harvest using the data available in this study were as follows: two years, 0; three years, 0.7; four years, 1.55; five years, 2.1; six years, 3.35; seven years 3.9; eight years, 3.85 and nine years 2.6. These calculations further sustain the theory that a period of successive years of low spat set will require between 6 to 8 years before the period of poor recruitment is reflected in com- mercial harvest. The 1975-1976 oyster season is nearing the end of the influence of high recruitment during 1964-65, followed by a boost in 1969 in some areas of the middle Eastern Shore. However, the Bay- wide reduction in spat set during the past eight years should begin to seriously affect harvest levels by late 1976. By inspection of the data in Figure 6 one could predict harvests of less than one million bushels of oysters for the next three to five years. However, any prediction of oyster harvest in Maryland can be invalidated by any changes in the length of season, in existing catch limits, in type of gear used, and by opening any areas previously closed to the fishery, or by annual changes in market demand for oysters. However, available oyster stocks are presently at the lowest level in recorded natural history for the Maryland portion of the Chesapeake Bay. Perhaps for the first time the supply of oysters will solely limit harvest regardless of social or economic changes. Therefore the expected harvest after 1976, should be significantly lower than the previous recorded low levels during the early 1960's (Figure 2). OYSTER MANAGEMENT STRATEGIES Several management practices are being employed by the Maryland Department of Natural Resources to sustain levels of oyster harvest thereby stabilizing the industry and social 58 G. E. KRANTZ AND D. W. MERITT structure supported by the Maryland seafood in- dustry. Transport of oysters from waters polluted by domestic wastes to clean waters for natural depuration provides several hundred thousand bushels of harvestable oysters annually. However, oyster populations in polluted waters have not received spat set for several years and numbers of oysters are being reduced. The planting of dredged oyster shell in seed areas (areas of traditionally high spat set) and subsequent transplantation of spat that attached to this substrate to growing areas, has maintained the harvestable populations of oysters in some locations. This technique was very successful during years of high spat set (1964-1965), but during the recent period of low spat set, a minimum number of harvestable oysters has been produced by the shell planting program. Because of the growth of algae and other estuarine fouling organisms and sedimentation of water born silt on planted shells, dredged shell must be planted annually to provide a suitable surface on which oyster spat may set. This necessitates removing shells planted in the previous year regardless of the number of spat col- lected on the shell. A recent attempt to cultivate the planted oyster shells (bagless dredging) to pro- duce clean surfaces has shown some success in years of good spat set. However, this technique is expensive in manpower and boat rental. The effec- tiveness of the bagless dredging procedure is still controlled by natural spat set in the Chesapeake Bay. Many of the subtle biological reasons for failure of oyster larvae to set in the Chesapeake Bay are currently unknown. Throughout the Bay, biolo- gists have noted an overall increase in turbidity in the water. This turbidity may be related to in- creased sediment loads in the Bay river systems and to an overall disappearance of rooted aquatic plants which trapped the silt in shallow waters. These rooted aquatic plants also bound energy from the sun, died, and decayed thereby releasing detritus and other plant materials that contributed to the nutrition of adult oysters and to the growth of algae used as food by oyster larvae. The decline in rooted aquatic plants may be linked subjectively to increased use of herbicides in "minimum-till-farming" as well as other in- dustrial practices that pollute the estuarine en- vironment with persistent chemicals. Levels of several pesticides and PCB's have increased in rivers throughout the United States and in Mary- land reached the highest levels in the summer of 1968 (Butler, 1973; Butler, 1976). Since that time these specific residues have declined but are still present in Chesapeake Bay biota. The impact created by these man-made "insults" upon the reproductive capacity of the oyster and other estuarine fauna is not known nor are the subtle natural chemical and physical changes found in the ever-changing Chesapeake Bay. Another technique that is being used to increase recruitment of natural oyster bars is to spawn and set young oysters in oyster hatcheries, then plant the spat in the natural environment to grow to market-size. Biological and engineering technolo- gy have advanced in the past 30 years to a state- of-the-art where oyster hatcheries can control spawning, increase growth of early stage spat and grow marketable oysters more rapidly in special systems than ever observed in nature. The cost of these procedures currently restricts use of the technology to growing oysters only in protected waters where survival of hatchery-reared spat is very high. Conditions on natural oyster bars, (open to public exploitation) are known to pro- duce high levels of mortality in spat and makes the use of hatchery-reared spat to sustain recruitment in natural waters of the Chesapeake Bay ques- tionable on an economic basis. Research and development studies in the University of Maryland Center for Environmental and Estuarine Studies experimental oyster hat- chery at Horn Point are directed toward increas- ing survival of spat on natural oyster bars, decreasing oyster spat production costs, improv- ing management practices for oyster bars, and producing enough oyster spat so that farming, not just fishing, for oysters in the Chesapeake Bay can become a practicality in the future. Presently the shellfish research hatchery at Horn point has a design capacity of 100 to 200 million oyster spat per year. This level of output is about 10 percent of the adult oysters harvested an- nually and would not be adequate to indefinitely sustain the existing Maryland oyster fishery. However, contributions of spat from this oyster hatchery can be of significant value in re- OYSTER SPAT TRENDS 59 habilitating over-harvested oyster bars during years when natural spat set did not occur. Oyster spat produced by this hatchery may be even more valuable if used to develop new biological and engineering technology, improve and demonstrate the oyster farming techniques for the Chesapeake and educate the public to potential impact of this "new" technology on the traditional oyster fishery. ACKNOWLEDGEMENTS The authors sincerely acknowledge the efforts of the many people who participated in the past oyster bar surveys and collected the data that made this report possible. Our task was merely to assemble and analyze the invaluable field observa- tions of others. Special recognition is given to those who envisioned the annual oyster bar survey, planned the sampling procedures, and dutifully supervised the programs through the years. Leadership and direction of these studies were provided by the staff of the Chesapeake Biological Laboratory — Dr. L. E. Cronin, Mr. G. F. Beaven, Mr. E. Dunnington and by members of Maryland Department of Natural Resources — Messrs. J. Manning, F. Sieling, R. Rubelmann, and H. Davis. LITERATURE CITED Anon, 1973. U. S. Department of Commerce Fishery Statistic of the U. S. Current Fisheries Statistical Digest No. 67. Anon, 1975. Impact of Tropical Storm "AGNES" on Chesapeake Bay. Baltimore District Corps, of Engineers. Department of the Army. Andrews, J. D. 1968. Oyster mortality studies in Virginia VII. Review of epizootiology and origin of Minchinia nelsoni. Proc. Nat. Shellfish. Assoc. 58:23-36. Beaven, G. F. 1946. Effect of Susquehanna River Stream Flow on Chesapeake Bay Salinities and History of Past Oyster Mortalities on Upper Bay Bars. Contribution No. 68, Annual Report. Maryland Board of Natural Resources, p. 123-133. Beaven, G. F. 1955. Twelfth Annual Report, Maryland Board of Natural Resources, 7 pp. Butler, P. A. 1973. Residues in Fish, Wildlife and Estuaries Pesticide Monitoring J. 6,4:238-326. Butler, P. A. 1976. Personal communication. Engle, J. B. 1946. Commercial Aspects of the Up- per Chesapeake Bay Oyster Bars in Light of the Recent Oyster Mortalities. United States Fish and Wildlife Report. Engle, J. B. 1955. Ten Years of Study on Oyster Setting In A Seed Area in Upper Chesapeake Bay. Proc. Nat. Shellfish. Assoc. 46:88-99. Ferguson, T.B., T. Hughlett., W.K. Brooks., and F. Winslow 1880. Report to the Commissioners of Fisheries of Maryland. 397 pp. Hidu, H., K. Drobeck, E. Dunnington, W. Roosenburg, and R. Beckett. 1969. Oyster Hat- cheries for the Chesapeake Bay Region. NRI Special Report No. 2, Contribution No. 382, 15 PP- Hidu, H. 1969. The feasibility of oyster hatcheries in the Delaware-Chesapeake Bay Region. Proc. of Contribution on Artificial Propagation of commercially valuable shellfish. College of Marine Studies. University of Delaware, Newark. NRI Contribution No. 396. Meritt, D. W. 1976. Oyster Spat Set on Natural Cultch in the Maryland Portion of the Chesapeake Bay (1939-1975) Special Scientific Report. UMCEES Ref. No. 76-26 HPEL. In Press. Sindermann, C. J. 1968. Oyster Mortalities with particular reference to Chesapeake Bay and the Atlantic Coast of North America. U.S. Depart- ment of Interior, Specs. Sci., Rept. Fisheries No. 569,10 pp. Yates, C. C. 1913. Summary of Survey of Oyster Bars of Maryland 1906-1912. Government Printing Office. ESTIMATION OF LOBSTER POPULATION SIZE AT MILLSTONE POINT, CONNECTICUT, BY MARK-RECAPTURE TECHNIQUES, 1975-1976 Gary H. Cole, Ronald L. Copp, and David C. Cooper BATTELLE COLUMBUS LABORATORIES WILLIAM F. CLAPP LABORATORIES DUXBURY, MASSACHUSETTS ABSTRACT Population estimations of the American lobster, Homarus americanus, in the Millstone point area of Long Island Sound were derived using the folly (1965) multiple recapture analysis method. Results to date indicate a population which ranges between 3,800 and 21,000 in an available habitat within the study area of approximately five square miles. Fishing pressure is slight and does not appear to exert a significant effect on size or distribution of the population, but is related to molting periodicity. INTRODUCTION In New England waters the American lobster, Homarus americanus, is the subject of an intensive commercial fishery and a less intensive recrea- tional fishery which are of major significance to many local economies. As in the case of pelagic fisheries, various studies have been conducted to determine stock density and the effects of fishing pressure on lobster populations in several localiz- ed areas (Wilder, 1947; Templeman, 1935; and Pa- loheimo, 1963). The approaches have varied, but mark-recapture techniques seem to provide the most reliable results since individual lobsters can be monitored over extended time periods and fair- ly large areas. Population estimation relying on mark-recap- ture techniques involves affixing a unique mark or tag to each member of a subset of a population, releasing the subset into the original population, and recapturing these marked subset individuals along with other organisms within the population. The population size is then a function of the ratio of unmarked to marked individuals within the population, and maximum likelihood estimations can be derived provided that several assumptions regarding longevity, emigration, and immigration are reasonable for the area and organism in ques- tion. Paloheimo (1965) discusses mark-recapture methods which are specifically relevant to lobster populations, allow for the varying catchability of tagged and untagged organisms, and allow for motility as related to temperature. At a more advanced level, Jolly (1965) devel- oped methods of analysis of mark-recapture data which provide estimates of instantaneous popula- tion size and allow for multiple recaptures of tagg- ed organisms, thus allowing the monitoring of in- dividuals over longer time periods than less sophisticated population estimation techniques. Likewise, confidence limits can readily be com- puted from the analyzed data which, in turn, por- tray a more realistic picture of population estima- tion in localized areas. These features render the Jolly method an ideal approach for the estimation of lobster populations within defined areas. The objective of this investigation has been to develop a statistically reliable estimate of lobster crops in a study area at Millstone Point, Connec- ticut, on the northern shoreline of Long Island Sound. Additionally, the design of the investiga- tion has allowed inferences regarding temporal and spatial variations in standing crops and 60 LOBSTER POPULATION SIZE 61 lobster movements inside and out of the study area to be made. Although the Millstone Point area is not gener- ally considered to be of primary importance for commercial or recreational lobstering, it was im- portant to assess lobster stocks in the area in order to provide a basis for determining the significance of a nearby bulk power facility. Assessment of lobster population densities is complex, especially with regard to immigration and emigration of lobsters within a study area. The extent and degree of lobster movements have not been clearly defined but appear to be functions of location, depth and temperature. Stewart (Personal com- munication) has indicated that some lobsters from the Long Island Sound area marked with sonic tags moved as much as 700 yards in one hour while other similarly marked lobsters moved little at all. Morrissey (1971) monitored tagged female lobsters and recorded movements along Cape Cod that averaged 26.1 km during about 39 days. Wilder (1963) and Scarratt (1970) evidenced little movement of tagged lobsters released near Prince Edward Island. Cooper et al (1975) showed no dis- cernible seasonal inshore-offshore movements of shallow (less than 75 feet) water lobster popula- tions, indicating to some degree that nearshore lobster populations may be nonmigratory; how- ever, there is no evidence indicating long term residency. METHODS AND MATERIALS Study Area. The study area is located about 10 kilometers southwest of New London, Connec- ticut, adjacent to Northeast Utilities' Millstone Point nuclear generating station. Extensive surveys of bottom types in this area have been conducted in conjunction with the preparation of environmental reports for steam electric stations at Millstone Point and have indicated only a few areas which are suitable for lobster habitation — rocky outcrops interspersed with patches of hard sand. These areas cover about two and a half square miles or about sixteen hundred acres. These locations as depicted in Figure 1 are general- ly south and east of Millstone Point. Most of the remainder of the study area has bottom types which are comprised of alluvial sediments derived from the Niantic River which enter Niantic Bay from the north. Lobster Acquisition. Several methods for cap- turing lobsters for tagging were attempted. Early attempts at setting out artificial burrow habitats and monitoring them regularly by SCUBA proved unproductive, as did removing them by hand from their natural burrows. The use of baited commercial pots set in trawls of five proved to be the most productive means of lobster acquisition. At the onset of this investigation pot trawls were set at several areas in Niantic Bay and around Millstone Point. No lobsters were taken in the pots in Niantic Bay and these pots were moved to more productive areas. No further attempts were made to fish unproductive areas north and west of Millstone Point because maximum catch per unit- effort was necessary to provide statistically reliable estimates. Beginning in September, 1975, pot trawls were set at each of the four areas near Millstone Point which were suitable for lobster habitation, baited with locally caught fish, and checked three times each week, every Monday, Wednesday, and Fri- day. Lobster Tagging. After restraining chelipeds with rubber bands, all lobsters were brought to a FIG. 1. The Study Area. Pot Trawl Locations Marked by Closed Circles. Lobster Release Points Marked by Open Circles 62 G. H. COLE, R. L. COPP, D. C. COOPER nearby field laboratory with a running seawater system and holding tanks for tagging. Lobsters with carapace lengths greater than 55 millimeters were marked with a sphyrion tag with a stainless steel anchor and No. 20 vinyl tubing. An iden- tification number and our organization's name were stamped on the tubing. Tagging procedures followed those outlined by Scarratt (1970) and allowed for continual security of the tag in spite of periodic molts. Tagged lobsters were retained in the holding tanks for several days in order to safeguard against any aftereffects of tagging, and released in the study area at points depicted in Figure 1. Although tag loss after release is an important consideration, no direct attempts were made to assess tag loss rates. Scarratt (1970) suggests a 56% survival rate, and an actual recapture rate of 46.7% for subcarapace sphyrion tags. These calculated limits were acceptable in the present study, and any unusual tag loss or mortality as a result of tagging could be indirectly measured from this data. Lobster Recapture. Tagging studies usually in- volve the tagging of legal size lobsters obtained from local lobster fishermen, recapturing of tag- ged lobsters by same, and tags being returned to the principal investigator. This approach gives estimates of only one segment of the population — legal sized lobsters — and it relys on and assumes accurate reporting by local lobster fishermen. This approach thus presents problems of inaccurate total catch reports and delayed tag returns which can bias population estimates. These factors were recognized at the time that the problem of lobster recapture was considered. Accordingly, only tag- ged lobsters which were recaptured in our own pot trawls were used in deriving population size estimates. These, in turn, were released im- mediately and returned to the population. However, tag returns from local (and in some cases, nonlocal) lobster fishermen were useful in determining lobster movements within and out- side the study area. To promote returns from local and nonlocal fishermen, letters were :ent to many Connecticut, eastern Long Island, and Fishers Island lobster- men. The tagging program was explained, and returns of tags from legal sized lobsters were re- quested. Tag identification numbers and point of capture was also requested from tagged sub-legal size lobsters. However, as explained in the letters, it was preferred that tagged sub-legal sized lobsters be returned to the water with the tag in- tact as this would enhance monitoring over a longer time period. A $2.00 reward was offered for each tag or tag number returned. Population Estimation. Population estimates were derived according to Jolly (1965). The Jolly model provides for an estimation of instantaneous population within any time interval specified, allows for a flexible sampling schedule, and pro- vides for death, emigration, and immigration. All were considered desirable for this investigation. The Jolly model, like most statistical methods, re- quires that certain assumptions regarding the population be made. Namely, the population must be single, that is, comprised of individuals that are free to move randomly through a defined area. All marked individuals released into the population must have equal likelihood of capture as unmarked individuals. The population may, however, consist of several classes of animals behaving in different ways. A key feature of this approach to population estimation is that the underlying assumptions of the Jolly model can and have been evaluated from the results of the tag- ging program itself. RESULTS AND DISCUSSION Validation of Jolly Model Assumptions. During the twelve-month period from September, 1975, through August, 1976, 3,811 lobsters were tagged and released. Figure 2 indicates the size frequency 60 70 SO 90 100 Carapace Length (mm) HG. 2. Size Frequency Distribution of Lobsters Caught During the Investigation LOBSTER POPULATION SIZE 63 distribution of all lobsters captured during this in- vestigation and suggests that a balanced subset of the population was acquired by our sampling pro- tocol. The average monthly rate of recapture of tagged to untagged lobsters was approximately nine percent (see Table 1), but varied somewhat between months. Nonetheless, all monthly recap- ture rates of lobsters tagged in the 1'th sample and subsequently recaptured remained fairly constant throughout the investigation (see Table 2). TABLE 1. Total Number of Lobsters Tagged, and Total Number of Recaptures Per Month Total Tagged Month & Released Recaptures Percent Oct. 1975 582 37 6.35 Nov. 1975 307 30 9.77 Dec. 1975 612 30 4.90 Jan. 1976 193 18 9.32 Feb. 1976 203 18 8.86 Mar. 1976 328 37 11.28 Apr. 1976 320 40 12.50 May 1976 214 37 17.28 Jun. 1976 373 49 13.13 Jul. 1976 268 25 9.32 Aug. 1976 227 3,811 25 11.01 Total 346 Catch per unit-effort, expressed as numbers of lobsters, tagged or untagged, (per 100-pot hauls) were calculated on a monthly basis and are listed in Table 3. A two-factor analysis of variance was conducted to assess differences between stations and between months. Since there was only one observation per cell, Tukey's test for additivity was used to determine if any interactions between stations and time occurred. These results are ex- pressed in Table 4. There were no significant dif- ferences between stations. However, there were differences in catch per unit-effort between months. Monthly differences in catch per unit- effort probably reflect varying degrees of cat- chability rather than changes in stock density, as population estimates do not show increases in population size coinciding with increases in catch per unit- effort. Tag returns from lobster fishermen indicate that there was little emigration from the Millstone study area during the investigation. There were 475 tags returned to us; 27 (5.7%) were taken from lobsters captured outside the study area. Us- ing the Jolly model, which provides for both emigration and immigration, movement was not seen as a major problem. These results do not suggest varying degrees of catchability for tagged or untagged lobsters, nor is there a suggestion of segregation of tagged or un- tagged lobsters. Therefore, it is assumed tagged TABLE 2. Lobsters Tagged in the i'th Sample and Subsequent!]/ Recaptured Month Tagged and Released Month 1975 1976 Recaptured Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Sep 1975 Oct 11 Nov 5 20 Dec 1 10 12 Jan 1976 0 1 5 9 Feb 0 2 4 6 3 Mar 1 5 3 8 7 10 Apr 0 1 4 6 14 3 6 May 1 0 5 8 4 1 5 11 Jun 1 3 4 6 1 4 4 9 16 Jul 1 0 2 3 2 0 0 2 0 11 Aug 1 2 0 2 0 2 0 2 1 3 6 64 G. H. COLE, R. L. COPP, D. C. COOPER TABLE 3. Lobster Catch Per 100-Pots Hauls at all Stations by Month Month Jordan Cove Intake Stations Effluent Twotree Total Jan Feb Mar Apr May Jun Jul Aug Totals 142 34 100 102 378 34 38 26 40 138 48 69 40 52 209 99 73 58 49 279 52 62 30 32 176 58 114 64 43 279 42 64 70 35 211 52 74 — 40 166 527 528 388 393 1,836 — = No data, pulled pots. TABLE 4. Analysis of Variance of Catch Per Unit-Effort (Two Factors, No Interaction) ' Source of Degrees of Sum of Mean Variation Freedom Squares Square F P Station Effect 3 1611.125 537.042 0.809 0.999 Month Effect 5 9371.708 1874.342 2.822 0.054 Residual Error 15 9963.125 20945.958 664.208 Totals 23 * Since there is only one observation per cell, we cannot test for interaction with Tukey's test for additivity. Using this model with no interaction means that if interactions are present, then the actual level of significance for testing mean effects is below the specified one. Tukey's test has F much less than 1, so conclude no interactions present so model fits. Now we can say that there are no differences between months or stations at x = 0.05; however, there is a difference between months at x = 0.055. TABLE 5. Monthly Population Estimates (Jolly) Method for October, 1975, through July 1976. Proportion Total Total Survival Number 95% Marked Marked Number Probability Joining Level I ALPHA M N PHI B ± Sep 1975 0.00 .9008 Oct .0178 165.75 9327.20 .3644 401.87 7656.15 Nov .0742 281.92 3800.32 1.0801 13714.05 2069.13 Dec .0358 638.25 17815.50 .4396 —57.23 9690.65 Jan 1976 .0711 552.71 7774.78 1.4221 4630.69 4736.04 Feb .0679 1064.75 15687.32 1.5778 —3227.32 10417.60 Mar .0932 2005.00 21524.26 .4127 1466.38 13303.05 Apr .0932 964.00 10348.82 .7935 —808.55 5591.17 May .1383 1023.62 7399.28 .5214 2346.48 4793.81 Jun .1043 646.86 6203.95 .6399 5274.06 4230.25 Jul .0709 655.83 9244.13 Aug .0742 LOBSTER POPULATION SIZE 65 lobsters mixed randomly with untagged lobsters had equal likelihood of capture as untagged lob- sters. If this were not the case, capture ratios of tagged to untagged lobsters would be signifi- cantly different over time and recaptures of any one group of tagged lobsters would be higher or more erratic. Population Estimation. Since the results do not suggest a violation of the Jolly model assump- tions, it was utilized in determining the size of the lobster population in the study area. Population estimates by month for the study period ranged from 3,800 in November, 1975, to 21,524 in March, 1976. Table 5 lists monthly estimates, and figure 3 depicts monthly mean estimates plotted with corresponding 95 percent confidence limits (two standard deviations about the mean). Considering the available lobster habitat in the study area is approximately two square miles, the approximate population density of legal and near legal size lobsters within the study area ranges from 1,500 to 10,000 lobsters per square mile; the mean population estimate for the study area com- puted from the entire data base is 10,912 or about 5,000 lobsters per square mile on an annual basis; the stock density ranges are comparable to those calculated by Paloheimo (1963) and Squires et al. (1975) for lobster grounds along Canada's Atlantic Coast and off the northwest coast of New- foundland. 30 o20 o o 10 U U 0 N 1975 F M A 1976 M Fig. 3. Jolly Model Lobster Population Estimates for Study Area, by Month. Closed Circles Designate Means. Brackets Designate Two Stan- dard Deviations About the Mean. Several parameters related to population char- acteristics and fishing pressure are listed in Table 6. Mean carapace length of all lobsters captured in this investigation is 76.3 millimeters, about 5 millimeters less than the legal size in Connecticut. May and August mean carapace lengths are TABLE 6. Parameters Related to Population Characteristics and Fishing Pressure Month Growth Via Local Lo Imminent Molt Tag Reti 5 4 15 — 28 — 48 — 31 — 20 — 16 13 56 54 54 102 86 66 103 56 40 Legal Size Lobsters From Our Pots Mean Carapace Length Sep 1975 Oct Nov Dec Jan 1976 Feb Mar Apr May Jun Jul Aug 4 3 20 43 16 16 22 34 33 69 55 38 77.6 79.5 78.0 71.6 79.0 76.9 71.5 — = No data. 66 G. H. COLE, R. L. COPP, D. C. COOPER TABLE 7. Correlation of Tag Returns with Carapace Length and Imminent Molt Factor Spearman Coefficient P Value Tag Returns Tag Returns Tag Returns Mean Carapace Length No. of Imminent Molts No. of Imminent Molts in Previous Months -.2796 .7143 .7857 .21 .06 .02 significantly (p = 0.05) lower than those for all other months. The number of tag returns from commercial lobster fishermen, an indication of fishing pressure, was highest during the summer months. In an effort to determine the relationships be- tween these parameters, Spearman rank correla- tion coefficients were computed for the variables. These are listed in Table 7 and indicate that fishing pressure is significantly tied to molting but is not related to the size distribution or the size of the population in a significant way. However, it is assumed that fishing pressure does contribute to fluctuations in lobster population density around Millstone Point. Emigration and immigration of lobsters may also have influenced population size estimates, but are not seen as major influences since the data does not suggest large scale movements of lobsters into or out of the study area during the investigation period. Factors such as available habitat, food, and intraspecific com- petition are probably more important influences on the size and size distribution of the population in the study area. ACKNOWLEDGEMENTS This research has been supported by Northeast Utilities Service Company under Purchase Order No. 040745 to Battelle Columbus Laboratories, William F. Clapp Laboratories, in an effort to develop a greater understanding of the marine biology of the Millstone Point area. REFERENCES Cooper, R. A., R. A. Clifford, and C. D. Newell. 1975. Seasonal abundance of the American lobster, Homarus americanus, in the Boothbay region of Maine. Trans. Amer. Fish. Soc. 104(4) :669-676. Jolly, G. M. 1965. Explicit estimates from capture- recapture data with both death and immigration — Stochastic Model. Biometrika 52:225-247. Morrissey, T. D. 1971. Movements of Tagged American Lobsters, Homarus americanus, Liberated off Cape Cod, Massachusetts. Trans. Amer. Soc. 1:117-120. Paloheimo, J. E. 1963. Estimation of catchabilities and population sizes of lobsters. J. Fish. Res. Bd. Canada 20:59-88. Scarrett, D. J. 1970. Laboratory and field tests of modified sphyrion tags on lobsters (Homarus americanus). J. Fish. Res. Bd. Canada 27:257-264. Squires, H. J., G. P. Ennis, and G. E. Tucker. 1974. Lobsters of the Northwest Coast of New- foundland, 1964-1967. Proc. Nat. Shellfish. Assoc. 64:16-27. Templeman, W. 1935. Lobster tagging in the Gulf of St. Lawrence. J. Biol. Bd. Canada l(4):266-278. Wilder, D. G. 1947. The effects of fishing on lobster populations as determined by tagging experiments. Fish. Res. Bd. Canada. Atlantic Prog. Rept. No. 37. Wilder, D. G. 1963. Movements, growth, and sur- vival of marked and tagged lobsters liberated in Egmont Bay, Prince Edward Island. Fish. Res. Bd. Canada. Prog. Rept. Atl. Coast 64:3-9. Proceedings of the National Shellfisheries Association Volume 67 — 1977 DETERMINATION OF SHELL CONDITION IN LOBSTERS (HOMARUS AMERICANUS) BY MEANS OF EXTERNAL MACROSCOPIC EXAMINATION G. P. Emu's DEPARTMENT OF FISHERIES AND ENVIRONMENT FISHERIES AND MARINE SERVICE NEWFOUNDLAND BIOLOGICAL STATION 3 WATER STREET ST. JOHN'S, NEWFOUNDLAND A1C1A1 ABSTRACT Shell condition determination by means of external, macroscopic examination of lobsters (Homarus americanus) is evaluated. The accuracy of the method is determined by comparing carapace length of tagged lobsters at release and recapture and by serum protein concentration of individual lobsters. Over 97% accuracy was achieved for those observations that could be confirmed. It is suggested that shell condition deter- mination by external examination can be used to estimate the proportion of the popula- tion molting. INTRODUCTION Determination of annual growth rates in lobsters requires data on molt increment and pro- portion molting at least for those sizes that molt no more than once annually. Using the "sphyrion" tag (Scarratt and Elson, 1965) molt increment data are relatively easy to acquire (Scarratt, 1970; Cooper, 1970; Ennis, 1972) but there are problems associated with obtaining reliable data on the pro- portion molting. Hepper (1965) discussed various methods of determining whether a lobster has molted and their inadequacies in estimating the proportion molting; he presented an alternate method based on histological examination of the integument of pre-molt lobsters. Aiken (1973) described proecdysis and a method of molt predic- tion in lobsters. The integument of a recently molted (new-shelled) lobster can be distinguished from that of a non-molted (old-shelled) lobster by histological means (personal communication with Dr. Frank Fifield). However, the difference does not persist long enough for early molters (equivalent to intermolt stages C2 — C, of Drach and Tchernigovtzeff, 1967) in a population to be distinguished from non-molters (stage C4) near the end or shortly after the molting period. Tag recapture data one year after release can be used to determine the proportion molting by compar- ing sizes of animals at release and recapture (Han- cock and Edwards, 1967). There is concern, however, that tagging might adversely affect the proportion molting. In lobster field studies, however, a much less time consuming, direct method of shell condition determination would be more useful and convenient than any of the above. Weber and Miyahara (1962) determined pro- portion molting in king crabs, Paralithodes camt- schatica, by means of external, macroscopic ex- amination of the shell. This method is subjective and depends on observer experience with the par- ticular species. If the accuracy of such observa- tions could be demonstrated by comparison with other types of evidence, it might be an acceptable 67 68 G. P. ENNIS method of determining proportion molting an- nually in lobsters. This paper describes the accuracy of using macroscopic observations to determine shell con- dition of lobsters from a Newfoundland popula- tion and discusses the application of this method in estimating the proportion molting annually in a lobster population. MATERIALS AND METHODS Fishing was carried out during July 1-9 im- mediately following the 1975 commercial lobster fishing season in the Arnold's Cove, Placentia Bay area and 387 lobsters were tagged with sphyrion tags in a procedure described by Ennis (1972). No evidence of molting was seen while SCUBA diving on July 18 but it was observed on July 23. All tag- ged lobsters were thus old-shelled at the time of tagging. Further fishing was carried out between September 17-23 by which time molting in the population had all but ceased and some tagged lobsters were recaptured. Shell condition of all lobsters caught was determined by external ex- amination and recorded as new (definite), which is equivalent to intermolt stages A2 — d; old (definite), which is equivalent to stage C4; new (uncertain) or old (uncertain), neither of which can be assigned an intermolt stage classification with certainty but most are probably stage C2 or C3. All evaluations of shell condition were made by the same individual and were based on several criteria: (1) shell firmness — slight pressure ap- plied with the thumb and fingers to each side of the carapace will result in buckling in new-shelled lobsters, in old-shelled lobsters the carapace is considerably more rigid and resistant to such pressure; (2) scratches — on the ventral surface generally, but particularly on the claws, of old- shelled lobsters darkly stained scratches are pre- sent and the color is usually more faded than dor- sally; the outside edges of the claws are particular- ly darkened; (3) color — in new-shelled lobsters the color is bright whereas in old-shelled lobsters the color is usually noticeably faded; (4) epizoa — various encrusting organisms are often found on the shells of lobsters, usually more abundantly on old-shelled lobsters. Comparison of carapace length measurements at times of release and recapture was used to con- firm the accuracy of shell condition classifications in tagged-recaptured lobsters. Serum protein con- centration in lobsters changes dramatically during the molting cycle (Ennis 1973) and this was in- vestigated as another means of confirming obser- vations on shell condition. Blood samples were taken for spectrophotometric analysis from 153 lobsters caught during the September 17-23 fishing period. Means were calculated for the serum pro- tein absorbance values of lobsters that had been classified old-shelled (definite) and those that had been classified new-shelled (definite). These were significantly different (P<.01). Absorbance values above the mean for the old-shelled group were considered proof that the lobster was old-shelled and values below the mean for the new-shelled group were considered proof that the lobster was new-shelled. RESULTS Only 19 of the 387 lobsters tagged just prior to the molting season were recaptured during the short (September 17-23) fishing period near the end of the season. All classifications of shell condi- tion by external examination made at the time of recapture were proven correct by comparing the carapace length at recapture with that at the time of tagging (Table 1). Of the 153 blood samples taken, 13 were from these recaptured lobsters. The absorbance values in this group of 13 indicated a fairly wide gap between old- and new-shelled lobsters. The lowest value for old-shelled lobsters was .219 and the highest for new-shelled lobsters was .132 (Table 1). This is a good indication that absorbance values above .284 (mean value for all blood samples from lobsters classified as old- shelled [definite]) and below .149 (mean value for all blood samples from lobsters classified as new- shelled [definite]) can be taken as proof of old- and new-shelled lobsters respectively. Of the 153 blood samples analyzed, 67 ab- sorbance values were above the old-shelled (definite) mean or below the new-shelled (definite) mean and only these are used to confirm the observations on shell condition for the lobsters from which these blood samples were taken. Ab- sorbance values for the remaining 86 blood samples are not used here. Of the 67 used all shell LOBSTER SHELL CONDITION 69 TABLE 1. Comparison of carapace length of "sphyrion" tagged lobsters at tagging and at recapture and validation of shell condition determinations. Shell condition Carapace length (mm) classification Serum protein At tagging At recapture Molted at recapture absorbance value 83 83 No Old (definite) .269 84 84 No Old (definite) .226 97 97 No Old (definite) .219 92 92 No Old (definite) .292 95 95 No Old (definite) .330 107 107 No Old (definite) .241 94 94 No Old (definite) .372 79 79 No Old (definite) .260 95 95 No Old (definite) .292 85 85 No Old (definite) 97 97 No Old (definite) 88 88 No Old (definite) 77 77 No Old (definite) 82 97 Yes New (definite) .132 80 92 Yes New (definite) .084 83 92 Yes New (definite) .097 90 106 Yes New (definite) .087 78 89 Yes New (definite) 71 82 Yes New (uncertain) TABLE 2. Numbers of serum protein absorbance values less than .149 (mean of all new-shelled [definitel) lobsters and greater than .284 (means of all old-shelled [definite]) lobsters for each shell condition classi- fication. Shell Condition No. No. of absorbance values < .149 No. of Absorbance values > .284 Results Old (definite) 29 Old (uncertain) 1 New (definite) 27 New (uncertain) 10 0 1 27 7 29 29 out of 29 correct 0 1 our of 1 incorrect 0 27 our of 27 correct 3 7 out of 10 correct 67 condition observations recorded as new (definite) and old (definite) were correct but mistakes were made in some recorded as new (uncertain) and old (uncertain) (Table 2). Comparison of absorbance values below .149 for the new (uncertain) group with those below .149 for the new (definite) group showed no significant difference (P>.9). This indicates that shell condition determinations for the new (definite) group can be made with the same con- fidence no matter on which side of the mean (.149) for the group the absorbance value falls. A similar comparison is not available for old-shelled lobsters but it is felt that the same is true for the old (definite) group. A total of 935 lobsters was caught during the September 17-23 fishing period; new (definite) and old (definite) shell classifications comprised 94.3% of the observations, new (uncertain) and old (uncertain) 5.7%. Mistakes were made only in the 70 G. P. ENNIS new (uncertain) and old (uncertain) groups. Since mistakes in one group are balanced somewhat by mistakes in the other, the overall effect on the ac- curacy of estimates of proportion molting would be slight. DISCUSSION In the field the most practical and convenient method of determining shell condition in the lobster, Homarus americanus, is by means of ex- ternal, macroscopic examination. The accuracy of such observations, however, needs to be demonstrated. For populations of lobsters that have a well- defined and relatively short annual molting period, shell condition determination by this method can be very accurate if sampling is carried out near the end of the molting period. In the pre- sent study, for those observations that could be confirmed, 97.2% accuracy was achieved. This is a clear indication that such observations can be used reliably for lobsters in the determination of proportion molting annually. ACKNOWLEDGEMENTS I am grateful to G. Dawe and A. Chafe for technical assistance in carrying out the fishing and tagging operations. G. Dawe also performed the spectrophotometric analysis of blood samples. REFERENCES Aiken, D. E. 1973. Proecdysis, setal development and molt prediction in the American lobster (Homarus americanus) . J. Fish. Res. Board Can. 30:1337-1344. Cooper, R. A. 1970. Retention of marks and their effects on growth, behavior and migrations of the American lobster, Homarus americanus. Trans. Amer. Fish. Soc. 99:409-417. Drach, P. and C. Tchernigovtzeff. 1967, Sur la methode de determination des stades d'intermue et son application generale aux crustaces. Vie Milieu Ser. A, Biol. Mar. 18:595-610. Ennis, G. P. 1972. Growth per molt of tagged lobsters, Homarus americanus, in Bonavista Bay, Newfoundland. J. Fish. Res. Board Can. 29:143-148. Ennis, G. P. 1973. Food, feeding, and condition of lobsters, Homarus americanus , throughout the seasonal cycle in Bonavista Bay, New- foundland. J. Fish. Res. Board Can.. 30:1905-1909. Hancock, D. A. and E. Edwards. 1967. Estimation of annual growth in the edible crab, Cancer pagurus L. Journal du Conseil 31:246-264. Hepper, B. T. 1965. Pre-moult changes in the structure of the integument of the lobster, Homarus vulgaris. ICES Rapp. et Proc. Verb. 156:7-14. Scarratt, D. J. 1970. Laboratory and field tests of modified sphyrion tags on lobsters (Homarus americanus). }. Fish. Res. Board Can. 27:257-264. Scarratt, D. J. and P. F. Elson. 1965. Preliminary trials of a tag for salmon and lobsters. J. Fish. Res. Board Can. 22:421-423. Weber, D. D. and R. Miyahara. 1962. Growth of the adult male king crab Paralithodes camt- schatica (Telesius). U.S. Fish and Wildlife Ser- vice, Fishery Bull. 200(62);53-75. Proceedings of the National Shellfisheries Association Volume 67 — 1977 AN EFFECT OF CHLORINATION ON THE HATCHING OF COON STRIPE SHRIMP EGGS: SO WHAT? T. O. Thatcher BATTELLE PACIFIC-NORTHWEST LABORATORIES MARINE RESEARCH LABORATORY SEQUIM, WASHINGTON ABSTRACT Groups of ten coon stripe eggs were subjected to chlorinated sea water in continuous-flow-through experiments. A 96-hr exposure to total residual oxidant levels as low as 0.16 mg'l resulted in delayed hatching of the eggs for 2 to 7 days compared to controls and lower toxicant concentrations. The observed delayed hatching of eggs exposed to chlorinated sea water occurred only after returning the eggs to clean water, in all cases except one. The possibility of both beneficial and detrimental effects of this delay are hypothesized, but the significance of this effect to individuals, populations, and ecosystems is yet to be determined. INTRODUCTION The increasing introduction of chlorine into marine environments is causing marine chemists and biologists a plethora of problems. The chemists face an enormous task of defining not on- ly the "simple" forms of the halogens (various species of chlorine, bromine and iodine, which together constitute what has come to be referred to as "free-and combined-available oxidant"), but also, at present, an innumerable quantity of in- termediate and end-product compounds that will form as a result of reactions involving potentially thousands of organic components (Vallentyne, 1957; Reuter, 1976). These reactions will occur under a variety of water quality parameters, which are suspected of influencing them. On the other hand, the biologist ultimately has the most important task of determining the answer to the big question that eventually must be applied to all such investigation of the synthesis and effects of chemical compounds in the marine environment, that is, SO WHAT? Part of the evaluation deals with observing whether the compounds of interest actually produce any effects in marine organisms. Another step is the determination of whether the observed effects are consequential or not at the in- dividual, population, species, or ultimately, the ecosystem level. The Ecosystems Department of Battelle-Northwest is investigating this complex problem of the chemistry and biology (i.e., the synthesis, effects and fate) of seawater chlorina- tion through a comprehensive program being coordinated at our Marine Research Laboratory at Sequim Bay, Washington. The study reported on here describes a sublethal effect of chlorination upon the hatching of coon stripe shrimp (Pandalus danae) eggs. Moreover, this report serves to il- lustrate the complexity of the biologist's problem of predicting whether an obvious effect is really of any consequence. MATERIALS AND METHODS One of the simplest methods of containing small organisms while performing aquatic flow-through toxicity tests has been in a 12.5 cm long cylindrical section of 3.8 cm diameter, white PVC plastic pipe, having one end covered with nytex netting of appropriate mesh size. These tubes are fastened 71 72 T. O. THATCHER in positions around the periphery of a glass 45-i aquarium. An automatic siphon, consisting of an inverted "U" tube on the effluent standpipe causes a continual 4 cm fluctuation in the solution surface level in the aquarium, and consequently within the tube containers, thereby insuring an exchange of solution within the cylinders. The flow rate through the aquarium is 0.5 i per minute. Eggs of coon stripe shrimp were used because it is an important food web and commercial species in the Pacific Northwest (Browning, 1974); it is likely to inhabit areas that will receive chlorinated effluents; and it has proven to be a good ex- perimental species in our previous laboratory studies. The shrimp were captured by otter trawl in Sequim Bay at depths generally between 15 and 40 meters. The water quality parameters at this site are essentially the same as at the laboratory. The egg- bearing females were segregated prompt- ly and held under laboratory conditions for a minimum of 2 weeks. The water temperature was 12. 5C ± 1.5°C; pH was 8.0 ± 0.2; salinity was at 30 0/00 ± 2.0, and dissolved oxygen was above 7.0 mg/1. Individuals with eggs nearing the hat- ching stage were then removed to 19-f glass aquaria in the experimental area where they were maintained in continuous-flowing conditions. Two or three of these shrimp were kept in each of several aquaria. All shrimp were fed daily on freshly-thawed Oregon Moist Pellet and minced clam at a rate of approximately 5% of their weight while kept at the laboratory. When the eggs on a shrimp were within several days of hatching (based on previous observa- tions), clusters of 75 - 100 were removed with fine- tipped forceps. These were then gently teased apart with dissecting needles, and 10 apparently healthy (with no visible defects) individual eggs were selected randomly and placed in each white plastic container. The eggs remained in the chlorinated sea water for 96 hours and were then transferred to clean water. These were observed daily for mortality, hatching, or abnormal development for approximately three weeks, in- cluding the 4-day exposure to the chlorinated sea water. Ten experiments were conducted from May As per personal communication with the Clorox Company, Oakland, California, and the USEPA Pesticide Registration Office, Seattle, Washington, May and June, 1975. through July, using eggs removed from several female shrimp. The two principal means of chlorinating water at industrial and municipal sites are by adding a solution of sodium hypochlorite (NaOCl) or by injecting chlorine gas. In the latter method, better than 99% converts to hypochlorite almost instan- taneously, so the reactions would be expected to be the same, regardless of the chlorine source (Dove, 1970; Morris, 1975). For these ex- periments, the chlorine source was a diluted NaOCl solution readily available as "Clorox." This solution was considered to be equivalent (but merely more diluted) to those currently in use by several west coast power plants. "Clorox" con- tains 90% water, 5.25% sodium hypochlorite, 4% sodium chloride, plus 0.75% pH buffers consisting of sodium hydroxide and sodium carbonate. It contains no other additives often referred to as "whiteners" or brighteners."1 A series of five chlorine exposure levels and one control was maintained by a proportional diluter (Mount and Brungs, 1967) that delivered toxicant solution to each 45-liter experimental aquarium at a rate of about 0.5 liter per minute. During the ex- periments, water quality parameters were iden- tical to those described previously for the 2-week holding period. Total residual oxidant levels, which are considered to be the most practical units currently available for analyzing the toxicant level (Brungs, 1973; Mattice and Zittel, 1976), were measured daily by an amperometric titration system employing 2 platinum electrodes (1 sta- tionary, 1 revolving) and using a polarograph for endpoint detection. Although such sensitivity was not necessary for these experiments, this system can detect total residual oxidant levels down to 1 ppb in sea water. Three different concentration intervals were employed during these experiments; the range of measured concentrations extended from 0.03 to 0.85 mg/1. These exposure levels represented ap- proximately 50% of the chlorine (CD added to the diluter, and the fate of the other 50% has yet to be determined. RESULTS The 96-hr LC50 values were found to be ap- proximately 0.5 mg/1 (ppm) for these near- hatch- COON STRIPE SHRIMP EGGS 73 ing eggs.2 However, early in the series of ex- periments, it was observed that some of the chambers, with sublethal concentrations of residual oxidant, contained eggs which hatched several days later than the eggs in the control and the lowest concentrations (Figure 1). The delayed hatching which occurred 2 to 7 days later than the control eggs or those exposed to low concentra- tions was observed at total residual oxidant levels as low as 0.16 ± .01 mg/1. In all cases except one (3-day delay at 0.17 mg/1), the reported delayed hatching occurred only after the eggs were remov- ed from the chlorinated solutions and returned to clean water. It is obvious from the data scatter (Figure 1) that there was no relationship between the concentra- tion of toxicant and the number of days of shrimp egg hatching delay. This may have been due to the inability to predict the exact date of hatching of the eggs when removed from the female. As a result, the age of the eggs used in one experiment could easily be several days older or younger than those in another experiment; however, the data from any one experiment were assumed not to be affected by variable age, due to the random selec- tion process. No additional effects were observed during the one to two week period following their hatching. 0 12 3 4 5 6 7 8 1) EXTRA DAYS REOUIRED FOR 50% Of THE EGGS TO HATCH FIG. 1. Relationship Between Coon Stripe Egg Hatching Delay and Total Residual Oxidant Con- centration in Chlorinated Sea Water. These data and their derivation by probit analysis are discussed in detail in a manuscript presently in preparation. DISCUSSION So what? What difference does it mean to the survival of coon stripe shrimp that some eggs hatch later than others? That question has not been answered by these preliminary experiments; however, several hypotheses are suggested: (a) the effect on the hatching process could be of significance to the survival of a population of shrimp that was continuously exposed to chlorinated sea water; (b) experiments have shown that newly hatched shrimp are more susceptible to chlorinated sea water than the egg stage and the delay in hatching could actually be of survival value; (c) if optimal larval releasing sites were avoided due to the presence of a toxi- cant, the selection of suboptimal sites could be detrimental to the population; and, (d) extending the early developmental period could result in greater predation losses. There are situations in which shrimp popula- tions may be continuously exposed to chlorinated water, e.g., mariculture systems employing power plant effluent cooling water, power plant effluent embayments or canals, embayments fed by sewage treatment plant effluents. Since it is not yet known whether the causative agent of the egg hatching delay is actually one of the components of the total residual oxidant fraction of the chlorinated sea water or a longer-lived by- pro- duct, the true potential of this effect can not presently be assessed. Further studies need to be conducted to determine whether the same effect is observed under intermittent chlorination schemes, and at what other stages during the reproductive cycle, if any, chlorinated water causes problems. If the egg-bearing females were stimulated to leave an area (exhibit avoidance) containing hatch-inhibiting levels of chlorinated sea water, it would appear to be beneficial to survival of the population for hatching not to occur until a "clean" water area was reached, since the newly hatched shrimp are more sensitive than the eggs. This, of course, makes the untested assumption that no latent or other deleterious effects result from the initial exposure. Also, if the hatching- in- hibitory chlorination levels were rather infre- quent, as a result of variable or intermittent chlorination, it would again appear to be of sur- vival value to the impacted populations to avoid 74 T. O. THATCHER hatching during periods of potentially lethal (to the newly hatched) conditions. If chlorinated sea water were to be released in an area which otherwise was the preferred site for female shrimp to release their young, such areas might be avoided by the females. Releasing the young in suboptimal areas could be detrimental to the shrimp populations. Another possible consequence of this delay would be that free-swimming forms would be available to predators for a longer period, if this delay in development carries on through the molting of the immature stages. Previous work relating to delayed egg hatching due to chlorination is not readily apparent. Alder- son (1973) observed a delay in hatching of flatfish eggs exposed to chlorinated sea water but did not specify the concentrations at which the effect was observed. Neff, et al. (1976) have also reported delays in the development of immature stages of crabs exposed to chlorinated compounds. In summary, another sublethal effect to shrimp has been identified which results from the chlorination of sea water. The significance of this effect to individuals, populations and the ecosystems has yet to be determined. Both negative and positive ramifications have been hypothesized. ACKNOWLEDGMENT This work was funded by the Energy Research and Development Administration under contract EY-76-C-06-1830. Their support is very much ap- preciated by the author LITERATURE CITED Alderson, R. 1973. Effects of low concentrations of free chlorine on eggs and larvae of plaice, Pleuronectes platessa L. In: Marine Pollution and Sea Life, M. Ruivo (ed.), pp. 312-315. Fishing News Ltd., Surrey, England. Browning, R. J. 1974. Fisheries of the North Pacific — History, Species, Gear, and Pro- cesses. Alaska Northwest Publ. Co., An- chorage. Brungs, W. A. 1973. Effects of residual chlorine on aquatic life. J. Wat. Poll. Contr. Fed. 45:2180. Dove, R. A. 1970. Reactions of small dosages of chlorine in sea water. Res. Rep. 42/70. File No. 0.308 0/ID. Job No. 10665. Central Electricity Generating Board. Southeastern Region, Scien- tific Services Dept., Univ. of Southampton. Mattice, J. S. and H. E. Zittel. 1976. Site- specific evaluation of power plant chlorination. J. Wat. Poll. Contr. Fed. 48:2284. Morris, J. C. 1975. Formation of halogenated organics by chlorination of water supplies. En- vironmental Health Effects Research Series EPA-600/ 1-75-002 (March 1975). U. S. En- vironmental Protection Agency, Washington, D. C. Mount, D. I., and W. A. Brungs. 1967. A simplified dosing apparatus for fish toxicology studies. Water Res. 1:21-29. Neff, J. M., R. B. Laughlin, Jr., and C. S. Giam. 1976. Effects of poly-chlorinated biphenyls, polychlorinated naphthalenes and phthalate esters on larval development of the mud crab Rhithropanopeus harrisii. Presented at: Biological Effects Program Workshop, NSE/IDOE, May 16-19, 1976, Texas A&M University, College Station Texas. Reuter, J. H. 1976. Organic matter in estuaries. Presented at: Chlorine Workshop, USEPA- Univ. of Md.-Md. Power Plant Siting Program, March 15-18, 1976, U. of Md., Chesapeake Bay Lab, Solomons, Md. Vallentyne, J. R. 1957. The molecular nature of organic matter in lakes and oceans, with lesser reference to sewage and terrestrial soils. J. Fish. Res. Board Can. 14:33-82. MORPHOLOGICAL VARIABILITY IN SEA SCALLOPS, PLACOPECTEN MAGELLANICUS (GMELIN) RELATED TO MEAT YIELD1 2 Herbert Hidu, Mark S. Richmond, Alison H. Price, II IRA C. DARLING CENTER UNIVERSITY OF MAINE AT ORONO WALPOLE, MAINE 04573 and MAINE COAST OYSTER CORP. BLUE HILL, MAINE ABSTRACT Morphological parameters were defined for the sea scallop, Placopecten magellanicus, which may contribute to large observed differences in muscle meat weight of uniform height classes. Relatively large variability in muscle width, as governed by variability in shell width, and muscle area (muscle scar) were major contributing factors. Age was a complicating factor with slight positive correlations between slow growth rate and relative size of muscle width and muscle area. INTRODUCTION This paper is an attempt to determine the im- portance of morphological differences in produc- ing large observed muscle weight variability in harvested populations of sea scallops, Placopecten magellanicus. It may provide information necessary for selecting superior animals for brood stock in projected mariculture operations of several species of scallops, perhaps allowing pro- duction of improved strains in controlled culture efforts. The relationship of shell height to meat yield has received considerable attention because it is used to set minimum size limits in the harvest of This work is a result of research sponsored in part by NOAA Office of Sea Grant, Department of Commerce, under Grant Nos. NG-40-72, 04-3-158-63, and 04-5-158-39. The U. S. Government is authorized to produce and distribute reprints for governmental purposes notwithstan- ding any copyright notation that may appear hereon. Ira C. Darling Center Contribution No. 101 natural populations (Haynes, 1966; Medcof, 1949; Baird, 1954). The minimum size limit of sea scallop drag mesh has generally been set at 4 in. or 10 cm because scallops are observed to increase muscle weight at accelerated rates at sizes over 4 in. The relationship of shell height to meat yield is obvious; however, data, especially those of Haynes (1966), reveal that other factors may be important as well, including geographical area and season of harvest related to gonad matura- tion. However, an analysis of shell height vs. muscle meat weight data taken at single locations during a single season reveals differences in meat yield from uniform sized scallops that require another explanation. For example, an inspection of Haynes' (1966) data, shows that scallops from Penobscot Bay, Maine, when divided into Vi cm shell height groups may vary by a factor as much as 3.1 (averaging over 2.3) with respect to meat (muscle) weight. These meat weight differences are not explained by differences in shucking 75 76 H. HIDO method since Haynes estimated only a 3% loss in shucking. Seasonal variability does not appear to have great importance, because scallops taken at Georges' Bank at distinct seasons and again divid- ed into V2 cm size classes reveal similar and greater (up to 7.2) variability in individual meat yield. The density of the muscle probably is not greatly variable. Thus, it would seem reasonable that meat weight differences in scallops of a given shell height might be governed by variability in "muscle area" at the base of attachment, and "muscle width" governed by the variability in overall width of the paired shells (Figure 1) and age as it might affect these factors. The muscle area and width components, of course, would be the deter- mining factors in muscle volume and thus weight. Presented here are data from a Maine popula- tion of scallops relating shell height to variability in shell length, muscle width (shell width less shell thickness), muscle area, muscle volume, and age. From these data it is possible to determine the relative importance of these parameters in explain- ing observed differences in scallop meat yield. MUSCLE Afi£A (LEFT VALVE) MUSCLE WIDTH SHELL WIOTH FIG. 1. Sea scallop, P. magellanicus, measured shell parameters. MATERIALS AND METHODS A collection of 151 randomly selected paired scallop shells was procured in February, 1971, from a commercial scallop boat working the in- shore commercial beds near Stonington, Maine. The following shell parameters were measured in the laboratory: Shell height is the maximum distance from shell margin to shell margin measured perpendicular to the external hinge line. Shell width is the maximum distance of the closed paired shells measured perpendicular to the com- missure plane. Muscle width is the shell width minus the thickness of each of the paired shells. Mean muscle area is the average of the area in cm2 of the muscle scars of the left and right valves measured by a planimeter. Muscle volume was calculated by multiplying mean muscle area by muscle width, assuming that the muscle of the closed scallop roughly approximated a cylinder. Age was figured by the method of Merrill et al. (1965) using banding patterns on the resilium and shell. To determine the reliability of our use of the technique, two technicians aged the shells in- dependently and a correlation coefficient, r, was calculated on the separate results. This value was 0.83, indicating considerable lack of confidence in our application of the aging technique of Merrill et al. However, since the determinations were un- biased, it is thought that some valid estimates of the effect of age on other parameters could be drawn. Regression analysis and correlation coefficients (r) were calculated using an IBM 370 computer. Shell height was regressed on shell length, muscle width, muscle area, age, and calculated muscle volume. Muscle width was also regressed on mean muscle area. To determine the relationship of age to muscle area and muscle width, and muscle area to muscle width, while minimizing the com- plicating effect of shell height, correlation coeffi- cients (r) were calculated within 1 cm height classes for each of the three paired factors. RESULTS The regression of calculated muscle volume on shell height (Figure 2) depicts variability similar to the reported data (Haynes, 1966) on meat weight vs. shell height. Therefore, we can assume that meat weight differences are largely the result of variability in parameters contributing to muscle volume, i.e., muscle area and muscle width. An inspection of the muscle width and area para- meters regressed on shell height (Figures 3 and 4) reveals that variability in both can contribute to significant differences in meat weight in scallops of a uniform shell height. These differences are greater than one would expect when examining the almost absolute uniformity of shell height vs. length (Figure 5). Confidence limits of the regres- VARIABILITY IN SEA SCALLOPS 77 sion lines in the muscle width component are es- pecially wide (Figure 3), indicating that this may be the major component contributing to meat weight differences. Given the relatively great variability in the mus- cle width and muscle area components, an impor- tant question is: 'Are the scallops with the relatively large muscle width the ones also with the large muscle area, regardless of the shell height of the scallop?" The spread in data points on the regression of muscle width vs. muscle area would Y= -78 192 . 10 562 X SHELL HEIGHT {CM} FIG. 2. Relationship between calculated muscle volume and measured shell height, and regression line with confidence limits at the .05 level. suggest that this is not true (Fig. 6). The very high correlation coefficient here (0.89) obviously results from the effect of a third parameter (shell height) on muscle area and muscle width. Further, the correlation coefficients of muscle area vs. mus- cle width, calculated within 1 cm height classifica- tions for all scallops in the sample, ranged bet- ween — .30 and .30 averaging .12 (Table 1). These data are slightly suggestive that a positive correla- tion exists, although a large sample of scallops of an absolutely uniform height would be needed to learn of the true relationship. Similarly, it would be important to know what effect age has on the shell parameters. There ap- pears to be a wide variation in age of especially the larger sized scallops (Fig. 7), although our lack of confidence in the aging technique certainly con- tributes to this variability. One wonders whether a slow growing scallop would have a relatively large muscle area and muscle width. Correlation coefficients of age vs. muscle width and muscle area for given size classes, averaging between .33 and .34 respectively (Table 1), would indicate that this might be the case to some degree although, again, a large sample of uniform height scallops would be necessary to substantiate these observa- tions. SHELL HEIGHT (CM) FIG. 3. Relationship between calculated muscle width and measured shell height, and regression line with confidence limits at the .05 level. 27- /. Y ■ -8 571 . 1 996 X . y / ' / 2*- s : / /. /S . • >x«\ -S- ' ' ' 21- ' S^. • S / . / : s • / " • x • '• ' / • s* / 18- s . \y\: ,,' s ' '. /^~ ■ '' ' 15- /, i\s"- '-~' 12- s- •' ,'y^' i s / S< s / . ./ ' y S j/1' / / • •** ' / 9- 1 s . ty*". ' y\ • 's' " 965 y\ .* / •• s 6: < »-. / 0 YV 1 1 i « r SHELL HEIGHT (CM) FIG. 4. Relationship between calculated muscle area and measured shell height, and regression line with confidence limits at the .05 level. 78 H. HIDO SHELL HEIGHT {CM) FIG. 5. Relationship between measured shell length and measured shell height, and regression line with confidence limits at the .05 level. DISCUSSION An improved scallop for mariculture no doubt should be fast growing, with a relatively large shell width (muscle width), and muscle area. This study suggests ways that such scallops might be found and the pitfalls to be avoided along the way. First, to select for improved growth form in scallops, it would be desirable to select several very distinct height classes within the population and then establish natural ranges in variability in shell width and muscle area and determine the ef- fect of age on these parameters. However, with P. magellanicus, at least, there would appear to be some difficulty in aging live scallops since the technique of Merrill involves counting of striations on the resilium of the hinge, in addition to lines on the shell surfaces. Our two technicians, working separately, using the cirteria of Merrill, frequently disagreed on the age of in- dividual scallops. Thus, aging live scallops using just the markings on the external shell surface would be even more difficult. To find potentially valuable specimens of P. magellanicus it would be desirable to measure separately for both shell width and muscle area, since there appears to be a very low correlation between "thick" scallops and the ones with large muscle scars within given height classes. And, again, since there appears to be some correlation 1 .. / . y . / / 4 5- Y" 786 ♦ 140 X . /,' / / / / .. . / ' . / •- / ■ 4.0- / ■ • X ' ".' X. .' o / ... /. - s T 6b- •/ ■/■•/■ n " / '. ' 'T/c '.' ". ''' * 30- / ■ -/ ■ ' , • - • •/ ■ / • o / '.'/..: ' vt 2 25- / / 20- / v 1 *S -iV « -^ — , — , — , — p~^ — ^ — , — , — , — , — , — , — , — 1 — 1 — , — 1 — , — . — ^-1 0 6 8 10 12 14 16 18 20 22 24 26 28 MUSCLE AREA (CM2) FIG. 6. Relationship between calculated muscle width and calculated muscle area, and regression line with confidence limits at the .05 level. between age and the parameters producing muscle volume, it would be valuable to select for the shell parameters, together with a relatively young age for the particular height class. Finally, the question of the role of environment vs. heredity in scallop growth form should be con- sidered. There is some thought that the relatively thick-bodied scallop might be free living and not fast in the bottom as others might be. We have made no observations on this, although it is ob- FIG. 7. Relationship between estimated age and measured shell height, and regression line with confidence limits at the .05 level. VARIABILITY IN SEA SCALLOPS 79 TABLE 1. The relationship of scallop age to muscle area; age to muscle width, and muscle area to muscle width expressed as correlation coefficient r for 1 cm height groups of a single population of 142 sea scallops. Muscle area Size Classes Ag e vs. Age vs. vs. (height cm) No. Scallops muscle area muscle width muscle width 8.0- 8.9 16 .71 .46 .38 9.0- 9.9 15 .45 .21 — .30 10.0-10.9 18 .06 .26 .26 11.0-11.9 22 .43 .57 .10 12.0-12.9 17 .53 .05 .30 13.0-13.9 18 .30 .26 .15 14.0-14.9 20 .59 .32 .10 15.0-15.9 16 .45 .50 — .07 142T X .33 .34 .12 viously an important point in selection. Heritabili- ty trials with selected animals and specimen collec- tion by SCUBA would determine whether or not this is true. LITERATURE CITED Baird, F. T., Jr. 1954. Meat yield of Maine scallops (Pectin magellanicus) . Research Bull. No. 16. Maine Dept. of Sea and Shore Fisheries, Augusta, Maine, 3 pp. Haynes, E. B. 1966. Length-weight relation of the sea scallop, Placopectin magellanicus (Gmelin). I.C.M.A.F. Bull. No. 3, 17pp. Medcof, J. C. 1949. Meat yield from Digby scallops of different sizes. Progress Repts. of the Atlantic Coast Stations, Fisheries Research Board of Canada, Bull. No. 44, pp. 6-8. Merrill, A. S., J. A. Posgay, and F. E. Nichy. 1965. Annual marks on shell and ligament of sea scallop (Placopectin magellanicus). Fishery Bulletin 65(2) :299-311. Proceedings of the National Shellfisheries Association Volume 67 — 1977 THE RELUCTANCE OF THE OYSTER DRILL (UROSALPINX CINEREA) TO CROSS METALLIC COPPER1 John E. Huguenin2 UNIVERSITY OF MASSACHUSETTS WAREHAM, MASSACHUSETTS ABSTRACT It was demonstrated statistically (highly significant) that oyster drills are extremely reluctant to cross metallic copper. This reluctance was shown to be due to some characteristic of the metal rather than to the effects of physical obstruction (statistically highly significant). It was also shown that the width of copper strip is an important parameter in preventing crossings (statistically highly significant). Copper barriers at least as wide as the largest animals are recommended for maximum effectiveness. In ad- dition, suggestions are presented for exploiting this phenomenon in practical applica- tions. INTRODUCTION The steady decline in oyster meat production in the U.S. from 231 million pounds (105,000 metric tons) in 1910 to 77 million pounds (35,000 metric tons) in 1951 (and presently somewhat less) has been blamed, at least in part, on ineffective con- trol of predation by oyster drills (Glancey, 1953). The oyster drill Urosalpinx cinerea (Say) is a small, predatory marine snail which is widely distributed in the coastal waters of North America and the British Isles. It is particularly destructive of young shellfish with their thin shells and greater vulnerability. A great deal of effort has been ex- pended in studying this predator and in trying to control its predation on shellfish stocks (Galtsoff, etui, 1937; Carriker, 1955). It is a common belief among some shellfish culturists in the Middle Atlantic and New England States that oyster drills avoid metallic copper and will not cross a barrier of copper. Past research The preparation of this paper has been supported by the International Copper Research Association under INCRA Grant #251. Current address: Groton Biolndustries, P.O. Box #133, Woods Hole, MA 02543. clearly indicates that this phenomenon exists both in the laboratory and in the field (Glude, 1956). Based on this effect, considerably efforts in the 1950s were expended trying to develop bottom mounted copper barriers to inhibit the movement of oyster drills and other gastropod predators on- to shellfish flats (Glude, 1956; Castagna, personal communication). The experiment described in this paper was designed to further explore the inhibiting effects of metallic copper on the movement of oyster drills. The intent was to carry out experimentation preliminary to a reevaluation of the practical pro- blems of trying to exploit this interesting phenomenon in commercial shellfish culturing. MATERIALS AND METHODS Fifty U. cinerea, ranging in length from 2 cm to 3 cm with a mean of 2.4 cm, were placed in a 60 cm x 30 cm x 30 cm deep glass aquarium with a useable volume of 50 L and a steady through flow of about 1.5 L/min of raw estuarine seawater. All the test animals were collected at low tide along the shore and within 30 m of the laboratory's seawater intakes in Wareham, Massachusetts. The bottom of the tank was covered with coarse sand 80 RELUCTANCE OF OYSTER DRILL 81 and three sides were covered on the outside with black contact paper to reduce the possibilities of inadvertent exterior stimuli. All drills were placed in the tank environment for at least three weeks prior to starting experimentation. Water temperatures were always in the range of 22-28 C and salinities were 15-30 0/00 during the ex- periments. Both parameters varied as a function of the tide and weather conditions. Two identical pedestals were made from stan- dard PVC (polyvinyl chloride) fittings. Each pedestal had a base consisting of a IV2 inch (nominal) socket coupling (4 cm long by 5.7 cm in diameter) joined to a length of l1 2 inch (nominal) plastic pipe (4.7 cm in diameter) forming a pedestal 13 cm high. The outside surfaces were roughened with coarse sand paper. Each pedestal had a IV2 inch (nominal) plug which formed a solid platform recessed 0.8 cm on the inside of the plastic pipe at the top. The pedestals each had four 1/16 inch diameter (0.16 cm) holes, to be used for securing experimental filaments, drilled at 0.6 cm intervals along their length and centered on the midpoint of the plastic pipe sections. The intent was to provide a section of constant diameter both before and after the "barrier" position. The overall dimensions of the pedestals were constrained by the depth of the water column. In order to stimulate drills to climb the pedestals, mussels, including both the blue mussel, Mytilus edulis and the ribbed mussel, Modiolus demissus, were used as "bait". Mussels were chosen over oysters, due to the apparent preference of drills for mussels when presented with prey animals of equal size (Carriker, 1955). In addition, both these mussel species were present in the area where the drills were collected. However, it should be noted that size is probably an even more important parameter than species in the food preference of drills, younger faster grow- ing bivalves being preferred (Haskin, 1950). Mussels were periodically placed on the platform at the top of the pedestals to accustom the drills to feeding from the pedestals during the early condi- tioning period. Drills did not appear to hesitate climbing up the pedestals or feeding in this man- ner. However, they were starved for about 3 weeks prior to the start of experimentation. Three experiments were conducted with the two pedestals. In experiment #1 (July 29 — Aug. 4, 1975), pedestal No. 1 was wrapped around the midpoint with a 2.3 m length of 20 gage copper wire (diameter of 0.8 mm). This produced a cop- per band 1.75 -- 2 cm wide including about a dozen evenly spaced wraps around the plastic pipe. The copper wire was leached in running seawater for more than a month before it was placed on the pedestal at which time it was greenish in color. The second pedestal was bare. Based on the exposed copper surface area-flow rate relationships and the range of corrosion rates for pure copper, it is clear that the stabilized cop- per concentrations in the water during these ex- periments, under the worst conditions, could not have been more than a doubling of the existing background level (6-8 PPB) (Huguenin & Ansuini, 1975). In experiment #2 (Aug. 5-15, 1975), pedestal No. 1 was unchanged but pedestal No. 2 was wrapped around the middle with nylon monofilament fishing leader material, also previously leached and of equivalent length and diameter to the copper wire of pedestal No. 1. The intent was to provide two pedestals of equal physical obstruction to the climbing drills, one of copper and the other of an "inert" material. In ex- periment #3 (Aug. 18-23, 1975), pedestal No. 1 continued unchanged while pedestal No. 2 had a single winding of the same copper wire producing a copper strip with a width equal to the diameter of the wire. The following procedures, involving frequent switching of bait and pedestal positions, were developed to cancel out any biases between the pedestals that might be present in the tank due to water circulation patterns, different lighting con- ditions, extraneous external stimuli or other posi- tional factors. Procedures also compensated for any learning on the part of the drills. Small live mussels of equal size (2-3 cm long), one of each species, were placed on top of the pedestals in the morning of each day of experimentation. Mussels on the pedestals were switched each day. On days data was collected, the tank was check- ed periodically to see if drills had climbed up either pedestal. These checks occured irregularly but usually at approximately 0.5 hr. intervals. If no drills were present on the pedestals, no record was made nor was any other action taken. 82 J. E. HUGUENIN However, if drills were on either or both pedestals a positive observation was recorded. The number of drills above and below the midpoint of each pedestal were noted. If no drills had proceeded past the midpoints, they were simply removed from the pedestals and placed to the sides of the tank. In contrast, if some had progressed past the midpoint, on either pedestal, not only were they removed but the pedestal positions were switched. All data were acquired during daylight hours. RESULTS Table 1 gives the numbers of drills noted per positive observation in each condition by experi- ment and a summary of the data. Utilizing an F test, the observed differences in oyster drill behavior between both pedestals in each of the three experiments were statistically analyzed using a one-way classification with equal numbers. Dif- ferences between pedestals in experiment §1, ex- periment §2, and experiment #3 were all statistically highly significant at the 1% level. DISCUSSION I observed during experiments #1 and #2 that drills tended to pile up directly under the copper pedestal No. 1 while not doing so on the other pedestal. However, activity levels of drills during the experiments varied considerably. They com- monly extended their proboscis over the copper strip to distances approaching the heights of their shells before backing off or letting go and falling off. This led to the idea that the dimension of the copper strip might be important, and resulted in experiment #3. In this experiment the single strand of wire, though stopping many drills, was clearly not as effective as the wider band of copper. Thus, in order to be most effective, the copper strip must be at least as wide as the height of the snails' shells. It was clear in experiment #1 that drills would TABLE 1. Number of Drills Observed During Experiments and a Summary of the Data # Drills # Drills Below Above Pedestal Pedestal Midpoint Midpoint Remarks EXPERIMENT 1 Pedestal #1 23 0 (Band of Copper (1.00)* (0) July 29- Wire) — 4 days of data taking, August 4, 23 positive observations 1975 Pedestal #2 6 30 over a six day period (Bare) (.26) (1.30) EXPERIMENT 2 Pedestal #1 62 9 (Band of Copper (1.05) (.15) — 9 days of data taking, August 5, - Wire) 59 positive observations August 15, over an eleven day period. 1975 Pedestal #2 18 79 (Leader Material) (.31) (1.34) EXPERIMENT 3 Pedestal #1 43 0 (Band of Copper (1.79) (0) August 18,- Wire) — 4 days of data taking, August 23, 24 positive observations 1975 Pedestal #2 17 16 over a six day period. (Single Strand of (.71) (.67) Copper Wire) ' Average number of drills per positive observation. RELUCTANCE OF OYSTER DRILL 83 not cross the copper wire. However, during the earlv phases of experiment #2 a few did cross. This might be explained by the fact that, while the cop- per wire had been leached for over a month prior to being installed on pedestal No. 1 for experiment #1, when the wire was being wrapped around the pipe the green oxide cracked in numerous places showing hairline cracks of golden copper colora- tion. During the 8 days between the start of ex- periment ffl and the start of experiment #2, these cracks had disappeared. The exact chemical condi- tion of the copper surface could be expected to in- fluence the degree of reluctance of oyster drills to cross copper. However, it isn't clear from the data whether or not this was the case in these ex- periments. The effects of copper can be better understood by examining the corrosion product films which form after exposure to seawater. The initial film is cuprous oxide which adheres tightly to the metal substrate but this film in time hydrolizes to form cuprous hydroxychloride which appears to be neither as toxic nor as tightly adherent (Efird, 1975). This second film is easily removed, such as by the continued slow action of abrasive particles carried by water currents. A higher corrosion rate to further increase the barrier effectiveness of the copper could be achiev- ed by coupling electrically to materials more noble than copper in the galvanic table. Unfortunately, copper tends to be more noble than most of the common metals and coupling to these materials would of course completely destroy the effec- tiveness of copper as an oyster drill barrier since the copper would no longer corrode at all. Materials that would be acceptable for this ap- plication include graphite, titanium and some of the high alloy specialty steels. It should be noted that the barrier effectiveness of copper is not due to the electropotential. This is substantiated by previous experiments (Castagna, unpublished) where comparable voltages were induced on car- bon with no response from the snails. There are interesting implications for large areas along the coast where traditional bottom culturing of shellfish can no longer be accomplished, or is carried out only with difficulty, because of serious predation by oyster drills. The work reported here, as well as previous related efforts (Glude, 1956; Castagna, unpublished), clearly show that copper is an effective barrier. The question revolves around how to exploit this phenomenon in practical applications. Previous efforts have concentrated on small bottom mounted barriers to protect traditional shellfish beds. These barriers were all very effective but cost and maintenance difficulties made them impractical (personal com- munication, Michael Castagna). However, since the basic repellent principle works well, there is a good chance that reevaluation, redesign and fur- ther efforts can overcome the problems. It should be possible to incorporate copper into bottom- mounted fences which are also designed to be ef- fective against other shellfish predators. Another alternative to achieving practical ap- plications is to consider a different approach, such as bottom-mounted frames which support off- bottom culturing of shellfish on strings or in trays. Oyster drills could easily be kept from climbing up the vertical supports by making these risers either wholly of a copper material or by using only small copper strips around the bases. As an example, a 4 inch (10.2 cm) diameter by 6 inch (15.2 cm) wide band of copper roof flashing 0.002 inch (0.55 mm) thick, when placed around each of the four legs of an off-bottom culture frame, would have a retail materials cost of only about four dollars and a useful lifetime in seawater of at least four years. Copper barriers may in time prove to be valuable, but alone they do not offer a long term complete solution. While drills do not have a pelagic stage, the young are noted for their ability to travel long distances by tacking on to floating and drifting objects (Carriker, 1957). Thus some movement of drills into the system, by drifting in on objects and by a few crossing the barriers, must be assumed. Other measures will be necessary to periodically remove these drills before they can build up to unacceptable levels. Therefore, a cop- per barrier may well form the core of an anti-drill system but will probably have to be supplemented by other techniques for it to fully realize its poten- tial. Considerable increases in yields of shellfish and new productive acreage could result if prac- tical systems can be developed. Under most likely circumstances involving estuarine applications, copper would not pose a threat to the quality or survival of the culture organisms, or their food organisms, although there are some potential pit- 84 J. E. HUGUENIN falls (Huguenin & Ansuini, 1975). These pitfalls include being aware of acceptable copper surface area-flow rate relationships and avoiding inadver- tent galvanic cells by coupling to other metals or by burying part of the copper material in the bot- tom. With care, copper additions to the water can usually be easily kept to concentrations several orders of magnitude below existing background levels. ACKNOWLEDGEMENTS The author wishes to acknowledge the many valuable comments, ideas and suggestions from Paul Chanley, Mike Castagna, Dr. Melbourne Carriker and Dr. Ken Tenore. Their interest and assistance with an early draft were very much ap- preciated. The author is especially grateful to Mike Castagna for access to his unpublished work in this area, which proved most helpful. Thanks are also extended to Terry Hammar for his help on the statistical analysis. LITERATURE CITED Carriker, M.R., 1955. Critical Review of Biology and Control of Oyster Drills Urosalpinx and Eupleura, U. S. Fish and Wildlife Service, Special Scientific Report — Fisheries No. 148, 150 pp. Carriker, M.R., 1957. Preliminary Study of Behavior of Newly Hatched Oyster Drills, Urosalpinx cinerea (Say), J. Elisha Sci., Soc, 73:328-351. Castagna, M. A Copper Barrier Fence for Protec- ting Oysters from the Predatious Snails Urosalpinx cinerea and Eupleura. Unpublished manuscript, 30 pp. Efird, K. D., 1975. Interrelation of Corrosion and Fouling for Metals in Seawater, Paper 124, presented at Corrosion '75, Toronto, Canada, National Association of Corrosion Engineers, 14 pp. Galtsoff, P.S., H. F. Prytherch & J. B. Engle, 1937. Natural History and Methods of Control- ling the Common Oyster Drills {Urosalpinx cinerea Say and Eupleura caudata Say) Fishery Circular No. 25. U. S. Bureau of Fisheries, 24 pp. Glancey, 1. B., 1953. Oyster Production and Oyster Drill Control. 1953 Convention Papers— National Shellfisheries Association, June 22-25, New Orleans, Louisiana, pp. 61-66. Glude, J. B. 1956. Copper, a possible barrier to oyster drills, Proc. Nat. Shellfish. Assoc. 47:73-82. Haskin, H. H., 1950. The selection of food by the common oyster drill, Urosalpinx cinerea Say. Proc. Nat. Shellfish Assoc. 41:62-68. Huguenin, J. E. & F. J. Ansuini, 1975. The Ad- vantages and Limitations of Using Copper materials in marine Aquaculture. Ocean '75 Conference Record, San Diego, CA, Sept. 22-24, 1975, jointly sponsored by the Marine Technology Society and the Institute of Elec- trical and Electronic Engineers, pp. 444-453. Proceedings of the National Shellfisheries Association Volume 67 — 1977 A STUDY OF THE LITTLENECK CLAM (PROTOTHACA STAMINEA CONRAD) AND THE BUTTER CLAM (SAX1DOMUS GIGANTEUS DESHAYES) IN A HABITAT PERMITTING COEXISTENCE, PRINCE WILLIAM SOUND, ALASKA. Richard B. Nickerson ALASKA DEPARTMENT OF FISH AND GAME F. R. E. D. DIVISION CORDOVA, ALASKA ABSTRACT Frequency of occurrence is stratified by tide level and some individuals survived a 5.5 feet (2m) landmass upheaval following the 1964 Good Friday earthquake having endured 8 years in abnormally high zones on the low tide terrace. Growth and age were determined from mark and recovery studies which subsequently yielded age- length- weight relationships and determination of critical size. Histological studies yielded spawning information. Population estimates were ob- tained by stratified random sampling and probability density function parameters. Harvesting with high pressure water jets appears practical and huge man-made pads of suitable substrate at optimum tide levels seem feasible on a crop rotation basis. INTRODUCTION Alaska possesses 33,904 miles of tidal shore line and extensive clam populations that have scarcely been touched commercially for human consump- tion. Fresh and frozen razor clams (Siliqua patula Dixon) from approved growing areas achieved in- terstate shipping status during 1975. Attention is now focusing on butter clams (Saxidomus giganteus Deshayes) and littleneck clams (Pro- tothaca staminea Conrad) for the same purpose. This study was conducted to obtain background data on the biology and yield of S. giganteus and P. staminea for management application and shellfish industry use when large-scale exploita- tion is initiated. Although S. giganteus and P. staminea are often found together on the same beach in various ratios (Bourne, 1967, Fraser and Smith, 1928) they are also found in the absence of the other (Fraser and Smith, 1928). I have recovered S. giganteus from marginal razor clam bearing substrate near Cor- dova, Alaska, but not P. staminea. Similarly, H. M. Feder (personal communication) found P. staminea at sites in Galena Bay and Landlocked Bay, Prince William Sound, but not S. giganteus; he observed the reverse situation at a site at the head of Port Fidalgo, Prince William Sound. Quayle (1974) describes P. staminea habitat as protected gravel-mud beaches and S. giganteus habitat as sand-gravel beaches. The habitat of the site examined in this study and its close proximity to Cordova, Alaska pro- vided a convenient location to study adequate densities (for statistical purposes) of P. staminea and S. giganteus simultaneously. MATERIALS AND METHODS Habitat. The study site was located near the 85 86 R. B. NICKERSON head of the west side of the east arm of Simpson Bay, Prince William Sound, Alaska, i.e., 60° 38' 22' No. Lat.; 145° 51' 52" W. Long. Distance from Cordova, Alaska to the site was 7.5 miles (12.07 km) straight line distance and 12.2 miles (19.63 km) by skiff. Prior to the 1964 Good Friday earthquake the study site formed the northwest shore of an island. The earthquake raised the landmass in the Cor- dova area 5.5 feet (2 m) (Reimnitz, 1966) which caused the island to become part of the mainland and created a small, very protected cove of the former northwest shore (Fig. 1). FIG. 1. Location of P. staminea and S. giganteus study site near Cordova, Alaska. The site was shaped somewhat like an amphi- theatre. Width of the beach at the —3 foot (—0.91 m) tide level was 37 feet (11.28 m) and became progressively wider at higher elevations; width at the +6 foot (+1.83 m) tide level was 111.5 feet (33.99 m). Slope distance from the +6 foot to the —3 foot tide level ranged from 62.25 feet (18.97 m) to 122.50 feet (37.34 m) averaging 89.58 feet (27.30 m). The upper limit of the beach merged with a rocky cliff. Large rocks 18 by 9 inches (46 X 23 cm) heavily capped the upper portion of the beach from the +15 (4.57 m) to the +9 foot (2.74 m) tide level. Size and density of rock capping diminished at lower beach levels. Cobbles 5 by 10 inches (13 X 25 cm) were common at the +6 foot tide level and were rare at the +2 foot (0.61 m) tide level. Small stones Vi to 3" (1.27 to 7.62 cm) capped the beach densely at the +4 foot (10.16 m) tide level and meagerly at the — 1 foot ( — 0.3048 m) tide level. The substrate became softer and muddier from the +1.5 to the —2.5 foot (+0.46 to —0.76 m) tide level, being completely mud at the latter level. The subsurface substrate, a damp mud-gravel, was very firm from upper tide levels down to the +2 foot level. Below the +1.5 foot tide level, consistency went from soft and yielding to a thick liquid, being entirely the latter at the — 2.5 foot tide level. At a depth of 1 foot (0.3048 m), a layer of organic matter resembling peat was occasionally encountered. Both 5. giganteus and P. staminea were found down to this depth residing on or just above, but not within or beneath, the organic layer. Frequency of occurrence by tide level. On July 11, 1972 eighteen tide levels were located by hand level and leveling rod in conjunction with local tide tables from +6.0 to —2.5 feet (+1.83 to — 0.76 m) relative to mean lower low water in 0.5 foot (0.15 m) increments. Each tide level was marked with two or more steel stakes, depending upon the contour. The stakes were appropriately labeled. Each tide level, therefore, represented a transect stratum. Strata were sampled by randomly plac- ing a square sampling frame enclosing an area of 1.0 ft.2 (0.0929 m2) along the beach n times. In- itially a 5 ft.2 (0.46 m2) frame was used, but was found to be too difficult to work with. All substrate was excavated to a depth of 1 foot (0.3048 m) which coincided with the maximum depth of the species involved at this site. Excava- tion was facilitated by use of small garden trowels and cultivator claws. Sampling from the —1 foot (—0.3048 m) to the —2.5 foot (—0.76 m) tide level required the sampler to sit on a 2-foot by 3-foot (0.61 X 0.91 CLAMS IN PRINCE WILLIAM SOUND 87 m) sheet of plywood after walking along boards placed on the beach to prevent sinking into the mud. Propane lanterns were used during night sampling. Large clams were removed as excavation pro- ceeded and placed in labeled containers. Clams smaller than 20 mm in total valve length were not as easily discerned, hence were recovered by washing the substrate through a screen containing 16 meshes per inch (2.54 cm) with water supplied by a Homelite Model XL pump. All clams from each 1 ft.2 sample were placed in labeled con- tainers and separated later at the laboratory. The average number of P. staminea and 5. giganteus per 1 ft.2 (0.0929 m2) by respective tide level was calculated. The frequency distribution of P. staminea by tide level was extremely leptokur- tic, that of S. giganteus less so. P. staminea data was smoothed by employing the gamma function: r(x) = ; ?> f-e-7dt whereas a function of the form: LnY = a + blnx+ c (lnx)2 +d (lnx)3 sufficied for fitting S. giganteus data. The above equations were employed to obtain estimates of proportions by tide level to serve as probability density function parameters in population estimates for clams residing in this particular tidal regime. Growth and Age. Assistance in determining growth and age of P. staminea and S. giganteus, collected during the "frequency of occurrence by tide level study" phase, was provided by a mark and recovery program. On January 18, 1973 67 S. giganteus and 200 P. staminea were captured at the Simpson Bay site and returned to the laboratory where they were kept in refrigerated buckets of sea water. On January 22, the clams were placed in a 20 p. p.m. solution of alizarin red and remained in the refrigerated solution (i.e. 33.5° F) (0.83° C) which was changed every few days, until February 9, 1973. The clams were a medium purple color when released from the solution. The same day they were returned to the Simpson Bay site. Prior to planting them in a marked, previously unex- cavated location between the previously ex- cavated+1 and + 1.5 foot (+0.3048 and +0.4572 m) tide level, each clam was marked by filing a narrow "V" at the ventral edge of the valves using a double extra slim taper file. The purpose of the file mark was to insure a permanent mark on the annulus that was being or had just formed lest the alizarin became too faded in time. The clams were recovered on April 11, 1974, 13 months later. From the aforementioned program, specific an- nuli which facilitated age and growth analysis of the stored valves of the "frequency of occurrence by tide level" specimens were easily identified. Early annuli on small specimens were determined with the aid of a dissecting microscope (10 x to 60 x). Length-age relationships were constructed for both species which, when subjected to the von Bertalanffy growth equation, It = foo(i-e-*e-"» (Ricker, 1958), the Walford ex- pression, it + 1 = i°°(l-k) + ki, (Ricker, 1958), Taylor's equations for maximum rate of natural mortality M = 2.996/A.,5 and life span to achieve 95 percent of the asymptotic length A.95 = t„ + 2.996/K, (Taylor 1958) and linear and curvilinear functions yielded the following: estimated growth rates; asymptotic length, f°°; asymptotic weight, w°°; maximum rate of natural mortality, M; the A.,5 value; the first differen- tials of absolute growth and biomass; and critical size. These statistics are of value in setting minimum legal size limits and for determining op- timum sustained yield. Length-weight relationship. Collections of P. staminea and S. giganteus were made biweekly to monthly, except for February, during 1972 and 1973. Greatest valve length was recorded to the nearest millimeter; total weight, shucked weight and trimmed weight (mantle muscles, adductor muscles and lower section of body — see Quayle, 1969, pp 52, 53) were recorded to the nearest hun- dredth of a gram. Prior to weighing, the whole clams were rinsed and brushed of adhering sub- strate, then dried with paper toweling. Shucked clams were blotted momentarily on paper towel- ing to remove excess liquids. Linear functions yielded length-weight relation- ships depicting change in weight by time. Age- length-weight relationships were obtained using data from the previous section. Spawning. Along with weight-length data, ad- ditional information obtained from the clams in- 88 R. B. NICKERSON eluded: sex; gonad appearance; relative degree of sexual maturity and ripeness based on gross obser- vation and microscopic inspection of living tissues; pH of the gonad using "pHydrion" paper; and representative diameters of ova. Blocks of gonad tissue were fixed in 10 percent buffered for- malin. The paraffin method of tissue preparation was followed. Hematoxylin staining procedures (progressive method) were employed using Dela- field's Hematoxylin and eosin Y counterstain. Tissues were placed in a dilute solution (about 0.5 percent) of parlodion in ether-absolute alcohol (50:50) for 2 minutes following the first absolute alcohol bath to prevent yolky tissues from loosen- ing from slides. Sections were cut 10 to 20 u thick, with S. giganteus sections averaging 15 u and P. staminea sections averaging 10 \x. During 1973 a 45-day Peabody-Ryan recording thermograph with sensory probe was placed 1 fathom (2 m) below mean lower low water at the study site. Seawater temperatures were related to gonad pH, and manifest changes in gonad ap- pearance with the passage of time were analyzed. Population estimation. Estimates of the stand- ing crop of P. staminea and S. giganteus for the entire beach at the site were made utilizing the stratified random sampling method (Cochran, 1963). In addition, employment of proportions by tide level yielding probability density function parameters (Nickerson, 1975) were used in con- junction with an index tide level for P. staminea for comparison with the stratified method. Methods of harvest. A hydraulic clam digger (Fig. 2) similar to that described by Bourne (1967) was tested at the Simpson Bay study site and at Observation Island, Orca Inlet, 3 miles (4.83 km) north of Cordova where a dense mat of blue mussels (M. edulis) covered a mud-cobble beach containing P. staminea and 5. giganteus. The dig- ger had a primary manifold composed of 2Vi inch (6.35 cm) i.d. copper tubing. Three secondary manifolds composed of 1 inch (2.54 cm) i.d. cop- per tubing extended from the primary manifold. Each secondary manifold was fitted with four V* inch (6.4 mm) i.d. nozzles on 3 inch (7.62 cm) centers. The odd shape of the digger permitted the handle to be reversed for digging razor clams to a depth of 18" (45.72 cm). A Homelite 2V* inch HG. 2. Hydraulic clam digger. (6.35 cm) high pressure pump delivering up to 105 pounds (47.63 kg) per square inch (6.4 cm2) and up to 12,500 gallons (47,3181) per hour was floated from a skiff. The screened intake was secured about 1 foot (0.3048 m) off the bottom. An in-depth study of the digger was not con- ducted; such a study was carried out by Bourne (1967) with butter clams. A pilot experiment by Feder and Paul (1973) wherein littleneck clams were excavated by hosing a jet of water over the substrate indicated that the method was produc- tive. The digger used at Simpson Bay and at Obser- vation Island was tested for apparent efficiency at various pressures parallel, perpendicular, and angular to the water's edge. RESULTS Frequency of occurance by tide level. P. staminea was distributed from the +5.0 to the —2.5 foot (+1.52 to —0.76 m) tide level, the mode occuring at the +1 foot (+0.3048 m) tide level (Fig. 3). 5. giganteus was distributed from the +3.5 foot to the —2.5 foot (+1.07 to —0.76 m) tide level, the mode occuring between the — 0.5 and the +0.5 foot (—0.15 to +0.15 m) tide levels (Fig. 4). The "estimated" curves shown in these figures have predictive value and will be used in a later section dealing with population estimation. Noteworthy are the two dips in the "observed" curves of both species located at the — 0.5 and —1.5 foot (—0.15 and —0.46 m) tide levels. The dips reflected the presence of two large, subsurface rocks occupying the area under respective sampl- ing frame locations. P. staminea obtained at the study site using a 5 ft.2 (0.46 m2) sampling frame at the +5 foot (1.52 CLAMS IN PRINCE WILLIAM SOUND 89 FIG. 3. Frequency of occurrence of P. Staminea by FIG. 4. Frequency of occurrence of S. giganteus by tide level, Simpson Bay, Prince William Sound, tide level, Simpson Bay, Prince William Sound, Alaska. Alaska. m) tide level (n = 4) ranged in approximate age from 13 to 15 years. Extreme disruption of growth occurred after achieving age 5 to 7 years. Max- imum valve length of the largest specimen was 30.70 mm. Growth disruption was attributed to land-mass uplift. Growth disruption of P. staminea was less noticable at the +4 foot tide level where the youngest specimen was approx- imately 10 ±1 years of age. At the +3.5 foot tide level 1 specimen of S. giganteus was collected; its age was approximately 16 ± 2 years and max- imum valve length was 34.75 mm. Post-earthquake recruitment of P. staminea was found at the +2.5 foot tide level and lower tidal zones. Post-earthquake recruitment of S. gigan- teus was found at the +2 foot tide level and lower tide zones. Growth and age. Age was determined by count- ing annuli of the "frequency of occurrence by tide level" specimens which were identified following the mark and recovery program. Age and growth rates for P. staminea and S. giganteus are presented in Tables 1 and 2, respectively. Walford lines, first differentials of absolute growth and biomass, asymptotic length and weight, critical size, A. ,5 value and maximum natural mortality rate for the above species are presented in Figures 5 and 6 respectively. Comparisons of growth rates between two tide levels each for P. staminea and S. giganteus reveal that growth rates are significantly greater at tide levels where modes of frequency of occurrence are located than at higher tide levels. P. staminea achieves a significantly greater rate of annual in- crement in valve length at the +1 foot (+0.3048 m) tide level than at the +2 foot (+0.61 m) tide level. Similarly, S. giganteus's rate of annual in- crement in total valve length is significantly greater at the 0 (MLLW) tide level than at the +1.5 foot (+0.46 m) tide level. Tables 3 and 4 present these data. Slower growth rates for P. staminea observed by Paul and Feder (1973) at Galena Bay are at- tributed to more severe winter conditions than observed at Simpson Bay. Inner Galena Bay, where Eater Beach and Indian Creek Flats (designated by Paul and Feder, 1973) are located, occasionally freezes and sea ice in excess of 1 foot (0.3048 m) thick is documented (Capt. H. Curran, M/V Montague, ADF&G, pers. comm.). Size of P. staminea and S. giganteus increases with depth of location within the substrate. That is, smallest clams were found near the surface; largest clams were found at the lower limit of cap- ture. 90 R. B. NICKERSON TABLE 1 Average greatest valve length and average whole weight of Protothaca from Simpson Bay. Prince William Sound. Alaska with corresponding relationship to age, and data for fitting a Walford line to length. Observed X Std. error Calculated Calculated Calculated t Age Number Length" Std. Dev. of Mean Length Weight (g) Weight (g) Annulus (Years) of Clams (mm) Sx (mm) Sx (mm) (mm)* 9-25-72c 7-17-73" 1 0.5 35 1.36 0.50 0.08 2 1.5 40 4.53 1.12 0.1b 3 2.5 49 10.26 2.72 0.39 10.64 0.37 0.32 4 3.5 4Q 18.56 4.00 0.58 10.28 1.04 1.96 5 4.5 40 26.63 3.96 0.57 26.09 4.49 4.88 6 5.5 40 32.13 4.08 0.58 31.46 7.56 8.59 7 0.5 40 35.85 4.40 0.63 35.70 10.74 12.50 8 7.5 42 38.52 4.83 0.75 30.04 13.78 16.50 9 8.5 28 41.74 4.56 0.86 41.67 16.52 20.10 10 9.5 8 45.97 2.48 0.88 43.75 18.92 23.29 11 10.5 45.30 20.96 26.03 12 11.5 46.68 22.66 28.33 13 12.5 47.70 24.06 30.25 14 13.5 48.50 25.21 31.81 15 14.5 49.14 2b. 14 33.09 "Obtained from the +1 toot ( +0.348 ml tide level 'Calculated trom the von Bertalanffy growth equation using final trail t°° =51.5 mm; r = 0.9987 'Calculated from Log Y = —3.2888 + 2.7823 LogX; r = 0.9984 The standard error of estimated Y. Sy'. is given by the following equation Sy1 = antilog \/~0 00005929 + (0.0226 log \ '' where X = departure from mean. Mx. Mx = 1 3822* N = 50 Calculated from log Y -3.5940 + 3.0233 Log X; r = 0.9941. Sy1 = antilog yr0 0001 123o + (0.0712 Log X)2 where X = departure from mean Mx M. 1 5276' N - 37 "From Ezekiel and Fox 1 1959) pp 287-288 Length-weight relationships. Significant changes (P<.05) in total body weight occur for P. statninea and S. giganteus throughout the year. Tables 1 and 2 present the greatest observed range in weight for P. statninea and S. giganteus, respec- tively, and also show correspondence with age and length. Table 5 presents total weight-shucked weight relationships for P. statninea. Tables 6 and 7 present total weight — shucked weight — trim- med weight for S. giganteus. The largest specimen of P. staminea collected at the Simpson Bay site measured 62 mm in total valve length and weighed 64.47 grams. The largest specimen of S. giganteus collected at the same site measured 98 mm in total valve length and weighed 258.94 grams. Spawning. Histological studies indicate that spawning is initiated by P. staminea between late May and mid-June with an accumulation of ap- proximately 1050 temperature units (i.e., the cumulative degrees fahrenheit of the maximum daily deviation of seawater temperature ±32°F. (0° C) observed from January 1 to the onset of spawning). Spawning of S. giganteus appears to begin between mid-June and early July when ap- proximately 1282 temperature units have ac- CLAMS IN PRINCE WILLIAM SOUND 91 TABLE 2. Average greatest valve length and average whole weight of Saxidomus from Simpson Bay. Prince William Sound, Alaska with corresponding relationship to age. and data for fitting a Walford line to length. (1) (2) (3) (4) (5) (6) (7) (8) (9) Observt d X Std. error Calculated Calculated Calculated t Age Number Length" Std. Dev. of Mean Length Weight (g) Weight (g) Annulus (Years) of Clams (mm) Sx(mm) Sx(mm) (mm)" l-18-73c 7-2-73d 1 0.5 22 1.65 0.67 0.14 0.0009 0.0012 2 1.5 50 6.37 1.67 0.24 0.05 0.07 3 2.5 50 14.82 2.16 0.31 0.68 0.85 4 3.5 47 23.32 4.15 0.61 2.66 3.27 5 4.5 40 30.62 5.45 0.86 6.05 7.36 6 5.5 33 38.74 6.29 1.09 12.30 14.82 7 6.5 26 45.41 0.63 1.30 19.85 23.78 8 7.5 16 48.95 6.77 1.69 44.28 24.90 29.74 9 8.5 8 51.26 5.79 2.05 49.78 28.61 34.12 10 9.5 54.72 34.84 41.44 11 10.5 59.15 44.05 52.26 12 11.5 63.13 53.61 63.44 13 12.5 66.70 63.28 74.73 14 13.5 69.90 72.88 85.92 15 14.5 72.78 82.31 96.00 lb 15.5 75.36 91.43 107.49 17 16.5 77.68 100.18 117.65 18 17.5 79.76 108.40 127.28 19 18.5 81.62 116.30 136.33 20 19.5 83.30 123.66 144.85 21 20.5 84.81 130.55 152.81 22 21.5 86.16 136.01 160.17 23 22.5 87.37 142.79 166.06 24 23.5 88.46 148.23 173.24 25 24.5 89.43 153.18 178.96 26 25.5 90.31 157.77 184.26 27 2o.5 91.10 161.97 189.10 28 27.5 91.80 165.75 103.46 20 28.5 92.44 169.26 197.50 30 29.5 93.01 172.43 201.15 "Obtained from the 0 foot (0.0 m i tide level. Mil W. "Calculated trom the von Bertalantty growth equation using final trial !°° = 98 mm: r = 0.9O84. Calculated from Log Y —3.6982 + 3.0149 Log X; r = 0.0978. Values for ages 0.5 to 8.5 obtained from Column (4). The standard error of estimated Y, Sy' is given by the following equation: Sy = antilogvAO. 00002561 + (0.02788 log x)-, where x = departure from mean \K Mx = 1.0685- N = 51 'Calculated from Log Y = —3.5585 + 2.9779 Log X: r = 0.9960. Values for ages 0.5 to 8 5 obtained from Column (41. Sy = antilog ^0.00002682 + (0.04464 logx)2. where x = departure from mean Mx, Mx = 1.7165* N = 37 'From Ezekiel and Foz ( 1959) pp 287-288. 92 R. B. NICKERSON Age 4 4 3 2 55 45 - !5 - 1 It = 51.5 (i-e-0.2375 ( t- 1. 5251 j 1/ wt = 48 (l-e-0.1025(t-5.0556) ,:/ 1/ for age 2/ for age .5 3.5 I I I 15 25 35 45 Length (mm) at age t L°° = 51.5 mm W» = 48 g A_ 95 = 14 . 14 years M =0.2119 Critical sire = 53.71 mm I 1- lst differential of absolute ;rowth From Table 1 1st differential of absolute biomass From Table 1 16 FIG. 5. Growth of P. staminea from Simpson Bay, Prince William Sound, Alaska. cumulated. Spawning continues into September, and by early October heavy proliferation of folli- cle cells is evidenced in ovaries of some females to the extent that lumina of the ovaries are complete- ly filled and the germinal epithelium is studded with ovogonia. Both developing ovocytes and residue ova are observed through winter months, and by late May ovaries of P. staminea and S. giganteus are filled with ripening ova 40 to 60^ in diameter. Proliferation of spermatogonia is in- itiated by mid-March with layers of gonia to four deep; by mid-April primary and secondary sper- matogonia are layered 10 to 15 cells deep and dense columns of developing sex cells are filling the lumina of the spermaries. The data suggest that spawning follows the at- tainment of extreme low gonad pH (acid) levels (i.e., 6.1) as with Siliqua patula Dixon (Nickerson, 1975). Seawater temperatures were significantly warmer (P<.01) during 1974 than in 1973 for the months April through September except for July (P>.05). Similarly extreme low gonad pH levels were attained at an earlier date in 1974 than evidenced for 1973 samples. This, in turn, may be reflective of nutrition and abundance of food. Both highly significant (P<.01) and non- signifi- cant (P>.05) differences were observed for gonad pH levels between S. giganteus and P. staminea for the same sampling periods during spring and summer months of 1973 and 1974. Interpretation of these similarities and differences is difficult without additional data. Speculation on these phenomena may lead to the hypothesis that P. staminea and S. giganteus achieve two or more peaks of spawning activity (Quayle, 1942) during average warm years with fluctuating seawater temperatures, but achieve only one peak of spawning activity during warmer than average years with less fluctuating seawater temperatures. Based on the sectioned samples the smallest sex- ually mature P. staminea was a female (ova to 50 CLAMS IN PRINCE WILLIAM SOUND 93 /i in diameter), 13 mm in total valve length. The smallest mature S. giganteus was a female (ova to 45 u in diameter), 31 mm in total valve length. Generally, these sizes correspond to 3 years of age for P. staminea and 5 years of age for S. giganteus. Both individuals contained very few ripe ova. Within the major period of tissue collection (1973 and 1974) maximum seawater temperatures at the Simpson Bay site did not exceed 53.6° F. (12° C) (on-site thermograph) during 1973 (August 24), and probably did not exceed 56.53 ± 1.79° F. (13.63 ± 0.99 ° C) (based on a correlation between 1973 seawater temperatures at the Cor- dova tide station and those at Simpson Bay, i.e., Y = —154.4176 +121.4868 log 10X; r = 0.9990; Sy.x = 0.2518) during 1974 (August 30). The following average seawater temperatures obtained from on-site thermograph data are presented below: April = 40.84 ±0.93° F. (4.91 ± 0.52° C) n = 19 May =42.15 ± 1.08° F. (5.64 ± 0.60° C) n = 26 June = 47.09 ± 2.67° F. (8.38 ± 1.49° C) n = 29 July = 51.12 ± 1.22° F. (10.62 ± 0.68° C) n = 31 August = 52.74 ± 0.84° F. (11.52 ± 0.47° C) n = 22 Sept. = 51.30 ± 0.48° F. (10.72 ± 0.27° C) n = 27 Population estimation. Tables 8 and 9 show population estimates for P. staminea obtained by the stratified random sampling method (stratify- ing on beach surface area) and by using pro- babilities and an index tide level. Probabilities were obtained from Table 10. As shown, con- fidence intervals are much narrower using the former method, but sampling by the latter method takes much less time. Table 11 shows population estimates for S. giganteus by employing the stratified random sampling method. Table 12 pro- vides proportion of 5. giganteus by tide level for persons interested in using the probability-index tide level(s) method. Methods of harvest. Hand-dug clams lead to a comparatively more expensive food product which cannot compete favorably with clam pro- ducts derived by reliable mechanized equipment (Nickerson, 1975). Use of the hydraulic clam harvester showed that littleneck and butter clams 10 t r 10 3 0 50 7 0 Length (mm) at age t 10 - L°° = 98 mm w°° = 310 g A. 95 = 29.6740 years M =0.1010 Critical size = 61.19 mm 1st differential of absolute growth Table 2 data smoothed by: LnY = 1.4881+1.7153 LnX-0.9239 LnX2 o o o 1st differential of abso- lute biomass Table 2 data smoothed by: LnY = -2.6956 + -*, .l',t)7LnX-1.8728LnX2 4 8 10 12 14 16 18 20 Annul i FIG. 6. Growth ofS. giganteus from Simpson Bay, Prince William Sound, Alaska. 94 R. B. NICKERSON TABLE 3. Comparison of growth rates of P. staminea from the + 1 foot ( + 0.304S m) ami the +2 foot ( + 0.61 m) tide levels, Simpson Bay. Prince William Sound, Alaska. 1972. + 1 foot tide leve 1 + 2 foot tide level Mean Greatest Mean Greatest "t" Annulus Valve Length Variance n Valve Length Variance n Value 2 4.5306 1.2656 49 3.8000 1.2089 35 2.96 3 10.2643 7.3772 40 8.b286 7.1225 35 2.74 4 18.5622 16.7085 49 14.9000 11.1442 35 4. 3o 5 26.6292 15.7125 40 21.5857 11.9191 35 o.Op 6 32.1316 16.6578 49 27.0857 15.1687 35 5.69 7 35.8510 19.3160 40 31.5278 10.2703 3b 4.48 8 38.5250 23.3492 42 35.4143 19.0052 35 2.03 9 41.7429 20.4820 28 38.7069 18.5804 20 2.54 10 45.9688 6.1271 8 42.2778 18.9182 18 2.15 Conclusion: growth rate is significantly greater IP< .05) at the + 1 toot tide level than at the +2 foot tide level for above age classes tested. can be harvested more rapidly and with less effort than hand digging with standard clam fork. Tests revealed that approximately 2500 clams per hour could be harvested by one person with the hydraulic device at zones of maximum density i.e., +1 foot ( +0.30 m) tide level. However, only 400 clams per hour could be harvested at the +3 foot ( +0.91 m) tide level. Due to drainage of the trench excavated by the nozzles, transects at a shallow angle from one tide level to a lower one gave better results than transects maintained on one contour. Transects perpendicular to the waters' edge drained rapidly, but yielded less than those at a shallow angle at optimum density tide levels due to the narrow width of this optimum density zone. In areas where beach slope would permit a wider optimum density zone, transects perpendicular to the waters edge would probably be most favorable. The trench averaged about 4 inches (10 cm) deep and 20 inches (51 cm) wide. Water pressure ranged from 30 to 60 psi (14 to 27 kg/6.4 cm2) depending upon density of clams; breakage was negligable. Qams left behind fell prey to sea stars, Evasterias troschelii and Pycnopodia helianthoides. Areas that were disturbed by the hydraulic harvester were clearly visible one year following its use, but TABLE 4. Comparison of growth rates ofS. giganteus from the 0 foot (0.0 m) and the +1.5 foot ( + 0.46 m) tide levels, Simpson Bay, Prince William Sound, Alaska, 1972. 0 foot tide level + 1.5 foot tide leve 1 Mean Greatest M ean Greatest "t" Annulus Valve Length Variance n V alve Length Variance n Value 2 6.3690 2.8023 50 5.4575 1.9685 20 2.12 3 14.8150 4.6598 50 11.2375 4.9226 20 6.12 4 23.3223 17.2604 47 19.2425 7.8235 20 3.0t, 5 30.6225 29.6626 40 26.9450 7.7379 20 2.79 6 38.7424 39.5372 33 33.5175 10.4695 20 3.38 7 45.4077 43.9699 26 39.3650 19.2745 20 3.44 8 48.9469 45.7968 16 43.5350 26.4837 20 2.64 9 51.2563 33.5246 8 46.5528 30.4472 18 1.89 Conclusion: Growth rate is significantly greater (P < .05) at mean lower low water than at the +1.5 foot tide level for age classes 2 to 8. CLAMS IN PRINCE WILLIAM SOUND 95 nearly undetectable after two years. Maximum pump pressure (105 psi) (48kg/6cm2) at Observation Island cut through a dense mat of mussels in a few minutes. Underlying substrate was a mud-cobble, and continued high pressure caused P. staminea and S. giganteus to bounce off the cobbles causing heavy damage to valves. When pressure was reduced to about 30 psi (14 kg/6 cm2) damage was minimal. DISCUSSION Earlier I mentioned that a land-mass upheaval of approximately 5.5 feet (2 m) occurred in the Simpson Bay area as a result of the 1964 Good Fri- day earthquake. Two points of interest occur here in regard to the distribution of P. staminea and S. giganteus on the low tide terrace, the first being that if the uppermost tide levels mentioned where post-earthquake recruitment were found are con- sidered to be the normal uppermost habitable zones of P. staminea and S. giganteus, respective- ly, for this tidal regime their presence at abnormal zones is evidence of remarkable survival capacity in a very hostile environment. The second point is that the uppermost post-earthquake recruitment zone of P. staminea coincides with findings of Feder and Paul (1973) at Galena Bay described by Roys (1971) as the "normal zone" where tectonic deformation ranged from —1 to +2 feet (-0.3048 to 0.6096 m). Approximately 34.39 percent of P. staminea collected at the Simpson Bay site were > 33.71 TABLE 5. Relationship of whole weight (X) to shucked weight (Y) for Protothaca, Simpson Bay. Pri)ice William Sound, Alaska. Shucked weight (grams) Y = a0 + a, X Date Whole weight (grams) 7-24-72" 8-26-72" 1-18-73' 1 0.71 0.21 1.10 5 1.58 1.51 1.93 10 2.67 3.12 2.96 15 3.75 4.74 4.00 20 4.84 6.35 5.04 25 5.92 7.97 6.08 30 7.01 9.58 7.12 35 8.10 11.20 8.15 40 9.18 12.82 9.19 45 10.27 14.43 10.23 50 11.35 16.05 11.27 a b c a„ = 0.49580311 a„ = 0.10899515 a„ = 0.88784667 a, = 0.21712986 a, = 0.32309154 a. = 0.20759088 r = 0.9756 r = 0.9903 r = 0.86O7 Sy.x = 0.4171 Sy.x = 0.4067 Sy.x = 0.9374 S0 = 0.20962217 S0 = 0 10996553 S„ = 0.55066632 S, = 0.01152195 S, = 0.00812550 S, = 0 02355937 n = 20 n = 33 n = 27 *a0, a, = regression coefficients, r = Coefficient of correlation. Sy.x = Standard error of estimate of Y on X So = Standard error of the regression coefficient a„ S, = Standard error of the regression coefficient a, 96 R. B. NICKERSON TABLE 6. Relationship of whole weight (X2) and shucked weight (X3) to trimmed weight (Xt) for Saxi- domus, Simpson Bay, Prince William Sound, Alaska, September 24, 1972. Whole weight (g)* X2 Shucked weight (g)* X3 10 20 30 40 50 60 70 80 00 100 110 120 130 140 150 200 250 300 4.57 8.01 11.44 14.88 18.31 21.75 25.19 28.62 32.06 35.49 38.93 42.36 45.80 49.23 52.67 69.85 87.03 104.20 Trimmed weight (g)* X, 1.43 3.11 4.78 6.46 8.13 9.81 11.49 13.16 14.84 16.51 18.19 19.86 21.54 23.22 24.89 33.27 41.65 50.03 X, = —0.67172 + 0.03911 X2 + 0.37399 X3 SI. 23 =1.15g. R1.23 = 0.9941 n = 23 To determine X3 from X2 use: X, =1.13649 + 0.34356 X2 r =0.9938 SX3.X2 = 2.47 grams So =0.71843 S, = 0.00836 To determine X, from X2 use: X, = —0.24669 + 0.16759 X2 r =0.9934 SX,.X2 = 1.24 grams So =0.36171 S, = 0.00421 "S1.23 = Standard error of estimate for multiple linear regression. R1.23 = Coefficient of multiple correlation. r = Coefficient of correlation SX,.Xi, SX,.Xi = Standard error of estimate. So, S, = Standard error of regression coefficients (or the regression line Y = a0 + a,x. CLAMS IN PRINCE WILLIAM SOUND 97 TABLE 7. Relationship of whole weight (X2) and shucked weight (X,) to trimmed weight (X,) for Saxi- domus, Simpson Bay, Prince William Sound, Alaska, January 18, 1973, Whole weight (g)* X2 Shucked weight (g)* X, Trimmed weight (g) x, 1.36 2.79 4.22 5.65 7.08 8.51 9.94 11.38 12.81 14.24 15.67 17.10 18.53 19.9b 21.30 28.55 35.70 42.86 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 200 250 300 3.74 6.67 9.60 12.53 15.46 18.39 21.32 24.26 27.19 30.12 33.05 35.98 38.91 41.84 44.77 59.43 74.08 88.74 X, SI. 23 = —0.5992 — 0.04865 X2 + 0.65423 X, = 0.07 grams Rl.23 = 0.9961 n = 32 To determine X3 from X2 use: X, =0.8083 + 0.2931 X2; r = 0.9937 SX,.X2 =1.70 grams; S0 = 0.4472; S, = 0.0062 To determine X, from X; use: X, =—0.0705 + 0.1431 X2 R =0.9849 SX,.X2 = 1.24 grams 'S1.23 = Standard error ot estimate tor multiple linear regression Rl-23 = Coefficient of multiple correlation r = Coefficient of correlation SXi.Xj. SX,.X2 = Standard error of estimate. Si. S, = Standard errors of regression coefficient tor the regression line Y = a„ + a,\ mm (my recommendation for minimum legal size). Applying this figure to the population estimate (Table 8) and length-weight data (Tables 1 and 5) reveals that approximately 25,000 lit- tleneck clams were available for harvest. They would have a minimum whole weight of approx- imately 505 pounds (229 kg) and a minimum drained edible meat weight of 157 pounds (71 kg). Considering that about 19 percent of the P. staminea production came from the +1 foot tide level stratum (greatest density) containing an area of 341.6 ft.2 (31.73 m2) an acre of similar tide level habitat would produce approximately 1.9 tons (1725 kg) of edible meat and 4.2 tons (3825 kg) of shell. Running through the same exercise with S. giganteus (Tables 2,' 7 and 11) using 61.19 mm as the minimum legal size we find that approximately 98 R. B. NICKERSON TABLE 8. Population estimate of Protothaca (all sizes pooled) at the Simpson Bay site, Prince William Sound, Alaska, by the stratified random sampling method. Tide Level N„ W»2S» h (ft) (ft.2) n* y*. w* y* V Wi W.S.2 1 + 5 400.41 9 1 0.07 0.11 0.07 0.0049 0.0049 2 + 4.5 441.33 8 1 0.07 0.13 0.12 0.0098 0.0084 3 + 4 422.18 9 18 0.07 2.00 2.46 0.1722 0.1722 4 + 3.5 450.26 8 10 0.08 1.25 1.93 0.2059 0.1544 5 + 3 553.28 9 43 0.09 4.78 14.90 1.7241 1.3410 6 + 2.5 549.10 8 59 0.09 7.38 22.56 3.0456 2.0304 7 + 2 510.40 0 184 0.09 20.44 179.83 20.8089 16.1847 8 + 1.5 441.00 8 290 0.07 36.25 500.86 40.9036 35.0602 9 + 1 341.60 9 361 0.06 40.11 473.93 24.3735 28.4358 10 + 0.5 285.09 8 240 0.05 30.00 67.40 2.8083 3.3700 11 0.0 320.11 9 121 0.05 13.44 23.23 0.8296 1.1615 12 —0.5 342.44 8 44 0.06 5.50 7.45 0.4470 0.4470 13 —1 294.84 8 51 0.05 6.38 13.99 0.5829 0.6995 14 — 1.5 242.50 8 47 0.04 5.88 26.73 0.7128 1.0692 15 —2 196.65 8 79 0.03 9.88 13.25 0.1988 0.3975 16 -2.5 207.50 8 76 0.03 9.50 27.98 0.4197 0.8394 N = 5998.69 n = 134 1.00 97.2476 y„ = 12.1278 clams per 1 ft.2 (0.0929 m2) for entire beach. S2(y„) = rhr (97.2476) — s^in^ (91. 3761 )= 0.7105 S(y„) = 0.8429 Estimated total number of clams = Ny„ = 72,751 Standard error of estimate = NS(y„) = 5056 NOTATION h =Stratum number. N» =Number or units in stratum. n* = Number of units in sample. yk, =Number of clams obtained in the ith unit. Wfc =i!Ji = Stratum weight. N "y* =Stratum mean. Sh' = Estimated variance. w,, =n = Sample weight - y„ =Mean number of clams for entire beach S!fy,,) = Variance of y„. S(y.,) =Standard error ofy.,. 91.3761 14.43 percent, or 3043 clams, were available for harvest. These would have had a minimum whole weight of approximately 327 pounds (148 kg) and a minimum trimmed weight of approximately 46 pounds (21 kg). Regression estimates indicate that the — 0.5 ft. ( — 0.15 m) tide level stratum contains the greatest densities, i.e., approximately 16 per- cent of the S. giganteus production. Propor- tionalizing again we find that an acre of similar tide level habitat would produce a minimum of approximately 0.47 tons (428 kg) of edible meat (trimmed weight), 0.56 tons (508 kg) of waste (gurry) and 2.30 tons (2087 kg) of shell. This trend of analysis leads to potential clam farming operations in Alaska by private aqua- culture groups. By creating huge pads (i.e., acre size) of suitable substrate, the top of which coin- cides with the optimum productive tide zone, CLAMS IN PRINCE WILLIAM SOUND 99 TABLE 9. Population estimate of Protothaca (all sizes pooled) at the Simpson Bay site, Prince William Sound, Alaska, utilizing probabilities from an index tide level ( + 1 foot)" relative to mean lower low water. Number of Clams Mean Standard deviation Standard error of mean per 1ft.2 40.1333 21 .7718 7.2573 P, = 0.1899 (1) (2) (3) Area ft.2 (4) (5) (6) Standard Strata Tide Level (ft.) Ai Pi Total Estimate deviation 1 + 5 400.41 0.0004 33.85 18.36 2 + 4.5 441.33 0.0007 65.29 35.42 3 + 4 422.18 0.0035 312.28 169.41 4 + 3.5 450.26 0.0133 1265.60 686.57 5 + 3 553.28 0.0333 3893.76 2112.31 0 + 2.5 549.10 0.0587 6811.92 3695.38 7 + 2 510.40 0.0969 10452.36 5670.27 8 + 1.5 441.00 0.1715 15983.90 8671.06 0 + 1 341.60 0.1899 13709.54 7437.25 10 + 0.5 285.09 0.1419 8549.57 4638.03 11 0 320.11 0.0728 4925.05 2671.78 12 —0.5 342.44 0.0612 4429.10 2402.73 13 — 1 294.84 0.0506 3152.95 1710.43 14 — 1.5 242.50 0.0417 2137.11 1159.30 15 -2 196.65 0.0346 1437.97 780.08 16 —2.5 207.50 0.0289 1267.35 687.52 5998.69 78427.59 ±14,867.58 (1) = Strata: (2) = Tide level, feet; (3* = Area. Ai, ft.2; (4) = Probability of disti ibution at the i'* tide level , Pi ^ P. A, . aA( PA, \2 — — -V * Vv "For conversion: 1 ft.2 = 0.0929 m- lft. = 0.3048 m many pads in a small area would reduce time and expense of routine sampling for paralytic shellfish poison and sanitary surveys. Several of these pads would be required for each species and a crop rotation plan would be adopted. Approximately 6 years would be required for P. staminea to achieve harvestable size; for S. giganteus approximately 11 years would be required to achieve harvestable size (i.e. realize the maximum benefit of body weight increase). Spat culture stations could form a branch of the aquaculture program, providing viable, robust seed stock for the "farms". ACKNOWLEDGEMENTS I would like to thank Mr. Robert Bynum, Fishery Technician, for field assistance; Mr. Tim Brown, Fish Culturist, for laboratory assistance; Mrs. Janice Shaw, Clerk Typist, for assisting in preparation of the text, figures and tables, and Mr. John McMullen, Dr. Bernie Kepshire, Dr. Roger Grischkowsky, and Mr. Ivan Frohne, for their editorial comments. 100 R. B. NICKERSON TABLE 10. Regression estimates of Protothaca frequency of occurrence by tide level on the low tide ter- race derived from a gamma distribution (lxy* = 0.9758; Sy.f(x) = 3.84) fitted to Simpson Bay data. Prince William Sound. Alaska. (1) Tide levels relative to mean lower low water Feet Meters (3) Density in clams per ft.2 Relative clam density (0.3048 m2) expressed in proportion 0.0877 0.0004 0.1441 0.0007 0.7418 0.0035 2.8166 0.0133 7.0287 0.0333 12.4054 0.0587 20.4870 0.0969 36.2482 0.1715 40.1315 0.1899 29.9981 0.1419 15.3802 0.0728 12.9365 0.0612 10.7042 0.0506 8.8201 0.0417 7.302Q 0.0346 6.1166 0.0289 + 5.0 + 4.5 + 4.0 + 3.5 + 3.0 + 2.5 + 2.0 + 1.5 + 1.0 + 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 1.52 1.37 1.22 1.07 0.91 0.7b 0.61 0.4b 0.30 0.15 0.00 -0.15 -0.30 -0.4b -0.61 -0.76 SUMS 211.3496 1.0000 'From Ezekiel and Fox ( 1959). p. 128 — index of correlation for curvilinear relations LITERATURE CITED Bourne, N. 1967. Digging efficiency trials with a hydraulic clam rake. Fish. Res. Bd. Can. Tech. report No. 151 22p. Cochran, W. G. 1963. Sampling techniques. John Wiley and Sons, Inc. New York. p. 87-113. Ezekiel, M. and K. Fox. 1959. Methods of Correla- tion and Regression analysis. John Wiley and Sons, Inc. p. 128, 288. Feder, H. M. and A. J. Paul. 1973. Abundance estimations and growth-rate comparisons for the clam Protothaca staminea from three beaches in Prince William Sound, Alaska with additional comments on size-weight relation- ships, harvesting and marketing. U. of Alaska, Inst. Marine Sci. Tech. Report No. R73-3. 34 p. Fraser, C. M. and G. M. Smith. 1928. Notes on the ecology of the littleneck clam, Paphia staminea Conrad. Trans. R. Soc. Can., Ser. 3, 22, Sect. V:249-269. Fraser, C. M. and G. M. Smith. 1928. Notes on the ecology of the butter clam, Saxidomus giganteus Deshayes. Trans. R. Soc. Can., Ser. 3, 22, Sect. V:271-284. Nickerson, R. B. 1975. A critical analysis of some razor clam (Siliqua patula Dixon) populations in Alaska. Alaska Dept. Fish & Game. 294 p. Paul A. J. and H. M. Feder. 1973. Growth, recruitment, and distribution of the littleneck clam, Protothaca staminea, in Galena Bay, Prince William Sound, Alaska. Fish. Bull. 71-3:665-677. Quayle, D. B. 1942. Sex, gonad development and seasonal gonad changes in Paphia staminea Conrad. J. Fish. Res. Bd. Can. 6(2):140-151. Quayle, D. B. 1969. Paralytic shellfish poisoning in British Columbia. Fish Res. Bd. Can., Bull. 168. 68 p. Quayle, D. B. 1974. The intertidal bivalves of British Columbia. British Columbia Provincial Museum. Handbook No. 17. 104 pp. CLAMS IN PRINCE WILLIAM SOUND 101 TABLE 11. Population estimate of Saxidomus (all sizes pooled) at the Simpson Bay site, Prince William Sound, Alaska by the stratified random sampling method. Tide Level W,2 S*J h (ft.) N„ ru y- w* y* S„2 wh W.S.2 1 + 3.5 450.26 8 1 0.10 0.13 0.12 0.0171 0.0120 2 + 3 553.28 Q 0 0.12 0.00 0.00 0.0000 0.0000 3 + 2.5 549.10 8 4 0.12 0.50 0.86 0.176Q 0.1032 4 + 2 510.40 0 12 0.11 1.33 2.37 0.3585 0.2607 5 + 1.5 441.00 8 25 0.09 3.13 8.41 0.9732 0.7569 6 + 1 341.60 9 44 0.07 4.89 23.43 1.4351 1.6401 7 + 0.5 285. 0Q 8 b3 0.06 7.88 33.87 1.7419 2.0322 8 0 320.11 0 56 0.07 6.22 14.75 0.Q034 1.0325 9 -0.5 342.44 8 33 0.07 4.13 1.85 0.1295 0.1295 10 — 1 294.84 8 5b 0.0b 7.00 7.73 0.3975 0.4638 11 —1.5 242.50 8 37 0.05 4.63 9.b7 0.3454 0.4835 12 — 2 10b. o5 8 40 0.04 5.00 8.58 0.1961 0.3432 13 -2.5 207.50 8 27 0.04 3.38 4.00 0.0914 0.1600 N = 4734.77 n = 108 1.00 6.7660 7.4176 Y„ = 3.6870 clams per 1 ft.2 (0.0929 rrr) for entire beach. S2(yJ = 7M (6.7660) - ttxtt? (7.4176) = 0.0611. S(y„) = 0.2471 Estimated total number of clams = N\\, = 17,457. Standard error of estimate = (NS(y„) = 1170. NOTATION h = Stratum number N, =Number of units in stratum nh =Number of units in sample y». = Number of clams obtained in the ith unit N, W» =. N Stratum weight yi, = Stratum mean Sh2 = Estimated variance wh =ni— Sample weight. "y„ = Mean number of clams tor entire beach. S'ty".,) = Variance of"y., S(y„) = Standard error of y^, Reimnitz, E. 1966. Earthquakes and the Copper River Delta, p. 127. In E. Reimnitz, late quater- nary history and sedimentation of the Copper River delta and vicinity, Alaska. Ph. D. Thesis. Univ. of Cal., San Diego. Ricker, W. E. 1958. Handbook of computations for biological statistics of fish populations. Fish. Res. Bd. Can. Bull. 119, 300 p. Roys, R. S. 1971. Effects of tectonic deformation on pink salmon runs in Prince William Sound. p220-237. In The great Alaska earthquake of 1964. Biology. Nat. Acad, of Sciences, Wash. D.C. Taylor, C. C. 1958. Temperature and growth — the Pacific razor clam. Bur. Comm. Fish. Lab. U. S. Fish and Wildlife Service, Woods Hole, Mass. p93-101. 102 R. B. NICKERSON TABLE 12. Regression estimates of Saxidomus frequency of occurrence by tide level on the low tide ter- race derived from a cubic function (ixy* = 0.97o5; Sy.f(x) = 1.16) fitted to Simpson Bay data Prince William Sound, Alaska. (1) (2) (3) Tidel evels : relative Density in clams to mean lower low water per ft.: Relative clam density Feet Meters (0.3048m2) expressed in proportion + 3.5 1.07 0.0946 0.0019 + 3.0 0.01 0.1O5O 0.0040 + 2.5 0.7o 0.4808 0.0098 + 2.0 0.61 1.1266 0.0230 + 1.5 0.4o 2.30O7 0.0471 + 1.0 0.30 4.0155 0.0820 + 0.5 0.15 5.8841 0.1201 0.0 0.00 7.306O 0.14O1 -0.5 —0.15 7.760Q 0.158d -1.0 —0.30 7.1610 0.14o2 -1.5 -0.4b 5.7940 0.1183 -2.0 — O.ol 4.1634 0.0850 —2.5 -0.76 2.6870 0.0548 SUMS 48.9903 1.0000 'From Ezekiel and Fox 1 1Q5<>>. THE RELATION OF SHELL LENGTH TO TOTAL WEIGHT, TISSUE WEIGHT, EDIBLE-MEAT-WEIGHT, AND REPRODUCTIVE ORGAN WEIGHT OF THE GASTROPODS NEPTUNEA HEROS, N. LYRATA, N. PRIBILOFFENSIS, AND N. VENTRICOSA OF THE EASTERN BERING SEA Richard A. Macintosh and A. ]. Paul NATIONAL MARINE FISHERIES SERVICE NORTHWEST AND ALASKA FISHERIES CENTER 2725 MONTLAKE BOULEVARD EAST SEATTLE, WASHINGTON 98112 INSTITUTE OF MARINE SCIENCE UNIVERSITY OF ALASKA SEWARD, ALASKA 99664 ABSTRACT The relations of shell length to total weight, tissue weight, edible-meat-weight, female-gonad-weight and male-penis-weight were determined for 214 Neptunea heros, 179 N. lyrata, 186 N. pribiloffensis, and 197 N. ventricosa from the eastern Bering Sea. Equations describing these relationships are presented. Edible-meat-weight was found to average 27.9% forN. heros, 30.5% forN. lyrata, 30.6% for N. pribiloffensis, and 27.9% for N. ventricosa. The plots of shell length vs. female-gonad-weight and male-penis- weight exhibit sudden increases in slope and become near vertical lines thus indicating a generalized size at which maturity is approached. In female N. heros, N. lyrata, N. pribiloffensis, and N. ventricosa this increase occurs at 110, 110, 105, and 102 mm respectively. Males of these species undergo similar increases at 95, 100, 90, and 87 mm respectively. INTRODUCTION Four species of large Neptunea (Gastropoda, Prosobranchia, see Figure 1), N. heros (Gray, 1850), N. lyrata (Gmelin, 1791), N. pribiloffensis (Dall, 1919), and N. ventricosa (Gmelin, 1791), are commonly encountered in the eastern Bering Sea (Fig. 2). These gastropods are currently harvested or show considerable potential for harvest in the eastern Bering Sea, and have been described by Macintosh1. 1 Macintosh, Richard A. 1976. A guide to the identification of some common eastern Bering Sea snails. Unpub. manuscr. Northwest and Alaska Fisheries Center, Natl. Mar. Fish. Serv., NOAA, Kodiak, AK 99615. Japan has commercially harvested snails, primarily Neptunea pribiloffensis, in the eastern Bering Sea since 1971. The fishery occurs east of 175° west longitude along the continental shelf around and northwest of the Pribilof Islands. As many as 28 vessels may be involved in the fishery (Kiyoshi Yoshihara, Dept. of Fisheries, Nihon University, 3-34-1 Shimoua Setagaya, Tokyo, 154 Japan, personal communication). However, Na- tional Marine Fisheries Service (NMFS) patrols in the eastern Bering Sea observed only 14, 5, 1, and 6 vessels fishing snails in the years 1971 through 1974, and no vessels in 1975 and 1976 respectively (James Branson, NMFS Law Enforce- ment Division, Box 1036, Kodiak, AK 99615, 103 104 R. A. MACINTOSH AND A. J. PAUL CM 1 CM 1 r« CM CM 1 FIG. 1. The four large Neptunea from the eastern Bering Sea, clockwise from upper left, Neptunea heros, N. lyrata, N. pribiloffensis, andN. ventricosa. personal communication). Catches from 1972 to 1975 have ranged from 3,000 to 3,574 metric tons (average 3,277 mt) of edible meat (data ob- tained by U.S. Embassy, Tokyo, Japan from the Japan Fishery Agency; provided to authors in June, 1976). Fishing vessels licensed to engage in the Japanese eastern Bering Sea snail fishery range in size from 96 to 490 gross tons, or about 25 to 50 meters in length (U. S. Embassy, Tokyo, Japan). Some vessels in the snail fleet operate independently while others fish for a factoryship. Processing consists of crushing the shells, briefly cooking the meats and removing the soft parts and shell fragments. The meats are graded by size and quality and quick frozen in trays. The small snails may be frozen whole (James Branson, person- al communication). The gear used in the fishery consists of conical pots which are similar in shape to those used in Japan's snow (Tanner) crab, Chionoecetes bairdi and C. opilio. fishery in the eastern Bering Sea. These pots are about 880 mm in diameter across the bottom, 450 mm across the top, and 363 mm in height. The diameter of the tunnel in the top of the pot varies from 120 to 150 mm. The webbing on the side of the pot changes from 60 mm mesh over the first 170 mm from the base, to 121 mm mesh on the remainder of the side (James Branson, pers. comm.). Some vessels in the fishery utilize 12 sets of pot gear, each set consisting of 500 pots on GASTROPODS IN BERING SEA 105 a common groundline. Four sets of pots, baited with fish, are picked and set every day making the average soak time three days (Nagai, 1975). Official figures on the total value of the fishery are not available; however, fishermen are reported to receive between one and two U. S. dollars per kg (Kiyoshi Yoshihara, personal com- munication) depending on product quality. Assuming an average value of $1.50 per kg, the average annual dockside value of the fishery is ap- proximately 4.9 million dollars. No data are available concerning the retail value of the catch; however, the meats are considered an expensive luxury item in Japan. The most common gastropod in the Japanese fishing area is Neptunea pribiloffensis, which in 1973 comprised about 70% of the total harvest by weight (Nagai, 1974). The other three species of Neptunea do not occur in the fishing area in large numbers (Nagai, 1974), but occur in dense ag- gregations in other parts of the Bering Sea (Kaim- mer et. al., 1976). Buccinum angulossum (Gray, 1839) and B. tenue (Gray, 1839) combined, ac- counted for an additional 20% of the 1973 catch. Descriptions of the egg capsules of Neptunea lyrata and N. pribiloffensis are available (Golikov, 1961; Ito, 1957). Similar data for the other two Neptunea are not available. No other literature concerning the basic biology of these gastropods has been found by the authors. The purpose of this investigation was to ex- amine the length vs. total weight, length vs. tissue weight and length vs. meat weight relationships of the large Neptunids which have actual or potential value to the eastern Bering Sea snail fishery. The relationships of shell length to female-gonad- weight and penis-weight were examined to deter- mine generalized sizes at sexual maturity. FIG. 2. Distribution of four large Neptunea in the eastern Bering Sea, clockwise from upper left, Nep- tunea heros, N. lyrata, N. pribiloffensis, N. ventricosa. 106 R. A. MACINTOSH AND A. J. PAUL TABLE 1. Locations and dates of collectionsof samples of Neptunea heros, N. lyrata, N. pribiloffensis, N. ventricosa, examined. (Locations of the sampling stations shown in Figure 2). Date of Sampling Location N. N. N. N. Collection Stations Latitude Longitude heros lyrata pribiloffensis ventricosa 11 8-20-75 L-24 58°39' 172 °17' — 8-23-75 K-23 58°10' 171 °52' — 8-24-75 J-22 57°51' 171 °06' — 8-24-75 H-24 57°12' 172 °13' — 8-24-75 O-01 59°40' 167°59' — 8-24-75 0-18 59°40' 168 °38' — 8-27-75 H-22 57°20' 170 °40' — 8-27-75 N-01 59°20' 167°26' — 8-27-75 N-02 59°20' 167°16' — 8-27-75 N-18 59°20' 168 °32' — 8-27-75 N-19 59°20' 169 °10' — 8-28-75 J-01 58°01' 167°43' 14 8-28-75 M-01 59°00' 167°54' — 8-28-75 M-18 59°00' 168 °34' — 8-28-75 M-19 59°00' 169 °09' 19 8-28-75 M-20 59°01' 169 °45' 36 8-28-75 N-20 59°20' 169°52' — 8-29-75 L-18 58°40' 168 °24' 10 8-29-75 L-19 58°40' 169 °05' 14 8-29-75 L-20 58°40' 169 °42' 10 8-30-75 K-01 58°20' 167°50' — 8-30-75 K-18 58°20' 168 °27' 14 8-30-75 K-19 58°20' 169 °05' — 8-31-75 F-18 56°40' 168°18' — 8-31-75 H-01 57°20' 167°41' — 8-31-75 1-01 57°41' 167°45' 30 8-31-75 K-18 57°41 ' 168 °22' — 8-31-75 1-18 58°00' 168 °23' — 9-02-75 C-04 55 30' 166 "04 ' — 9-16-75 D-07 56°00' 164 '01 ' — 9-17-75 A-03 54 c59 ' lt>6:16' — 9-18-75 B-03 55 14' 166 39 ' — 9-24-75 E-06 56 '20' 164 °35' — 9-24-75 E-07 56°11' 164^16' — 0-24-75 F-06 56°40' 164 35 ' — 10-00-75 1-11 58 00' 161 29 ' — 10-06-75 1-12 57 50 lol 08 — 10-0b-75 K-13 58 09' 160 09' — 10-13-75 1-02 58 00 167 08' 32 10-14-75 1-03 58 01 ' 166 30' — 10-14-75 J-04 58 00 165 51 — 13 — 21 — 14 — 22 — — 20 — 8 5 — — 32 — 3 — 13 — 10 — 2 — 4 — 4 — 5 78 5 — 3 — 3 1 2 33 10 7 10 2 — 1 14 8 26 21 3 5 1 14 39 28 15 5 2 12 — — 7 11 — 35 — — 4_ Total 214 179 186 197 GASTROPODS IN BERING SEA 107 METHODS The size-weight relationships of 214 Neptunea heros, 179 N. lyrata, 186 N. pribiloffensis, and 197 N. uentricosa were examined. The specimens were collected in the course of the 1975 NMFS synoptic trawl survey of the eastern Bering Sea shelf. Location and dates of collection are given in Table 1. All collections were made with an Eastern otter trawl constructed with 10.2 cm mesh on the wings and body and 8.8 cm mesh in the intermediate sec- tion of the codend. The codend was lined with 3.2 cm mesh web. Very few snails less than 4.0 cm in length are retained by this gear. Specimens were taken from random samples collected and preserved in 10% formalin aboard the vessel. In the laboratory, individual shell lengths (the distance from the apex of the spire to the end of the siphonal canal) were determined. Each shell was cleaned of all encrusting material and total weight recorded to the nearest gram. The animals were removed from their shells and total tissue, female gonad and accompanying connec- tive tissue, penis, and edible meat were weighed to the nearest tenth of a gram. Digestive organs, respiratory tissue, gonad and opercula were removed from each specimen to obtain edible- meat-weight. Weights were taken with a precision balance, and plots, regression lines and regression equations were determined and plotted by com- puter. The Gauss-Jordan method was used in the solution of all normal equations (Cooley and Lohnes, 1962; Ostle, 1954). RESULTS The equations describing the relationships be- tween shell length and total weight, tissue weight, edible-meat-weight, female-gonad-weight and penis-weight are presented in Table 2 (also see TABLE 2. Size-weight relationships of four species of Neptunea from the eastern Bering Sea. See Figs. 3 to 18 N. heros N. lyrata N. pribiloffensis N. ventricosa Total Wt., g = Tissue Wt., g = Edible Meat Wt., g = Female gonad Wt., g = Penis Wt., g = 0.0347 L"363557 0.0256 L'85089 0.0225 L39'420 0.0098 L12""" 0.0101 L'272427 0.0411 L309406 0.0267 L35'768 0.0238 L3 S99M 0.0098 L9284'9 0.0095 L'° ,389 0.0381 L'-"344 0.02b6L'9923J 0.0241 L3 83347 0.0109 L"124" 0.0102 L'063264 0.0439 L32232' 0.0298 L37'237 0.0253 L3 8073S 0.0113 L'490757 0.0115 L2802691 "L= total shell length (mm) Figs. 3 to 18). Formalin preserved edible-meat- weight was found to average 26.8% (± 4.5%), 30.5% (±6.1%), 30.6% (±5.2%), and 27.9% (±4.4%) for Neptunea heros, N. lyrata, N. pribiloffensis, and N. ventricosa respectively. In all of the diagrams of shell length vs. female gonad or penis weight some scattering among points representing the larger specimens is ap- parent (Figs. 11 to 18); however, a generalized size at which sexual maturity is approached can be determined by the curves. Female Neptunea heros N. lyrata, N. pribiloffensis, and N. ventricosa ex- hibit sudden increases in gonad-weight when their shell lengths are approximately 110, 110, 105, and 102 mm respectively. Males of all four species display similar increases in the relationship of penis-weight to shell length. These increases occur for males at approximately 95, 100, 90, and 87 mm for N. heros, N. lyrata, N. pribiloffensis, and N. ventricosa respectively. 108 R. A. MACINTOSH AND A. J. PAUL Neptunea hetos 45C 315 1 £ IOC 22S 150 75 1—4"^ TOTAL SmELL LENGTH IMMI FIG. total from 3. The relationship of total shell length to weight and tissue-weight for Neptimea heros the eastern Bering Sea. 20 HO 60 80 100 120 140 160 TOTAL SHELL LENGTH IMMI FIG. 5. The relationship of total shell length to total weight and tissue-weight for Neptunea pribiloffensis from the eastern Bering Sea. FIGA.The relationship of total shell length to total weight and tissue-weight for Neptunea lyrata from the eastern Bering Sea. Nepiuneo venfncoso 12D mo 160 TOTAL 5HELL LENGTH IMHI FIG. 6. The relationship of total shell length to total weight and tissue-weight for Neptunea ven- tricosa from the eastern Bering Sea. DISCUSSION Rae Baxter (Alaska Dep. Fish and Game, Bethel, AK: personal communication), working with Neptunea lyrata from an isolated Bristol Bay population, reported an edible meat recovery of 22% for specimens with an average length of 55 mm. Individuals of this small size were unavail- able for this study. No other data is available con- cerning the percentage of recoverable meats for in- dividual species of eastern Bering Sea gastropods; however, Japan Fishery Agency data (U. S. Em- bassy, Tokyo, Japan) for both total weight and recovered meat weight of the 1974 harvest in- dicates an edible meat recovery of 27% . Although this is an average of the recoveries of ten or more GASTROPODS IN BERING SEA 109 species, the value compares favorably with those generated for the four Neptunea examined. Possi- ble differences between formalin preserved edible- meat-weights used in our study and the fresh weights that Japanese data are based upon have not been critically examined. Females of all four species of Neptunea examin- ed appear to approach sexual maturity at shell lengths of 10 to 15 mm larger than do males of the same species (Figs. 11 to 18). Similar observations have been made for N, antigua (L.) from Danish waters (Pearce and Thorson, 1967). Female Nep- ISO Neptunea heros 125 l I OC I 75 J 50 \M: 25 n 60 BO 100 ISO TOTAL SHELL LENGTH IMM1 FIG. 7. The relationship between total shell length and edible-meat-weight for Neptunea heros from the eastern Bering Sea. 100 120 TOTAL SHELL LENGTH 1HH1 FIG. 8. The relationship between total shell length and edible-meat-weight for Neptunea lyrata from the eastern Bering Sea. i- - > Neptunea pribitoffensts • ./ ' /: " so J:<:. H0 ;0k- ' 30 '*y.'1. 20 ( . . 10 n TOTAL SHELL LENGTH (MM! FIG. 9. The relationship between total shell length and edible-meat-weight for Neptunea pribiloffen- sis from the eastern Bering Sea. Neptunea ventncoso : si. ■ ■ 0 20 10 60 B0 100 120 140 160 TOTAL SHELL LENGTH 1MH! FIG. 10. The relationship between total shell length and edible-meat-weight for Neptunea ven- tricosa from the eastern Bering Sea. 110 R. A. MACINTOSH AND A.J. PAUL .- 15 5 Neptunea heros 1(0 60 80 100 TDlflL SMELL LENGTH IHM1 FIG. 11. The relationship between total shell length and female-gonad-weight for Neptunea heros in the eastern Bering Sea. Neptunea lyroto 60 80 100 TOTAL SHELL LENGTH IMM1 FIG. 12. The relationship between total shell length and female-gonad-weight for Neptunea Iyrata in the eastern Bering Sea. Neptunea pr.piloffensis 1)0 60 80 100 !20 140 160 TOTAL SHELL LENGTH (MHt Neptunea ventncoso 1)0 60 80 100 120 1D0 160 TOTAL SHELL LENGTH IMM1 FIG. 13. The relationship between total shell FIG. 14. The relationship between total shell length length and female-gonad-weight for Neptunea and female-gonad-weight for Neptunea ventricosa pribiloffensis in the eastern Bering Sea. in the eastern Bering Sea. GASTROPODS IN BERING SEA 111 tunea also tend to reach larger sizes than males of the same species (Pearce and Thorson, 1967; Nagai, 1974; Kaimmer et. a\., 1976). Nagai (1974) reported the average live weights of Neptunea pribiloffensis captured in the pot fishery to be 107 g for females and 92.5 g for males. Female N. pribiloffensis of this weight would have approximate shell lengths of 103 mm and males 99 mm (Fig. 5). Individuals of this size may be just approaching maturity (Figs. 13, 17) and may not have spawned before capture. Neptunea heros 0 JO 10 60 80 100 120 110 160 I BO T0TM. SHELL LENGTH IHM1 FIG. 15. The relationship between total shell length and penis-weight for Neptunea heros in the eastern Bering Sea. Neptunea l/rato 60 80 100 120 TOTfiL SHELL LENGTH IHH] FIG. 16. The relationship between total shell length and penis-weight for Neptunea lyrata in the eastern Bering Sea. Neptunea pribiloffensis 60 80 100 TOTAL SMELL LENCTH IHH1 FIG. 17. The relationship between total shell length and penis-weight for Neptunea pribiloffen- sis in the eastern Bering Sea. Neptunea vent- 60 60 100 TOTAL SHELL LENCTH IHHI FIG. 18. The relationship between total shell length and penis-weight for Neptunea ventricosa in the eastern Bering Sea. 112 R. A. MACINTOSH AND A. J. PAUL ACKNOWLEDGMENTS We would like to thank the following IMS and NMFS staff members: Carol Bennie, LoHama Schaeffer, and Marilyn Buker for typing and editing; Rosemary Hobson for computer program- ming assistance; Howard Feder for computer time under Grant 04-5-158-41; Judy Paul and Alan Spalinger for general assistance. LITERATURE CITED Cooley, W. W. and P. R. Lohnes. 1962. Multivariate Procedures for the Behavioral Sciences. John Wiley and Sons, New York. 211 P- Golikov, A. N. 1961. Ecology of reproduction and the nature of egg capsules in some gastropod molluscs of the genus Neptunea (Bolten). (Transl. from Russian). Zool. Zhurnal 407(7):997-1009. Ito, K. 1957. A few observations of the spawning habits of Neptunea intersculpta. Yume-hama- . gun. 89:11-16. (Japanese). Kaimmer, S.M., J. R. Reeves, D. R. Gunderson, G. B. Smith, and R. A. Macintosh. 1976. Baseline information from the 1975 OCSEAP survey of the demersal fauna of the eastern Ber- ing Sea, Sect. IX, p. 157-368. In Demersal fish and shellfish resources of the eastern Bering Sea in the baseline year 1975. NOAA, NMPS, NWFC, Seattle, Wa., 619 p. Nagai, T. 1974. Studies on the marine snail resources in the eastern Bering Sea. I. Species composition, sex ratio, and shell length com- position of snails in the commercial catch by snail-basket-gear in adjacent waters of Pribilof Islands, 1973. (Transl. from Japanese). Far Seas Fish. Res. Lab. Bull. 10:141-156. Nagai, T. 1975. On the variation of catch per unit effort. An analysis of the snail fishing data in the eastern Bering Sea. I. Far Seas Fish. Res. Lab. Bull. 12:121-135. Ostle, R. 1954. Statistics in Research. The Iowa State Univ. Press, Ames, Iowa, 487 p. Pearce, J. B. and G Thorson. 1967. The feeding and reproductive biology of the red whelk, Neptunea antigua (L.) (Gastropoda, Prosobran- chia). Ophelia 4:277-314. Proceedings of the National Shellfisheries Association Volume 67 — 1977 SEA ANEMONE PREDATION ON LARVAL OYSTERS IN CHESAPEAKE BAY (MARYLAND) Clyde L. MacKenzie, Jr. U.S. DEPARTMENT OF COMMERCE NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION MIDDLE ATLANTIC COASTAL FISHERIES CENTER SANDY HOOK LABORATORY HIGHLANDS, NEW JERSEY 07732 ABSTRACT Diadumene leucolena, a sea anemone, is indicated as a biological controlling feature of oyster populations throughout the Maryland portion of Chesapeake Bay, where significant predators had not been reported. A survey showed that D. leucolena den- sities (no./m1) averaged 103.5 on beds of live oysters and 176.2 on shell beds. D. leucolena was observed to be a voracious predator of oyster larvae. The findings sug- gest that D. leucolena substantially limits abundance of the oyster. INTRODUCTION Diadumene leucolena, sea anemone, has widespread distribution in Maryland portion of Chesapeake Bay, yet little is known of its ecological significance, especially its possible role as a predator of pelagic larvae of mollusks such as the oyster (Crassostrea virginica). According to Thorson (1950), coelenterates prey on animal lar- vae in Europe. Orton (1922) states that a scypho- zoan (Aurelia aurita) ingests European oyster (Ostrea edulis) larvae. Anthozoan feeding on in- vertebrate larvae has not been examined in North America. However, it may be a significant limiting feature of many invertebrate populations, including commercial mollusks. Oysters are widely distributed in Chesapeake Bay. They grow on beds of various sizes and con- centrations in most rivers, creeks, embayments, and straits on both the western and eastern shores of Chesapeake Bay, extending northward to about Baltimore and southward to the Virginia Capes (Lippson, 1973). Maryland is believed to be essen- tially free of predators that control or limit the oyster populations. The most serious oyster predators of eastern North America, i.e., oyster drills (Urosalpinx cinerea and Eupleura caudata) and starfish (Asterias forbesi) are scarce or absent because salinities below 15 parts per thousand are too low for their existence. The Maryland portion of the bay shows salinities below this point for much of the year. The blue crab (Callinectes sapidus) is present, but is not a significant oyster predator (Van Engel, 1958). The flatworm (Stylochus ellipticus) is also present, but is believ- ed to be a minor oyster predator (Webster and Medford, 1961; Christensen, 1973). Mud crabs (family Xanthidae), that have been shown by MacKenzie (1970, a) to prey on oyster spat, are abundant in Maryland, and may also be minor oyster predators. Oysters have supported a substantial commercial fishery in Maryland since the mid-1880's (Engle, 1956; Lyles, 1969; Sieling, 1970). The objective of this study was to determine the role of Diadumene leucolena as a predator of oyster larvae and to estimate its significance in controlling abundance of oysters. MATERIALS AND METHODS The density of D. leucolena was determined on 14 beds in six major oyster areas throughout 113 114 C. L. MACKENZIE, JR. Maryland between July 8 and July 11, 1974 (Figure 1). Normally, early July is the beginning of the oyster larvae setting season that extends into September (Beaven, 1955). The areas selected to determine D. leucolena density were those in- dicated by Beaven (1955) and Lippson (1973) to be good oyster setting areas that are where most commercial oystering activity occurs. The follow- ing three bed types were examined: (1) beds of seed and market oysters; (2) beds of blank shells spread in 1971, 1972 and 1973; and (3) beds of shells dredged from beneath the bottom of the bay and spread in spring, 1974. Water depth over the beds ranged from 2.7 to 5 m. Population density estimates of D. leucolena were made once on each bed by scuba divers who collected all individual anemones in a number of small measured areas. The anemones along with oysters, shells, and other material were put in a bag and brought to the surface for immediate counting while they were live. About 1.4 cl (2 pecks) of material were gathered from each bed. The D. leucolena found were uniformly large and therefore considered to be fully grown adults. No small presumably 0-age group individuals were observed. In the laboratory, fully-developed, ready-to-set oyster larvae were exposed to anemones. Ten to 30 larvae were placed in 1-liter plastic containers with sea water containing individual anemones. D. leucolena behavior in relation to the larvae that contacted their tentacles was observed through a compound microscope. FIG. 1. Maryland portion of Chesapeake Bay in- dicating location of survey sites. Arrows point to six oyster areas surveyed. FIG. 2. A group of sea anemones, Diadumene leucolena, attached to a cluster of oysters in Chesapeake Bay (from Klingel and Culver, 1955). RESULTS Survey Results. Diadumene leucolena was found in all the areas surveyed. It grew on beds of both live oysters and blank shells that had been on a bed for at least a year (Figure 2), but none oc- curred on shells that had been spread in spring, 1974. Thus, it is probable that the larval settle- ment phase of the D. leucolena reproductive cycle had not begun by early July. D. leucolena grew on all sides of oyster and shell surfaces in contact with the water. The most dense populations were found in Harris Creek, Little Choptank River, Tar Bay, and Holland Strait, and fewer grew in Eastern Bay and St. Marys River (Table 1). SEA ANEMONE PREDATION 115 TABLE 1. Average density of sea anemones on various oyster beds in Maryland, water depth, salinity, and bed type. Under bed type, years denote dates when shells were spread on beds by the State of Mary- land. Average Number of Depth Salinity Anemones/m2 (m) (PPO Bed Type Eastern Bay Parson's Island 1. 36 5.0 11.0 Live oysters 2. 19 1971-73 shells Harris Creek Seaths Point 127 3.3 — Live oysters Middle Ground 16 3.3 — 1971-73 shells Mill Point 0 2.7 — 1974 shells Little Choptank River Casson's Bar 103 5.0 11.5 Live oysters Ragged Point 0 — 11.7 1974 shells Tar Bay Bluff Point 118 5.0 13.7 Live oysters Windmill Point 508 2.7 13.8 1971-73 shells Holland Strait Church Rock 219 3.3 13.9 Live oysters Chain Shoal 332 5.0 14.5 1971-73 shells St. Mary's River Thompson's Shore 18 3.3 13.5 Live oysters Church Point 6 4.0 13.5 1971-73 shells Thompson's Shore 0 3.3 13.5 1974 shells Population densities (no./m2 averaged 103.5 (range, 18 to 219) anemones on live oysters, and 176.2 (range, 6 to 508) anemones on shells spread in 1971, 1972, and 1973. Fewer anemones grew on live oysters than on shells, because in the process of feeding, probably oysters interfere with the set- tlement behavior of D. leucolena, thereby reduc- ing its numbers. Laboratory Observations. Diadumene leucolena captured and consumed all oyster larvae that contacted a tentacle. At the contact point, the tentacle bent vigorously and sharply and in- voluted longitudinally, locking the larva against it. Immediately, the bent tentacle section that held the larva moved toward the anemone mouth, that in turn opened in the direction of the tentacle (Figure 3). The larva was then released into the mouth, and passed down the gullet, while the ten- tacle returned to its original position. Individual D. leucolena were able to capture more than one larva per minute, and once three larvae were captured simultaneously. DISCUSSION Diadumene leucolena probably has general distribution in Chesapeake Bay, certainly where environmental conditions are similar to those found in this study. For instance, Merrill and Boss (1966) found it on a bed in the lower Patuxent River in 3.3 m of water, while Klingel and Culver (1955) also found it in the Virginia portion of the Chesapeake Bay. D. leucolena is dependent on a firm substratum for its existence and grows on stones and pier pilings, as well as oysters and shells. The anemone is absent on bottoms con- sisting of mud or sand. D. leucolena has high abundance and occupies much space on the commercial oyster beds of Maryland, which means it probably removes quantities of macroplankton near the bottom. D. leucolena is an opportunistic feeder and would in- 116 (I \ f § . FIG. 3. Sea anemone holding captured oyster lar- va (see arrow) on tentacle. Note the mouth open- ing in direction of larva. gest molluscan larvae, including the oyster and soft-shell clam (Mya arenaria), also important in commerce (Shaw and Hamons, 1974). During scuba observations, the anemone was the only visible predator of oyster larvae on the bottom. Earlier observations (unpublished) showed that the oyster and barnacles (species not identified), two common filter feeders, do not consume fully- developed oyster larvae. Possible presence of predators in the water was not examined. I believe that D. leucolena may be highly destructive to oyster larvae because they congregate near the bottom before setting. If so, the anemone would greatly reduce abundance of the oyster through- out the Maryland portion of Chesapeake Bay. Accordingly, control of D. leucolena on oyster and shell beds should increase oyster abundance. Application of quicklime (CaO), that effectively kill the starfish without harmful environmental side effects in more saline waters (MacKenzie, 1970, a, b), might also kill the anemone. Reduc- tion in abundance of the anemone should be followed by greater abundance of oyster. ACKNOWLEDGMENTS I thank F. William Sieling, III, and Frank Nelson, Department of Natural Resources, State of Maryland, for assistance in collecting field samples; and Warren S. Landers, National Marine Fisheries Service, Milford, Connecticut, for sup- plying oyster larvae. LITERATURE CITED Beaven, G. F. 1955. Various aspects of oyster set- ting in Maryland. Proc. Natl. Shellfish. Assoc. 45:29-37. Christensen, D. J. 1973. Prey preference of Stylochus ellipticus in Chesapeake Bay. Proc. Natl. Shellfish. Assoc. 63:35-38. Engle, J. B. 1956. Ten years of study on oyster set- ting in a seed area in upper Chesapeake Bay. Proc. Natl. Shellfish. Assoc. 46:88-99. Klingel, G. C. and W. R. Culver. 1955. One hun- dred hours beneath the Chesapeake. Natl. Geogr. Mag. 57:681-696. Lippson, A. J. 1973. The Chesapeake Bay in Maryland. An atlas of natural resources. The Johns Hopkins Univ. Press, Baltimore and Lon- don. 55 p. Lyles, C. H. 1969. Historical catch statistics (shellfish). U. S. Fish Wildl. Serv., C. F. S. No. 5007, 116 p. MacKenzie, C. L.. Jr. 1970a. Causes of oyster spat mortality, conditions of oyster setting beds, and recommendations for oyster bed management. Proc. Natl. Shellfish. Assoc. 60:59-67. MacKenzie, C. L., Jr. 1970b. Oyster culture in Long Island Sound 1966-69. Commer. Fish. Rev. 32(l):27-40. MacKenzie, C. L., Jr. 1970c. Oyster culture modernization in Long Island Sound. Am. Fish Farmer 1(6):7-10. Merrill, A. S. and K. J. Boss. 1966. Benthic ecology and faunal change relating to oysters from a deep basin in the lower Patuxent River, Maryland. Proc. Natl. Shellfish. Assoc. 56:81-87. Orton, J. H. 1922. The mode of feeding of the jelly-fish, Aurelia aurita, on the smaller organ- SEA ANEMONE PREDATION 117 isms in the plankton. Nature (Lond.). 110:178-179. Shaw, W. N. and F. Hamons. 1974. The present status of the soft-shell clam in Maryland. Proc. Natl. Shellfish. Assoc. 64:38-44. Sieling, F. W. 1970. Brief history of shell dredging in Maryland. Chesapeake Bay Affairs 3(1):1. Thorson, G. 1950. Reproduction and larval ecology of marine bottom invertebrates. Biol. Rev. (Camb.) 25:11-45. Van Engel, W. A. 1958. The blue crab and its fishery in Chesapeake Bay. Part 1 — Reproduc- tion, early development, growth, and migra- tion. Commer. Fish. Rev. 20(6):6-17. Webster, J. R. and R. Z. Medford. 1961. Flatworm distribution and associated oyster mortality in Chesapeake Bay. Proc. Natl. Shellfish. Assoc. 50:89-95. ABSTRACTS OF THE TECHNICAL PAPERS PRESENTED AT THE 1976 NSA CONVENTION GROWTH OF MARKED RANGIA CLAMS IN THE POTOMAC RIVER Aven M. Andersen Michael D. Bilger Inland Environmental Laboratory Center for Environmental and Estuarine Studies University of Maryland College Park, Maryland 20742 As part of a pilot study on reproduction and growth of brackish-water clams, Rangia cuneata. in Maryland, a total of 456 clams were dug, mark- ed, and planted into five square-meter plots near Lower Cedar Point. Then, from 3 months to 2 years later, attempts were made to recover them. About 25% have never been recovered, about 13% were dead when recovered, and about 62% were recovered live at least once. A few were recovered as many as five times. Most of those that were alive when recovered had grown little or not at all. A few grew considerably. The greatest percentage increase in length was for a clam that grew from 23 mm to 41 mm (78%) in 277 days. The general lack of growth in these marked clams is supported by the failure of modal lengths of monthly random samples of wild clams to change much with time. The irregular and poor growth of the marked clams casts doubt on the usefulness of shell markings for aging Rangia in this area. MANILA CLAM RESEEDING PROSPECTS IN WASHINGTON STATE Kenneth K. Chew College of Fisheries University of Washington Seattle, Washington 98195 During the past two and half years, the Washington State Departments of Fisheries, Parks and Natural Resources and the National Marine Fisheries Service have joined with the College of Fisheries and the Washington Sea Grant in studies centered around the planting of hatchery-reared seed Manila clams (Venerupis japonica) on Puget Sound beaches to investigate the feasibility of the culture and to refine the techniques of planting and site selection. The state-of-the-art for techni- que used for planting seed clams, recovery and substrate sediment size and predation problems were discussed. The probability of successful harvesting seeded clams at this point is still uncertain. Results from all of our studies tend to indicate that in most beaches tested, one might expect between 0-25% recovery at harvest. Most of the studies monitored after one year reveal recovery of less than 10%. However, enclosure cage studies with V4 inch mesh screens consistently provided high recovery and the potential reason for this is discussed. GROWTH AND SURVIVAL OF CULTCHLESS SPAT PLANTED IN NOMINI AND LOWER MACHODOC CREEKS IN 1973 Dexter S. Haven Virginia Institute of Marine Science Gloucester Point, Virginia 23062 Cultchless spat were planted in Lower Machodoc and Nomini Creeks in October, 1973, by the Virginia Marine Resources Commission and were subsequently monitored for growth and survival by the Virginia Institute of Marine Science. The study ended in July 1975. About 12 acres were planted at a density of one 118 PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION 119 million spat per acre. By July 1975 about 80% had died in Lower Machodoc Creek; in Nomini Creek mortality was 73%. There was an initial mortality of small spat (20%) less than J 2 inch long shortly after planting. However, the cause of death of the larger spat which died later was not apparent. From October 1973 to July 1975, spat grew from an average size of 0.9 inches to 2.1 inches. AGE AND GROWTH OF PROTOTHACA STAMINEA (CONRAD) AND SAXIDOMUS GIGANTEUS (DESHAYES) AT KIKET ISLAND, WASHINGTON Jonathan P. Houghton Dames & Moore Seattle, Washington As part of an intensive survey of the intertidal ecology of Kiket Island, Washington, a detailed investigation was made of the distribution, abun- dance, age, and growth of the two recreationally important venerid clams, the native littleneck, Protothaca staminea Conrad, and the butter clam, Saxidomus giganteus Deshayes. Collections were made on eight transects en- compassing a variety of bottom types about the island. I examined a series of 0.25 m2 quadrats randomly located along fixed beach contours over two years' time to evaluate the abundance of distribution of these species. Von Bertalanffy growth equations were fitted to length data for clams for various locations about the island. Tests for differences in growth rate of P. staminea showed growth to be best near mean lower low water and less rapid at higher and lower tide levels. Growth was also better on the north side of the island than at the same tidal level on the south side. Several hydrographic features such as a higher and more stable regime of temperature and salinity on the north side may account for this. Age frequency curves and "cumulative sur- vivorship'' curves were constructed and used to predict numbers of clams expected in each cohort at each location. These figures, combined with the average weight gained per year for each cohort were summed for a smoothed estimation of pro- duction and standing crop of P. staminea and S. giganteus. About ninety percent of the total stand- ing crop of P. staminea (3,319 kg) was on the north side of the island. A high correlation was found between the loca- tion of areas with a high diversity and richness of invertebrates in general, and areas with high den- sities, good growth and substantial production and standing crop of P. staminea and S. giganteus. It is suggested that the same environmental factors control all of these phenomena. THE SYSTEMATIC IDENTIFICATION OF COMMERCIALLY USEABLE SOURCES OF NATURAL OYSTER SPAT IN EASTERN CANADA Rene E. Lavoie Environment Canada Fisheries & Marine Service Halifax, Nova Scotia In Canada, the American oyster Crassostrea virginica lives in the southwest portion of the Gulf of St. Lawrence through part of the provinces of New Brunswick, Prince Edward Island and Nova Scotia. Most of the landings come from fishing the natural populations. A leasehold industry is gradually developing but it is dependent on the natural populations for its stocking and seed oysters. A five year oyster spatfall monitoring program was conducted between 1971 and 1975 to identify reliable and commercially useable sources of natural oyster spat for the benefit of the leaseholders. The program involved 134 stations distributed in 39 bays and estuaries in the pro- vinces of New Brunswick and Prince Edward Island. The stations were visited weekly from mid- June to early September. The program identified eleven bays and estuaries which regularly produc- ed a commercial set between 1971 and 1975. This paper also examines some of the physical parameters which appear to affect the relative spat production success of three of the commercially useable sources identified by the spatfall monitor- ing program. Further habitat and population 120 ABSTRACTS studies were conducted in Caraquet Bay and the Buctouche River in New Brunswick, and in the Bideford River in Prince Edward Island. Physical factors of the environment, biological parameters of the oyster populations as well as different ratios expressing the relationship between the oyster populations and their environment were cor- related with the average oyster set recorded during the spatfall monitoring program. This study showed a significant negative correlation between tidal exchange in the estuaries and spatfall success. The study also demonstrated a very significant positive correlation between the number of flow diversions per unit area in the estuaries and oyster spatfall success. ANNUAL STRUCTURAL CHANGES IN THE INNER SHELL LAYER OF GEUKEN5IA (^MODIOLUS) DEMISSA Richard A. Lutz Department of Oceanography University of Maine Walpole, Maine Seasonal cycles of growth were found to be reflected in structural patterns within the inner shell layer of the Atlantic ribbed mussel, Geuken- sia (^Modiolus) demissa, providing a method of age and growth rate determination. Monthly samples of mussels were obtained over a two-year period from both a natural intertidal population and an experimentally rafted population in the Damariscotta estuary, Lincoln County, Maine. Examination of acetate peels prepared from polished and etched longitudinal shell sections revealed the presence of at least two, and often four, distinct crystalline structural types within the inner shell layer. Scanning electron microscopic examination of both the inner layer growth surface and fractured shell sections reveal- ed that the deposition of nacreous structure was restricted to the relatively warm months between May and September. During the remainder of the year, several crystalline patterns were observed all of which, when viewed in longitudinal sections, present a prismatic- like appearance, structurally very similar to myostracal prisms found within the shells of numerous bivalves. Examination of G. demissa specimens from geographically isolated populations from Maine to Florida suggests a correlation between the mean annual temperature range and the structural com- position of the inner shell layer. The shell struc- ture of this species can potentially be used for determining paleotemperatures. OBSERVATIONS OF SEA SCALLOP STOCKS ON GEORGES BANK AND MIDDLE ATLANTIC SHELF IN 1975 Clyde L. MacKenzie, Jr. and Arthur S. Merrill Sandy Hook Laboratory Middle Atlantic Coastal Fisheries Center National Marine Fisheries Service Highlands, New Jersey Two resource assessment surveys (R/V Albatross IV) for sea scallops from Georges Bank southward to Cape Hatteras, North Carolina were made in 1975. Scallops were most abundant on Northern Edge and Northeast Peak on Georges Bank, and south of Long Island and east of the northern New Jersey coast on Middle Atlantic Shelf. A widespread scallop set occurred on Georges Bank and Middle Atlantic Shelf in 1972. The set was sufficiently abundant to predict a significant increase in commercial scallop landings in the near future. A COMPARISON OF GROWTH AND SURVIVAL OF SUBTIDAL CRASSOSTREA VIRGINICA (GMELIN) IN FOUR SOUTH CAROLINA SALT MARSH IMPOUNDMENTS John J. Manzi and Victor G. Burrell, Jr. Marine Resources Research Institute Charleston, South Carolina 29412 Four South Carolina salt marsh impoundments and their associated tidal creeks or estuaries were assessed as tracts for the culture of subtidal Crassostrea virginica. The impoundments were PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION 121 chosen primarily for their diversity and ranged from old large impoundments with appreciable tidal exchange and surrounded by extensive low marsh, to new small impoundments with little tidal exchange and surrounded by maritime forest. Floating and bottom hardware cloth trays (1.22 x 0.61 x 0.14 m) each holding 200 seed oysters (initial y = 43.8 mm) were placed at each location and sampled monthly for growth and survival. Coincidental monthly estimates of primary pro- duction (14C), phytoplankton concentrations and total organic carbon were performed. Ancillary data collected biweekly at all locations included standard hydrographic information (temperature, salinity, pH, and turbidity) and nutrient deter- minations (nitrates, nitrites, orthophosphates and silicates). Results indicated that over a six month period (October — April) growth at all locations was significantly (a = 0.001) greater in ponds than in adjacent creeks and greater in floating than bot- tom trays. Growth means ranged from as little as 1.00 mm month"' in the Wando River, to median values of approximately 2.25 mm month"1 in tidal creeks, to a relatively high growth rate of 3.11 mm month"' in Blue Heron Pond (Kiawah Island). Survival was high in all areas ranging from 85.0% at Blue Heron Pond to 94.5% at Kiawah Creek. There were no significant dif- ferences in survival in comparisons between all locations. A direct correlation between growth in oysters and primary production and phytoplank- ton biomass was established. This relationship was reiterated by indications of an inverse correla- tion between nutrient concentrations and growth. AN EPIZOOTIC OF "DERMO" DISEASE IN OYSTERS IN THE MARYLAND PORTION OF THE CHESAPEAKE BAY. Sara V. Otto and George E. Krantz Maryland Department of Natural Resources Fisheries Administration State Laboratory Oxford, Maryland 21654 University of Maryland Center for Environmental & Estuarine Studies Horn Point Environmental Labs. Cambridge, Maryland 21613 An outbreak of "Dermo" disease of epizootic (epidemic) proportions has been documented in the Maryland portion of the Chesapeake Bay. Prevalence of this disease has reached 100% on specific bars. Badly affected oyster bars have ex- perienced 25% to 60% mortality in one year. Historical data and past oyster bar surveys (1960 to 1971) are compared to the geographic distribu- tion, disease intensity, and oyster mortality in 1975. Some observations suggest that a new strain of "Dermo" — highly adapted to the low salinity Maryland estuarine environment — may be a fac- tor in the upsurgence of this disease. Some possi- ble management practices to control the disease and maintain commercial oyster production are discussed. EFFECTS OF THREE TOXICANTS ON OYSTERS (CRASSOSTREA VIRG1NICA) EXPOSED CONTINUOUSLY FOR TWO YEARS Patrick R. Parrish1, James M. Patrick, Jr.2 and Jerrold Forester2 1EG&G, Bionomics Marine Research Laboratory Route 6, Box 1002 Pensacola, Florida 32507 2U. S. Environmental Protection Agency Gulf Breeze Laboratory Sabine Island Gulf Breeze, Florida 32561 Three separate populations of oysters were ex- posed continuously for 104 weeks in flowing, natural sea water in the laboratory to 0.01^g/f of Aroclor® 1254 or p,p'-DDT and its metabolites or dieldrin. Maximum residues (based on ^g of toxicant per g of tissue) occurred after 8 weeks of exposure; average whole-body residues (wet weight) of five oysters from each treatment analyzed individually were: Aroclor® 1254, 1.65 Mg/g; DDT (and metabolites DDD and DDE), 0.46 Mg/g; and dieldrin, 0.08 ug/g. Seasonal patterns of accumulation and loss of the three tox- icants were similar and were apparently related to 122 ABSTRACTS spawning. Toxicant residues decreased 45% to 81% in early July and late October, 1972, and 44% to 91% in late October, 1973. (Low spat sets in water adjacent to the laboratory indicated minimal spawning in wild oyster populations dur- ing the spring of 1973.^ Growth rate (height and in-water weight) of exposed oysters was not significantly different from that of control oysters (Student's t-test; P<0.05) after 72 weeks of ex- posure. Mortality was not significant (<9%) in any group during the entire study. ® Registered trademark, Monsanto Company, St. Louis, MO. Hatching consistency of the Canadian cysts was inferior to that of the other tested strains when 48 hours was allowed for hatching and when hat- ching procedures were similar for all three strains. The hatching medium and procedures recom- mended by the supplier did not give results superior to those obtained with diluted sea water. However, longer hatching time may provide satisfactory hatch rates for the Canadian Artemia. If hatching consistency can be improved with modified hatching methods, this new source may be a satisfactory and very welcome supplement to presently limited supplies. EVALUATION OF A SULPHATE LAKE STRAIN OF ARTEMIA AS A FOOD FOR LARVAE OF THE GRASS SHRIMP, PALAEMONETES PUGIO Anthony J. Provenzano, Jr. and Joseph W. Goy Institute of Oceanography Old Dominion University Norfolk, Virginia 23508 Artemia nauplii from a newly available commercial supply originating in a Canadian sulphate lake were tested against Artemia from two other sources for effectiveness in supporting larval development of the grass shrimp, Palaemonetes pugio. Groups of 24 shrimp larvae pooled from three females were reared in isola- tion, each larva in 25 ml of static artificial sea water at 25 °C and 17 p.p.t. Water was changed daily and newly hatched nauplii were provided daily at a concentration of 5-35 nauplii/ml for the Canadian strain and 40 nauplii/ml for the other two strains. Survival rate to metamorphosis for larvae fed with Canadian Artemia (71%) was as good as that for larvae fed with San Francisco Artemia (63%) and clearly superior to that for larvae fed with Great Salt Lake Artemia (29%). No significant dif- ference was found in mean number of stages re- quired to metamorphosis or in mean time to metamorphosis between groups fed the three strains. SOUTH CAROLINA'S HYDRAULIC ESCALATOR HARVESTER FISHERY Raymond J. Rhodes, Willis J. Keith andV. G. Burrell, Jr. South Carolina Marine Resources Center S. C. Wildlife and Marine Resources Department Charleston, South Carolina 29412 Beginning in 1974 the delta area of the North and South Santee River in South Carolina were experimentally opened to hydraulic escalator harvester operations. During the 1974-75 clam season in South Santee and North Santee river, 31,538 "bags" (250 individuals per bag) of hard clams, mostly Mercenaria mercenaria, were harvested by 9 vessels. In the recent 1975-76 season, 25,948 bags were harvested from the North Santee Bay by 7 vessels. The estimated mean hourly catches in the South Santee River, North Santee River and North Santee Bay were 10.1, 10.8 and 19.8 bags per hour respectively. Besides regulatory constraints on vessel permits and operating days, exvessel prices, equipment malfunctions and weather conditions limited the fishing time. The effectiveness of fishing input (e.g. vessel characteristics, crew size, captain skills, etc.) will be discussed. PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION 123 THE EFFICIENCY OF "NITROGEN" TRANSFER IN ARTIFICIAL UPWELLING MARICULTURE. I. THE CONVERSION OF DEEP-SEA WATER DISSOLVED NITRATE TO PHYTOPLANKTON PROTEIN TO TAPES SEMIDECUSSATA MEAT-PROTEIN IN A FULLY MANAGED SYSTEM.1 Oswald A. Roels, Thomas E. Dorsey, Kenneth Rodde, Scott Laurence, Richard Lyon and Paul W. McDonald The University of Texas Marine Science Institute, Port Aransas Marine Laboratory, Port Aransas, Texas 78373 The efficiency of conversion of deep-sea water dissolved nitrate to phytoplankton protein to Tapes semidecussata meat-protein was determined in the St. Croix Artificial Upwelling mariculture system, during a five-week experiment. Chaetoceros curvisetus (STX-167, a centric diatom) and an unidentified naked flagellate (S-l, isolated by Guillard from the Sargasso Sea) were individually grown in continuous culture in con- crete tanks on shore, using nutrient-rich sea water containing 31 ^eq NO/ liter"1 pumped from 870-m depth. During the five week experimental period, 69% of the deep water nitrate-nitrogen was converted into algal protein-nitrogen. Twen- ty percent of the incoming deep water nitrate was not converted by the phytoplankton at the 24- hour turnover rate used in this system. The 11% of deep-water nitrate-nitrogen, unaccounted for in the study, was probably present in the form of non-protein nitrogen in the cells and possibly as dissolved organic nitrogen in the medium. The two algal cultures were mixed and the mixture was fed at 1 ml/sec. to 4-liter shellfish containers holding 35, 70 or 140 g of 13-mm long T. semidecussata. Simultaneously, the same mixture was fed at 2 ml/sec to shellfish containers holding 35, 50, 70, 100 or 140 g of 13-mm long T. semidecussata. Each shellfish treatment was run in duplicate. At nine-day intervals the shellfish in each treatment were culled back to the starting weight. At the 1 ml/sec flow rate, the efficiency of conversion of phytoplankton protein to shellfish meat protein averaged 33% . At the 2 ml/sec flow rate, this efficiency averaged 30%. An average of 75% of phytoplankton protein nitrogen entering the shellfish tanks could be accounted for as clam meat and shell protein nitrogen, particulate pro- tein nitrogen in the effluent from the shellfish tanks, particulate protein nitrogen deposited in the tanks, and dissolved ammonia and nitrate formed in the tanks. Particulate protein deposited in these tanks as fecal ma^er, debris, wall growth, etc., appeared to be a function of the amount of algae entering the shellfish tanks rather than the mass of shellfish in the tanks. The maximum shellfish weight gain was obtain- ed with the 100-g 2-ml/sec treatment. The lowest weight gain was registered by the 35-g 1-ml/sec shellfish. The fastest increase in length of in- dividual shellfish was achieved at the 35-g 2- ml/sec treatment. The slowest increase in length occurred in the 140-g 1-ml/sec group. In this study, the fastest growing clams were the least ef- ficient food converters. Detailed analysis of the results of this experi- ment provides valuable guidance for the design of a shellfish maricultve system where the trade-off between efficiency of food conversion and the time required for the animals to reach market size will be guided by economic considerations. This work was supported by Sea Grant 04-5-158-59 from NOAA, U. S. Department of Commerce, and by matching funds from the G. Unger Vetlesen Foundation. PRELIMINARY OBSERVATIONS ON A SHORT-CLAW GROWTH FORM OF THE MALAYSIAN PRAWN, MACROBRACHIUM ROSENBERGII (DE MAN) Paul A. Sandifer and Theodore I. J. Smith Marine Resources Research Institute Charleston, South Carolina 29412 During the 1975 harvest of ponds used for ex- perimental cultivation of the Malaysian prawn, Macrobrachium rosenbergii, in South Carolina, a short-claw growth form was noted. The propor- tions of short-claw prawns in the pond popula- tions were not determined, but several preliminary observations were made. First, the short claws were generally golden in color as op- 124 ABSTRACTS posed to the dark blue color of the claws in the "normal" long-claw form. Second, the short- (golden) claw form was noted almost exclusively among males. Third, the claws of the short-claw form were smaller by about 30-70% in length and 80-250% in weight as compared to normally claw- ed males of similar total length. Fourth, because of their smaller claws, short-claw males generally tended to be slightly smaller in total weight than long-claw animals of identical total length. Fifth, the tail comprises about 5-8% more of the total body weight in the short-claw form. Thus, for the same total weight of prawn, the short-claw form yields more tail meat than does the long-claw form. Sixth, the short-claw animals appear to be somewhat less aggressive than long-claw specimens. However, this observation requires careful verification. It is now important to deter- mine if the short-claw form is an inherited or en- vironmentally controlled trait and if it is sex link- ed. In the future manipulation of prawn popula- tions to produce more short-claw animals might reduce mortalities due to aggressive interactions as well as increase the yield of marketable prawn tails. The rate of change has been accelerated in areas of intense industrial and residential development such as the Barataria Basin, where over 9,000 acres of marshland have been lost by dredging. Salinity data compiled for a twenty-year period have documented increasing salinity in the Barataria estuary. Salinity has increased at an average monthly rate of 0.01 ppt at St. Mary's Point, a station in upper Barataria Bay. During the past thirty years, natural oyster spatfall has occurred further and further inland, in bayous and lakes that previously had been too fresh to sup- port oyster growth. The encroachment of highly saline Gulf water into present oyster-growing areas leads to mortalities resulting from the conch, Tliais haemostoma, and the fungus, Labyrin- thomyxa marina. In essence, the oyster industry in Barataria Basin is steadily being squeezed between encroaching salinity from the south and en- croaching pollution from the north. As these two forces continue, availability of areas suitable for oyster production will decline. BARATARIA BASIN: SALINITY CHANGES AND OYSTER DISTRIBUTION Virginia R. Van Sickle, Barney B. Barrett and Ted B. Ford Center for Wetland Resources, Louisiana State University, Baton Rouge, La. 70803. Division of Oyster Water Bottoms and Seafoods, Louisiana Wildlife and Fisheries Commission, P. O. Box 14526, Baton Rouge, La. 70808. Natural processes such as erosion, subsidence, and rise in sea level have contributed to a gradual alteration of salinity regimes in coastal Louisiana. SOME SPATIAL AND NUTRITIONAL EFFECTS ON THE CULTURING OF THE LARVAE OF CRASSOSTREA VIRGINICA, THE AMERICAN OYSTER N. T. Windsor and John L. Dupuy Virginia Institute of Marine Science Gloucester Point, Virginia Experiments were carried out on the spatial and nutritional aspects of the culturing of oyster lar- vae. Results indicate spatial requirements for the larvae to culture volume influences the percent yields of pediveligers and metamorphosed larvae. The successful development of an algal diet con- sisting of three species which consistantly produce pediveligers in 7-10 days is described. In addition the effect of various algal densities on the larval growth rate and percent yield of pediveligers is also considered. PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION 125 NSA PACIFIC COAST SECTION DEVELOPMENT OF UNIVERSITY OF WASHINGTON'S EXPERIMENTAL OYSTER HATCHERY John (Hal) Beattie University of Washington College of Fisheries, Seattle, Washington For the past five years the College of Fisheries at the University of Washington has been studying the problems of summer mortalities of Pacific Oysters in Washington state. Researchers have succeeded in simulating mortality situations in the laboratory and in isolating bacteria associated with the kill. The purpose of the development of this experimental shellfish hatchery is to in- vestigate the possibility of using genetics to help reduce or eliminate summer kill. The methods in- clude using surviving adult oysters of a laboratory challenge as brood stock, rearing the offspring on selected oyster beds in Washington state and challenging these oysters against Japanese oyster seed of the same age. In addition the tissues of parents and offspring will be tested using elec- trophoresis in order to examine inheritance pat- terns. The hatchery is a result of cooperation be- tween the College of Fisheries, National Marine Fisheries Service, E. P. A., Washington State Department of Fisheries, several growers and the four shellfish hatcheries in Washington state. ALGAL CHEMOSTATS AND OYSTER LARVAE W. P. Breese, P. L. Donaghay, R. E. Malouf, and L. F. Small Oregon State University Corvallis Oregon A method of prolonged, continuous algal culture is described. The vessel contains 16 i of algal culture with a density of from 3 to 10 million cells per ml. Sterility is established by using ozone gas and is maintained by using chlorine (3 to 5 ppm) in the dilution water. Prolonged continuous algal culture may affect the food quality of the algae. Under certain culture conditions, Pseudoisochrysis paradoxa can become a highly superior food yielding enhanced growth and survival of Pacific oyster larvae. However, under different culture conditions this same species lost up to 90% of its food value. The poor food quality characteristic of this food type applied not only to Pacific oyster larvae, but also to brine shrimp. The reason for this phenomenon is not yet known, but is under investigation. These findings strongly suggest that although con- siderable potential exists for improving larval culture techniques, application of continuous culture techniques at this time may be premature. MUSSEL STUDIES IN SEABECK BAY AND CLAM BAY Linda Chaves-Michael College of Fisheries University of Washington Seattle, Washington A Mytilus edulis culture project was initiated during 1974 in Puget Sound by the College of fisheries at the University of Washington. The 126 ABSTRACTS methods used were to approximate Spanish raft culture methods. The initial goals were to deter- mine a suitable setting substrate, time of setting during the year, growth time to market size and yield. Synclove, a synthetic resembling manila rope was more promising than manila or oyster strings for catching mussel seed, yield, and durability. Market size (50 mm) was reached after only one year's growth. During 1975 and 1976, mussel set prediction has been attempted by the examination of plankton samples for mussel larvae. The 1975 results were inconclusive as low water temperatures resulted in sporadic and minor mussel setting. The 1976 plankton samples are more promising as the presence of larvae in the water and mussel setting have been greater than during 1975. The Mussel Culture Studies are a cooperative effort between the University of Washington, Na- tional Marine Fisheries Service and Sea Grant. A similar Sea Grant study is being conducted in the New England states. The results of the two studies will be compared. PRELIMINARY FINDINGS ON A RECENT SUMMER KILL OF PACIFIC OYSTERS Kenneth K. Chew College of Fisheries University of Washington Seattle, Washington Significant mortalities of Pacific oysters (Crassostrea gigas) have been recorded in Puget Sound, Willapa Bay, Washington and Humboldt Bay, California in the 1960s. Since 1970, the mor- talities have subsided although there have ap- parently been isolated cases of low background mortalities from time to time during the summer. There recently was a significant mortality discovered at Rocky Bay during the month of August, recording a mortality of 20-30% of the population. This occurred after a week of very warm temperatures and a low tide series. The mortalities revealed what appeared to be similar symptoms of mortalities occurring in the 1960s. Observations were also made at Mud Bay, where 5-10% mortality occurred, although this may be considered background mortality. Samples of oysters were taken from Rocky Bay and Mud Bay and challenged with water temperatures above 20° C. The results of these tests revealed similarities with some of our earlier laboratory challenges. The results of these tests and the approaches used to seek a better understanding of problems related to the stress conditions and the identification of potential pathogens are discussed. RECENT CLAM STUDIES IN OREGON'S ESTUARIES Thomas F. Gaumer Oregon Department of Fish and Wildlife Newport, Oregon Intertidal and subtidal clam surveys have been conducted in ten of Oregon's estuaries since 1973. During these surveys we examined more than 1.1 million feet (325,280 m) of transect line and made 7,250 observations on the distribution, abundance and species composition of clams. In addition data were collected on substrate material and vegeta- tion at each of the sample stations. Commercial quantities of subtidal gaper clams, Tresus capax, were located and mapped in Tillamook, Yaquina and Coos bays. As a result, considerable interest has been generated in the commercial harvest of these clams. Special com- mercial harvesting permits were issued for Ya- quina and Coos bays to evaluate the effects of a commercial fishery on the habitat and clam resources. The fishery failed to develop in Ya- quina Bay due partially to poor marketing condi- tions. In Coos Bay the fishery produced over 55,000 pounds (25,166 kg) of gaper clams. Statistical tests showed the gaper clam could most reliably be aged by counting the annuli in the chondrophore. Butter, Saxidomus giganteus, cockle, Clinocardium nuttallii, and littleneck, Venerupis staminea, clams were aged by counting the annuli on the exterior surface of the shell. Ag- ing studies showed spawning or survival of clam set to be highly sporadic. Plankton samples showed gaper clam larvae to be widely and generally evenly scattered PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION 127 throughout the channel areas and across the tideflats. Two periods of peak spawning were recorded in Yaquina Bay; both associated with the lunar cycle with maximum production of larvae at the periods of greatest tidal range. A haplosporidian infection of the gaper clam was documented subtidally from five of Oregon's estuaries. Incidence ranged as high as 89% and ap- peared to widely distributed through the estuarine range of the clam. The infection appeared to be age dependent with older clams more heavily parasitized. BIOCHEMICAL GENETIC IDENTIFICATION OF SPECIES AND HYBRIDS OF THE BERING SEA TANNER CRAB, CHIONOECETES BAIRDI AND C. OPILIO. W. S. Grant, L. Bartlett* andF. M. Utter U. S. Dept. of Commerce National Marine Fisheries Service Northwest Fisheries Center Seattle, Washington *Kodiak Facility The eastern portion of the Bering Sea is an area of overlap in the distributions of the tanner crab, Chionoecetes bairdi and C. opilio. In this area the species hybridize, and recent abundance data in- dicate that hybrids make up about 20% of the total tanner crab population. The presence of the hybrid creates problems in resource assessment studies and is potentially problematical in com- mercial catch regulation because it is often dif- ficult to distinguish the hybrid from its parent species. Morphological characters integrade and there is often disagreement as to which traits characterize the hybrid. This report is an exten- sion of research by Allyn Johnson, which describ- ed proteins using electrophoresis that were diagnostic of each species and the hybrid. Using starch gel electrophoresis we assayed for 23 genetic loci. The average observed heterozygo- sities of C. bairdi, C. Opilio and the hybrids were 4.9, 6.3 and 12.9% respectively. The higher level of heterozygosity found in the hybrids resulted from the crossing of two separate strains. Using Rogers' (1972) index of genetic similarity (which ranges from 0 to 1.0) our biochemical data in- dicate that C. bairdi and C. opilio are very closely related (S = 0.907) and it is not surprising that hybrids form between them. Our results confirm the validity of using general proteins as a diagnostic character to distinguish the two species and the hybrid from each other. THE ECONOMIC FEASIBILITY OF BRINE SHRIMP CULTURE UNDER SEMI-CONTROLLED CONDITIONS Richard S. Johnston and Larry O. Rogers Oregon State University Corvallis, Oregon During the summer of 1974, the authors cultured artemia s. on the campus of Oregon State University in outdoor ponds. This pre-pilot pro- duction scheme was part of an attempt to analyze the economic feasibility of producing a. salina eggs under semi-controlled conditions. We observed, in contrast to other reports in the literature, that artemia over-wintered not only as eggs, but as both adults and juveniles as well. Preliminary findings tend to indicate that artemia egg production alone, under the particular semi-controlled conditions chosen and at pro- jected world prices, would not be economically feasible when associated with the high costs of land acquisition, pond construction, and pond maintenance which presently prevail. However, if artemia eggs are produced as a bi-product of solar salt manufacturing, it would appear that such an enterprise, using semi-controlled conditions, would be lucrative, being able to attract and retain resources. The techniques used in this study may prove useful to aquaculturists who use artemia as feed for various cultured organisms. THE DEMAND FOR PACIFIC OYSTERS: A PRELIMINARY REPORT Richard S. Johnston and A. Nelson Swartz Oregon State University Corvallis, Oregon 128 ABSTRACTS Using highly aggregated data, we find the de- mand for Pacific oysters to be relatively price- elastic, in contrast to the price-inelastic relation- ship for all oysters uncovered by other research- ers. Consumer demand appears to respond positively to increases in real income levels. Addi- tional data on retail prices and shipments to par- ticular geographical markets, inventory holdings and exports are needed for a more complete analysis. SEA URCHINS - WASHINGTON'S NEWEST FISHERY -PRESENTS SOME PRICKLY-PROBLEMS Chris Jones Washington State Department of Fisheries Brinnon, Washington Washington's newest fishery is for red sea ur- chins (Strongylocentrotus franciscanus) which are harvested by divers in the waters of the Strait of Juan de Fuca and in the San Juan Islands. The gonads are extracted and cleaned, and exported, preferably fresh, to Japan. Through May of 1976, approximately 600,000 lb of urchins were harvested. The fishery was closed June through August because of low yield, caused by spawning in the spring. Gonad yield averages about 10% although it can range from about 5% to over 20% depending on spawning cycle and food availabili- ty. If the total harvest reaches 1-million lb this year, as expected, the value to the fishermen will be around $75,000 and the export value will ex- ceed $200,000. Continued growth of the fishery will depend on the market, the number of divers in the fishery, and management decisions by the Washington Dept. of Fisheries designed to achieve a sustained level of harvest. Urchin populations are characterized by slow growth and recruitment; also the presence of large adults is believed to be a condition for either setting or early survival or ur- chins. In order to prevent overfishing, a minimum size of 3.75 inches has been imposed along with a rotation of areas open to fishing. Current studies are concentrating on estimating the extent of the resource and the time and conditions required for repopulation of a fished area. MANILA CLAM RESEEDING STUDIES IN PUGET SOUND Mark Miller, Charles D. Magoon, Lynn Goodwin and Chris Jones University of Washington, Seattle, Washington Department of Natural Resources Seattle, Washington Washington State Department of Fisheries Olympia, Washington Over the past two years the College of Fisheries and the Washington State Departments of Fisheries and Natural Resources with cooperation from the Washington State Department of Parks and the National Marine Fisheries Service, have taken part in studies involving the planting of hatchery-spawned Manila clams on Puget Sound beaches. Various reseeding experiments are being performed with small seed clams which may lead to a potential for rehabilitating overused clam beaches. Factors such as tidal height, planting density, time of planting, size of seed on planting and substrate type are being considered. Other studies are being performed to determine important Manila clam predators. An effort is being made to protect seed clams from predators or environ- mental conditions by use of wire cages and plastic netting. Also, the experimental culture of Manilas in a large suspended sediment tray is being carried out. At all study areas recovery of seed clams generally has been relatively low over periods of one to two years; however, recovery at a par- ticular tide level may be appreciable. Until better understanding of the losses of planted clams is at- tained, large scale reseeding should not be at- tempted on most Puget Sound beaches. DUNGENESS CRAB MORTALITY FROM CHANNEL MAINTENANCE DREDGING IN GRAYS HARBOR, WASHINGTON Herb Tegelberg and Ron Arthur Determination of the distribution of Dungeness crabs (Cancer magister), and some effects of chan- nel maintenance dredging on crabs were part of a PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION 129 dredging effects study in Grays Harbor funded by the Corps of Engineers. Sampling showed year- round residence of juvenile crabs in the ship chan- nel east to Cow Point, and supported a previous assumption that Grays Harbor is an important "nursery" area for Dungeness crabs. Maintenance dredging annually removes about 1 million yards by pipeline dredging of the upper reaches of the 30-foot-deep ship channel, and another 1 million yards by hopper dredging of the outer reaches. Screening of dredged material near the pipeline dredge discharge was an effective means of sampling for crab mortality, although debris and unstable bottom made the task diffi- cult. Based on sampling of approximately 0.2 per- cent of the discharge during two-thirds of the dredging period, total mortality was estimated to be approximately 17,000 crabs in the 1975 pipe- line dredging of 767,000 yards. Direct sampling of a portion of the dredge material pumped by the hopper dredges Pacific and Biddle could not be accomplished during this study. This resulted in considerable experimenta- tion with methods. Screening the hopper overflows and beam trawling in the disposal site were ineffective. A single airlift sampler used in the hopper had a potential of sampling 0.4% of the dredged volume pumped by the Biddle. Forty- four Dungeness crabs were captured in a volume of dredging material estimated to be equivalent to 3 minutes dredging time. Crab survival was poor with the exception of very small crabs (20-35 mm carapace width). More extensive sampling is need- ed, but the findings indicate that hopper dredging kills large numbers of Dungeness crabs in Grays Harbor. This would be expected to apply to other west coast estuaries. The sampling indicates that pipeline dredging causes less direct mortality than hopper dredging. CURRENT STATUS OF SHELLFISH HARVEST PROBLEMS IN WASHINGTON STATE Ronald E. Westley Washington State Department of Fisheries Brinnon, Washington Many changes in laws and regulations effecting the commercial harvest of shellfish have either oc- curred or are being proposed. Some are good, some are creating problems. Some of these changes directly prohibit shellfish harvest, and some indirectly prohibit harvest by setting a lengthy and complex review system. All result in some cost in obtaining needed permits, and all create some amount of uncertainty. In most in- stances, the standards of acceptance appear vague and it is difficult to judge if permits can be obtain- ed. INFORMATION FOR CONTRIBUTORS TO THE PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION Original papers given at the Annual Association Convention and other papers on shellfish biology or related subjects will be considered for publica- tion. Manuscripts will be judged by the Editorial Committee or by other competent reviewers on the basis of originality, contents, clarity of presen- tation and interpretations. Each paper should be carefully prepared in the style followed in the 1972 PROCEEDINGS (Volume 63) before submission to the Editorial Committee. Papers published or to be published in other journals are not acceptable. Manuscripts should be typewritten and double- spaced; original and two copies are required to facilitate reviews. Tables, numbered in arabic, should be on separate pages with the title at the top. Scientific names should be underlined. Il- lustrations preferably should be 8 x 10 inch prints which can be reduced to a size of 6V4 x 8 inches or smaller. Glossy photographs are preferred to originals. Illustrations smaller than a page should be carefully oriented and loosely attached to plain white paper with rubber cement. Legends should be typed on separate sheets and numbered in arabic. Authors should follow the style prescribed by Style Manual for Biological Journals which may be purchased from the American Institute of Biological Sciences, 1401 Wilson Blvd., Arlington, VA. 22209. American Standard for Periodical Ti- tle Abbreviations, available through American National Standards Institute, 1430 Broadway, New York, New York 10018, should be followed in the "Literature Cited" section. Each paper should be accompanied by an abstract which is concise yet understandable without reference to the original article. It is our policy to publish the abstract at the head of the paper and to dispense with a summary. A copy of the abstract for submission to Biological Abstracts will be requested when proofs are sent to the authors. The author or his institution will be charged $25.00 per printed page. If figures and/or tables make up more than ¥3 of the total number of pages there will be a charge of $30.00 for each page of this tabular material (reckoned on the ac- tual amount of page space taken up) in excess of the set limit, regardless of the total length of the article. Reprints and covers are available at cost to authors. Page type will be retained for three months after publication. When proof is returned to authors, information about ordering reprints will be given. The present agency from which authors may obtain reprints is The Memorial Press Group, 23 Middle Street, Plymouth, Massa- chusetts 02360. Contributions are accepted at any time. However, for inclusion in the PROCEEDINGS of the current year, all manuscripts should reach the Editor by October 1, prior to the Annual Conven- tion. Send manuscripts and address all cor- respondence to the Editor, Dr. Robert E. Hillman, Battelle, Duxbury, Massachusetts 02332. G.P. Ennis & Determination of Shell Condition in Lobsters r%/* (Homarus americanus) by Means of External Macroscopic Examination 67 T.O. Thatcher An Effect of Chlorination on the Hatching of Coon Stripe Shrimp Eggs: So What? 71 Herbert Hidu, Mark S. Richmond, and Allison H. Price, II Morphological variability in Sea Scallops, Placopecten magellanicus (Gmelin) Related to Meat Yields 75 John E. Huguenin The Reluctance of the Oyster Drill (Urosalpinx cinerea) to Cross Metallic Copper 80 Richard B. Nickerson A Study of the Littleneck Clam (Protothaca staminea Conrad) and the Butter Clam (Saxidomus giganteus Deshayes) in a Habitat Permitting Coexistence, Prince William Sound, Alaska 85 Richard A. Macintosh and A.J. Paul The Relation of Shell Length to Total Weight, Tissue Weight, Edible-Meat-Weight, and Reproductive Organ Weight of the Gastropods Neptunea heros, N. lyrata, N. pribiloffensis, and U. ventricosa of the Eastern Bering Sea 103 Clyde L. MacKenzie, Jr. Sea Anemone Predation of Larval Oysters in Chesapeake Bay (Maryland) 113 Abstracts: NSA Annual Meeting NSA Pacific Coast Section 118 PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION CONTENTS Volume 67 — June 1977 List of Abstracts by Author of Technical Papers Presented at 1976 NSA Annual Meeting, Miami, Florida v Michael Castagna and John N. Kraeuter Mercenaria Culture Using Stone Aggregate for Predator Selection 1 Robert E. Malouf and Wilbur P. Breese Food Consumption and Growth of Larvae of the Pacific Oyster, Crassostrea gigas (Thunberg), in a Constant Flow Rearing System 7 Kwang H. Im and Don Langmo Economic Analysis of Producing Pacific Oyster Seed in Hatcheries 17 Joseph G. Loesch and John W. Ropes Assessment of Surf Clam Stocks in Nearshore Waters along the Delmarva Peninsula and in the Virginia Fishery South of Cape Henry 29 Victor G. Burrell, Jr. Mortalities of Oysters and Hard Clams Associated with Heavy Runoff in the Santee River System, South Carolina in the Spring of 1975 35 Raymond J. Rhodes, Willis J. Keith, Peter J. Eldridge and Victor G. Burrell, Jr. An Empirical Evaluation of the Leslie-DeLury Method Applied to Estimating Hard Clam, Mercenaria mercenaria, Abundance in the Santee River, South Carolina 44 George E. Krantz and Donald W. Meritt An Analysis of Trends in Oyster Spat Set in the Maryland Portion of the Chesapeake Bay 53 Gary H. Cole, Ronald L. Copp, and David C. Cooper Estimation of Lobster Population Size at Millstone Point, Connecticut, By Mark-Recapture Techniques, 1975-1976 60 MBL WHOI LIBRARY III UH 1ABB $