JOURNAL OF SHELlinSH RESEARCH The Joutnal of Shellfish Research (formerly Proceedings of the Nationai Shellfisheries Association) is the official publication 'of the National Shellfisheries Association Editor Dr. Sandra E. Shumway Department of Marine Resources and Bigelow Laboratory for Ocean Science West Boothbay Harbor Maine 04575 EDITORIAL BOARD Dr. Monica Bricelj Marine Sciences Research Center State University of New York Stony Brook, New York 11794-5000 Dr. Anthony Calabrese National Marine Fisheries Service Milford, Connecticut 06460 Dr. Kenneth K. Chew College of Fisheries University of Washington Seattle, Washington 98195 Dr. Charles Epifanio College of Marine Studies University of Delaware Lewes, Delaware 19958 Dr. Paul A. Haefner, Jr. Rochester Institute of Technology Rochester, New York 14623 Dr. Robert E. Hillman Battelle Ocean Sciences New England Marine Research Laboratory Duxbury, Massachusetts 02332 Dr. Herbert Hidu Ira C. Darling Center University of Maine Walpole, Maine 04573 Dr. Lew Incze Bigelow Laboratory for Ocean Sciences McKown Point West Boothbay Harbor, Maine 04575 Dr. Louis Leibovitz Marine Biological Laboratory Woods Hole, Mass. 02543 Dr. Roger Mann Virginia Institute of Marine Science Gloucester Point, Virginia 23062 Dr. Gilbert Pauley College of Fisheries University of Washington Seattle, Washington 98195 Dr. Les Watling Ira C. Darling Center University of Maine Walpole, Maine 04573 Journal of Shellfish Research Volume 7, Number 1 ISSN; 00775711 June 1988 Journal uj Shellfish Research. VoL 7, No. 1. 1-6, 1988. BIVALVE LARVAL RESEARCH, IN TRANSITION: A COMMENTARY MELBOURNE R. CARRIKER College of Marine Studies University of Delaware Lewes. DE 19958 U.S.A. Substantial progress has been made in the complex, but rewarding study of the biology of planktonic molluscan bi- valve larvae during the last four decades. This brief com- mentary is written at a time when, of necessity, we are moving from field to closed system molluscan mariculture. Presented in the interest of providing historical continuity and encouraging further creative research, 1 summarize some of the highlights of research, point out voids in knowledge, and touch on suggestions (some perhaps far out) for further inquiry on the biology of molluscan bivalve larvae. Earliest ecological studies of bivalve larvae in the western hemisphere were conducted by Julius Nelson, Bi- ologist at the New Jersey Experiment Station, with the as- sistance of his son, Thurlow, in estuaries on the east coast of New Jersey. Their pioneer work on seasonal distribution and movements of larvae oi Crassostrea virginica (Gmelin) was begun in 1908 (Nelson 1909). The first obstacle to be overcome by Julius Nelson was development of a method for sampling larvae. The proto- type of the present day plankton net was probably first in- troduced by Johannes Miiiier in 1846 (Sverdrup et al. 1942). Nelson, although familiar with the prototype, was unable to use such a net because "the finest cloth has just about space enough between its meshes to permit the pas- sage of the fry when they are full grown" (Nelson 1909). With characteristic ingenuity. Nelson went to the use of Lautenschlager filter paper folded inside a chemical funnel. Early in the morning when the air was stillest, and the sur- face of the coves calmest, he dipped water from the surface into the glass funnel. After the water had almost filtered out, he held a tumbler below, and punched a hole through the bottom tip of the filter allowing the residue to drop into the glass. He then washed sediment and larvae from the filter paper through the hole into the tumbler. Although this was an effective procedure, it was extremely time con- suming and limited sampling to surface waters. By 1913 bolting cloth, fine enough to retain all stages of oyster larvae, was available from the Simplex Company, and Nelson started towing a plankton net. In addition, he devised a method for pouring measured quantities of sur- face water, as well as bottom water taken with a Moore sampler (a brass cylinder with a valve at each end), through Marine Biological Laboratory 1 LIBRARY JUN241988 Woods Hole, Mass. the net to determine the vertical distribution of the larvae. Julius and Thurlow Nelson were thus able to make impor- tant contributions to the ecology of oyster larvae, and com- menced identification of larvae of other bivalves in New Jersey estuaries as well. These technological advances may seem modest enough, yet without them investigation of the larval ecology of oysters would not have been possible. Technological advances for the study of the distribution and movements of bivalve larvae in the field have been meager (Carriker 1967: Kasyanov et al. 1975; Scheltema 1975; Turner 1975; Mann 1986b). Spatial and temporal distribution are still monitored by pumping seawater into plankton nets from different strata in the water column. The laboriousness of this approach was amply demonstrated by Wood and Hargis (1971) in their ambitious, large scale, multiship study of bivalve larval transport in the James River. Continuous sampling with the Hardy plankton re- corder is impractical near shore because of shallow turbid waters, and because the sampler screens primarily hori- zontal strata. Coulter counter-like devices could be used in conjunction with a pump to sample subsurface waters, but are not useful because they do not distinguish different species, and likewise sample only a single stratum per in- strument. The research of Price et al. (1977) on automatic sorting of zooplankton could suggest a means for sorting bivalve veligers, but would be totally impractical in the identification of larvae even to the genus or family. The admirable investigations of Lutz et al. (1982) on the identi- fication of bivalve larvae using hinge structures will insure accuracy, but will not permit rapid identification of species even in the laboratory. Counting plantigrade bivalves is still done by laborious examination and sorting under the microscope in sediments excised from the bottom. A method for "automated" sorting is urgently needed. Some of us in the past have experimented with techniques using density gradients (see also Sellmer 1956) and washing and screening (Coffin and Welch 1964), without startling re- sults; and others are now experimenting with density gra- dient centrifugation, also not simple in practice. Perhaps an entirely different approach based on behavioral responses shortly after the time of collection in the field, can be de- vised to separate young bivalves from sediment samples. The use of larval traps (Carriker 1961) for determining the 1 Carriker density of pediveliger sets in the field has not received gen- eral acceptance. Innovations are critically needed in these areas of veligei and plantigrade samplings! In the course of evolution behavioral mechanisms un- doubtedly evolved that are based on estuarine circulation systems for retention of bivalve larvae within estuaries (Carriker 1967). Yet portions of estuarine larval popula- tions can be transported onto the continental shelf by spring ebbings tides (Carriker 1961), carried on flooding tides into adjacent estuaries, swept from the shelf out into the deep ocean, or even dispersed long distances in the ocean (Mann 1983, 1986b). Scheltema (1971), for example, indicated that shipworm larvae can be carried by ocean currents across ocean basins. The question of what mechanisms stimulate larvae, like those of Spisula solidissima (Dillwyn) and Arctica islamlia (Linne), to settle on specific bottoms of the continental shelf is still unanswered. These are intri- cate zoogeographic problems requiring investigation, par- ticularly with reference to the fate of displaced larvae. The behavioral and physiological mechanisms by which bivalve larvae maintain their vertical positions in the water are yet undefined. Evidence for mechanisms that could trigger selective swimming have not been experimentally demonstrated (Wood and Hargis 1971). The belief that velar activity continues throughout planktonic existence without cessation does not seem to be true in all cases. For example, the gas bubble described by Nelson (1928) sug- gests that at least one flotation device could exist that allows periodic rest; however, no one has yet shown how it is produced and maintained. Furthermore, although trocho- phores appear to swim continuously, many veligers (perhaps all?) alternate between periods of active upward swimming and periods of passive sinking with the velum either trailing or retracted within closed values (Mann and Wolf 1983; Mann 1986a; Carriker 1986). Vertical distribution of swarms of larvae and patches of food microorganisms do not necessarily coincide in the water column. Whether larvae can identify and swim to- ward food particles has been questioned for some time. There is growing evidence, however, that some post-meta- morphosed bivalves can preferentially ingest algae and re- ject particulate inorganic material in the pseudofeces (Ki0rboe et al. 1980; Newell 1982), and can qualitatively discriminate among different kinds of food particles through rejection tracts on the palps (Shumway et al. 1985); whether bivalve veligers also can, is an intriguing possibility that awaits exploration. Mann (1986b) noted that the speed of active vertical (upward) movement of dif- ferent species of bivalve larvae can range approximately from 1 to 10 mm/sec, which suggests that veligers could move relatively rapidly if attracted to metabolites from spe- cific food microorganisms. Multivariate experimental studies determining larval response to levels of two or more simultaneously applied environmental factors have been carried out (see. for example, Cain 1973 on combined ef- fects of temperature and salinity, and Bayne 1964 on light and gravity); that conditions of environmental factors at which parents develop gonads and spawn can influence the tolerance of embryos and larvae to these factors is unre- solved (Cain 1973) and bear examination. Acutely needed for field investigations is a method for continuously recording, three-dimensionally, the vertical position and density of bivalve larvae. On first thought, something like a computerized optical-electronic system might serve the purpose; I am told by colleagues, however, that looking for larvae acoustically in the field is not prac- tical because of problems of size, appropriate frequency, and transmission qualities of sea water. A new approach being developed by M. Pleass and D. Dey (personal com- munication, College of Marine Studies, University of Dela- ware) employing laser doppler for studies with microscopic planktonic organisms does appear promising — we await further developments with much interest. Another critical requirement in the field is for tagging large populations of bivalve larvae in order to follow their movements in the water and settlement onto the bottom. Tagging of laboratory spawned embryos with isotopes and discharging them in selected estuaries probably is feasible; but following tagged larvae in the open water could prove very difficult even with the resources of modern tech- nology, because of dilution problems resulting from dis- persal and losses from predation (Carriker 1961). Submers- ibles suggest the use of motion picture cameras with magni- fying optical systems for investigating the behavior of pediveligers while settling. There are serious problems with this, though, resulting from the compromise of magnifica- tion and depth of field in camera systems — to say nothing of the extraordinary problem of discriminating among dif- ferent species of bivalve larvae in mixed natural popula- tions. Perhaps all that can be done until further technolog- ical advances are made is to study pediveliger behavior in the laboratory in artificial systems (Prytherch 1934; Cran- field 1973b). employing both pediveligers captured in the field (as did Prytherch 1934) and those cultured in the labo- ratory. These "simple" studies have been highly informa- tive, and many more are worth conducting. Evidence is growing to show that behavioral swimming responses probably guide bivalve larvae into the vicinity of their natural benthic habitat, where settling behavior directs them to available favorable sites. Random settlement of bi- valve larvae on unsuitable substrata thus probably occurs less commonly (Carriker 1967; Crisp 1976; Gray 1974; Meadows and Campbell 1972). Crisp (1976) goes so far as to suggest that dispersal and habitat searching are part of an integrated pattern in which maintaining itself at a certain depth the larvae increases its chance of being carried in a favorable direction, light and gravity, among other factors, giving the larvae a reference orientation. Comparative field and laboratory studies on habitat selection (or on the be- havior that results in bivalves being found in particular Bivalve Larval Research places at particular times in nature — at best a difficult area of research) will have major importance in applied malaco- logical ecology (Meadows and Campbell 1972). The same may be said of investigations in controUed-en- vironment (closed) mariculture (Carriker 1976). Before bi- valves can be cultured efficiently and cost effectively their choice of habitat (Meadows and Campbell 1972; Crisp 1976). chemical and physical requirements (Bolton 1982; Gallagher and Mann 1986). preferred foods (Walne 1970; Epifanio 1976), extent of gregariousness (Hidu et al. 1978), and spawning (Morse et al. 1977) and settling (Morse et al. 1979) must be much better understood. Un- fortunately, very few comprehensive studies have been made on even commercial species. Also it's possible there could exist species of bivalves other than those currently utilized that could serve as well, or better, m closed culture (Harry 1985). Lack of the planktonic larval stage in a species, for example, could be a major practical advantage; two species that come to mind are Tiostrea liitcria; and T. chilensis. which are ready to attach to the substratum upon release from the parent, thus bypassing the planktonic stage (Chanley and Dinamani 1980). Mann (1983) urges, how- ever, that those contemplating culture of non-native species should consider carefully the potential spread of pest and disease organisms. Many species of bivalves, as shown by extensive re- search on benthic communities are restricted to certain types of substrata. Nevertheless, little research has been done, especially among infaunal bivalves, to identify the chemical and physical features that attract pediveligers to specific substrata during settlement and metamorphosis. Initial studies were carried out by Keck et al. (1974) on factors influencing settling by Mercenaria mercenaria (Linne) and by Veitch and Hidu (1971 ) on the properties of a partially purified settling factor for Crassostrea virginica. Studies by Morse et al. (1977, 1979) have increased the predictability of spawning (using hydrogen peroxide) and settlement and metamorphosis (using chemical inducers). These chemicals are effective for a number of bivalves. There is a paucity of information on the optimal velocity gradients under which pediveligers settle, or their orienta- tion with reference to currents during settling. We know that pediveligers search substrata for variable periods of time alternately crawling and swimming (Carriker 1967; Cranfield 1973b), but information is lacking on the time spent in these activities relative to different substrata, cur- rent velocity, temperature, salinity, suspended sediment, and competitors. Some species of plantigrades secrete a long byssus that carries young bivalves on relatively slow currents (Sigurdsson et al. 1976; Prezant and Chalermwat 1984). In other species, like Ostrea ediilis Linne (Cranfield 1973b). the pediveliger in the laboratory employs an elabo- rate pattern of byssal attachments prior to cementing its shell to the substratum; whether a similar behavior is ex- pressed in the field is unknown. Uncommon success has been achieved in the laboratory culture to settlement of the larvae of many bivalve species (for example, Chanley and Andrews 1971). Primary credit for the pioneer investigations in this field in the United States goes especially to Loosanoff and Davis (1963), in Great Britain to Cole (1937), Bruce et al. (1939), and Walne (1963), and in Japan to Imai and Hatanaka (1949). A substantial part of this achievement was contingent upon the successful culture of several species of algae for feeding the larvae. It is probable that there exist other species of microscopic algal foods better suited for specific maricul- tural purposes than those currently employed; a high-tem- perature tolerant tropical flagellate, for example, has been found recently and is in current use (Ewart and Epifanio 1981). The advent more recently of plastic containers, screens, seawater tubing and other products greatly simpli- fied the routine technological aspects of their work (Mann 1983). Invaluable, also, was the discovery that some bi- valves can be conditioned to spawn out of season (Price and Maurer 1971; Gallager and Mann 1986b) permitting culture of larvae throughout much of the year. This practice is now applied in many hatcheries and laboratories with, for ex- ample, oysters, clams, and scallops. Bivalves not spawned by standard stimuli (thermal shock, addition of gametes, salinity and pH changes, exposure to hydrogen peroxide) can be induced to spawn by injection of serotonin (Matsutni and Nomura 1982), a procedure used successfully by Gibbons et al. ( 1983) to spawn the refractory Arctica islan- dica (Linne). These important collective successes have paved the way for the operation of hatcheries and the cul- ture of bivalves in more or less controlled closed systems away from the sea (Dupuy and Rivkin 1972; Epifanio 1976, 1982; Dupuy et al. 1977; Castagna and Kraeuter 1981; Webb and Chu 1982; Wilson et al. 1984). Important research is in progress on identification of additional nutri- tional species of algae, combinations of nutritional species, manipulation of the chemical composition of algae through changes in cultural techniques, attempts to produce nutri- tional, formulated, encapsulated food particles (Epifanio 1976; Langdon 1983; Langdon and Bolton 1984). and the nutritional role of dissolved organic matter particularly free amino acids (Stephens 1982; Manahan and Crisp 1983). No artificial foods that nourish bivalve larvae as well as algae are yet available. 1 believe artificial feeds must be produced economically before commercial closed shellfish maricul- ture is successful. Addition of limited concentrations of particulate inorganic matter to laboratory cultures of bi- valve larvae has been shown experimentally to improve growth in oysters (Ali 1982). The mechanism whereby this occurs is unclear but merits investigation in view of the ubiquitous suspension of particles that surrounds bivalve larvae in nature particularly in estuaries (Carriker 1986) and the suggested potential importance of silts in bivalve larval culture. Successful laboratory cultivation of many species of bi- Carriker valve larvae has opened unprecedent opportunities for mul- tivariate experimentation in laboratory systems on behavior and movements in the water column of different bivalve larval stages in response to such ecological factors as gravity, salinity, light, temperature, pressure, external me- tabolites of algal food species, pheromones from parental adults, and chemicals from various kinds of bottom sub- strata. Such studies have begun to appear (for example, Haskin 1964; Bayne 1963, 1964, 1973; Hidu and Haskin 1978; Mann and Wolf 1983). This is a rich field for phys- ioecological, behavioral experimentation, the results of which could be highly beneficial to closed as well as field mariculture of bivalves. Cultured larvae are also excellent subjects for the investigations of the effect of various frac- tions of oils, pesticides, and heavy metals (Mileikovsky 1970; Calabrese et al. 1973). The emerging technology of closed aquatic systems should make possible long-range studies of the effects of sublethal concentrations of these chemicals. Information thus gained, however, should be applied to natural field systems with caution until we know better whether laboratory culture modifies the tolerance of larvae to pollutants. Significant studies of the macromorphology, micro- structure, physiology, behavior, and biochemistry of bi- valve larvae are appearing with increasing frequency. No- table, by way of examples, are publications of Cranfield (1973a, b, c, 1974, 1975) on glands of the foot and mantle folds of the pediveliger relative to settlement; Cragg and Nott (1977) on statocysts; Carriker and Palmer (1979) and Waller (1981) on the valves; Elston (1980) on soft tissues; Crisp (1967) on chemical factors and settling; Holland and Spencer (1973) on biochemical changes; Bayne et al. (1975) on effects of stress; and Gallager and Mann (1981) on lipid staining as an indicator of health. These investiga- tions have established a noteworthy trend for structural- functional studies in bivalve larvae which will enhance basic and applied larval malacology. Since the pioneer investigations of the Nelsons during the early years of the century, new substantive knowledge on bivalve larvae has been appearing at scientific meetings and in technical journals at an accelerating rate. This augurs well for future advances required to accomodate the difficult, but necessary, shift from field to closed system bivalve mariculture — a move necessitated by the growing inhospitality of conditions in coastal waters. This urgent trend requires essential new biological and technological information beyond what is now available. Studies touched upon briefly in this commentary suggests some areas defi- cient in information as well as some new directions for re- search. 1 have drawn attention to them to stimulate further creative basic-applied research on bivalve larvae and asso- ciated technology requisite for a successful transition in bi- valve molluscan mariculture. REFERENCES CITED Ali, S. M. 1982. Effect of natural silt on oyster growth. In: Proceedings of the Second International Conference on Aquaculture Nutrition: Bio- chemical and Physiological Approaches to Shellfish Nutrition. CD. Pruder, C. J. Langdon and D. E. Conklin (Eds.). Louisiana State Uni- versity, Baton Rouge, pp. 431-432. Bayne, B. L. 1963. Responses of Mytilus edulis larvae to increases in hydrostatic pressure. Nature. Lond. 198:406-407. Bayne, B. L. 1964. The responses of the larvae of Mytilus edulis L. to light and gravity. Oikos 15:162-174. Bayne, B. L. 1973. The responses of three species of bivalve mollusc to declining o.xygen tension at reduced salinity. Comp. Biochem. Physiol. 45A:793-806. Bayne, B. L.. P. Gabbott & J. Widdows 1975, Some effects of stress in the adult on the eggs and larvae of Mytilus edulis L. J. Mar. Biol. Ass. U.K. 55:675-689. Bolton, E. T., Ed. 1982. Intensive Marine Bivalve Cultivation in a Con- trolled Recirculating Seawater Prototype System. University of Dela- ware Sea Grant College Program, Newark, Delaware, DEL-SG-07-82. 165 pp. Bruce, J. R., M. Kmght & M. W. Parke. 1939. The reanng of oyster larvae on an algal diet. J . Mar. Biol. Ass. U.K. 24:337-374. Cain, T. D. 1973. The combined effects of temperature and salinity on embryos and larvae of the clam Rangia cuneate. Mar. Biol. 21:1-6. Calabrese, A., R. S. Collier, D. A. Nelson & J. R. Macinnes. 1973. The toxicity of heavy metals to embryos of the American oyster, Crassos- trea virginica. Mar. Biol. 18:162-166. 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. Carriker, M. R. 1967. Ecology of estuarine benthic invertebrates: a per- spective. In: Estuaries. G. H. Lauff (Ed.). Publ. No. 83, Amer. Ass. Adv. Set.. Washington. DC, pp. 442-486. Carriker, M. R. 1976. Opening comments. In: Proceedings of the First International Conference on Aquaculture Nutrition. K. S. Price, W. N. Shaw, and K. S. Danberg (Eds.). College of Marine Studies, University of Delaware, Newark, pp. 7-12. Carriker, M. R. 1986. Influence of suspended particles on biology of oyster larvae in estuaries. Am. Malacol. Bull.. Spec. Ed. No. 3:41-49. Carriker, M. R. & R. E. Palmer. 1979. Ultrastructural morphogenesis of prodissoconch and early dissoconch valves of the oyster Crassostrea virginica. Proc. Nat. Shellfish. Ass. 69:103-128. Castagna. M. & J. N. Kraeuter. 1981. Manual for growing the hard clam Mercenaria. Spec. Rep. Applied Mar. Sci. Ocean Engineer. No. 249. Virginia Institute of Manne Science, Gloucester Point. 1 10 pp. Chanley, P. & J. D. Andrews. 1971. Aids for identification of bivalve larvae of Virginia. Malacologia 1 1:45- 1 19. Chanley, P. & P. Dinamani. 1980. Comparative descriptions of some oyster larvae from New Zealand and Chile, and a description of a new genus of oyster, Tiostrea. New Zealand J. Mar. Freshw. Res. 14:103-120. Coffin, G. W. & W. R. Welch. 1964. A technique for separating small moUusks from bottom sediments. Proc. Nat. Shellfish. Ass. 53:175- 180. Cole, H. A. 1937. Experiments in the breeding of oysters {Ostrea edulis) in tanks, with special reference to the food of the larva and spat. Fish. Invest.. London (Ser. 2) 15:1-28. Cragg, S. M. & J. A. Nott. 1977. The ultrastructure of the statocysts in the pediveliger larvae of Pecten maximums (L.) (Bivalvia). J. Exp. Mar. Biol. Ecol. 27:23-36. Bivalve Larval Research Cranfield. H, J. 1973a. A study of the nioiphology. ultrastmcture. and histochemisti7 of the foot of the pediveliger of Ostreu edulis. Mar. Biol. 22:187-202. Cranfield. H. J. 1973b. Observations on the behavior of the pediveliger of Ostreci ediilis during attachment and cementing. Mar. Biol. 22:203- 209. Cranfield. H. J. 1973c. Observations on the function of the glands of the foot of the pediveliger of Oslrea edulis during settlement. Mar. Biol. 22 21 1 -223 Cranfield. H. J. 1974. Observations on the morphology of the mantle folds of the pediveliger of Oslrea edulis L. and their function dunng settlement. J. Mar. Biol. Ass. U.K. 54:1-12. Cranfield. H.J. 1975. The ullrastructure and histochemistry of the larval cement of Ostrea edulis L. J. Mar. Biol. Ass. U.K. 55:497-503. Crisp. D. J. 1967. Chemical factors inducing settlement in Crassostrea virginiia (Gmelinl. J. Anim. Ecol. 36:329-335. Crisp, D. J. 1976. Settlement responses in marine organisms. In: Adapta- tion to Environment: Essays on the Physiology of Marine Animals. R. C. Newell (Ed.). Butterworths. London, pp. 83-124. Dupuy. J. L. & S. Rivkin. 1972. The development of laboratory tech- niques for the production of cultch-free spat on the oyster. Crassostrea i(ri,'(/»((/. Chesapeake Sci . 13:45-52. Dupuy, J. L., N. T. Windson & C. E. Sutton. 1977. Manual for design and operation of an oyster seed hatchery for the American oyster Crassostrea virginica. Spec. Rep. Applied Mar. Sci. Ocean Engineer. No. 142. Virginia Institute of Marine Science, Gloucester Point. 104 pp. Elston. R. 1980. Functional anatomy, histology, and ullrastructure of the soft tissues of the larval American oyster. Crassostrea virginica. Proc. Nat. Shellfish. Ass. 70:65-93. Epifanio, C. 1976. Culture of bivalve moUusks in recirculating systems: nutritional requirements. In: Proceedings of the First National Confer- ence on Aquacultiire Nutrition. K. S. Price. W. N. Shaw, and K. S. Danberg (Eds.). College of Marine Studies, University of Delaware, Newark, pp. 173-194. Epifanio, C. E. 1982. Phytoplankton and yeast as foods for juvenile bi- valves, a review of research at the University of Delaware. In: Pro- ceedings of the Second International Conference on Aquactdture Nu- trition: Biochemical and Physiological Approaches to Shellfish Nutri- tion. G. D. Pruder, C. J. Langdon. and D. E. Conklin (Eds.) Louisiana State University, Baton Rouge pp. 292-304. Ewart, J. W. & C. E. Epifanio. 1981 . A tropical flagellate food for larval and juvenile oysters, Crassostrea virginica Gmelin. Aquaculture 22:297-300. Gallager, S. M. & R. Mann. 1981. Use of lipid-specific staining tech- niques for assaying condition in cultured bivalve larvae. J . Shellfish. Res. 1:69-73. Gallager, S. M. & R. Mann. 1986. Growth and survival of larvae oi Mer- cenaria mercenaria and Crassostrea virginica relative to brookstock conditioning and lipid content of eggs. Aquactdture 56:105-121. Gibbons, M. C, J. G. Goodsell, M. Castagna & R. A. Lutz. 1983. Chemical induction of spawning by serotonin in the ocean quahog Arctica islandica (Linnc). J. Shellfish Res. 3:203-205. Gray, J. S. 1974. Animal-sediment relationships. Oceanogr. Mar. Biol. Ann. Rev. 12:223-261. Harry, H. W. 1985. Synopsis of the supraspecific classification of living oysters (Bivalvia: Gryphaeidae and Ostreidae). Veliger 28:121-158. Haskin, H. H. 1964. The distribution of oyster larvae. In: Symposium on Experimental Marine Ecology. N. Marshall, H. P. Jeffries, T. A. Na- pora and J. M. Sieburth (Eds.). Occ. Publ. No. 2. Graduate School of Oceanography, University of Rhode Island, pp. 76-80. Hidu, H. & H. H. Haskin. 1978. Swimming speeds of oyster larvae Crassostrea virginica in different salinities and temperatures. Es- tuaries 1:252-255. Hidu, H., W. G. Valleau & F. P. Veitch. 1978. Gregarious setting m European and American oysters-response to surface chemistry vs. wa- terboume phenomena. Proc. Nat. Shellfish. Ass. 68:11-16. Holland, D. L. & B. E. Spencer. 1973. Biochemical changes in fed and starved oysters, Oslrea edulis L., during larval development, meta- morphosis and early spat growth. J. Mar. Biol. Ass. U.K. 53:287- 298. Imai, T. & M. Hatanaka. 1949. On the artificial propagation of Japanese common oyster, Ostrea gigas Thunberg, by non-colored naked fiagel- lates. Btdl. Inst. Agr. Res. Tohoku Univ. 1:1-7. Kasyanov, V. L. & V. A. Kulikova. 1975. Reproduction of marine mol- luscs; a review of Soviet works. In: The Ecology of Fouling Communi- ties. J. D. Costlow (Ed). Duke University Manne Laboratory, Beau- fort, North Carolina, pp. 111-129. Keck, R., D. Maurer & R. Malouf. 1974. Factors influencing the setting behavior of larval hard clams, Mercenaria mercenaria. Proc. Nat. Shellfish. Ass. 64:59-67. Ki0rboe, T., F Mohlenberg & O. N0hr 1980. Feeding, particle selection and carbon adsorption in Mvtilus edulis in different mixtures of algae and resuspended bottom sediment. Ophelia 19:193-205. Loosanoff, V. L. & H. C. Davis. 1963. Reanng of bivalve mollusks. Adv. Mar. Biol. 1:1-136. Langdon, C. J. 1983. Growth studies with bacteria-free oyster (Crasso- strea gigas) larvae fed on semi-defined artificial diets. Biol. Bull. 164:227-235. Langdon, C. J. & E. T. Bolton. 1984. A microparticulate diet for a sus- pension-feeding bivalve mollusc, Crassostrea virginica (Gmelin). J. Exp. Mar. Biol. Ecol. 82:239-258. Lutz, R.. J. Goodsell, M. Castagna, S. Chapman, C. Newell, H. Hidu, R. Mann, D. Jablonski, V. Kennedy, S. Siddall, R. Goldberg, H. Beattie, C. Falmagne, A. Chestnut & A. Partndge. 1982. Preliminary observations on the usefulness of hinge structures for identification of bivalve larvae. J. Shellfish. Res. 2:65-70. Manahan, D. T. & D. J. Cnsp. 1983. Autoradiographic studies on the uptake of dissolved amino acid by bivalve larvae. J. Mar. Biol. Ass. U.K. 63:673-682. Mann. R. 1983. Bivalve mollusc hatcheries: a critical appraisal of their development and a review of their potential value in enhancing the fisheries of developing nations. Mcms. Asoc. Latinoam. Acuicult. 5:97-105. Mann. R. 1986a. Arctica islandica (Linnel larvae: active depth regulators of passive particles. Am. Malacol. Bull.. Spec. Ed. 3 (19861:51-57. Mann. R. 1986b. Sampling of bivalve larvae. Canadian Spec. Publ. Fish. Aquatic Sci. 92:107-116. Mann. R. & C. C. Wolf. 1983. Swimming behaviour of larvae of the ocean quahog Arctica islandica in response to pressure and tempera- ture. Mar. Ecol. Progr. Ser. 13:211-218. Matsutani. T. & T. Nomura. 1982. Induction of spawning by serotonin in the scallop, Patinopecien yessoensis (Jay). Mar. Biol. Lett. 3:353- 358. Meadows, P. S. & J. I. Campbell. 1972. Habitat selection by aquatic in- vertebrates. Adv. Mar. Biol. 10:271-382. Mileikovsky, S. A. 1970. The influence of pollution on pelagic larvae of bottom invertebrates in marine nearshore and estuarine waters. Mar. Biol. 6:350-356. Mileikovsky, S. A. 1973. Speed of acfive movement of pelagic larvae of marine bottom invertebrates and their ability to regulate their vertical position. Mar. Biol. 23:11-17. Morse. D. E.. H. Duncan. N. Hooker & A. Morse. 1977. Hydrogen per- oxide induces spawning in mollusks. with activation of prostaglandin endoperoxide synthetase. Science 196:298-300. Morse. D. E.. N. Hooker. H. Duncan & L. Jensen 1979. Gamma ami- nobutyric acid, a neurotransmitter, induces planktonic abalone larvae to settle and begin metamorphosis. Science 204:407-410. Nelson. J. 1909. Report of the Biological Department of the New Jersey Agricultural College Experiment Station for the year 1908. New Brunswick. New Jersey, pp. 149-178. Carriker Nelson, T. C. 1928. Pelagic dissoconchs of the common mussel, Mytilus ediiUs, with obsei-vations on the behavior of the larvae of allied genera. Biol. Bull. 55:180-192. Newell, R. I. E. 1982. Molluscan bioenergetics — a synopsis. Pro- ceedings of the Second International Conference on Aquaciiltiire Nu- trition: Biochemical and Physiological Approaches to Shellfish Nutri- tion. G. D. Pruder, C. J. Langdon and D. E. Conklin (Eds.). Loui- siana State University, Baton Rouge, pp. 252-271. Prezant. R. S. & K. Chalermwat. 1984. Flotation of the bivalve Cor- bicula fluininea as a means of dispersal. Science 225:1491-1493. Price, C. A., J. M. St. Onge-Bums, J. B. Colton, Jr. & J. E. Joyce. 1977. Automatic sorting of zooplankton by isopycnic sedimentation in gradients of salica: performance of a "rho spectrometer". Mar. Biol. 42:225-231. Price, K. & D. Maurer. 1971. Holding and spawning Delaware Bay oysters (Crassostrea virginica) out of season. II. Temperature require- ments for maturation of gonads. Proc. Nat. Shellfish. Ass. 61:29-34. Prytherch, H. P. 1934. The role of copper in the setting, metamorphosis, and distribution of the American oyster, Ostrea virginica. Ecol. Monogr. 4:47-107. Scheltema, R. S. 1971. Dispersal of phytoplankton shipworm larvae (Bi- valvia: Teredinidae) over long distances by ocean currents. Mar. Biol. 11:5-11. Scheltema, R. S. 1975. The significance of pelagic larval development to marine fouling organisms. In: The Ecology of Fouling Communities. J. D. Costlow (Ed.). Duke University Marine Laboratory, Beaufort, North Carolina, pp. ll-Al . Sellmer, G. 1956. A method for the .separation of small bivalve molluscs from sediments. Ecology 37:206. Shumway, S. E., T. Cucci, R. C. Newell & C. M. Yentsch. 1985. Par- ticle selection, ingestion, and absorption in filter-feeding bivalves. J . E.xp. Mar. Biol. Ecol. 91:77-92. Sigurdsson, J. B., C. W, Titman & P. A. Davies. 1976. The dispersal of young post-larval bivalve mollusks by byssus threads. Nature 262:386-387. Stephens, G. C. 1982. Dissolved organic material and the nutrition of marine bivalves. In: Proceedings of the Second International Confer- ence on Aquaculture Nutrition: Biochemical and Physiological Ap- proaches to Shellfish Nutrition. G. D. Pruder, C. J. Langdon and D. E. Conklin (Eds.). Louisiana State University, Baton Rouge, pp. 338-357. Sverdrup, H. U., M. W. Johnson & R. H. Fleming. 1942. The Oceans, their Physics. Chemistry, and General Biology. Prentice-Hall, New York. 1087 pp. Turner. R. D. 1975. Bivalve larvae, their behavior, dispersal and identifi- cation. In: The Ecology of Fouling Communities. J. D. Costlow (Ed). Duke University Marine Laboratory, Beaufort, North Carolina, pp. 23-25. Veitch, F. P. & H. Hidu. 1971. Gregarious setting in the Aniencan oyster Crassostrea virginica Gmelin: I . Properties of a partially purified '"setting factor". Chesapeake Sci. 12:173-178. Waller, T. R. 1982. Larval settlement behavior and shell morphology of Malleus candeanus (d'Orbigny) (Mollusca: Bivalvia). In: The Atlantic Barrier Reef Ecosystem at Carrie Boy Cay. Belize. 1. Structure and Communities. K. Riitzler and I. G. Macintyre (Eds.). Smithsonian Contr. Mar. Sci. 12:489-497. Walne, P. R. 1963. Observations on the food value of seven species of algae to the larvae of Ostrea edulis. 1. Feeding e.xpenments. J. Mar. Biol. Ass. U.K. 43:767-784. Walne, P. R. 1970. Studies on the food value of nineteen genera of algae to juvenile bivalves of the genera Ostrea. Crassostrea. Mercenaria. and Mytilus. Fishery Invest.. Land. Ser. 2, 25:62 pp. Webb, K. L. & F. E. Chu. 1982. Phytoplankton as a food source for bivalve larvae. In: Proceedings of the Second International Confer- ence on Aquaculture Nutrition: Biochemical and Physiological Ap- proaches to Shellfish Nutrition. G. D. Pruder, C. J. Langdon and D. E. Conklin (Eds). Louisiana State University, Baton Rouge, pp. 272-291. Wilson, J., J. Simons & E. Noonan. 1984. A manual for the construction and operation of a simple oyster hatchery. Aquaculture Tech. Bull.. Ireland 8:75 pp. Wood, L. & W J. Hargis, Jr. 1971. Transport of bivalve larvae in a tidal estuary. In: Fourth European Marine Biology Symposium. J. D. Crisp (Ed.). Cambridge University Press, pp. 29-44. Journal of Shellfish Research. Vol. 7, No. 1 . 7- 10. 1988. FIELD STUDIES OF BIVALVE LARVAE AND THEIR RECRUITMENT TO THE BENTHOS: A COMMENTARY ROGER MANN Virginia Institute of Marine Science School of Marine Science College of William and Mary Gloucester Point. VA 23062 ABSTRACT A list of factors influencing the recruitment of bivalve larvae might include, but not be limited to. the following: egg quality, physical environment, food availability, loss to predation and disease dunng larval development, interplay of passive dis- persal (honzontally) by water currents and depth regulation by active swimming, proximity of suitable and available substratum as metamorphic competency is achieved, and availability of sufficient metabolic reserves to complete metamorphosis to the benlhic form. While tractable methods exist to quantify aspects of certain members of the above list, the focus of such work has usually been biased towards laboratory experiments or hatchery production. The purpose of this commentary is to suggest that a refocussing of efforts in bivalve larval biology on natural systems is both timely and needed. KEY WORDS: Bivalve larvae, recrtiitnient COMMENTARY There is no question that laboratory work has allowed us to make many advances in the understanding of bivalve larval ecology; however, this work has often focussed on the culture of bivalves for economic purposes rather than examination of interesting ecological questions per se. In- deed, it was the intent of the original '" laboratory "" work of Brooks (1890) not to provide greater understanding of the ecology of oyster larvae but to provide a culture method for that species so that a repopulation of the Chesapeake Bay could be effected. Nonetheless the products of years of lab- oratory work suggest that we should again seriously con- sider field prograins to examine larval biology and recruit- ment to the benthos. After several years of editing the Journal of Shellfish Research and even more years of reading the literature relating to bivalve larval biology an impression remains that the option for intensive field work is usually countered by comments such as: "too difficult, too much variability, too little control, and too time con- suming." Consequently, we remain in the laboratory. A list of factors which influence recruitment — here de- fined as successful metamorphosis from the pelagic pedive- liger larva (sensu Carriker 1961) to the benthic, generally attached, feeding juvenile — of bivalve larvae might in- clude, but not be limited to, the following: egg quality as influenced by the availability of food to the parent or- ganisms, physical environment and food availability during larval development, the interplay of passive dispersal (hori- zontally) by water current and depth regulation by active swimming, loss to predation and disease, proximity of suit- able and available substratum as metamorphic competency is achieved, and availability of sufficient metabolic re- serves to complete metamorphosis to the attached benthic form. The list is not intended to be definitive or infer that the factors are listed in order of importance. It is, however. comprehensive. I wish to proceed through this list and demonstrate that we have the ability to quantify (to a vari- able degree) all of these factors. Consequently, it seems reasonable to suggest that we attempt such a quantification in a known field situation as part of a comprehensive exam- ination of larval survival. I know of no case where this has been attempted for bivalve larvae. The first item to be considered is egg quality. It has been documented for some time that a strong relationship exists between broodstock condition and larval viability in the flat oyster, Ostrea edulis L. (Helm, Holland and Stephenson 1973). More recently, Gallager and Mann (1986) have demonstrated similar strong relationships between brood- stock condition and lipid contents of eggs in both Merce- naria mercenaria L. and Crassostrea virginica Gmelin. We offer a simple technique, based on the lipid specific stain Oil-Red-0, for assessing egg lipid content. Although developed and used in both the laboratory and commercial hatcheries, there is no reason why this cannot be used for field collected specimens. Indeed, we examined bivalve larvae using this technique in a preliminary manner during a field study of larval distribution on the Southern New England Shelf in 1981 and found it tractable and informa- tive. The second item is the physical environment during larval development. There is a considerable volume of lit- erature on this subject although it is not always presented in a manner that is easily interpreted when attempting to apply the laboratory generated data sets to field situations. The data should be examined and used in models of field situa- tions. Here, I offer two such examples. Lough (1974) ex- amined the data of Brenko and Calabrese (1969) on the influence of temperature and salinity on Mytilus edulis L. larvae using response surface techniques. Immediately evi- dent from this approach is the optimal physical environ- ment for growth and survival. Yet. this approach is rarely 8 Mann used. It is simple to inteq^ret these data in concert with temperature and sahnity values from the field. By contrast the tabular data of Davis and Calabrese ( 1964) for Crassos- trea virginica Gmelin and Mercenaria mercenaha L., al- though informative, are considerably more difficult to use. An alternative approach, one that 1 have used in modelling occurrence and growth oi Arctica islandica L. larvae on the New England Shelf (Mann, 1986a), involves stepwise inte- gration of such data into more complex models. 1 will ad- dress this in my discussion of larval dispersal later in the text. When discussing the physical environment for devel- oping larvae it is also relevant to include the presence of toxic materials. These may originate from natural sources, for example the exudates of blooms of the microorganism Phaeocystis pouchetii. or from waste disposal activities. In coastal areas adjacent to urban development the latter can be alarming in volume and variety of composition. None- theless progress is being made by toxicologists in quanti- fying the impact of selected toxic materials on larval mol- luscs. The third item is food availability. Even though we can culture larvae in the laboratory on diets of phytoplankton, there is still no definitive statement on what larvae can and cannot eat in the field. How do we determine if enough food is present? Examine a worst case scenario; exclude dissolved organic carbon (D.O.C.), which Manahan (1983a, 1983b) and Manahan and Crisp (1982, 1983) have shown to be available for use by invertebrate larvae, and exclude non-phytoplankton particulate organic carbon (P.O.C.). The latter may be considerable: for example, work by Hugh Ducklow at the University of Maryland has shown that in the upper Chesapeake Bay bacterial biomass may be equal to that of phytoplankton. This leaves only phytoplankton in our examination. If larvae can survive on this, they can certainly survive when all the other carbon sources are also made available to them. Mann (1985) offers a series of calculations examining food availability at a station on the New England Shelf — a station where chlo- rophyll a concentration is probably well below that of in- shore and estuarine regions where oyster and clam larvae are expected to grow and metamorphose. The calculation is simple and the result suggests that an estimated standing stock of cell concentrations in the range 0.54 cells/fjil (ob- tained using very conservative conversion factors) to 67.7 cells/|JLl (using more reasonable conversion factors) is present during the summer and fall in the waters of the shelf environment. With the exception of the lowest esti- mates (0.54 cells/|xl) of food concentration there is gener- ally enough food present for larval development based upon laboratory estimates of bivalve larval requirements (see Walne 1965; de Schweinitz and Lutz 1976; Lutz et al. 1982). In essence we need worry only about atypical rather than typical events with respect to food impacting larval survival. As an example here, I offer the "brown tide" phenomena which Southern New England and Long Island have recently experienced — essential monocultures of ap- parently unpalatable phytoplankton. My point, however, is that it is generally difficult to make an argument that food quantity is ever limiting to larval growth. The fourth item is larval dispersal. Is this an active or passive process? I have recently addressed this subject (Mann 1986a) and reviewed the literature (Mann 1986b). In regions of intense vertical mixing the weak swimming ability of larvae is overwhelmed and dispersal is passive. Consequently, if you want to know where the water (and therefore the larvae) is going you must consult your friendly, local physical oceanographer. To quote Andrews (1979); "Usually hydrographic regimes have not been known or appreciated to plan sampling of larvae." Fortu- nately, the trend toward active development of programs in collaboration with physical oceanographers is changing rapidly. In coastal systems seasonal stratification can be in- tense irrespective of whether estuaries or the inner shelf is being examined. In such regions, larval behaviour, a com- ponent that can be easily quantified in the laboratory, can be important and is amenable to modelling. The models can also be tested for validity in the field. The point that 1 wish to make is that we can use simple laboratory experiments in conjunction with field data, both physical and biological, to build testable computer models of larval dispersal. Physical scientists are progressing in the development of three di- mensional, finite difference models of currents and sedi- ment transport in coastal regions (see Sheng 1983). The modelling of sediment particle dynamics has many analo- gies with the modelling of larval behaviour. The problem is large but tractable and we, as bivalve ecologists, should address it. The fifth item is disease and predation. We have a host of methods to examine disease in the stressful environment of a commercial hatchery operation (see Elston 1984 and references therein). While not all of these can be easily uti- lized on field collected specimens, due to small numbers of larvae collected, observational techniques such as electron microscopy can be used and draw upon the data provided by laboratory culture procedures. Castagna (personal com- munication) comments that in laboratory cultures signifi- cant numbers of larvae fail to metamorphose or develop very slowly. In the field these larvae would have increased susceptibility to predation. In a review by Gibbons and Blogaslawski (in press) a listing of predators on larvae in- clude Aurelia, Balanus, Brevortia, Chrysaora, Chtha- lamus. Diadume, Mnemiopsis. Noctiluca, Polydora, Sphaeroides and a host of filter feeding bivalves and fish. Such impacts are potentially quantifiable using a combina- tion of laboratory experiments and field collections. It would be particularly profitable here to coordinate efforts with larval fish ecologists (whose activities are consider- able in the coastal regions) interested in fish feeding, diet and stomach contents. The sixth item is substratum availability. Certain bi- Recruitment of Bivalve Larvae 9 valves, notably oysters, exhibit substratum specificity. While the practice of provision of substratum to enhance settlement of commercially valuable bivalves can be traced back to Roman times and the writings of Plinius, and has been practiced extensively since the 1850's on the U.S. east coast, surprisingly (appallingly) little quantitative in- formation exists on the fate of that substratum, over time, and its availability as a substratum to oysters in the face of competition for that substratum by what we term " 'fouling" " species. In 1985 Richard Rheinhardt and I attempted to quantify the temporal and spatial development of fouling communities on clean shell substratum in the James River, Virginia. Our focus was, in part, to provide managers with a time window for optimal planting of shell to maximize oyster larval settlement and minimize prevention of settle- ment by fouling organisms. We used point sampling tech- niques to quantify our data — again illustrating that we must be prepared to look outside of our classical discipline to seek guidance from others in developing our field. The resultant manuscript is in review; however, to summarise, we illustrate that differences in rate of development and extent of areal coverage of fouling communities can be quantified. We also demonstrated that changes in the com- munity structure could be elucidated using detrended corre- spondence analysis (Hill and Gaugh 1980). Examination of the predominant fouling species over time can give some insight into their potential impact on settlement of bivalves on adjacent, available substratum. The final item is the assessment of whether or not mor- phologically competent-to-metamorphose larvae have suf- ficient energy reserves to complete that same metamor- phosis. It is now accepted that metamorphosis is an energy consuming and thus critical period of the life cycle for a multitude of marine fishes and invertebrates. The impor- tance of lipid reserves lo this process in bivalves has been reported by Gallager, Mann and Sasaki (1986). As with egg quality we demonstrate that larval quality, including pediveliger larvae, can easily be assayed using a lipid spe- cific stain. As I noted earlier, this technique is both simple and quantifiable. It can and has been used in the field and on other species. We have no excuse not to examine the viability of larvae in the field. In summary then, I hope that this commentary has con- vinced you that we have at our disposal viable methods to examine many of the factors influencing larval survival and recruitment in the field. It is time to address the problem at hand. ACKNOWLEDGMENTS Preparation of this document was supported by the Vir- ginia Institute of Marine Science. The stimulus to finally write on the subject came as an invitation to present a paper at the Seventh Annual Shellfish Workshop at Milford, Con- necticut in March. 1987. Jay D. Andrews. Michael Cas- tagna, Mary Gibbons and David Stilwell critically reviewed an early draft of the text. Thanks are also given to NOAA Sea Grant. ONR and NSF Biological Oceanography who supported much of the work described above and in which I was fortunate enough to participate. Contribution Number 1430 from the Virginia Institute of Marine Science, Col- lege of William and Mary. REFERENCES CITED Andrews, J. D. 1979. Pelecypoda: Ostreidae. In: A. C. Giese and J. S. Pearse. eds., Reprodiiclion of Marine Invertebrates. Volume 5: Pele- cypods and Lesser Classes. Academic Press. N.Y. p. 293-339. Brenko. M. H. & A. Calabrese. 1969. The combined effects of salinity and temperature on larvae of the mussel. Mytiliis ediilis. Mar. Biol. 4:224-226. Brooks, W, K. 1890. The Oyster. Johns Hopkins Press. Baltimore. 225 pp. Camker, M. R. 1961. Interrelation of functional morphology, behaviour and autecology in early stages of the bivalve Mercenaria mercenaria. J. Elisha Mitchell Sclent. Soc. 177:168-242. Davis, H. C. & A. Calabrese. 1964. Combined effects of temperature and salinity on development of eggs and growth of larvae of Mercenaria mercenaria and Crassostrea virginica. U.S. Fish. Wildl. Sen-. Bull 63:643-655. Elston, R. A. 1984. Prevention and management of infectious diseases in intensive mollusc husbandry. J. World Marie. Soc. 15:284-300. Gallager, S. M. & R. Mann. 1986. Growth and survival of larvae of Mer- cenaria mercenaria and Crassostrea virginica relative to broodstock conditioning and lipid content of eggs. Aquaculture 56(2):I05- 121. Gallager, S. M., R. Mann & G. C. Sasaki. 1986. Lipids as an index of growth and viability in three species of bivalve larvae. Aquaculture 56(2):8I-103. Gibbons. M, C, & W J, Blogaslawski. (in press) Predators, pests, para- sites and diseases. In: J. J. Manzi and M. Castagna, eds.. Clam Cul- ture in North America. Elsevier, N.Y. p 167-200. Helm. M. M., D. L. Holland & R. R. Stephenson. 1973. The effect of supplementary algal feeding of a hatchery breeding stock of Oslrea edulis L. on larval vigour. J. Mar. Biol. Assoc. U.K. 53:673-684. Hill. M. O. & H. G. Gaugh, Jr. 1980. Detrended correspondence anal- ysis: an improved ordination technique. Vegetatio 42:47-58. Lutz. R. A., R. Mann, J. G. Goodsell & M. Castagna. 1982. Larval and early post larval development of the ocean quahog Arctica islandica. J. Mar. Biol. Assoc. U.K. 62:745-769. Lough, R. G. 1974. A re-evaluation of the combined effects of tempera- ture and salinity on survival on growth of Mytilus edulis larvae using response surface techniques. Proc. Natl. Shellfish. Assoc. 64:73-76. Manahan. D, T. 1983a. The uptake and metabolism of dissolved amino acids by bivalve larvae. Biol. Bull (Woods Hole) 164:250-263. Manahan, D. T. 1983b. The uptake of dissolved glycine following fertil- ization of oyster eggs {Crassostrea gigas Thunberg). J. Exp. Mar. Biol. Ecol. 68:53-58. Manahan, D. T. & D. J. Cnsp. 1982. The role of dissolved organic mate- rial in the nutrition of pelagic larvae: Amino acid uptake by bivalve veligers. Am. Zoot. 22:635-646. Manahan. D. T. & D. J. Crisp. 1983. Autoradiographic studies on the uptake of dissolved amino acid by bivalve larvae. J. Mar. Biol. Assoc. U.K. 63:673-682. 10 Mann Mann, R. 1985. Seasonal changes in the depth distribution of bivalve larvae on the Southern New England Shelf. J. Shellfish Res. 5(2):57- 64. Mann, R. 1986a. Arclica islaiulua (Linnel larvae: Active depth regulators or passive particles? ,4m. Mai. Bull. Spec. Eel. No. .?:5l-57. Mann, R. 1986b. Sampling of bivalve larvae. /«.■ G. S. Jamieson and N. Bourne, eds.. North Pacific Workshop on Stock Assessment and Man- agement of Invertebrates. Can. Spec. Pub. Fish. Aquat. Sci. 92:107- 116. de Schweinitz, E. H. & R. A. Lutz. 1976. Larval development of the northern horse mussel Modiolus modiolus (L.) including a comparison with the larvae of Mytilus edulis L. as an aid in planktonic identifica- tion. Biol. Bull. (Woods Hole) 150(3): 348-360. Sheng, Y. P. 1983. Mathematical modeling of three dimensional coastal currents and sediment dispersion: model development and application. Tech. Rep. CERC-83-2, Contract No. DACW39-80-C-0087, U.S. Army Corp. Engineers, Washington, D.C. 288pp. Walne, P. R. 1965. Observations on the influence of food supply and temperature on the feeding and growth of the larvae of Ostrea edulis L. Fishery Imesi.. Lond., Ser. 2, 24(1). 45pp. Journal of Shellfish Research. Vol. 7, No. 1, 11-18. 1988. AN EVALUATION OF HEMOLYMPH DIAGNOSIS FOR DETECTION OF THE OYSTER PARASITE HAPLOSPORIDIUM NELSONI (MSX) SUSAN E. FORD AND SHEILA A. KANALEV Shellfish Research Lahoraioiy New Jersey Agricultural Experiment Station Cook College. Rutgers University Port Norris. New Jersey 08349 ABSTRACT Use of hemolymph to diagnose Huplosporidium nelsoni (MSX) infections in the oyster Crassoslrea virginica was evaluated by compaiing results obtained from both fresh and fixed/stained hemolymph preparations with the standard histological method employing tissue sections. Fixed/stained hemolymph preparations detected 95-98'7t of all systemic infections, 58-64'7( of all subepithelial/local infections, and 30-33% of all epithelial infections that were found by tissue histology. Fresh hemolymph results were as reliable as fixed results for diagnosing advanced infections, but detected only 69% of light systemic infections and were only half as accurate as fixed hemolymph examination for more localized infections. Light infections were more difficult to diagnose in hemolymph samples taken during the early part of the infection period (August) than they were later (November), but there were no differences in detection accuracy associated with resistance or susceptibility to mortality caused by the parasite, KEY WORDS: Parasite, diagnosis, hemolymph. Haplospondium nelsoni. MSX. oyster, Crassostrea virginica rNTRODUCTION Diagnosis of the parasite Haplosporidium nelsoni (MSX) (Haskin, Stauber. and Mackin 1966) in the Amer- ican oyster Crassostrea virginica (Gmelin) is usually done by microscopic examination of fixed, stained tissue sec- tions (Andrews 1966; Farley 1968: Newman 1971; Krantz et al. 1972; Ford and Haskin 1982). Although histologic detection provides considerable information about both host and parasite, it is time-consuming, expensive, and re- quires destructive sampling of oysters. A more rapid and less expensive method for detecting H. nelsoni would be a valuable alternative when speed is necessary, when a com- plete histology laboratory is not available, or when there is reason to keep the host alive for further sampling or experi- mentation. In histologic section, plasmodial stages of//, nelsoni are seen in the circulatory system after they proliferate from initial sites of infection in the gill epithelium, indicating that the parasite should also be detectable in hemolymph samples and suggesting that hemolymph samples might provide a suitable means for diagnosis. In fact, H. nelsoni has been described in hemolymph collected from infected oysters (Farley 1968; Myhre 1969), but, to date, no sys- tematic assessment of hemolymph diagnosis has been made (also, see Burreson et al. 1988). To evaluate hemolymph diagnosis as an alternative to histology, we examined fresh and fixed/stained hemolymph samples for the presence and abundance of H. nelsoni and compared results with standard histological diagnosis of the same oysters. In addition, we determined the influence of infection intensity, season of collection, and level of resis- tance to //. /!e/5o«/-caused mortality on the effectiveness of hemolymph diagnosis. MATERIALS AND METHODS Oyster Collections Naturally infected oysters were collected in lower Dela- ware Bay at three times during the 1984-85 infection cycle (see Ford and Haskin 1982): August 1984 — onset of new infections; November 1984 — winter prevalence peak with estab- lished infections; May 1985 — spring prevalence peak with residual infections from previous summer's exposure. Oysters were also categorized according to whether they had been selected for resistance to mortality caused by //. nelsoni (Haskin and Ford 1979): Unselected — highly susceptible stocks, imported from outside Delaware Bay, or their offspring; Delaware Bay natives — naturally selected wild stock of inter- mediate resistance; Selected — laboratory-reared strains selected for resistance to //. nelsoni-cau^ed mor- tality. Oysters were kept in recirculating sea water at 12-14°C and 18-20 ppt salinity for 1-2 days before hemolymph was collected. Hemolymph Diagnosis The shell of each oyster was notched and hemolymph was collected from the adductor muscle sinus using a 1-ml II 12 Ford and Kanaley tuberculin syringe and a 25-gauge needle (Feng et al. 1971; Ford 1986). Hemolymph volumes of 0.1 or 0.2 ml were diluted to 1 .0 ml in isosmotic sea water (19 ppt) and placed in a chamber constructed of a plastic embedding ring fas- tened by an elastic band to a glass slide (C. A. Farley, per- sonal communication). Cells were allowed to settle for 20-30 minutes at which time the sample was scanned at 200 X using an inverted microscope. A rough quantifica- tion of parasite abundance was made at this time: Rating 1 2 3 4 H. nelsoni None noted Present, but sparse Up to 10 per field 10 to 20 per field More than 20 per field After the fresh preparation was examined, the superna- tant was drained, and the chamber removed. Attached cells were fixed in 3% glutaraldehyde or Davidson's fixative for 5 minutes and stained with Wright's stain. A minimum of 100 H. nelsoni plasmodia were counted in random 200 x fields; however, at least 5 fields were counted regardless of number of parasites. Parasite concentration per milliliter of hemolymph was estimated from the number of fields exam- ined, the settling chamber area, and the volume of hemo- lymph collected. Histological Diagnosis Preparation and examination of tissue sections followed Ford and Haskin (1982), and the infection rating scheme was a modification of theirs. Briefly, one 6 ixm transverse section through each oyster was examined and infections were scored according to parasite abundance and location in the tissues: Systemic — Parasites found in all tissues. Advanced — Averaging more than one parasite per 1000 X oil immersion field; Light — Averaging fewer than one parasite per 1000 X field but more than 100 per sec- tion; Local — Parasites subepithelial, but localized in certain tissues, primarily the gill. Light — Averaging fewer than one per 1000 x field but more than 100 per section: Rare — Averaging fewer than 100 parasites per section; Epithelial — Parasites found only in epithelium, usually gill or palp. Light — Same as Local; Rare — Same as Local; None — No parasites found. Statistics A total of 602 oysters was examined and classified ac- cording to month of collection, selection background, his- tologically determined infection level, parasites per milli- liter of hemolymph as determined from fixed hemolymph preparations and/or the rating (0-4) obtained from fresh hemolymph examination. Accuracy of hemolymph diagnosis was determined by comparing the number of cases in which hemolymph exam- ination agreed with histological results in detecting either the presence or the absence of parasites. Results were fur- ther evaluated by comparing parasite concentrations in he- molymph from oysters with different histologically deter- mined infection levels. The effects of season and selection for resistance on detection accuracy were assessed for fixed hemolymph samples using "G" tests of independence (Sokal andRohlf 1981). RESULTS Appearance of Haplosporidium nelsoni in Hemolymph Samples Histological sections show H. nelsoni in hemolymph vessels to be approximately spherical with clearly distin- guishable nuclei (Figs. 1-3). In fixed hemolymph slides stained with Wright's stain, H. nelsoni plasmodia generally appeared as dense dark blue or purple bodies, approxi- mately spherical, but occasionally showing extreme vari- ability in shape (Figs. 4 and 5). They were clearly distin- guishable from hemocytes in size, shape, and staining quality. Parasite nuclei were not often visible except in large plasmodia that had been flattened and disrupted by the cover slip. Parasites in fresh hemolymph were also generally spher- ical, but sometimes displayed non-spherical forms (Figs. 6-8). They had a textured surface, a distinct plasma mem- brane, and were usually much larger than hemocytes. When parasites were of the same size as granulocytes, they could be distinguished because of the refractile granules in the latter (Fig. 7). Although they were capable of frequent and relatively rapid shape changes, parasites did not pro- duce pseudopodia or filipodia, nor did they display loco- motion characteristic of hemocytes. Outlines of parasite nuclei became visible after preparations had settled for half an hour or more. Detection Accuracy in Hemolymph Compared to Histologic Preparations Infections diagnosed as advanced by histology were de- tected in 977c to 98% of specimens also examined by fresh or fixed hemolymph (Table 1). Detection of Light/Systemic infections was equally good (95%) in the fixed hemolymph slides, but fell to 69% in fresh preparations. Localized in- fections, either Light or Rare, were detected with approxi- mately 60% accuracy using fixed hemolymph, but only 17-32% accuracy in fresh preparations. Light or Rare/Epi- thelial infections were diagnosed in only 30% of the fixed hemolymph and 10-20% of the fresh hemolymph slides. Diagnoses of no parasites in fixed hemolymph agreed with histology in 80%- of the cases, whereas this figure was 90% for fresh hemolymph preparations. Hemolymph Diagnosis for MSX 13 Figures 1-3. Histological section showing//. nWsoni Plasmodia in hemolymph vessels. Figure 1. 100 x Figure 2. Same Tield as Figure 1. 320 x Figure 3. Several Plasmodia with distinct, "capped" nuclei. 1000 x Parasite Concentration in Hemolymph Compared to Tissue Infection Levels To determine how hemolymph parasite concentration related to tissue infection level, oysters in each tissue cate- gory were classified according to the estimated number of parasites per milliliter of hemolymph. Each class repre- sented one log (base 10) increment ranging from to 10' (Fig. 9 Fixed). Tests of independence were then performed comparing, between tissue categories, the fractional distri- bution of parasites in each hemolymph concentration class (Table 2). Advanced and Light/Systemic infections were significantly different from each other and from all other ratings. Differences in hemolymph counts became less sig- nificant as tissue infection intensities decreased, and no differences were found among the lightest categories and the None rating. For further examination, Light and Rare infections were combined in the Local and Epithelial cate- gories, because there were no statistical differences be- tween them. The concentration of parasites in hemolymph. as esti- mated by fixed hemolymph slides, clearly paralleled their abundance in histologic section (Fig. 9 Fixed). In the None tissue category, 80% of the oysters had no detectable he- molymph parasites and an additional \09c had fewer than 100 per ml. Of the latter, most counts were extrapolated from only 1-2 parasites (or parasite-like cells) per slide, and could have been incorrect diagnoses. Nearly 10%, however, had concentrations of more than 100 parasites per ml, and several were in the lO"* class. Seventy percent of oysters with infections classed as Ep- ithelial by histology, had no detectable parasites in hemo- lymph, and when they were found, virtually all concentra- tions were below lOVml. The proportion of undetected in- « ^ Figures 4 and 5. Plasmodia (p) and hemocytes (g) in a fixed, stained hemolymph preparation. 320 x 14 Ford and Kanaley / «t. Figures 6-8. Plasmodia in fresh preparation. Figure 6. Large, spherical Plasmodia contrast in size and shape with smaller hemocytes, which are generally spread out on the slide. Nomarski differential interference contrast. 100 x Figure 7. Spherical and non-spherical Plasmodia. Note contrast between refractile granulocytes (g) and non-refractile Plasmodia (p). Hoffman differential interference contrast. 320 x Figure 8. Nu- merous Plasmodia including a clover-shaped one in upper right corner (arrow). Hoffman differential interference contrast. 320 x fections fell to somewhat less than 40% in the case of Local infections, and again, positive samples generally had fewer than lO-'/ml. Among oysters with Light/Systemic infec- tions, hemolymph parasite concentrations were about evenly divided above and below 10^/ml, while Advanced cases were mostly above 10'*/ml. In the 2-3% of cases in which no parasites were found in the hemolymph of oysters diagnosed as having Advanced infections, notes taken at the time indicate that these individuals were gaping, that the hemolymph contained few cells and much debris, or both. A similar increase in hemolymph parasite concentration with heavier tissue infection was found for fresh prepara- tions although the pattern was less regular in the two Sys- temic categories than was found for fixed slides (Fig. 9 Fresh). Association of Season and Selection with Detection Accuracy To evaluate the effect of season and selection for resis- tance to mortality on detection accuracy, histological cate- gories were combined as Systemic, Local, or Epithelial re- gardless of parasite abundance in the sections because per- cent agreement was statistically the same (p > 0.05) within each of these groupings (see Table 1). Within each com- bined category, percent agreement between hemolymph and histology was approximately the same regardless of se- lection history (Table 3). Local infections appeared to be least accurately detected in hemolymph in August, when infections were relatively new, and most accurately diag- nosed in November when infections were better established (Table 3). Agreement for Systemic and Epithelial infec- tions was approximately the same for all collection periods. Agreement for negative ratings was greatest in the earliest stages of infection (August) and least during the final stages (May). Three-way "G" tests of independence were performed using percent agreement, tissue infection category, and ei- ther selection or season as variables (Table 3). Significant association was found between tissue infection intensity and both selection (p < 0.01) and season (p < 0.05), and between agreement and tissue intensity (p < 0.01), but not between agreement and either selection or season. There was, however, significant interaction between agreement, tissue intensity, and season (p < O.OI). TABLE I. Agreement of fresh and fixed hemolymph preparations with histology for diagnosis of Haplosporidium nelsoni infections. Hemolymph Diagnosis Histological Diagnosis Fixed Fresh Ratio' Percent Ratio' Percent Svslemic Advanced 55/56 98 33/34 97 Light Local 35/37 95 11/16 69 Light 59/92 64 15/46 32 Rare 7/12 58 1/6 17 Epithelial Light Rare 14/43 38/126 33 3/15 30 6/60 20 10 All infections 208/366 57 69/177 39 None 121/152 80 66/73 90 Total Examined 329/518 64 135/250 54 Number of cases agreeing with histology/total in histological category. Hemolymph Diagnosis for MSX 15 FIXED FRESH 10 10^ 10^ lO'* 10^ parasites/ml 12 3 4 INFECTION RATING Figure 9. (FIXEDl Frequency distribution of Haplosporidiiim nelsoni concentration in hemolymph. estimated from flxed/stained slides and grouped according to parasite levels in tissue sections. (FRESH) Frequency distribution of//, nelsoni infection ratings obtained from fresh hemolymph and grouped according to parasite levels in tissue sections. DISCUSSION The reliability of hemolymph diagnosis for detection of the parasite Huplosporidium nelsoni depends, not unex- pectedly, on the severity of infection. Advanced infections can be detected with almost 1007c accuracy, and equally well in fresh or fixed hemolymph samples. Hemolymph diagnosis will fail to detect a heavy infection only when an oyster is moribund, with inadequate hemolymph circulation to keep hemocytes and parasites in suspension. Lighter, more localized infections are more difficult to detect be- cause there are few or no circulating parasites. In these cases, examination of fixed/stained hemolymph samples is generally about twice as accurate as that of fresh material. The lower accuracy of fresh hemolymph examination was because it was intended for rapid screening so less time was spent searching for parasites. Also, parasites were easier to distinguish from hemocytes when stained. Since the para- sites were obviously present in the fresh preparations, addi- tional time spent examining them, and perhaps increased recognition ability of the observer, would undoubtedly in- crease the reliability of this technique. In a parallel study of oysters infected by H . nelsoni in lower Chesapeake Bay. Burreson et al. (1988) reported al- most the same levels of agreement between hemolymph diagnosis and tissue histology as we found. Cooper et al. (1982) compared hemolymph screening with histology for diagnosis of soft clam, Myci arenaria. neoplasia and found that accuracy of diagnosis was related to disease severity, although overall correspondence of the two methods was 16 Ford and Kanaley TABLE 2. Results of "G" tests for independence to determine whether hemolymph concentrations' of Haplosporidium nelsoni differ according to tissue infection level. Advanced Systemic Light Systemic Light Local Tissue Infection Category Rare Light Local Epith. Rare Epith. None Advanced Systemic N = 93 G = 34.68 p <0.01 N G P = 147 = 114.66 <0.0I N = 68 G = 35.40 p < 0.01 N = 98 G = 95.40 p<0.01 N = 182 G = 163.17 p < 0.01 N = 208 G = 185.02 p <0.01 Light Systemic N G P = 128 = 42.05 < 0.01 N = 49 G = 15.95 p < 0-01 N = 79 G = 48.08 p < 0.01 N = 163 G = 84.50 p< 0.01 N = 189 G = 99.29 p < 0.01 Light Local N = 103 G = 0.65 NS N = 133 G = 14.85 p < 0.01 N = 217 G = 34.92 p < 0.01 N = 243 G = 48.90 p < 0.01 Rare Local N = 54 G = 3.99 NS N = 138 G = 4.92 NS N = 164 G = 7.53 NS Light Epithelial NS N = 170 G = 7.97 NS N = 196 G = 10.38 Rare Epithelial N = 278 G = 6.26 NS None ' Parasite hemolymph concentrations were grouped in log (base 10) from to 10' for each tissue category, then the distribution in each tissue category was compared by a "G" test for independence with the distribution in every other category. much higher (94%) than was the case for H. nelsoni infec- tions. Farley et al. (1986) found higher prevalences for the neoplastic condition using hemolymph diagnosis than when using histology, except in dead animals, from which suit- able hemolymph samples could not be obtained. The higher sensitivity of hemolymph diagnosis for this disease com- pared to H. nelsoni is because neoplastic cells are of host origin. They may, in fact, stem from hematopoietic tissue (Yevich and Barszcz 1976), whereas H. nelsoni infective stages are encountered during feeding, and lodge and pro- liferate in the gill epithelium for variable periods before entering the circulation. The choice of diagnostic technique for H. nelsoni de- pends on how the results will be used. For management purposes, either hemolymph method is entirely adequate for detecting infections that are advanced enough to cause immediate problems (i.e., days), but both have limited use- fulness in finding early infections that would permit longer-range predictions (i.e., weeks or months) (see Ford and Haskin 1988). If hemolymph diagnosis is to be used in a monitoring program, results should first be evaluated against histology and occasionally validated in the same way during the course of the project. This is particularly important since the ratio of different infection types, as well as the accuracy of hemolymph diagnosis for each kind of infection, changes seasonally (Ford and Haskin 1982; this study). For most purposes, we suggest that hemolymph diag- nosis be made from fixed/stained slides, which are rapid and inexpensive to make compared to tissue sections. They do not have to be examined immediately, they can be pre- pared quantitatively, and can be saved for documentation. In certain cases, however, fresh preparations are entirely adequate and far more convenient. For instance, we rou- tinely use this method to monitor the development of infec- tions in groups of oysters to be used as a source of live H. nelsoni and to select individuals with sufficient numbers of parasites for in vitro experiments. To speed the process, we do not use chambers, but scan slides holding drops of he- molymph from several oysters. It was not unexpected to find parasites in hemolymph samples from a number of individuals diagnosed as having no parasites in histological section, because we have always assumed that histological sections miss some light infections (Ford and Haskin 1982). In these cases, infec- tions are detected by hemolymph diagnosis because the volume of hemolymph examined is much larger than the volume represented on a tissue section. The same explana- tion can be made for the parasites found in the hemolymph of 13% of the oysters with infections classed as Epithelial Hemolymph Diagnosis for MSX 17 TABLE 3. Effect of selection for resistance to Haplosporidium nelsoni-caused mortality and season of collection on agreement of hemolymph diagnosis with tissue histology. Selection Status Month of Collection Infection Unselected' Selected^ Del. Bay^ August November May Category N< % N % N % N % N % N % Systemic 24/25 96 36/37 97 32/33 47 31/32 47 31 33 94 3(1 30 100 Local 11/17 65 23/37 62 31/50 62 27/56 48 27/31 87 11/17 65 Epithelial 13/39 33 24/75 32 16/58 28 31/91 34 14/50 28 8/31 26 None 38/50 76 58/75 77 25/27 93 62/72 86 38/50 76 21/30 70 Total 86/131 66 141/224 63 104/168 62 151/251 60 110/164 67 70/108 65 ' Unselected = highly susceptible stocks, imported from outside Delaware Bay, or their offspring. - Selected = laboratory-reared strains selected for resistance to H. nelsoni. ' Delaware Bay = native oysters with intermediate resistance. •* Number of cases in which blood and histological diagnoses agreed/total in histological category. Results of three-way tests of independence. R = Percent agreement; C = Selection or Season; A = Tissue category. DF Selection Season G P G P R X C X A independence 17 191.61 NS 192.24 NS A X C independence 6 29.74 <0.01 13.30 <0.05 R X A independence 3 157.17 <0.01 157.17 <0.01 R X C mdependence T 0.47 NS 2.18 NS R X C X A interaction 6 4.24 NS 19.59 <0.01 by histology. Obviously, some infections that appear to be localized in tissue section are really systemic. Restriction of parasites to localized lesions is characteristic of strains selected for resistance to H. nelsoni-causcd mortality in Delaware Bay (Myhre and Haskin 1970; Ford and Haskin 1987; this study). It was important to find no selection-re- lated difference in correspondence of hemolymph and tissue diagnosis at any infection level, indicating that he- molymph diagnosis would give equivalent results for both selected and unselected strains. The use of hemolymph in parasitological work is not limited to diagnosis, but may provide a means of obtaining live parasites for in vitro study. In such cases, the investi- gator would want to increase chances for collecting large numbers of parasites from heavily infected animals. The maximum concentration of parasites found in our study was 7.8 X 10^ per milliliter of hemolymph. Feng (1965) dem- onstrated a linear increase in heart rate and numbers of cir- culating hemocytes with increased temperature, finding nearly twice the cell concentration at 22°C as at 12°C. We kept oysters at 12-14°C because reduced temperature pro- longs survival of infected individuals; however, Feng's (1965) results predict that more than a million parasites per milliliter could be harvested from some individuals allowed to acclimate at higher temperatures before bleeding. In fact, preliminary tests in this study, performed on oysters collected in May and kept at 20°C or higher, found several individuals with parasite concentrations of more than one million per milliliter. Since the volume of hemolymph that can be collected from an oyster ranges up to several milli- liters, several million parasites could be obtained from a single large, heavily infected individual. Histology is still the most reliable detection method for H. nelsoni and provides more detailed information on the disease process than does hemolymph diagnosis; however, hemolymph diagnosis can be a very useful tool provided one understands its limitations. Hemolymph preparations can, in fact, contribute greatly to the study of host-parasite interactions, especially in vitro. Finally, hemolymph diag- nosis is especially valuable because it permits repeated sampling from the same individual (Cooper et al. 1982; Farley et al. 1986; Ford 1986), so that the disease process can be studied and manipulated experimentally with min- imum numbers of animals and with increased statistical control over individual variability. ACKNOWLEDGMENTS We thank C. Rizzo, D. O'Connor, and C. Peterson for helping to prepare and read slides; and W. Canzonier, B. Barber, H. Haskin, and D. Kunkle for helpful comments 18 Ford and Kanaley on the manuscript. This is New Jersey Agricultural Experi- ment Station publication No. D-32504-1-87. supported by state funds; by the New Jersey Commission on Science and Technology, Fisheries and Aquaculture Technology Exten- sion Center; and by the New Jersey Department of Environ- mental Protection. REFERENCES CITED Andrews, J. D. 1966. Oyster mortality studies in Virginia. V. Epizooti- ology of MSX, a protistan parasite of oysters. Ecology 47:19-31 . Burreson, E. M.. M. E. Robinson, and A. Villalba. 1988. A companson of paraffin histology and hemolymph analysis for the diagnosis of Ha- plosporidiiim nelsoni (MSX) in Crassostrea virgiiiica (Gmelin). J. Shellfish Res 7:19-23. Cooper, K. R., R. S. Brown and R, W. Chang. 1982. Accuracy of he- molymph cytological screening techniques for the diagnosis of a pos- sible hematopoietic neoplasm in the bivalve mollusc, Mya arenaria. J . Inverlebr. Pathol. 39:281-289. Farley, C. A. 1968. Minchinia nelsoni (Haplosporida) disease syndrome in the American oyster Crassostrea virginica. J. Protozoal. 15:585- 599. Farley, C. A., S. V. Otto, and C. L. Reinisch. 1986. New occurrence of epizootic sarcoma in Chesapeake Bay soft shell clams. M\a arenaria. Fish. Bull. 84:851-857. Feng, S. Y. 1965. Heart rate and leucocyte circulation in Crassostrea virginica (Gmelm). Biol. Bull. 128:198-210. Feng, S. Y., J. S. Feng, C. N. Burke, and L. H. Khairallah. 1971. Light and electron microscopy of the leucocytes of Crassostrea virginica (Mollusca: Pelecypoda). Z. Zellforsch. 120:222-245. Ford, S. E. 1986. Effect of repeated hemolymph sampling on growth, mortality, hemolymph protein and parasitism of oysters. Crassostrea virginica. Comp. Biochem. Physiol. 85A:465-470. Ford, S. E., and H. H. Haskin. 1982. History and epizootiology of Ha- plosporidium nelsoni (MSX), an oyster pathogen, m Delaware Bay, 1957-1980. J. Invertebr. Pathol. 40:118-141. Ford, S. E., and H. H. Haskin. 1987. Infection and mortality patterns in strains of oysters Crassostrea virginica selected for resistance to the parasite Haplosporidium nelsoni (MSX). J. Parasit. 73:368-376. Ford, S. E., and H. H. Haskin. 1988. Management strategies for MSX [Haplosporidium nelsoni) disease in oysters. In: Disease Processes in Marine Bivalve Molluscs. W. S. Fisher, Ed. American Fisheries So- ciety Special Publication (in press). Haskin, H. H.. and S. E. Ford. 1979. Development of resistance lo Min- chinia nelsoni (MSX) mortality in laboratory-reared and native oyster stocks in Delaware Bay. Mar. Fish. Rev. 41 (l-2):54-63. Haskin, H. H., L. A. Stauber and J. A. Mackin. 1966. Minchinia nelsoni n. sp. (Haplosporida, Haplosporidiidae): causative agent of the Dela- ware Bay oyster epizootic. Science 153:1414-1416. Krantz. E. L., L. R. Buchanan, C. A. Farley, and A. H. Carr. 1972. Minchinia nelsoni in oysters from Massachusetts waters. Proc. Natl. Shellfish. Assoc. 62:83-88. Myhre. J. L. 1969. Nucleic acid and nucleic protein patterns in vegetative stages of the haplosporidan oyster parasite, Minchinia nelsoni ( Haskin, Stauber and Mackin). Proc. Natl. Shellfish. Assoc. 59:52-59. Myhre, J. L. and H. H. Haskin. 1970. MSX infections in resistant and susceptible oyster stocks. Proc. Natl. Shellfish. Assoc. 60:9. Newman. M. W. 1971. A parasite and disease survey of Connecticut oysters. Proc. Natl. Shellfish. Assoc. 61:59-63. Sokal, R. R. and F. J. Rohlf. 1981 . Biometry. W, H. Freeman, San Fran- cisco. 859 pp. Yevich. P. P. and C. A. Barszcz. 1976. Gonadal and hematopoietic neo- plasms in Mya arenaria. Mar. Fish. Rev. 38:42-43. Journal of Shellfish Research. Vol. 7. No. I, 19-23, 1988. A COMPARISON OF PARAFFIN HISTOLOGY AND HEMOLYMPH ANALYSIS FOR THE DIAGNOSIS OF HAPLOSPORIDWM NELSONI (MSX) IN CRASSOSTREA VIRGINICA (GMELIN) EUGENE M. BURRESON, M. ELIZABETH ROBINSON AND ANTONIO VILLALBA' Virginia Insilitute of Marine Science School of Marine Science College of William and Mary Gloucester Point. Virginia 23062 ABSTRACT Diagnosis of the oyster pathogen Haplosporidiwn nelsom (MSXl by paraffin histology is compared with a technique in which hemolymph drawn from the oyster adductor muscle smus is examined for parasite plasmodia. Oysters from seed beds of the James River, Virginia imported to an MSX endemic area in May, 1986 were sampled monthly through December, 1986 and in February, 1987. A sample of 25 oysters was bled each month and then processed for sectioning. Of the 200 oysters sampled. 89 (44.5%) were diagnosed as infected using histology and 61 (.10.5%) were diagnosed as infected using hemolymph examination. All the heavy and moderate infections diagnosed by paraffin histology were also diagnosed by hemolymph, but only 64.3% of the light infections and only 43.5% of the rare infections were diagnosed by hemolymph analysis. However. 92.3% of the undetected rare infections and 60.0%^ of the undetected light infections were localized in gills and plasmodia had not entered the circulatory system. The hemolymph technique, which takes about 4 h. detected 89.7% of the systemic infections diagnosed by paraffin histology. KEY WORDS: Haplosporidium nelsoni. MSX, oyster, diagnosis, techniques INTRODUCTION Traditionally, protozoan parasites of oysters have been diagnosed by paraffin histological techniques (Farley 1967; Andrews and Frierman 1974; Ford and Haskin 1982), This technique is accurate, but time consuming when rapid diagnosis is needed. Although promising new diagnostic procedures utilizing enzyme immunoassay, monoclonal an- tibody techniques and nucleic acid probes are under devel- opment in most areas of the world where oyster diseases significantly reduce the harvest, none are currently avail- able. A simple, rapid technique for diagnosis of bivalve diseases has been developed by C. Austin Farley. The tech- nique relies upon the presence of parasites in host hemo- lymph that occurs in systemic infections. The purpose of this study was to compare diagnosis of Haplosporidium nelsoni (Haskin, Stauber and Mackin) through an annual infection period using traditional paraffin histology and hemolymph analysis. Because of a prolonged drought and record high salinity in the Chesapeake Bay, Virginia during 1986. oysters had the highest levels of H. nelsoni ever recorded. These high prevalences allowed good comparisons of the two techniques. MATERIALS AND METHODS Oysters were dredged in May, 1986 at Horsehead Rock in the James River. Virginia, Oysters from this rock are known to be highly susceptible to MSX and have been used as controls for over 25 years for disease monitoring pro- grams conducted in Chesapeake Bay by the Virginia Insti- Present address: Cumarsa, Punta Moreira, Reboredo, O'Grove, Ponte- vedra, Spain. tute of Marine Science and in Delaware Bay by Rutgers University, A 0,6 m by 1.2 m (2 by 4 feet) tray containing 378 oysters was suspended from a pier at VIMS in the lower York River, an MSX endemic area. Two additional trays of 400 oysters each were placed at the usual moni- toring location about 1 km upriver. The MSX infection pe- riod typically begins in May each year (Andrews and Frierman 1974), therefore oysters for disease monitoring are transplanted to trays each year at that time. A sample of 25 oysters analyzed for H. nelsoni at the time of transplan- tation was negative for the parasite. Samples of 25 oysters were taken from the pier tray in late May and approxi- mately every 30 days through November, 1986. No oysters remained in the tray after the November sample was re- moved so an additional sample in February, 1987 was taken from one of the other monitoring trays. All oysters were analyzed for the presence of H. nelsoni by both he- molymph analysis and paraffin histology, Hemolymph Analysis Immediately after sampling oysters from the tray, each oyster was numbered and the shell notched opposite the adductor muscle with a hand-held grinding tool. Using a 22 ga needle, 0. 1 to 0,2 ml of hemolymph was drawn from the adductor muscle sinus into a 3 cc syringe containing 2,0 ml of cold 15 ppt artificial seawater containing 0.05 gm/1 phenol red. If too much hemolymph is withdrawn from the oyster a thick cell layer results after settling and the slide may be difficult to diagnosis. Contents of the syringe were gently expressed into Farley chambers and hemocytes al- lowed to settle for one hour. These chambers were devel- oped by C, Austin Farley, NMFS, Oxford, Maryland and consisted of plastic tissue-embedding rings sanded tlat to 19 20 BURRESON ET AL. prevent leakage and held to numbered microscope slides with an elastic "pony-tair' band. Rubber bands are not sat- isfactory because they are easily cut by the edges of the glass slide. After one hour, chambers were removed and the slide with attached cell monolayer was fixed for 5 min in Dietrich's AFA. Slides were stained with Harris" hema- toxylin and eosin, coverslipped, and examined for the pres- ence of Plasmodia. Slides were initially scanned at 100 x in their entirety or until Plasmodia were observed. Later, plasmodia were counted in each of 5 lOOx fields randomly sleeted using a microslide field finder. In heavy infections five 450 x fields were counted and converted to 100 x counts by mul- tiplying by a factor of 20. Average counts per five 100 x fields are given in Table 1 ; a -t- symbol indicates that Plas- modia were observed on the slide, but none was present in the randomly selected fields. Plasmodia counts could not be standardized to number of hemocytes because of high variability in number of hemocytes, even in uninfected oysters, and because heavy infections increased hemocy- tosis. Generally, there were relatively more hemocytes in heavy infections. Paraffin Histology After hemolymph was withdrawn, oysters were opened and an approximately 5 mm thick section of tissue through the visceral mass that included mantle, gills, stomach, in- testine and digestive diverticula was excised and fixed in Davidson's AFA for 24 h. Tissue was embedded in par- affin, sectioned at 6 |xm and stained with Harris' hematox- ylin and eosin. Oyster tissue was not trimmed before or after embedding and sections from only one oyster were placed on each slide. Sections were diagnosed, without ref- erence to hemolymph preparations, by technician Juanita Walker who has been responsible for MSX diagnosis at VIMS for over 20 years. Rating of infection intensity was as follows: rare (R)- less than 10 plasmodia in entire section, not limited to gill epithelium; rare localized (RL)-rare to light infection re- stricted to gill epithelium; light (L)-less than 2 plasmodia per 450 X field but greater than 10 in entire section; light localized (LL)-many plasmodia but infection restricted to gill epithelium; moderate (M)-2 to 5 plasmodia per 450 x field; heavy (H)-greater than 5 plasmodia per 450 x field; sporulation (S)-any infection when spores were present. TABLE 1. Comparison of prevalence and intensity of H. nelsoni in C. virginica determined by paraffin histology (P) and hemolymph analysis (A) in samples of 25 oysters. Hemolymph analysis values are average number of plasmodia per five 100 x microscope fields. Oyster 27 May 86 27 Jun 86 29 July 86 28 Aug 86 25 : Sep 86 27 Oct 86 1 Dec 86 27 Feb 87 number P A P A P A P A p A P A P A P A 1 U u u U R + U L U 2 U u LL L 10 u U U H 55 3 u u U U RL R + L 16 L 1 4 u u u u L RL U U 5 u u u RL H 64 L + u M 130 6 u u u RL U U RL L + 7 u u u U u M 34 LL L 3 8 u u H 384 u u U L 3 U 9 u u H 528 M 8 u R 1 RL L 4 10 u u u R + L U U U 11 u u u U L 3 RL L 18 M + 12 u u u L M 7 U U RL 13 u u u U U U u U 14 u u LL u H 161 u L 2 L + 15 u u u u L 1 R 3 R + L + 16 u u LL M 28 L 1 L 4 U M 6 17 u LL L 3 M 20 L H 1816 H 176 L + 18 u u U U U U U L 1 19 u u M 15 H 736 H 41 R 3 U M 1 20 u u L + L 5 L 1 R 2 RL L 3 21 u u U L 2 R R -1- LL U 22 u u M + RL S + M 52 LL L + 23 u u U L 1 H 496 U L -1- U 24 u u LL U U R + U u 25 u u RL L U RL LL M + Total No. infected 1 11 6 14 9 16 11 15 12 15 7 17 16 Prevalence (%) 4.0 44.0 24.0 56.0 36.0 64.0 44.0 60.0 48.0 60.0 28.0 68.0 64.0 H = heavy; L = light; LL = 1 ight, localized; M = moderate; R = rare; RL = rare, localized; S = spore! s; U = uninfected; -1- = Plasmodia present on slide, but none in random fields; = no plasmodia on slide. Comparison of MSX Diagnosis Techniques 21 RESULTS Of the 200 oysters analyzed for H. nelsoni, 89 (44.5%) were infected based on paraffin histology (Table 1 ). Preva- lence of infection gradually increased through September and then remained at approximately 60% through fall and winter. Hemolymph analysis detected 61 infections in the 200 oysters (30.5%), or 68.5% of the total detected by par- affin histology. Hemolymph analysis detected all the heavy and moderate infections as determined by paraffin his- tology, but only 64.3% (27/42) of the light infections and 43.5% (10/23) of the rare infections. However, of the 15 light infections not detected by hemolymph analysis, 9 cases (60.0%) were localized infections and of the 13 un- detected rare infections, 12 (92.3%) cases were localized infections (Table 1). Thus, hemolymph analysis detected 37 of the 44 (84.1%) rare and light systemic infections. As expected, no localized infections were detected by hemo- lymph analysis because plasmodia had not entered the cir- culatory system. If the localized infections are removed from consideration, hemolymph analysis detected 61 of the 68 (89.7%) systemic infections. Plasmodia counts from hemolymph analysis sorted rela- tively well into rare-light, moderate and heavy infections, although there was some overlap (Table 2). Counts from hemolymph analysis for rare and light infections as deter- mined by histology ranged from 1 plasmodium on the slide to about 5 Plasmodia per lOOx field, with much overlap between rare and light counts. Counts for moderate infec- tions ranged from a few plasmodia on the slide to 130 Plas- modia per 100 X field and counts for heavy infection ranged from 41 to over 1800 plasmodia per 100 x field. The oyster with spores in the digestive diverticula had a very low number of plasmodia in the hemolymph. Size of Plasmodia in hemolymph preparations depended TABLE 2. Summary of hemolymph plasmodia counts for the four intensity categories determined by paraffin histology. Heavy Paraffin histology intensity category Moderate Light Rare 1816 736 528 496 384 176 161 55 41 130 18 1 3 52 16 1 3 34 10 + 2 28 5 + 1 20 4 + + 15 4 + + 8 3 + -1- 7 3 + -1- 6 3 + 1 3 + + 2 + 2 -1- 1 1 upon the intensity of infection. In rare and light infections, Plasmodia were usually small, between 5 and 20 |xm in diameter, with 2 to 15 nuclei per plasmodium. In moderate and heavy infections plasmodia ranged from 5 to 70 |j.m in diameter and large plasmodia often had over 100 nuclei (Figure la). Most nuclei were about 2 |xm in diameter with an obvious eccentric endosome and dark-staining bar of "Kernstab'", but large metaphase nuclei, up to 11 \i.m long, were also present in many plasmodia. Plasmotomy appeared to be occurring in all large plasmodia and in many small Plasmodia. This process appeared to commence with gradual concentration of cytoplasm in from two to seven areas at the periphery of the plasmodium (Figure lb) with subsequent fragmentation of the original plasmodium into smaller plasmodia (Figure Ic). Plasmodia less than 15 ixm in diameter were often observed in phagocytes. DISCUSSION Results of this study suggest that hemolymph diagnosis of//, nelsoni may be an acceptable and preferable alterna- tive to paraffin histology diagnosis, depending upon the objective of the diagnoses. If a rapid survey of a large number of oysters is required for oyster mortality predic- tions, then hemolymph analysis is probably the technique of choice. All heavy and moderate infections and 84.1% of rare and light systemic infections were diagnosed by this technique. Similar results were obtained by Ford and Kan- aley (1988) using hemolymph diagnosis. The hemolymph technique did not detect localized gill epithelial infections, but investigators disagree on the fate of these infections. If, as some believe, gill epithelial infections always develop into systemic infections, then the hemolymph technique may seriously underestimate the actual prevalence, and po- tential mortality, because localized gill infections may ac- count for almost 50% of the infections in certain months (Table 1, July and December, for example). If, as others believe, gill epithelial infections do not develop into sys- temic infections, failure to detect localized infections may not be a serious disadvantage if the goal is to predict mor- tality. Unfortunately, until the life cycle of MSX is solved, there is no way to know if all systemic infections begin as localized infections in the gills. Thus, periodic calibration with paraffin histology should be incorporated into any monitoring program using hemolymph diagnosis. The main advantage of the hemolymph technique is rapid diagnosis of H. nelsoni — approximately 4 h for a sample of 25 oysters as compared to more than 48 h for paraffin histology. The hemolymph technique may also de- tect other parasites, such as Perkinsus marinus, but no comparisons with paraffin histology or thioglycollate cul- ture have been made for this parasite. The disadvantages of the hemolymph technique are primarily related to the fact that no permanent section of oyster tissue is obtained as it is in paraffin histology. Thus, there is no record of oyster tissue response to H . nelsoni infection and no record of 22 BURRESON ET AL. • ' — * f? f \ ^. - ---I 1 B * .%i|l # m ^1^ * Figure 1. MSX Plasmodia in liemolymph preparations. (A) Low magniflcation of heavy H. nelsoni infection illustrating various sizes and shape! of Plasmodia. (B) A large Plasmodium in the process of fragmenting. (C) Plasmodium that has almost completed separation into two smallei Plasmodia. (D) Phagocytized Plasmodia (arrows). The phagocyte on the left has engulfed at least tvto plasmodia. All bars = 20 ^m. Comparison of MSX Diagnosis Techniques 23 other parasites present. These may or may not be important considerations, depending upon the objective. The process of plasmotomy described may be an artifact of the technique in which plasmodia are allowed to settle onto a glass slide, but it clearly indicates that plasmodia are capable of fragmentation as suggested by Farley ( 1967). ACKNOWLEDGMENTS This study could not have been completed without the diagnostic expertise of J. Walker and without the assistance of J. Andrews in the field. Virginia Institute of Marine Science Contribution No. 1431. Andrews, J. D. & M. Frierman. 1974. Epizootiologv of Minchiiiia nel- soni in susceptible wild oysters in Virginia. 1959 to 1971. J. Invert. Pathol. 24:127-140. Farley, C. A. 1967. A proposed life cycle of Minchinia nelsvni (Haplo- sporida, Haplosporidiidae) in the American oyster Crussostrea vir- ginica. J . Protozoal. 14:616-625. REFERENCES CITED Ford, S. E. & H. H. Haskin. 1982. History and epizootiology of Haplo- sporidium nelsoni (MSX). an oyster pathogen in Delaware Bay, 1957-1980. J. Invert. Pathol. 40:118-141. Ford, S. E. and S. A. Kanaley. 1988. An evaluation of hemolymph diag- nosis for detection of the oyster parasite Haplosporidium nelsoni (MSX). J. Shellfish Res. 7:. Journal of Shellfish Research. Vol. 7, No. 1, 25-31, 1988. EFFECTS OF THE PARASITE MSX (HAPLOSPORIDIUM NELSONF) ON OYSTER (CRASSOSTREA VIRGINICA) ENERGY METABOLISM. I. CONDITION INDEX AND RELATIVE FECUNDITY BRUCE J. BARBER, SUSAN E. FORD, AND HAROLD H. HASKIN Department of Oyster Culture New Jersey Agricultural Experiment Station Cook College. Rutgers University Shellfish Research Laboratory P.O. Box 687 Port Norris. New Jersey 08349 ABSTRACT The effects of the endoparasite MSX iHaplosponJium nelsoni) on condition index and fecundity in the American oyster. Crassoslrea virginica. were examined between May and November, 1985. On most sampling dates, mean condition index and mean relative fecundity were related to MSX infection intensity (uninfected individuals > epithelially infected individuals > systemi- cally infected individuals). Overall, oysters with MSX infections confined to gill epithelium had a condition index that was \39c (P < 0.02) lower and a relative fecundity that was 35% lower than uninfected oysters. Oysters with systemic (general) MSX infections had a condition index that was 31% lower (P < 0.001) and a relative fecundity that was 81% lower (P < 0.01) than uninfected oysters. Reduced fecundity was manifested primarily as a reduction in the number of mature eggs produced rather than in the size of individual mature eggs. The observed reduction in fecundity is most likely the result of metabolic stress in which MSX reduces food intake and competes for energy reserves (reduces condition) which in uninfected individuals would be used for gamete production. Even at sublethal levels, MSX affects oyster fisheries by reducing meat yield and recruitment potential. KEY WORDS: Oyster, parasite, MSX, condition, reproduction, Crassoslrea virginica. Haplosporidium nelsoni INTRODUCTION In oyster biology, "condition" is a general term refer- ring to meat quality that is most often assessed for eco- nomic reasons. However, condition is of physiological im- portance as well, being defined by Mann (1978) as "the ability of an animal to withstand an adverse environmental stress, be this physical, chemical or biological." The ener- getic advantage displayed by individuals in good condition is due to the fact that condition is directly proportional to the amount of glycogen contained in the tissues (Galtsoff 1964; Walne 1970). Glycogen is the primary energy storage substrate in oysters, providing energy for many physiological processes (Bayne 1976; Gabbott 1976, 1983). Stored glycogen is clearly important as a source of en- ergy for reproduction. Numerous studies have shown that glycogen is stored when food is abundant and later utilized in the production of gametes (Bayne 1976; Gabbott 1976, 1983). Fecundity, or the amount of gametogenic material produced, is directly related to the amount of glycogen ac- cumulated prior to gametogenesis (Loosanoff 1965; Bayne 1975). Thus any stress that reduces condition (glycogen content) would also reduce fecundity (Bayne 1975). The protozoan parasite MSX (Haplosporidium nelsoni [Haskin, Stauber and Mackin 1966]) has been responsible for mortalities of the American oyster, Crassostrea vir- ginica (Gmelin), in Delaware and Chesapeake Bays since the late 1950"s (Andrews 1966; Haskin et al. 1966; Haskin and Ford 1982). A typical pattern of infection leading to oyster mortalities has been described by Ford and Haskin (1982) and Ford (1985) as follows. New infections are ac- quired from June through October each year. Infections are initially confined to gill epithelium, where the plasmodia divide and proliferate, eventually breaking through the basement membrane. At this point infections rapidly be- come systemic as parasites are spread via the circulatory system. Ensuing mortalities occur in late summer and early fall as infections intensify. Infection levels are usually high over the winter, but mortalities lessen, most likely because of reduced metabolic activity of both host and parasite. Re- newed parasite proliferation and oyster mortality accom- pany rising water temperatures the following spring. The high prevalence levels often recorded in late spring are from infections acquired the previous year. Even though the epizootiology of MSX is fairly well un- derstood, little is known as to how MSX affects the energy metabolism of individual oysters, eventually causing death. Newell (1985) found that oysters infected with MSX had significantly lower clearance rates than uninfected oysters and that this was correlated with a decline in condition index. Ford and Figueras (1988) described the pathological effects of MSX on gametogenesis. This paper reports the seasonal relationships that exist between condition, relative fecundity, and intensity of MSX infection in C. virginica. MATERIALS AND METHODS Oysters used in this study were from the 1980 and 1981 year classes produced as part of an ongoing experimental breeding program at Rutgers University (Haskin and Ford 1979; Ford and Haskin 1987), and were maintained in trays 25 26 Barber et al. intertidally at the Rutgers oyster hatchery located on lower Delaware Bay. Thirty individuals were examined six times between May and November, 1985 for condition index, relative fecundity, and degree of infection by MSX. Three additional samples were taken for the determination of rela- tive fecundity and MSX level only. This interval of time was chosen because it is the period during which both re- productive and MSX activities are greatest. The condition index used was that of Walne ( 1970): C.l. Dry Tissue Wt. (g) x 1000 Shell Cavity Vol. (ml) Oysters were cleaned of fouling organisms, measured (shell height), and the volume of water displaced by the whole oyster was obtained. A standard transverse (dorsal) tissue section of each oyster across gill, stomach, intestine, and digestive diverticula was removed and weighed (wet) prior to fixation for histological processing. The remaining tissue was weighed wet and then dried so that a dry wt./wet wt. ratio could be obtained and used to calculate the dry weight of the entire animal. The displacement volume of the two valves was measured and subtracted from the total dis- placement volume to obtain the shell cavity volume. The histological procedures used for the diagnosis of MSX are those of Ford (1985). Tissue sections were fixed in Davidson's solution, dehydrated, cleared, and embedded in paraplast. Six-jjim sections were stained with iron hema- toxylin, acid fuchsin, and aniline blue. Each oyster was then categorized with respect to MSX infection intensity as either uninfected, epithelially infected (gills only), or sys- temically infected (subepithelial, general). Since the oyster gonad is confined to a layer surrounding the digestive gland, relative fecundity (amount of gameto- genic material) was obtained from the histological section as the ratio of [gonadal area/area of entire visceral mass] x 100, using an image analysis system (Bioquant). Measure- ments were made only on sexually differentiated indi- viduals. This morphological approach to quantifying gonad production in oysters has been used previously (Mori 1979), and is more precise than measuring just the thick- ness of the gonadal layer (Loosanoff 1965). In addition to the determination of relative fecundity, the diameters of 350 mature ova from oysters (collected 12 June) both uninfected (n = 7) and infected (n = 3) with MSX (both epithelial and systemic infections) were ob- tained from the histological sections using the Bioquant image analysis system. Both condition index and relative fecundity values were i2 30 CO D ■D 25 > ""? 20 O 15 n 10 E Uninfected Epithelial Systemic M J J A S O N Month-1985 Figure 1. Number of individuals (Crassoslrea virginica) within eacli MSX infection category at various times between May and December, 1985. MSX Effects on Oyster Condition and Fecundity 27 transformed (arcsin) prior to analysis (Zar 1974). Condition and fecundity were examined with respect to infection in- tensity and sampling date using a two-factor analysis of variance. To determine overall effects of infection intensity on both condition and fecundity, the means of differences (all sampling dates) between epithelially infected and unin- fected oysters and between systemically infected and unin- fected oysters were examined using a paired t-test (Zar 1974). Mean egg diameters were compared using a t-test. RESULTS Mean shell height ranged from 93 mm to 107 mm, and little growth occurred over the course of the study. Regres- sion analysis demonstrated that over this size range, condi- tion index and relative fecundity were unrelated to shell height. The distribution of oysters within the respective infec- tion categories reflected a typical progression of MSX in- fection in 1985 (Figure 1). Between May and September, the number of uninfected oysters within a sample de- creased, and the number of oysters with both epithelial and systemic infections increased. Infections found in May and June were most likely acquired during the previous summer-fall infection period, or earlier. New infections be- came apparent in the July sample as an increase in the number of oysters with epithelial infections. Infection in- tensity generally increased from July through October. In November MSX prevalence and intensity was reduced. Cumulative oyster mortality increased steadily throughout the study period (Figure 2). Mortalities occur- ring in May, June, and July were most likely the result of chronic infections. By November, cumulative mortality was starting to level off at about 50%. Two-factor analysis of variance showed significant dif- ferences in condition index associated with both sampling date and infection category (P < 0.005). Mean condition index within each of the three infection categories generally increased between May and November 1985 (Table 1). In all months but August, when condition index was the same for both uninfected and epithelially infected groups, there was a reduction in mean condition index with increase in MSX infection intensity (i.e., the mean condition index of uninfected oysters > mean condition index of epithelially infected oysters > mean condition index of systemically infected oysters). Multiple comparison (Student-Newman- Keuls test) revealed that significant differences in mean 50 > 4-^ 40 c5 o 30 > 20 E ^o] o M J J A S O Month-1985 Figure 2. Cumulative mortality (%) of experimental oysters [Crassostrea virginica) between May and December. 1985. 28 Barber et al. TABLE 1. Mean condition index ( ± Isd) of oysters within each MSX infection category. Values not significantly different are connected by the same line TABLE 2. Mean relative fecundity ( ± Isd) of oysters within each MSX infection category. Values not significantly different are connected by the same line. Infection Category Date Infection Category Uninfected Epithelial Systemit Uninfected Epithelial Systemic Date X sd X sd A% X sd A% X sd X sd A% X sd A% 21 May 75 5 56 4 -25 48 4 -36 21 May 21 5 9 7 -58 -100 25 June 71 2 62 3 -13 51 1 -29 12 June 25 June 30 18 1 5 19 8 2 4 -37 -56 9 4 8 7 -71 22 July 103 1 84 2 -19 64 4 -37 -76 22 August 91 1 92 4 + 1 66 6 -28 9 July 6 6 12 5 + 55 -100 19 September 107 — 100 3 -7 74 6 -31 22 July 8 6 6 5 -25 4 4 -51 6 November 119 4 102 5 -15 87 5 -27 9 August 22 August 19 September 6 4 4 4 3 5 -30 2 3 3 -71 — 21 — 2 -97 1 -100 condition i ndex occurred between infection 1 categories ; for 6 November — all dates (Table 1 ) A comparison of the differences in means between in- fection categories over the whole study period indicated that condition index was 13% lower in oysters with epithe- lial infection than in uninfected oysters (P < 0.02). Differ- ences varied from to 25% with smallest differences oc- curring in August and September, when most epithelial in- fections were of recent origin. Oysters with systemic infections had a condition index that was 29% lower than uninfected oysters (P < 0.001). These differences were rel- atively consistent (27-38%) over the whole period. Two-factor analysis of variance also revealed significant differences (P < 0.005) in relative fecundity between in- fection category but not sampling date. Nonetheless, at all three levels of infection, relative fecundity was greatest in early June as maximum ripeness was attained and gradually decreased afterward as spawning proceeded (Table 2). The only exception was the single uninfected oyster in Sep- tember, which had a ripe gonad. All reproductive activity had ceased by November in all infection categories. As was the case with condition index, relative fecundity was gener- ally reduced with increasing infection intensity (all but 9 July and 22 August samples). It should be noted that the reduction in relative fecundity measured in this study (as a ratio) was due to a decrease in actual gonadal area (numer- ator) rather than due to an increase in total area (denomi- nator). Multiple comparison (Student-Neuman-Keul) indi- cated that significant differences in mean relative fecundity between infection categories occurred on 21 May, 12 June, and 19 September (Table 2). Comparing the differences between means of uninfected oysters and both categories of infected oysters for the entire study period indicated that relative fecundity was 35% lower in oysters with epithelial infections. Relative fecun- dity was 81% lower in oysters with systemic infections (P < 0.01). The mean diameter of mature ova was 37.6 jjim (s.d. = 5.9) in uninfected oysters and 35.9 [xm (s.d. = 6.1) in oysters with MSX infections. These means were not statis- tically different (P > 0.05). DISCUSSION The indication that parasites stress their host, affecting a wide variety of biochemical and physiological functions is well documented (Lauckner 1983; Newell and Barber 1988). Although host/parasite relationships vary widely in their specifics, the stress to the host caused by parasitism generally has a nutritional basis (Thompson 1983). This can be manifested as altered food consumption, digestion, assimilation, and/or energy storage. With less energy avail- able, host growth, including reproduction, is often im- paired. The primary metabolic substrate in marine bivalves is glycogen, and condition index is representative of the ener- getic state or health of an individual. A reduction in the condition of several bivalve species associated with para- sitism has been reported. Infestation by Polydora caused a reduction in condition index in both C. virginica and My- tilus edulis (Lunz 1941; Kent 1979). Crassostrea gigas in- fected with Mytilicola had a lower condition index than un- infected individuals (Katkansky et al. 1967). Cole and Savage (1951) found that the flesh weight of A/, edulis was reduced in relation to the number of parasites (Mytilicola) present. This was supported by Bayne et al. (1979) who found that high numbers of Mytilicola caused a depression in feeding rate and a reduced scope for growth in M. edulis, "which would result in time in a decline in condition." The present study demonstrates that MSX reduces con- dition in C. virginica and that the reduction is related to the severity of the infection. Even though condition index varies seasonally, its proportional reduction in systemically infected oysters remains fairly consistent when measured MSX Effects on Oyster Condition and Fecundity 29 against uninfected oysters. The same is not true for oysters with gill epithelial infections, which have little or no de- pressing effect on condition index when they are newly ac- quired (August and September), but are associated with considerable reduction in condition as gill infections be- come older and more numerous (November-July). Newell (1985) demonstrated that MSX infection was re- sponsible for a significant reduction in the feeding rate of oysters. A reduction in incoming energy would in itself ne- cessitate the utilization of stored reserves to meet metabolic requirements and result in a lower condition index (Newell 1985). A possible loss of feeding capability caused by MSX in epithelial infections (confined to gill tissue) may in itself be enough to lower condition index to the extent noted in the fall and spring samples. Another possibility is that infections diagnosed as epithelial may actually be sys- temic, with parasites located outside of the plane of section (see Ford and Kanaley 1988). The chances of this hap- pening increase as gill infections become established, as they would be in the fall and spring samples. However, oysters with systemic (general) infections would be af- fected not only by a continued decrease in incoming food but by the metabolic burden of an increasing number of circulating parasites. Evidence of disrupted energy metabo- lism in oysters systemically infected with MSX is provided by Ford (1986) who found a drop in total serum protein concentration with increased infection intensity and Eble (1966) who found a considerable reduction in levels of di- gestive gland enzymes associated with MSX infection. Under prolonged negative energy balance, the result of re- duced food intake, decreased digestive activity, and in- creased parasite metabolism, heavily infected oysters die at a greater rate than uninfected oysters. Mortalities could presumably result directly from the effects of MSX or indi- rectly from a decreased ability to resist predation or envi- ronmental stress. By diminishing the total amount of nutrients available to the host at any time (both incoming food and stored re- serves), parasitism might also reduce fecundity, or the amount of gametogenic material produced (Bayne 1975). Parasitism has been shown to affect bivalve reproduction both directly and indirectly. Cases of castration in which gametogenesis is directly inhibited has been reported for Pecten alba, M. edulis, and Brachidontes recurvus. all in- fested with trematode larvae (Uzmann 1953; Hopkins 1954; Sanders 1966). In other cases, effects are more subtle. The incidence of hermaphroditism in C. virginica increased as the result of infection with Bucephalus sp. (Tripp 1973). The gametogenic cycle of M. edulis was retarded, but oth- erwise unimpaired by Mytilicola (Williams 1969). Al- though not measured directly, a reduction in fecundity in M. edulis infected with Polydora was inferred by Kent (1979). The results of this study show that fecundity in C. vir- ginica, as measured by the ratio of germinal area to total area in a cross section of visceral mass, is reduced in rela- tion to MSX infection intensity, and that this relationship is generally consistent over the entire MSX infection cycle, irrespective of reproductive state. A similar finding was re- ported by Ford and Figueras (1988) for oysters on the planted grounds of Delaware Bay. Ford and Figueras (1988) presented evidence suggesting that infected oysters showing clear inhibition of gametogenesis early in the re- productive period can recover from the disease and com- plete a delayed gametogenic cycle. The reduction in condi- tion associated with parasitism shown by the present study, however, suggests that "recovered" oysters would still have relatively low energy reserves and might produce fewer gametes than individuals that had never been in- fected. Stress has been shown to reduce fecundity in bivalves as the result of competition between maintenance metabolism, somatic production, and gamete production for the finite amount of energy made available from the diet (Bayne et al. 1985). Bayne et al. (1978) have shown that M. edulis produce fewer eggs having a lower energy content under conditions of temperature and nutritive stress, and that the reduction in fecundity is proportional to the level of stress. In the case of C. virginica, MSX reduces fecundity in rela- tion to intensity (and duration) of infection. Egg size is probably reduced on average in individuals infected with MSX as a result of the delaying effect on the gametogenic cycle caused by MSX (Ford and Figueras 1988). However, there was no difference in the mean diameter of mature eggs (those most likely to be spawned) between oysters in- fected and uninfected with MSX, suggesting that the re- duction in fecundity found in this study is primarily the result of a decrease in the number of eggs produced. A similar finding was reported by Barber et al. (1988) for a deep-water population of scallop, Placopecten magel- lanicus. thought to be nutritionally stressed. Although MSX is often found in gonadal tissue as the result of systemic infection, only rarely does MSX directly destroy ripe gametes (Ford and Figueras 1988). Thus it ap- pears that MSX affects reproduction in oysters indirectly by reducing available energy. This is most likely the result of a reduction in both feeding rate (Newell 1985), which re- duces the ingested ration, and in reserves stored prior to the initiation of gametogenesis. With less energy available for gametogenesis (as reflected by reduced condition), fewer gametes are produced. Oysters with epithelial infections in May and June when maximum ripeness is attained not only have a reduced food intake but may have been more heavily infected the previous fall when nutrients were being stored. The greater duration of infection and associated metabolic burden presented by systemic infections reduce relative fe- cundity to an even greater extent. The sublethal effects of MSX on oyster energy metabo- lism have potential implications for the oyster fishery. The loss in condition translates into a loss of yield (meat 30 Barber et al. weight). Even though condition is generally increasing in the fall when traditional harvesting begins, MSX preva- lence can be high at this time and act to reduce condition. Secondly, Bayne (1975) has shown that larval growth and survival is adversely affected when the parents are stressed. Thus MSX may be responsible not only for a reduction in the number of gametes produced, but in the viability of those that are produced. In lower Delaware Bay. where MSX is enzootic (Ford and Haskin 1982), an associated reduction in spatfall might be expected. However, attempts to correlate MSX prevalence with spatfall have failed to detect any impact (Ford and Figueras 1988). This would be the case if most present-day larval production is from oysters in the upper seed beds, where protection from MSX is gained from reduced salinity (Haskin and Ford 1982). Even though we are currently unable to determine larval origins, the results of the present study suggest strongly that the contribution of lower bay (planted) oysters to overall recruitment in Delaware Bay is considerably dimin- ished by the sublethal effects of MSX. ACKNOWLEDGMENTS We thank D. O'Connor and L. Ragone for the histo- pathological analyses, and R. I. E. Newell for inspiration and reviewing the manuscript. This is NJAES publication D-32504-1-88, supported by state funds and by the New Jersey Department of Environmental Protection (Bureau of Shell fisheries) and the New Jersey Commission on Science and Technology (Fisheries and Aquaculture Technology Extension Center). REFERENCES CITED Andrews. J. D. 1966, Oyster mortality studies in Virginia. V. Epizooti- ology of MSX, a protistan pathogen of oysters. Ecology 47:19-31. Barber, B. J., R. Getchell, S. Shumway & D. Schick. 1988. Reduced fecundity in a deep-water population of the giant scallop Placopecten magellaniciis in the Gulf of Maine, USA. Mar. Ecol. Prog. Ser.: 42:207-212. Bayne. B. L. 1975. Reproduction in bivalve molluscs under environ- mental stress. Vemberg. F. J., ed.. Physiological Ecology of Es- tuarine Organisms. Columbia, SC: University of South Carolina Press, p. 259-277. Bayne, B. L. 1976. Aspects of reproduction in bivalve molluscs. Wiley, M. L., ed., Estuarine Processes. New York, NY: Academic Press, p. 432-448. Bayne, B. L., D. L. Holland. M. N. Moore. D, M. Lowe & J. Widdows. 1978. Further studies on the effects of stress in the adult on the eggs of Mytilus eciulis. J. Mar. Biol. Assoc. UK 58:825-841. Bayne, B L, J. M. Gee, J. T. Davey & C. Scullard. 1979. Physiological responses of Mytilus eiliilis L. to parasitic infestation by Mylilicola inleslinalis. J. Cons. int. E.xplor. Mer 38:12-17. Bayne, B. L., D. A. Brown, K. Bums, D. R. Dixon, A. Ivanovici, D. R. Livingstone, D. M. Lowe, M. N. Moore, A. R. D. Stebbing & J. Widdows. 1985. The Effects of Stress and Pollution on Marine An- imals. New York, NY: Praeger Publishers. 384 p. Cole. H. A. & R. E. Savage. 1951. The effect of the parasitic copepod, Mylilicola inleslinalis (Steuer) upon the condition of mussels. Parasi- m/og.v 4:156- 161. Eble. A. F. 1966. Some observations on the seasonal distribution of se- lected enzymes in the American oyster as revealed by enzyme histo- cheinistry. Proc. Natl. Shellfish. Assoc. 56:37-42. Ford, S. E. 1985. Chronic infections oi Haplosporidium nelsoni (MSX) in the oyster Crassoslrea virginica. J. Invertebr Pathol. 45:94-107. Ford, S. E. 1986. Comparison of hemolymph proteins from resistant and susceptible oysters. Crassoslrea virginica. exposed to the parasite Haplosporidium nelsoni (MSX). J. Invertebr. Pathol. 47:283-294. Ford, S. E. & A. J. Figueras. 1988. Effects of sublethal infection by the parasite Haplosporidium nelsoni (MSX) on gametogenesis. spawning. and sex ratios of oysters in Delaware Bay, USA. J Aquai. Diseases: In press. Ford, S. E. & H. H. Haskin. 1982. History and epizootiology oi Haplo- sporidium nelsoni (MSX), an oyster pathogen, in Delaware Bay, 1957-1980. J. Invertebr. Pathol. 40:118-141. Ford, S. E. & H. H. Haskin. 1987. Infection and mortality panems in strains of oysters Crassoslrea virginica selected for resistance to the parasite Haplosporidium nelsoni (MSX). J. Parasii. 73:368-376. Ford, S. E. & S. A. Kanaley. 1988. An evaluation of hemolymph diag- nosis for detection of the oyster parasite Haplosporidium nelsoni (MSX). J. Shell/. Res.: 7:. Gabbott, P. A. 1976. Energy metabolism. Bayne, B. L., ed., Marine Mussels. Cambridge: Cambridge University Press, p. 293-355. Gabbott, P. A. 1983. Developmental and seasonal metabolic activities in marine molluscs. Hochachka, P. W.. ed.. The Mollusca. Vol. 2. New York, NY: Academic Press, p. 165-217. Galtsoff, P. S. 1964. The American Oyster {Crassoslrea virginica). U.S. Fish Wildl. Serv. Fish. Bull. 64. 480 p. Haskin, H. H. & S. E. Ford. 1979. Development of resistance to Min- chinia nelsoni (MSX) mortality in laboratory-reared and native oyster stocks in Delaware Bay. ^.5. Nail. Mar. Fish. Sen\ Mar. Fish. Rev. 41:54-63. Haskin. H. H. & S. E. Ford. 1982. Haplosporidium nelsoni (MSX) on Delaware Bay seed oyster beds: a host-parasite relationship along a salinity gradient. J. Invertebr. Pathol. 40:388-405. Haskin. H. H.. L. A. Stauber & J. G. Mackin. 1966. Minchinia nelsoni n. sp. (Haplosporida. Haplosporidiidae): causative agent of the Dela- ware Bay oyster epizootic. Science 153:1414-1416. Hopkins. S. H. 1954. Cercaria brachidonlis n. sp. from the hooked mussel in Louisiana. J. Parasii. 40:29-31. Katkansky. S. C. A. K. Sparks & K. K. Chew. 1967, Distribution and effects of the endoparasitic copepod. Mylilicola orienialis. on the Pa- cific oyster. Crassoslrea gigas. on the Pacific coast. Proc. Nail. Shellfish. Assoc. 57:50-58. Kent. R. M. L. 1979. The influence of heavy infestations of Polydora ciliata on the flesh content of Mytilus edulis. J. Mar. Biol. Assoc. UK 59:289-297. Lauckner. G. 1983. Diseases of mollusca: bivalvia. Kinne. O.. ed.. Dis- eases of Marine Animals. Vol. II. Hamburg. West Germany: Biolo- gische Anstalt Helgoland, p, 477-962. Loosanoff, V. L. 1965. Gonad development and discharge of spawn in oysters of Long Island Sound . Biol. Bull. ( Woods Hole) 1 29:546- 56 1 . Lunz. G. R. 1941. Polydora. a pest in South Carolina oysters. J. Elisha Mitchell Scient. Soc. 57:273-283. Mann. R. 1978. A companson of morphometric. biochemical and physio- logical indexes of condition in marine bivalve molluscs. Thorpe. J. H. and J. W. Gibbons, eds.. Energy and Environmental Stress in Aquatic Systems. Washington. DC: Technical Information Center. U.S. Dept of Energy, p. 484-497. Mori. K. 1979. Effects of artificial eutrophication on the metabolism of the Japanese oyster Crassoslrea gigas. Mar. Biol. (Berl.) 53:361- 369. MSX Effects on Oyster Condition and Fecundity 31 Newell. R, I, E. 1985. Physiological effects of the MSX parasite Haplo- sporidium netsom (Haskin, Stauber & Mackin) on the American oyster Cnissostrea virginica (Gmelin). J. Shellfish Res. 5:91-95. Newell. R. I. E. & B. J. Barber. 1988. A physiological approach to the study of bivalve molluscan diseases. Fisher, W. S., ed.. Disease Pro- cesses in Mannc Bivalve Molluscs. American Fisheries Society. Spe- cial Publication: In Press. Sanders. M. J. 1966. Parasitic castration of the scallop Pecien alha (Tate) by a bucephalid trematode. Nature 212:307-308. Thompson, S. N. 1983. Biochemical and physiological effects of meta- zoan endoparasites on their host species. Comp. Biochcm. Physiol. 74B:183-211. Tripp, M. R. 1973. Hermaphroditism in 6Hcf/)/i 16%f) and the size of the oyster har- vest the previous season. In general, the 4 years of fair recruitment between 1966 and 1979 were characterized by high salinity while salinity during the poor years was low. Above average salinity prevailed during 1980-82 and 1985-86, and high recruitment was observed in many areas of Chesapeake Bay in Maryland (Krantz et al., 1982; Krantz and Davis, 1983; W, Outten, Maryland Department of Natural Resources, Annapolis. Maryland, pers. comm.). Although oysters are found throughout much of Chesa- peake Bay, low population density or harvesting difficulty may limit the extent to which some areas can be worked. The Flag Pond oyster bar on the western shore of Chesa- peake Bay near the Calvert Cliffs Nuclear Power Plant (CCNPP) was such an area prior to the early 1980s. The bar was included in the 1906-1912 Survey of Maryland Oyster Bars (Yates, 1913) and consists of 275 ha (680 acres), much of which is now shifting sand (inshore) or soft mud (offshore, depth >9 m), both unsuitable for oysters. Studies in 1968 (ANSP, 1968) and 1979 (Abbe. 1980) esti- mated a population size of 5-7 x 10' bu (based on 350 oysters per bushel) on about 81 ha of productive bottom (60-85 bu ha '). In some areas of the bar where oysters were found, however, many were attached to large rocks, effectively reducing the actual catchable density. In 1980, 102 X 10' bu of shell were planted in the dis- charge area of CCNPP. and in 1982 another 197 x 10' bu were planted near Camp Conoy, both within the boundaries of Flag Pond bar. Subsequent spat sets near the plant were higher than recent averages, although 1980 (25-50 spat bu"') was just slightly better than during 1975-79 (0-10 spat bu"') (Davis et al., 1981). In 1981, the spat density was 140 bu"' of material from natural bottom and 1060 bu"' on planted shell (Krantz et al.. 1982); the 1982 den- sity was also over 200 spat bu" ' (Krantz and Davis, 1983). Because higher spat densities can lead to increased popula- tion densities and ultimately to higher yields of oysters, we began a program to determine the success of the shell planting and the newly settled spat over a multi-year pe- riod. This paper presents the results of spring and fall surveys of Flag Pond bar during 1979 and from 1983 to 1986. It documents temporal changes in population structure and relates shell planting efforts to larval recruitment and sub- 33 34 Abbe Figure 1. Location of Flag Pond oyster bar in central Chesapeake Bay adjacent to the Calvert Cliffs Nuclear Power Plant (CCNPP) Shell Planting and Increased Oyster Recruitment 35 sequent commercial activity. It also demonstrates the value that shell planted at least 2 years can have for recruiting larvae. MATERIALS AND METHODS Description of Study Area Flag Pond oyster bar is located 12 km north of the mouth of the Patuxent River on the western shore of Chesapeake Bay (Fig. 1 ); it is 4.4 km long and about 1 . 1 km wide at its widest point. The potential productive oyster bottom was estimated in 1979 to be 81 ha or 29% of the total area (Abbe, 1980). Much of the inshore bottom was littered with rocks eroded from the 3()-m-tall cliffs along the shore, and oysters on this rocky bottom were difficult to harvest by conventional methods (hand tonging, patent longing, and dredging). Water depth over the bar ranges from less than 1 m to 10 m, but most of the oysters are confined to depths less than 8 m. Water temperatures range from near 0°C in winter to about 28°C in summer, and annual mean salinities gener- ally range from about 10 to \57c( (Abbe, 1983). Dissolved oxygen (DO) concentrations in summer range from 5 to 9 mg Ojl"' in water less than 8 m deep, but are often 1 mg Oil"' or less in depths greater than 9 m. DO concentrations less than 3 mg O2I"' in shallow water occasionally occur for a day or so as a result of upwelling caused by west or southwest winds. Sampling The study was conducted from the 42-ft research vessel JOSEPH LEIDY during May-June and October-No- vember 1979 and 1983-86 in the four areas outlined in Figure 2 selected from earlier mapping studies. Area 2 lay in the mimediate discharge area of the CCNPP. It consisted of about 1 1 .4 ha of shell-covered hard clay bottom and had shell spread on part of it in 1980. Area 4 (Camp Conoy) was 16.4 ha in size and had a sand, rock, and shell bottom; it received shell in 1982. Area 5 was a 19.5-ha continuation of Area 4. Its bottom was sand and shell, and it also had shell planted in 1982. A 4-ha tract at the common boundary of Areas 4 and 5 received an additional 70 x 10^ bu of shell in 1984. All three areas were surveyed twice each year, and Area 6 was added to the 1983-86 surveys. Area 6 was a narrow 3.9-ha strip adjacent to and inshore of Areas 4 and 5. The bottom consisted of rock on hard clay, shell on sand, and sand. No shell was planted in this shallow area during the study. Each area of study was outlined with buoys positioned with the aid of radar and LORAN C. Within each area a series of transects was made, with several points sampled along each transect. The LEIDY was anchored at a sam- pling location and three steel squares (each 0.33 m^) con- nected in series by 2-m lines were thrown from each side so they spread apart and fell to the bottom. Water depth and 1980 2000 Meters Figure 2. Outlines of areas sampled on Flag Pond during 1979 and 1983-86. Shaded areas show approximate areas and dates of shell planting. turbidity prevented sight of the bottom from the LEIDY and thus prevented the bias that could have existed if the squares were deployed in clear water. These methods were similar to the random quadrat sampling by May (1971) in his survey of the Alabama oyster resources. Divers removed all materials including oysters, shells, boxes (two attached empty shells), and rocks from each square and put them into nylon-mesh catch bags which were brought to the surface for examination. Shells and boxes were counted (except in 1979), and oysters were counted by size class: legal or 76 mm (3 inch) and over, sublegal, and spat. Sublegal oysters were those less than 76 mm that had set at least the prior year, and spat were oysters that had set during the present setting season (July- September). After all counts were recorded the LEIDY was moved to a new location and the procedure repeated. The number of sample locations in each area varied ac- cording to areal size, but ranged from 9 (18 m-) to 31 (62 m^). May (1971) determined that a sample density of 2.06 m^ ha^' (1 yd- acre"') was sufficient to adequately sample the Alabama oyster population. During 5 years of the 36 Abbe present study we averaged 25 1 ni'^ sampled annually on 5 1 ha (4.92 m^ha"'). Statistical Analysis Data was analyzed using a nested analysis of variance (General Linear Model Procedure) (SAS Institute, Inc.. 1982) in which all factors were assumed to be random. The model takes the form, log (7,jki + 1) = 11 + a. + Pj(a,) + 7kOj) + 5,(7k) where 7ijki = parameter measurement (number of shells, boxes, or oysters) for area,, locationj, positioHij, and replicatej. |x = overall mean, a, = area effect, p. = location-within-area effect, 7i; = position (port or starboard)-within-location effect, and 8] = replicate-within-position effect When significant differences among areas were detected (p < 0.01), a Student-Newman-Keuls (SNK) test was used to rank the areas according to their mean densities and de- termine where the differences occurred. RESULTS Densities of shells and boxes were highly variable from year to year, with shells increasing markedly after shell planting; e.g., in 1984, shells increased from 82 to 228 m"- (Table 1). with some individual samples yielding 1000 m~^ or more. Boxes appeared numerous in 1983 (Table 1 ). but this resulted from one unusual sample (1 m^) in Area 2 ( 1075 boxes) and three samples from Area 4 (249, 137, and 1 14 boxes) where very high densities of shells and sublegal oysters were found. Otherwise, the density of boxes in 1983 was similar to those of other years. In 1986 the den- sity of boxes increased sharply from spring to fall during the second successive year of high salinity (Fig. 3). Less variable than shells or boxes were the densities of legal oysters which increased each spring from the previous spring (Table 1 ) due to growth of sublegal oysters which reached 76 mm in length. Legal densities also increased from spring to fall each year for the same reason. Only in 1986 did the density of legal oysters decrease from spring to fall. Legal density of 1.7 m"^ in 1979 increased to 4.3 m~^ by 1983 as oysters that set in 1980 and 1981 began to reach legal size. Legal density peaked at 5.4 m"- in 1985 before decreasing slightly in 1986. Sublegal densities were fairly stable from fall of one year to spring of the next when recruitment was poor (1983 and 1984). When recruitment was good, however, as in 1985, the density of sublegals increased markedly the fol- lowing spring as spat became part of the sublegal popula- tion (Table 1). Densities less than 1 m^- in 1979 reached 33 m^- in spring 1983 (Table 1 ) as a result of successful recruitment in 1981 and 1982, then decreased to 22 m- by fall 1983 as some oysters reached 76 mm and entered the legal class and others died. The increase in legal density from spring to fall and the decrease in sublegal density during the same time seen in Table 1 was exhibited in each of the four individual areas (Table 2). Areas 4, 5 and 6, which were adjacent to each other, were similar in legal densities (4.6 to 5.4 m"'), but far different in sublegal densities which averaged 34.3 m~-^ in Area 5, 17.4 m"^ in Area 4, and only 4.7 m"^ in Area 6. In Area 2, adjacent to the CCNPP discharge, the mean TABLE 1. Densities of shells, boxes, and live oysters per m^ collected during spring and fall 1979 and 1983-86 from Flag Pond bar. Area Surveyed Legal Sublegal (m^) Shells m -2 Boxes m ^ Oysters m"^ Oysters m ~ ^ Spat m ^ Spring 1979 125 * * 1.7** 0.8** — 1983 105 64.3 18.9 2.7 33.2 — 1984 110 82.2 3.3 4.0 22 2 — 1985 182 135.2 1.4 4.9 17.2 — 1986 150 123.8 2.5 5.8 17.7 — Mean 134 101.3 6.5 3.8 18.2 Fall 1979 112 * * 1.7** 0.6** 0.2** 1983 112 73.2 6.2 5.8 22.3 1.8 1984 110 228.2 2.4 5.3 18.3 1.1 1985 132 243.6 2.1 6.1 8.0 23.4 1986 119 143.0 9.3 4.0 19.2 36.8 Mean 117 172.0 5.0 4.6 13.7 12.7 * Shells and boxes not counted ** Areas 2, 4. and 5 only Shell Planting and Increased Oyster Recruitment 37 20 / \ /I ^^ 8 / \ / \ .. / \ / °, 15 \ \. / \^ / > 1- ? 10 \ \ / \ -— / /' N \ / \ / _l < CO 5 1 1 1 1 r 1 1 1 1 i 1 1 1 1 1 1 1 1970 1972 1974 1976 1978 1980 1982 1984 1986 Figure 3. Annual salinity means and ranges from 1969 to 1986 near Calvert Cliffs, Maryland. legal density was 2.3 m ^ and mean sublegal density was 6.0 m-2. Spat densities were similar in Areas 4 and 5 (18.2 and 18.4 m"-. respectively), and lower in Area 2 (9.0 m"- (Table 2). Spat density was much lower in Area 6 (1.4 m"-, only 8% of the densities in Areas 4 and 5 and 16% of the density in Area 2. Although spat density was much lower in Area 6 than in Areas 4 and 5, the ratio of spat per shell was not. Combining 1985 and 1986 data revealed 0.16, 0.13, and 0.11 spat per shell in Areas 4, 5. and 6, respectively. Oysters counted as spat in the fall were considered sub- legals the following spring and eventually became legal (if they survived), yet the sublegal density was only half the spat density in Area 2. about equal to spat density in Area 4, and twice the spat density in Area 5. The sublegal den- sity of Area 6 was 3.5 times spat density. The low ratio of sublegals to spat in Area 2 compared to the high ratio in Area 6 indicates poorer survival of spat in Area 2 (dis- charge area of CCNPP) than in an undisturbed area. The ratio of legal oysters to sublegals was only 0. 13 in Area 5. but was 0.31 and 0.38 in Areas 4 and 2, respec- tively. In Area 6. where sublegals were less abundant and harvesting was minimal, the ratio was 1.0. Data analysis revealed no significant differences among areas in 1983 (Table 3a). but in the fall of 1984. the density of sublegal oysters in Area 5 was significantly greater (p < 0.01) than in Areas 2 or 6 (Tables 3a and 3b). In spring 1985, shell densities were greater in Areas 2 and 5 than in Area 6, but legal oyster densities were greater in Areas 4 and 6 than in Area 2 (Table 3b). In the fall of 1985, the spat TABLE 2. Mean density of legal, sublegal, and spat oysters per m^ in four areas of Flag Pond bar for 5 years. Area 6 mean is for 1983-1986 only. Legal Sublegal Spat Total Area 2 Spring 1.8 7.2 Fall 2.8 4.7 Mean 2.3 6.0 Area 4 Spring 5.2 22.1 Fall 5.6 12.7 Mean 5.4 17.4 Area 5 Spring 4.1 36.1 Fall 5.1 32.5 Mean 4.6 34.3 Area 6 Spnng 4.6 5.2 Fall 4.8 4.2 Mean 4.7 4.7 9.0 18.2 18.4 1.4 9.0 16.5 27.2 36.6 40.2 56.0 9.7 10.3 38 Abbe TABLE 3a. Summary of F values testing the ratios of area variances to location variances from an analysis of the four areas of Flag Pond bar. Legal Sublegal df Shells Boxes Oysters Oysters Spat 1983 Spnng 3/49 1.55 1.48 0.77 0.34 — Fall 3/52 2.65 1.93 1.48 0.33 0.33 1984 Spring 3/51 1.26 1.01 3.22 2.31 — Fall 3/51 1.72 0.75 1.35 4.42** 2.58 1985 Spring 3/87 4.58** 0.36 4.39** 2.69 — Fall 3/62 3.15 2.07 0.74 2.34 4.42** 1986 Spnng 3/71 4.94** 0.87 3.95 3.58 — Fall 3/56 1.60 3.39 2.81 2.04 3.36 ** F statistic significant at p =£ 0.01 density in Area 6 was significantly less than elsewhere (Table 3b). Only shells differed among areas in the spring of 1986, with the density in Area 5 significantly greater than in Areas 4 or 6 (Table 3b). Densities in Table 3b were computed using means of log values, and thus differ some- what from the arithmetic means in Table 1 . Although dif- ferences among areas were few, differences among various locations and positions were common. DISCUSSION Shell densities were unknown prior to shell planting in 1980 and 1982 since shells were not counted in 1979, but they were probably similar to the mean densities in Area 6 from 1983 to 1986 (19.4 m"-^) because no shells were ever planted there. During 1983 and early 1984, overall shell densities were 60-80 m - (Table 1), but more shell planting in June 1984 increased the density nearly three- fold. Although larvae set wherever they could find hard TABLE 3b. Results of Student-Neuman-Keuls tests applied to significant differences in Table 3a. Densities are per m^, and areas not significantly different are connected. 1984 (Fall) Sublegal oysters Area 5 59.6 Area 4 11.5 Area 2 3.9 Area 6 3.2 Area 2 43.8 Area 6 4.0 1985 (Spring) Shells Area 5 43.6 Area 4 3.1 Area 4 18.8 Area 6 7.1 Legal oysters Area 5 2.1 Area 2 1.0 Area 4 13.0 Area 5 79.0 Area 2 10.2 Area 2 27.4 1985 (Fall) Spat Area 5 6,9 Area 4 17.5 Area 6 1.6 1986 (Spring) Shells Area 6 11.5 surfaces (shells, rocks, or oysters), most set where shell had been planted at densities of 1000 m~^ or more. In 1983. densities of sublegal oysters up to 600-800 m~^ were found on shell planted in 1982. Adjacent to the planted areas sublegal densities were only 5-20 m~- or less. The shell planted in 1984 (Fig. 2) caught very few spat that year, but salinity was low (Fig. 3), and recruitment was poor throughout much of the Bay. In 1985 and 1986 mean salinity was above 17%o, and oyster recruitment was similar to that of the early 1980s. There is little doubt about the correlation between salinity and recruitment as sug- gested by Ulanowicz et al. (1980), but whether increased recruitment was solely because of higher salinity or be- cause of other factors associated with lack of rainfall, such as reduced runoff carrying pollutants into the Bay, is un- clear. It is clear, however, that salinities were lower in the mid Bay during the 1970s (mean 12.0%c) than they were from 1980 to 1986 (mean 15.4%f; Fig. 3). Three excellent recruitment years occurred in 1981, 1985, and 1986 when salinity ranged from at least 13%c to over 20%f, and aver- aged above 17%c (Fig. 3). Overall mean spat densities in Areas 2, 4, and 5 were 15, 43, and 25 m^^, respectively, in 1985 and 28, 43, and 62 m~^, respectively, in 1986. These means, however, included samples from shell- planted bottom as well as barren bottom. Sample densities on planted shell alone were as high as 277 m"- in 1985 and 510 m~^ in 1986. Spat in Area 6, where shell was never planted, were 2 and 3 m"^ in 1985 and 1986, respectively. Had cultch been planted in Area 6, it too might have had spat densities of 15-60 m"' instead of only 2-3 m'^'^. Earlier studies (Beaven, 1948; Engle, 1956) indicated that more larvae set on newly planted shell than on old shell, and Shaw (1967) recommended planting shell in Broad Creek. Maryland during the first week of July to serve as cultch for larvae that would begin setting soon thereafter. These studies imply that cultch should be planted as close to the time of larval setting as possible, so that setting is not impeded by fouled cultch. This is still a wise management practice, as far as it is practical to do so. Yet in 1984, the newly planted shell on Flag Pond caught very few spat because few larvae were available to set. In October 1985 and 1986, mean densities of 10- to 25-mm spat on this 1984 shell were 172 and 260 m"^, respec- tively; and the highest sample density in 1986 (510 m"^) was on cultch planted in 1982. These high densities indi- cate that planted shell can still be highly valuable as cultch even after 4 years. Although additional cultch will improve recruitment in some areas, it will not do so everywhere. It will not guar- antee good setting in historically poor recruitment areas such as the Patuxent River in Maryland (Sieling, 1950; Kennedy and Breisch, 1981). Nor will it improve setting where the density of brood stock is inadequate to supply enough larvae (Andrews, 1983; Haven and Fritz, 1985). Shell Planting and Increased Oyster Recruitment 39 Planted shell was obviously critical to increased recruit- ment on Flag Pond, but shells were planted too heavily in some locations resulting in loss of oysters from sedimenta- tion and biodeposition. Where shell was spread in a layer 30 cm or more thick in 1982, larvae set throughout the en- tire mass. In 1983 the densities of 30- to 60-mm oysters in this mass were 400-800 m"-'. By 1984 these densities were reduced to 50-60 m"'^, all confined to the upper cultch layer; oysters below these had been smothered by sediments and biodeposits. Haven et al. (1981) indicated increased sedimentation in the James River in Virginia over the last three decades and suggested it could have affected recruitment. Biodeposition, however, can pose an even greater threat than sedimentation. Haven and Morales- Alamo (1966) determined that oysters deposited filtered material seven times faster than it settled by gravity. Lund ( 1957) calculated that oysters covering a hectare of bottom could produce 18.7 metric tons (dry weight) of fecal mate- rial in 11 days. Once this material settles into the crevices of a thick shell layer it is difficult for currents to remove. Thus it builds up and eventually smothers the oysters below the surface layer. Thinner layers of shell planted over larger areas may reduce mortalities from smothering and result in larger harvestable populations. The density of legal oysters on Flag Pond bar peaked in fall 1985 at 6.1 m"'. and declined only slightly to 5.8 m"- in spring 1986. despite heavy fishing pressure by oys- termen. By fall 1986, legal density could have reached the highest level during this study as large numbers of oysters recruited in 1981 and 1982 became legal. Unfortunately, the high salinity which allowed the excellent recruitment in 1985 and 1986 may also have allowed •"MSX" caused by Haplosporidium nelsoni or "Demio"" caused by Perkinsus marimis to invade and kill large numbers of oysters. Dredging in Areas 2, 4, and 5 from August to October 1986, as part of another study, indicated two or more boxes for each live legal oyster, and the present data (Table 1 ) confirm this ratio (2.3:1). Small oysters seemed less af- fected. Newly set spat and oysters less than 2 years old appear less susceptible to infection from Dermo than older oysters (Andrews and Hewatt, 1957; Andrews, 1966). but MSX may infect and kill all sizes (A. Farley, NMFS, Ox- ford, Maryland, pers. comm). Ratios of legal to sublegal density in the four areas re- flected commercial activity. These ratios were low in Areas 2. 4, and 5 where oysters were removed from the popula- tion as they became legal. Little harvesting occurred in the rocky shallows of Area 6 which allowed legal oysters to reach higher densities relative to small oysters than else- where. Based on total densities determined by summing legals, sublegals. and spat (Table 2). Area 5 has the best commer- cial potential of all four areas. Area 4's potential is lower, but its immediate prospects are best with a legal density of 5.4 m'". The potential of Area 2 is third best, but its im- mediate prospects are less than those of Area 6. which has the poorest commercial potential of all (because of low sublegal and spat densities). The low overall density of Area 6 may be unimportant from a commercial viewpoint, however, since it is a small area, and little activity occurs there; but because the legal density is similar to that of Areas 4 and 5, the oysters could be an important source of reproductive material for other areas. Spat were scarce at all locations in 1979. a condition perhaps similar to that occurring during much of the 1970s; however, larval recruitment was vastly improved in the early 1980s. Evidence of this improvement comes from the surveys of Krantz et al. (1982), Krantz and Davis (1983). and from the increase in sublegal oysters observed in this study from 1979 to 1983. Low spat densities of only 1.8 and 1.1 m~^ occurred in the wet years of 1983 and 1984 (Fig. 3), but the dry years of 1985 and 1986 resulted in spat densities of 23.4 and 36.8 m"^, respectively. The outlook for this area of the Bay is better now than at any time in the past two decades, provided that disease mortality is not too great. The statistical differences common among locations and between positions probably resulted because of patchy dis- tribution of oysters within larger areas. Even when areal densities were similar, some locations supported large com- ponent densities while others supported low densities or were barren. Patchy distribution of spat is typical on bottom consisting of scattered rocks and shells, because larvae are limited by where they can set. Even where favor- able substratum exists, e.g.. where shells have been planted, oysters may set gregariously as suggested by Crisp (1967). Hidu (1969) demonstrated in lab experiments that the presence of spat on cultch attracts more larvae and stim- ulates setting. The oyster population of Flag Pond bar was estimated to be 7 X 10-^ bu in 1979 based on 200 legal oysters or 400 sublegals per bushel (spat were not included in the esti- mate). By 1983, the population estimate had increased to 54 X 10^ bu. Although population size decreased during 1984-85 because of natural mortality and heavy pressure from commercial harvesting, the population was estimated at 43 X 10' bu in 1986, more than six times greater than in 1979. With overall sublegal densities more than 18 m"- in 1986 and spat at nearly 37 m"^, the prospects for increased harvests from Flag Pond bar over the next few years are good, provided lower (normal) salinities reduce the sus- pected disease problem. Flag Pond is not unlike many other oyster bars in central Chesapeake Bay. and if similar re- cruitment has occurred on them over the past 2 years, Maryland oyster harvests should begin to improve from the poor season of 1986-87 with increased catches for the next few years. A comparison of Areas 4 and 5 with Area 6 may best illustrate the value of shell planting, even after several 40 Abbe years. Portions of Areas 4 and 5 were planted with siiell in 1982 and 1984; Area 6 received no shell. The mean sub- legal density in Areas 4 and 5 was 25.8 m"^: the mean sublegal density in Area 6 was 4.7 m~^. The mean spat density in Areas 4 and 5 was 18.3 m"-. Area 6 had 1.4 spat m"^. These differences in oyster density between planted and unplanted areas are considerable and result from the availability and absence, respectively, of suitable sub- stratum on which larvae can set. The number of larvae that can set is proportional to the area of clean shell surface (MacKenzie, 1983). Numbers of spat per shell were much more similar among these areas than spat densities. During years of low larval recruitment, shell planting at first appears to waste resources, but recruitment success is difficult if not impossible to predetermine. If planting is conducted annually, recruitment during some years will surely be inadequate to increase population, e.g., most of the 1970s. Shell planting during poor reproductive years, however, can still be a valuable practice as shown by the high spat densities of samples in 1986 from 4-year-old planted shell. Planted shell may be of little value, however, if buried by shifting bottom or spread in water deeper than 9 m which may become anoxic in summer (Taft et al., 1980; Officer et al.. 1984). but when put down prior to above average recruitment, it can result in much greater oyster densities than would otherwise occur. Shell planting therefore continues to be a major means to increase oyster production in Maryland. ACKNOWLEDGMENTS Sincere appreciation goes to W. L. Yates. Jr., E. M. Newman, T. A. Thoman, K. R. Braun, and T. R. Poe for their efforts underwater collecting samples. Special thanks also go to E. S. Perry for his statistical analyses and to J. G. Sanders and J. N. Kraeuter for their critical reviews of the original manuscript. This study was supported throughout by the Baltimore Gas and Electric Company. LITERATURE CITED Abbe, G. R. 1980. Oyster population survey at Calvert Cliffs. Maryland. Report prepared for Baltimore Gas and Electric Company. 16 pp. Available from: The Academy of Natural Sciences, Philadelphia, PA. Abbe, G. R. 1983. Blue crab (Callinecles sapidus Rathbun) populations in mid-Chesapeake Bay in the vicinity of the Calvert Cliffs Nuclear Power Plant. 1968- 1981. J. Shellfish Res. 3:183- 193. Abbe, G. R. 1986. A review of some factors that limit oyster recruitment in Chesapeake Bay. Amer. Malac. Bull.. Spec. Ed. No. 3:59-70. Academy of Natural Sciences of Philadelphia (ANSP). 1968, A survey of oyster density on the upper portion of Flag Pond Oyster Bar, Chesa- peake Bay. Maryland. 4 pp. Available from: The Academy of Natural Sciences, Philadelphia, PA. Andrews, J. D. 1966. Oyster mortality studies in Virginia. V. Epizooti- ology of MSX, a protistan pathogen of oysters. Ecology 47:19-31. Andrews, J. D. 1983. Transport of bivalve larvae in James River, Vir- ginia. J. Shellfish Res. 3:29-40. Andrews. J. D. and W. G. Hewatt. 1957. Oyster mortality studies in Vir- ginia. II. The fungus disease caused by Dermocyslidiiim inaruuim in oysters of Chesapeake Bay. Ecol. Monogr. 27:1-26. Beaven, G. F. 1948. Observations on fouling of shells in the Chesapeake area. Proc. Natl. Shellfish. Assoc. (1947):1 1- 15. Crisp, D. J. 1967. Chemical factors inducing settlement in Crassostrea virginica (Gmelin). J. Animal Ecol. 36:329-335. Davis. H. E.. D. W. Webster, and G. E. Krantz. 1981. Maryland oyster spat survey, fall 1980. Maryland Sea Grant Tech. Rept. UM-SG-TS- 81-03. University of Maryland, College Park. MD. 22 pp. Engle, J. B. 1956. Ten years of study on oyster setting in a seed area in upper Chesapeake Bay. Proc. Natl. Shellfish. Assoc. 46:88-99. Haven. D. S. and L. W. Fritz. 1985. Setting of the American oyster Crassostrea virginica in the James River. Virginia, USA: temporal and spatial distribution. Mar. Biol. 86:271-282. Haven. D. S.. W. J. Hargis. Jr.. and P. C. Kendall. 1981. The oyster industry of Virginia: its status, problems and promise. A comprehen- sive study of the oyster industry in Virginia. 2nd edition. Spec. Papers Mar. Sci. No. 4. Virginia Institute of Marine Science. Gloucester Point. VA. 1024 pp. Haven. D. S. and R. Morales-Alamo. 1966. Aspects of biodeposition by oysters and other invertebrate filter feeders. Limnol. Oceanogr. 11:487-498. Hidu. H. 1969. Gregarious setting in the American oyster, Crassostrea virginica Gmelin. Chesapeake Sci. 10:85-92. Kennedy. V. S. and L. L. Breisch. 1981. Maryland's oysters: research and management. Maryland Sea Grant No, UM-SG-TS-81-04. Uni- versity of Maryland. College Park, MD. 286 pp. Krantz. G. E.. H, A, Davis and D, W. Webster, 1982, Maryland oyster spat survey, fall 1981. Maryland Sea Grant Tech. Rept. UM-SG-TS- 82-02, University of Maryland, College Park. MD. 14 pp. Krantz, G. E. and H. A. Davis. 1983. Maryland oyster spat survey, fall 1982, Maryland-Sea Grant Tech. Rept, UM-SG-TS-83-OI. University of Maryland. College Park. MD. 14 pp. Lund. E, J. 1957. A quantitative study of clearance of a turbid medium and feeding by the oyster, Piihl. Inst. Mar. Sci. University of Texas, 4:296-312. MacKenzie. C. L.. Jr. 1983. To increase oyster production in the north- eastern United States. Mar. Fish. Rev. 45(3): 1-22. May. E. B, 1971. A survey of the oyster and oyster shell resources of Alabama. Alabama Mar. Res. Bull. No. 4. Alabama Dept. Conserv.. Dauphin Island. AL. 53 pp. Officer. C. B.. R. B. Biggs. J. L. Taft. L. E. Cronin, M. A. Tyler, and W. R. Boynton. 1984. Chesapeake Bay anoxia: origin, development. and significance. Science 223:22-27. SAS Institute. Inc. 1982. SAS users guide to statistics. 1982 ed. SAS Institute. Inc., Cary. NC. 584 pp. Shaw. W. N. 1967. Seasonal fouling and oyster setting on asbestos plates in Broad Creek, Talbot County. Maryland. 1963-65. Chesapeake Sci. 8:228-236. Sieling, F. W. 1950. Intensity and distribution of oyster set in Chesapeake Bay and tributaries. Proc. Nail. Shellfish. Assoc. (1949):28-32. Taft. J. L.. W. R. Taylor. E. O, Hartwig. and R. Loftus. 1980. Seasonal oxygen depletion in Chesapeake Bay. Estuaries 3:242-247. Ulanowicz, R. E.. W. C. Caplins and E, A, Dunnington, 1980. The fore- casting of oyster harvest in central Chesapeake Bay. Esiuarine Coastal Mar. Sci. 11:101-106. Yates. C. C. 1913. Summary of survey of oyster bars of Maryland (1906-1912), U.S. Coast and Geodetic Survey. Washington, DC. 81 pp. Journal of Shellfish Reseiinh. Vol. 7, No. 1. 41-45. 1488. THE APPLICATION OF HYDROACOUSTICS TO THE MAPPING OF SUBTIDAL OYSTER REEFS JOSEPH T. DEALTERIS Department of Fisheries, Animal and Veterinary Science University of Rhode Island Kingston. Rhode Island 02882 ABSTRACT Hydroacoustic techniques were used to delineate the extent of subtidal oyster reefs in the three tributaries of Chesa- peake Bay. The instruments included a precision survey echo sounder operating at 2nO kHz. and a side scan sonar system operating at 100 kHz. The sounder proved to be a useful and economical tool to remote sense bottom type and topographic features along a narrow path (<2 m) across the estuary. The sonar mapped a considerably wider swath (200 m) along vessel track, providing 100 percent coverage of the bottom when adjacent echograms were mosaicked. The sonar did not provide additional capability in identifying bottom type or resolving types of oyster reefs, but produced data that was used to assess sedimentary processes and environmental impacts to the oyster reefs. In this regard, bottom scars with a total extent of almost 20 km were observed on Wreck Shoal in the sonar records. These are attributed to the passage of commercial vessels over the shallow oyster reef. KEY WORDS: Hydroacoustic mapping, oyster reefs INTRODUCTION High frequency hydroacoustic instruments (20-200 tcHz) have been used to delineate and estimate the abun- dance of finfish resources since the 1950's (Johannesson and Mitson 1983. Thome 1983). Low frequency hydro- acoustic techniques (3-10 kHz) have been utilized to investigate the subbottom character of the oyster reefs in Texas (Bouma 1976) and Virginia (DeAlteris 1987). The published literature is devoid of an evaluation of the appli- cation of high frequency hydroacoustics to the mapping of subtidal oyster reefs. Galtsoff (1964) described an oyster reef as an aggregation of live oysters and empty shells oc- cupying the bottom of an estuary. An oyster reef is further defined herein to include estuarine bottoms consisting of live oysters and shells with densities of live oysters from sparse and scattered (10/m-) to very concentrated (1000/ The purpose of the hydroacoustic surveys described in the sequel was to: 1 . delineate the limits of the subtidal oyster reefs from the adjacent bottom, 2. investigate in de- tail any bottom topographic features and textural patterns that might provide insight into the estuarine sedimentation processes operating on the oyster reefs, and 3. identify evi- dence of damage to the oyster reefs that might be associated with commercial activities on the estuary such as dredging or transportation. In coastal and estuarine waters, oystermen and re- searchers have used a variety of techniques to detect sub- tidal oyster reefs. Moore (1910) used a chain attached to a cable and dragged over the bottom to identify the character of the bottom in the James River, Virginia. When passing over an oyster reef, the vibrations of the chain dragging over shells were transmitted up the cable to the operator tending the gear. Tongs were used to periodically sample the bottom, ground truthing the results of the remote sensing technique. This technique is still used by oys- termen searching for uncharted oyster reefs to harvest. The public oyster reefs of the Virginia portion of Chesapeake Bay and its tributaries were mapped during the 1970's using an electronic version of Moore's technique (Haven et al. 1979, Haven and Whitcomb 1983). The presence or ab- sence of shells and/or oysters was monitored continuously with an underwater microphone mounted in a steel frame and dragged from a cable behind the vessel. The sounds made by the microphone bouncing over shells and oyster or sliding over sand and mud were amplified and broadcasted. The intensity and frequency of the sounds and the per- centage of time the microphone was impacting on shells or oysters or other bottom types were recorded by the oper- ators. These data were periodically compared to ground truth samples taken with tongs. Data collected from oyster reefs in three tributaries of the Chesapeake Bay are used to evaluate the effectiveness of two hydroacoustic mapping techniques. Wreck Shoal is a subtidal oyster reef located in the James River estuary, Virginia. The James River is a major tributary of the southern portion of the Chesapeake Bay. The estuary is approximately 80 km long and varies in width from 3 to 10 km. Channel depths range from 6 to 28 m. The Wreck Shoal study area is in the middle of the James River estuary with water depths ranging from 3 to 9 m, and encompasses an area of approximately 8 km^. Two significantly different types of oyster reefs are found in ad- jacent areas of Wreck Shoal study area. The hard-rock reefs are characterized by a relatively thick shell layer (>2 m), higher densities of live oysters (91/m^), a coarser interstitial sediment (37% gravel. 40% sand, 23% silt-clay) and a neg- ligible sediment cover. In contrast, the mud-shell reefs are characterized by a very thin oyster shell layer (<10 cm), considerably lower densities of live oysters (28/m-^), a finer interstitial sediment (8% gravel, 25% sand, 67% silt-clay), and a 1-2 cm layer of very fine sediments covering the reef. The hard-rock reefs are flourishing with respect to 41 42 Dealteris oyster productivity and shell deposition, and are non-accre- tional with respect to fine sediments. The mud-shell reefs are marginal in oyster productivity and shell deposition, and are accretional with respect to fine sediments (DeAI- teris 1986). The Back River is a small, shallow tributary of lower Chesapeake Bay, 9 km in length, with a mean depth of 2 m. The axis of the main channel bifurcates into the North- west and Southwest Branches of the Back River at approxi- mately the midpoint along its length. The Back River study area included oyster leases in both branches of the estuary. The mouth of the Back River is located 10 km north of the mouth of the James River. In contrast to the large, public domain natural oyster reefs of the James River, the bottom of the Back River has been divided into private lease areas for the cultivation of oysters. The productivity of these areas is dependent upon the investment made by the lease- holder in shell stock and seed oysters. The natural bottom in the major portion of the Back River is a soft mud. Areas that are actively cultivated have thick shell layers (10-20 cm) and oyster densities of \0-50/nrr distributed over sev- eral hectares. The Wicomico River is a narrow meandering tributary of the upper Chesapeake Bay, Maryland, and is more than 30 km in length. Oysters are produced on both public and private bottom in this estuary. The mouth of the estuary is relatively wide (2.5 km) and shallow (2 m). The Wicomico River study area included two lease areas and was approxi- mately 3 km upstream from the mouth of the estuary, with water depths of 2-3 m. MATERIALS AND METHODS Two acoustic methods are evaluated as to their ability to distinguish the bottom hardness and roughness and the to- pographic features associated with an oyster reef. Both in- struments operate at relatively high frequencies. The conventional method was echo sounder, which em- ploys a vertical axis acoustic beam. A Raytheon DE-719B precision survey fathometer operating at 200 kHz was uti- lized. The transducer employed 10 deg beam width di- rected toward the bottom. In shallow water, the strength of the reflection is determined principally by the reflectivity of the target (target strength), in this case the bottom. In less than 10 m of water, the width of the swath of bottom cov- erage is less than 2 m. Multiple reflections may appear on the echogram depending on the gain setting of the receiver- amplifier. By setting the gain so as not to produce a second reflection on a soft, non-reflective bottom, when the sounder passes over a harder, rougher bottom with a higher target strength, a second reflection appears on the echo- gram. Thus, by tuning or effectively adjusting the gain set- ting of a sounder, the instrument can be used to easily dis- tinguish between the hard, rough bottom of an oyster reef and a soft mud bottom. The other method, a side scan sonar, used an acoustic beam with its main axis slightly below horizontal. The beam is very narrow in the horizontal plane, yet sufficiently broad in the vertical plane to obtain echoes from a point on the bottom directly below the transducer to points 100-200 m abeam of the transducer. The combination of the beam shape and the very short length pulse length gives the sonar the capability to resolve small topographic irregularities and differences in roughness in the sea floor. As the trans- ducer is towed below the survey vessel, the reflected or backscattered echoes are graphically recorded in a form that approaches a topographic or plan view map. Projec- tions above the bottom and acoustically rough surfaces are good acoustic backscatters, and therefore darken the sono- gram. Depressions of the bottom or relatively smooth bottoms are represented by a lightening of the sonogram. A Klien Hydroscan System was used, operating with 100 kHz transducer frequency at the 100 m range, and displaying on a dual channel, analog, wet-paper recorder. Navigation control for the Wreck Shoal acoustic surveys was provided by LORAN C, using a Northstar 7000 system. The navigation unit was point calibrated at a known location at the beginning and end of each survey day. Fix marks were noted every 100 m along each tran- sect. Navigational accuracy was approximately ± 20 m, or ± 0.1 micro-second of time difference in LORAN C signals. Position determination in the Wicomico River and Back River acoustic surveys was accomplished using visual ranges to shore mounted monuments. The Wreck Shoal study area was 4 km long by 2 km wide. The track lines for the Klien sonar and the precision bathymetry survey were spaced 91 m apart. The Klien sonar was operated at the 100 m range, which resulted in approximately a 50 percent overlap in the records. The Back River study area included almost the entire estuary and transect lines for the echosounder were spaced so as to bisect individual lease areas for the identification of ac- tively cultivated areas, that could be further investigated with conventional sampling techniques. The Wicomico River study area was limited to two private lease areas near the mouth of the river. Three acoustic survey transect lines were run with the echo sounder over each oyster ground to identify the patch oyster reefs and to search for evidence of an alledged grounding by a large commercial vessel. The Wreck Shoal surveys were conducted in June, July 1984, and March 1985. The Back River survey was con- ducted in May 1981, and the Wicomico River survey was conducted in April 1985. RESULTS The hydroacoustic survey of the Back River included more than 25 private lease areas, only two of which are discussed herein. In the transect across the N.W. Branch of the Back River (Figure 1. upper echogram), the echo sounder began in the nearshore area at Langley A.F.B., crossed the channel with a maximum depth of 2.4 m, and Hydroacoustic Mapping of Subtidal Oyster Reefs 43 FT {0 M) 10 FT (3.05 M) KLIEN SIDE SCAN SONAR 26 JUNE 1985 RAYTHEON FATHOMETER 21 MAY 1981 FT (0 M) - — z^^ """ ita^ s P^ X r" i' « 10 FT (3.05 M) mt n lif -, (i 1 ^_ *-Tr '">».''?"'•' -H S tr- ff^ ' ^•^ 1 ^ - ,;_-_ -; ' 1 ^ -So - •^ --:- i '-si^ , "/ — > _ 1 -V- -^ RAYTHEON FATHOMETER 21 MAY 1981 Figure I. Echograms from the Back River, Virginia, showing tran- sects across lease tracts in each hranch of the estuary. reached the opposite shore of Tin Shell Point. The bottom type between fix marks 32 and 33 was a soft mud and be- tween fix marks 33 and 34 graded into and out of hard- shelled oyster reef. The location of this oyster reef was in- dicated by the presence of the second and third reflections in the echogram, and was confirmed by tong sampling. The mean density of oysters on this reef was 43/m-. This was a planted oyster reef, and was not perceptable above the nat- ural bottom elevation. During the transect across the S.W. Branch of the Back River (Figure 1 , lower echogram), the echo sounder passed over large topographic high, with second and third reflections evident in the echogram. Sam- pling with tongs confirmed the identification of an oyster reef with a density of 65/m'. An extensive hydroacoustic survey of the Wreck Shoal study area was conducted using both an echo sounder and side scan sonar simultaneously. Two subareas were se- lected for detailed analysis and comparison of the results of the sounder and sonar surveys. Subarea B is located along the flank of the natural channel on the inshore portion of the study area. The preci- sion bathymetry along the center line of the area and the original record of the Klien side scan sonar are shown in Figure 2. Fix mark 3.8 denotes the center of the main FT (0 M )_ 10 FT (3.05 M)-^^ RAYTHEON FATHOMETER 26 JUNE 1985 Figure 2. Echogram and sonogram from Subarea B, Wreck Shoal, James River, Virginia, showing the channel axis, the north bank of the channel, and the shallow hard-rock oyster reef. The fix marks 3.8 to 3.4 were made at 100 m intervals. channel, with step rising bank to fix mark 3.7. The channel bank rises from 15.3 to 6.1 m in a 30 m distance or about 10 m to 30 m, for a slope of about 20 deg. Along the edge of the bank, there are a series of features of nearly uniform wavelength of 10 m and a height of 0.3 to 0.9 m. From fix mark 3.7 and beyond, there is a strong second reflection in the echo sounder record, indicating a hard, rough bottom. Ground truth data collected with oyster tongs in this area confirms the presence of a dense oyster reef (91/m-). On the shallow hard oyster reef portion of the study area, beyond fix mark 3.5, a bottom scar 0.3 m in depth passes the center of the sonogram. There is another distinct scar visible in the upper part of the sonogram. Subsequent surveys of the area in July 1984 and March 1985 with an E.G. & G. side scan sonar reveal identical features on the bottom. Subarea C is located in the central portion of the Wreck Shoal study area, in the relatively deep trough zone that makes up the mud-shell oyster reef environment. Water depths range from 4.0 to 5.5 m. The echogram suggests two different bottom types in subarea C (Figure 3). From fix marks 2.0 to 1.8, the presence of a weak second reflec- tion indicates a mud-shell oyster reef. In contrast, from fix- marks 1.7 to 1.4, the absence of a second reflection indi- cates a soft mud bottom type. Ground truth data collected with oyster tongs confirmed a mud-shell oyster reef (oyster density of 28/m') in the downstream portion of the area and a soft mud bottom with very scattered shells and oysters (oyster density of 4/m') in the upstream portion of the area. 44 Dealteris FT (0 M) i^ 10 rr (3.05 M) KLIEN SIDE SCAN SONAR 25 JUNE 1984 FT (0 M ) "^1 " 10 FT (3.05 M) RAYTHEON FATHOMETER 25 JUNE 1984 Figure 3. Echogram and sonogram from Subarea C, Wreck Shoal, James River, Virginia, showing a transition zone between a mud-shell oyster reef and a mud bottom. The fix marks 2.0 to 1.4 were made at 100 m intervals. Unfortunately, it is impossible to distinguish between the mud-shell reef and the mud bottom in the sonar record. The sonogram does show two segments of a continuous bottom scar that passes beyond this record, turns around, and passes back over this record. When adjacent sonograms were mosaicked, the length of this bottom scar was greater than 2 km. The center line of this transect passes over the scar between fix marks 1.8 and 1.7. and the record indi- cates the depth of the scar is 0.2 m. Subsequent surveys of the area of July 1984 and March 1985 with an E. G. & G. side scan sonar indicate identical features on the bottom, but again there was no distinction between bottom types. Six transects from the Wicomico River were taken on two private lease areas. A large commercial vessel with a draft of 2.9 m had allegedly run aground in the lease areas and caused damage to the oyster reef. The echosounder survey data of this area was corrected for tide elevation on the instrument, and therefore indicated depths relative to high water, to coincide with the tide elevation at the time of the alleged grounding. The echogram (Figure 4) indicates water depths in the lease area range from 1.8 to 2.9 m at high water, precluding the possibility of grounding a vessel on all but the outer edge of the lease area. The record also indicates that bottom type varies from hard oyster reef to a mud bottom. Tong sampling on the larger offshore oyster reef indicated an oyster density of 1 1 8/m^. Diver inspection of the bottom also confirmed a lack of recent sediment de- posits on the oyster reef that would suggest a vessel had grounded and then attempted to power-off from that area. RAYTHEON FATHOMETER 14 APRIL 198e Figure 4. Echogram from the Wicomico River, Maryland, showing a transect across a lease tract. DISCUSSION Based on the results presented previously it is clear that an echo sounder is capable of delimiting subtidal oyster reefs from the adjacent estuarine bottom. Depending on the type of the reef, the echogram may indicate the reef as a topographic high relative to the adjacent bottom. However, the definitive test for identification of the reef is the pres- ence of second and third reflections in echogram of a prop- erly tuned echo sounder. Recall that proper tuning is the reduction of the gain in the receiver-amplifier so as to pro- duce only a single reflection on the echogram from a mud bottom. In addition, it appears that an echo sounder is ca- pable of discerning between types of oyster reefs, such as the hard-rock and mud-shell types. However, the distinc- tion is much more subjective than a simple presence or ab- sence test. The interpretation of the side scan sonar record is not as clear as the echo sounder. Under some circum- stances (data not presented here), the sonar has demon- strated the capability of distinguishing between mud bottoms and oyster reefs. Yet, in other cases, as in the Wreck Shoal study area, the interpretation is not as clear. In fact, it is impossible to delineate between either the pres- ence or absence of an oyster reef or between types of oyster reefs in these data. The secondary objective of the hydroacoustic surveys was to identify any textures or patterns on the bottom of the study areas that might assist in assessing the sedimentation processes operating in the areas. In that regard, the results of both the echo sounder and sonar surveys are most useful. Both instruments reveal scars in the bottom when present. There are several possible explanations for the scars on Wreck Shoal. Perhaps the most plausible would be an an- chor or dredge dragging across the bottom, or possibly the keel of a boat. The width of the tracks or scars (5-10 m) was directly measured from the side scan sonar record. The precision echo sounder records the vertical dimension of the scars when they pass directly under the survey vessel. The depth of the track varies from 0.1 to 0.9 m. Water Hydroacoustic Mapping of Subtidal Oyster Reefs 45 depths in the areas of the scars range from 2.7 to 5.5 m. Because of the width and depth of the bottom scars and the water depth, it is doubtful that any of the above explana- tions reasonably account for these marks. An alternative explanation is that the marks have been caused by the pro- peller wash of large commercial vessels operating out of the navigation channel, crossing Wreck Shoals. Following a methodology developed by Liou and Her- bich (1976), it can be shown that shallow draft commercial vessels in the James River are capable of generating pro- peller wash exit velocities in excess of 10 m/sec. This jet of water moving downstream from the propeller would gen- erate maximum bottom velocities of 10.7. 3.4 and 2.0 m/sec in water depths of 3. 1 , 4.3. and 5.5 m, respectively. These water flows are certainly capable of scouring the bottom and creating the wide trench observed on both the echograms and sonograms. It is the longevity of these scars on the bottom that is of interest, with respect to sediment transport processes on Wreck Shoal. The sonar indicated no observable change in the majority of the bottom scars between two surveys, a period of nine months. This implies that there is not much active sediment transport occurring on Wreck Shoal. These trenches would have otherwise filled and become unrecognizable. Future surveys of Wreck Shoal will be able to determine exactly how long these fea- tures remain. The sonar surveys identified about 19+ km of bottom scars with the width of the scars about 10 m including the adjacent banks. Therefore, the resulting area impacted is 2 X lO'm- or about 20 hectares. In terms of the total area of the Wreck Shoal study area (more than 800 hectares), the area affected by the bottom scars is less than two percent. ACKNOWLEDGMENTS Support for the field work in this project was provided by the Virginia Institute of Marine Science through the James River Seed Oyster Bed Project. The University of Rhode Island provided support through the Summer Fac- ulty Fellowship Program and the College of Resource De- velopment, Agricultural Experiment Station. This is contri- bution number 2390 of the University of Rhode Island. College of Resource Development. Agricultural Experi- ment Station. Kineston, R.I.. U.S.A.. 02881. REFERENCES Bouma. AH. 1976. Subbottom characteristics of San Antonio Bay. Bouma. A. H., ed. Shell Dredging and Its Influence of Gulf Coast Environments. Gulf Pub!.. Houston. TX; 132-148. DeAlteris, J. T. 1986. The sedimentary processes and geomorphic history of Wreck Shoal, an oyster reef of the James River. Virginia. Ph.D. Dissertation. College of William and Mary in Virginia. 205 p, Galtsoff, P. S. 1964. The American Oyster, Crassostera Virginia. Gmclin U.S. Fish Wildt. Sen. Fish. Bull. 64:1-480. Haven. D. S. & J. P. Whitcomb. 1983. The origin and extent of oyster reefs in the James River. Virginia. Jour. Shellfish Res. 3:141-151. Haven, D. S.. J. P. Whitcomb, J. M. Zeigler & W. C. Hale. 1979. The use of sonic gear to chart locations of natural oyster bars in lower Chesapeake Bay. Proc. Natl. Shellfish. Assoc. 69:11-14. Johannesson, K. A. & R. B. Mitson. 1983. Fisheries Acoustics. FAO Fisheries Technical Paper No. 240:249 p. Liou, Y. C. & J. B. Herbich. 1976. Sediment movement induced by ships in restricted waterways. TAMU-SG-76-209:85 p. Moore. H. F. 1910. Condition and extent of oyster beds m the James River. U.S. Bur. Fish. Doc. No. 729.'83 p. Thome, R. E. 1983. Hydroacoustics. Neilsen, L. A. and D. L. Johnson. eds. Fisheries Techniques. Amer. Fish. Soc: 239-259. Journal of Shellfish Research. Vol. 7. No. 1. 47-50. 1988. OVERWINTERING AMERICAN OYSTER SEED BY COLD HUMID AIR STORAGE HERBERT HIDU', SAMUEL R. CHAPMAN^ AND WILLIAM MOOK^ ^Department of Animal and Veterinaiy Sciences Ira C. Darling Center University of Maine Walpole. Maine 04573 ^Ira C .Darling Center University of Maine Walpole. Maine 04573 ^Mook Sea Farms Damariscotta, Maine 04543 ABSTRACT Young-of-the-year hatcherv' produced American oysters were successfully humid air stored at 0° to 6°C for monthly periods up to 6 months. Survival of oysters between b and 55 mm length was uniformly high, over 80%, and equaled or exceeded that of seawater stored controls in nearly all cases. There is an indication that premature replacement back to ambient sea water before April was detrimental to survival. Measurement of growth and survival after the subsequent summer season revealed continued high survival, but those returned to water after April had grown less. The decreased growth rate is probably due to lack of exposure to the complete growth season. Tnals with local industry indicate that air storage of juvenile oysters may be practical on a commercial scale. KEY WORDS: Overwintering, air storage, oysters, American oysters INTRODUCTION In recent years shellfish hatcheries and associated field nursery operations have been prominent in North America. In Maine. American oyster seed (Crassostrea virginica L) is economically reared to a 25-60 mm size in its first season in floating tray fields (Clime & Hamill 1979). After the first year, however, tray deployment becomes unwieldy and uneconomic, making bottom culture the only alterna- tive. Thus the handling of large quantities of seed oysters over the first winter has become a major problem. Holding oysters in trays in the field exposes them to silting and weather related loss. Onshore sea water systems are expen- sive and unreliable. Planting, especially small 25 mm clutchless seed oysters, in early winter has resulted in sub- stantial loss due to predation by eider ducks (R. Clime per- sonal communication). Cold humid air storage is a poten- tially attractive strategy for overwintering seed. Acceptable winter survival of seed would afford nursery- growout operations greater profit and the hatcheries could more efficiently expand their production season and offer yearling seed for sale the following spring. Air storage of American oysters has received earlier at- tention from a marketing but not a culture perspective (Friedman, 1933, Needier, 1934; Medcof, 1960), The cri- teria for success with market oysters are palatability and salability. The criteria for continued culture would merely be survival and subsequent ability to grow. Anecdotal evi- dence (Medcof, 1960) indicates that humid cold air stored market oysters were edible, presumably alive, after 4. 6, and even 8 months of air storage. Thus the potential exists for long term storage viability. Market oysters at 34°F (/°C) showed negligible mortality at 3 months but incurred 40% mortality after 5 months, (Medcof, 1960), Mortality was higher at storage temperatures of 41-60°F at 5 months. This study reports survival and subsequent growth of several sizes of American oyster seed overwintered in humid air up to 6 months at several temperatures. Results of scale up commercial overwintering trials are described. Methods 1986 Experiments Young-of-the year cultchless seed oysters produced from Maine stock were procured from ambient (Figure 1) sea water upwelling units at Mook Sea Farms, Walpole, Maine, in late December 1985, These were screen graded into two size groups (x = 7 mm S = 1 .2 and (x = 9 mm; S = 1,2), Each group of oysters was subdivided into sub- groups of 200 and each placed in 100 mm x 15 mm dis- posable plastic petrie dishes. The dishes were lined with several sheets of brown paper toweling wetted with sea water at approximately 30 salinity. Petrie dish bottoms were drilled to drain any accumulated water. Appropriate sets, (6 sets of 5 petrie dishes with no replicates) were placed in tightly sealed plastic containers, also containing damp toweling, to be periodically removed for sampling. Container sets were then placed in 3 constant temperature units at 0, 3 and 6°C (±0,5° approx,) Oysters were re- moved from the Petri dishes and placed in ambient sea water in mesh bags at t = months (control), 1 . 2. 3. and 4 months (Figure 1), After 4,6 months, the last group of oysters were removed and it and all other oysters were in- cubated for one month in 400 liter rearing vessels at 15°-20°C with cultured algal mixtures Isochrysis galbana and Chaetocerus sp, fed commensurate with clearance 47 48 HiDU ET AL. AIR STORAGE OF SEED OYSTERS AlR STORED AIR STORAGE FOR : r MONTH 2 MONTHS 3 MONTHS 4 MONTHS 4.6 MONTHS — AMBIENT WATER 1986 15 - 20"C , ALGAL FED 1 MONTH - 2 MONTHS- 3 MONTHS- 4 MONTHS- 5 MONTHS- 6 MONTHS- ' 1986 EXPERIMENT • 1987 EXPERIMENT « ; . « DEC I JAN I FEB 1 MAR I APR ^ MAY I JUNE I JULY I Figure 1 . Experimental regimes in the liumid air storage of American oyster cultchless seed in 1986 and 1987 with associated ambient sea- water temperatures. rates. Vessels were drained and cleaned three times per week. After the incubation period, determinations of living and dead oysters were made under a dissecting microscope. Three categories were evident: dead (shells permanently agape), live (new shell growth); and "presumably live" (shells tightly closed but no shell growth). To determine viability of the "presumably live" animals, a single group (9 mm held at 3° for all time periods) was held an addi- tional month in warmed sea water and fed as above. The additional month allowed the oysters to either add new shell or die. With differential survival subsequently shown in the different monthly groups, it would have been highly desirable to incubate all oysters for an additional month. However, this limited information helped us with the de- sign of the second year confirmatory results. 1987 Experiments The second year trials were modified in accordance with first year's experience (Figure 1). These experiments in- cluded two groups of young of the year seed oysters, 12 mm (S = 1.4) and 40 mm (S = 5.0) and were humid air stored at 0°C with a second group of the 12 mm oysters air stored in a home refrigerator at 4 to 6°C as insurance against equipment failure. Oysters after 0-3 months air storage were placed directly in ambient sea water before the warm water incubation period (Figure 1). Oysters stored for 4 to 6 months were placed directly in warmed algal fed baths as above. The warm sea water-algal fed incubation period was lengthened to 1.5 months minimum to allow determinations of viable oysters. A failure of the 0° cold unit on May 15 at 4.5 months prompted us to remove all 5 month groups to ambient sea water 15 days early and to place the 6 month groups in air storage at 4-6°C for the remaining 1 .5 months of storage. The failed unit registered 16°C and may have been at that temperature up to 3 days before discovery. All of the 40 mm group were then held over the re- mainder of the summer in field nursery lantern nets to de- termine the effects of the treatments on subsequent growth and longer term viability of the living oysters. Mean lengths and total wet weights with standard errors of the mean ( ± ) 2 Sg^) of this group were determined on Sep- tember 30, 1987. RESULTS 1986 Experiments initial determinations of viability after 1 month of incu- bation under optimal conditions revealed a high "ap- parent" survival of the 7 and 9 mm oysters regardless of length and temperature of air storage (Figure 2). Lack of replication within treatments did not permit statistical anal- ysis but it is apparent that survival in the 7 mm oysters was 5 to 10% poorer than the 9 mm oysters. The anomolously low survival of the 7 mm seed oysters air stored for 4 months is no doubt due to error in determining live and dead oysters. Culture of the 9 mm 3°C "presumably live" group re- C- 'C -tl.- 3 -C -D-6 "C % SURVIVAL OF AIR STORED SMALL (7 MM) SEED OYSTERS MONTHS AIR STORED % SURVIVAL OF AIR STORED LARGE (9MM) SEED OYSTERS : - -D-6 MONTHS AIR STORED Figure 2. "Apparent" survival of 7 mm and 9 mm seed air storage up to 4.6 months in 1986. These figures are high since live oysters which did not show growth 1 storage suffered later mortality. oysters after erroneously month post Overwintering of Seed Oysters 49 vealed an apparent relationship between the length of storage and viability (Figure 3). Those held in air storage the longest 3, 4, and 4.7 months had survivals of 70% to 95% whereas those held in air storage for 1 and 2 months had less than 65% survival. This finding prompted us to modify our experimental protocol in the second year. 1987 Experiments Survival of seed oysters after 1 to 5 months of air storage was high, generally over 80%, and increased mor- tality at 6 months related to size. (Figure 4). The 40 mm oysters experienced a 95 + % survival to 6 months despite the fact that the last two groups experienced a 16° tempera- ture shock at 4.6 months. The smaller 12 mm oysters sur- vived well but at reduced levels than the larger oysters. Temperature here, 0° vs 4-6°, had little apparent effect on survival. There is a suggestion, again, m the 0° group that an avoidance of early placement in cold ambient seawater enhances survival. Survival in the oysters removed from air storage in May and June approximated 95% whereas the controls and those removed earlier approximated 85%. Six months of air storage appears to approach the limit for the 12 mm oysters as evidenced by the reduction to 80% sur- vival in the 4-6°C group. The drastic reduction in survival down to 25% in the 0° group of 12 mm oysters has no doubt been enhanced by the 16° temperature shock experi- enced at 4.5 months. Air storage did not affect the ability of seed oysters to survive and grow through the subsequent summer season. The 40 mm seed oysters of 1987 continued to survive over 957f through the end of September. Ability to grow was not affected (Figure 5) however it is obvious that those which were air stored the longest, and thus deprived of increasing portions of the growth season, grew the least. This is ap- parent particularly in the September total weights of those 2 3 MONTHS AIR STORED 100 90 80 70 60 > > 5 =) ^ 40 30 20 I X X— . = 12 mm. OYSTERS X = 40 mm, OYSTERS 0° C 4- 6° C ~~1 Jan r 1987 I ~I APR I MAY 3 4 I JUNE I 5 6 Figure 3. "Actual" survival of 9 mm seed oysters damp air held at 3°C up to 4.6 months in 1986. Oysters were held 2 months in warmed algal-laden water post storage before viability determination. 2 Months Air Stored Figure 4. Survival of 12 and 40 mm seed oysters at two temperatures after up to 6 months air storage. Cold unit failure at 4.5 months in the 0°C groups necessitated premature sampling of 5 month groups and placement of 6 month groups at 4-6°C. air stored for 4, 5 and 6 months (through April. May and June.) Shell lengths, although there was a declining trend, were not affected as drastically as weight. DISCUSSION The ability of seed oysters to withstand humid air condi- tions for a period of six months represents remarkable ability to withstand adverse conditions. The compelling questions are: what are the physiological mechanisms al- lowing prolonged survival and indeed how long can an oyster remain alive out of water? We tested only to 6 months in the second year because practical application of our results requires only 3-5 months storage. The first year's experiments were terminated at 4.6 months because a sample of oysters upon dissection appeared dead. The shells were tightly closed but the meats were in a highly desiccated condition. When these were immersed in sea water there was no evidence of heartbeat and the mantle tissue showed no response to probing with a needle. When the remaining intact oysters were subsequently incubated in warmed algal-laden sea water, they resumed feeding and over 80% survived in nearly all cases. 50 HiDU ET AL. E E J 40 T T :^^ T I T .x^ 1^ LENGTH T ^X 1 T/ yi\ \I — ~— • 1 i"^ -^T T INITIAL LENGTH • -^^ — K JAN. 1 1 i\ WEIGHT 1 35 30 25 20 I 10 12 3 4 5 6(Jjne30) AIR STORAGE (months) Figure 5. Shell length and total weight of 40 mm air stored seed oysters sampled September 30, 1987 after one growth season post air storage. Initial weights were not recorded. Survival in all groups ex- ceeded 95%. The physiological mechanisms of such survival should be investigated. Although we made no measurements of glycogen utilization, the reduced survival of those placed back in winter sea water after only 1 and 2 months air storage would suggest that glycogen utilization rates are somehow lessened with cold air storage. The air stored sample dissected in May, although again in an extremely desiccated condition, appeared to carry some remaining glycogen reserve. If glycogen utilization shuts down under cold air conditions then it might be possible to air store oysters for an indefinite period. Our follow-up studies will air store oysters up to 1 year. Scale-up commercial application is cautiously under way in Maine. Mook Sea Farms hatchery has constructed a 4' X 6' X 4' deep wood lined underground pit for over winter storage of hatchery seed. In cooperation with Dodge Cove Marine Farm, Walpole, Maine, approximately 70,000 six to ten mm and 50,000 twenty to thirty mm American oyster seed were overwintered from December 1986 to May 1987 with over 90% survival. Oysters were in 1" plastic mesh bags in damp layered burlap and seaweed. Temperatures ranged from 1° to 10°C throughout. Other commercial trials, however, should proceed strictly on an experimental basis because of possible effects of such un- tested variables as physiological condition, geographical race of the stock, and variability in storage conditions such as temperature and percent humidity. These results suggest other commercial use of air storage of shellfish as a culture technique. For example, it may be possible to avoid an infective period of a parasite by relatively short air storage periods. This might be partic- ularly valuable with yearling oyster nursery stock in MSX, Haplosordiion nelsoni, endemic areas. The possibilities for overwinter storage of other species, particularly hard shelled clams Mercenaria mercenaha, are self-evident and are under investigation. In summary, the old idea and prac- tice of cold air storage of market oysters to prolong via- bility and shelflife should be investigated and expanded for its use as a culture technique as an avoidance procedure against adverse environmental components. ACKNOWLEGMENTS Thanks for the support of the University of Maine's Fisheries and Aquaculture Group of the Maine Agricultural Experiment Station "External Publication No. 1266." Thanks also to Les Watling for use of constant temperature apparatus. Linda Kindblom graciously provided technical work. LITERATURE CITED Friedman, M, H. 1933. The freezing and cold storage of live clams and oysters. Biol Bd. Canada. Ann. Rep't. for 1932:23-24. Medcof, J. C. 1958. Studies on stored oysters (Crassostrea virginica). Proc. Nan Shellfish Ass n. Vol. 49:13-28. Needier, A. W. H. 1934. The storage of oysters in the shell. Fish. Res. Bd. Canada. Bull. 44:1-4. Clime, R. and D. Hamill. 1979. Growmg oysters and mussels in Maine. Coastal Enlerprises, Inc., Middle Street, Wiscasset. Maine. 46 p. Journal of Shellfish Research. Vol. 7, No. I, 51-55. 1988. DEVELOPMENT AND EVALUATION OF TECHNIQUES TO STUDY ACQUIRED IMMUNITY TO PERKINSUS MARINUS IN THE OYSTER, CRASSOSTREA VIRGINICA (GMELIN) FU-LIN E. CHU Virginia Institute of Marine Science School of Marine Science The College of William and Mary Gloucester Point, Virginia 23062 ABSTRACT This paper describes a radiometnc technique developed to measure phagocytosis of Perkinsus marinus zoospores by oyster hemocytes. The spores o{ P . marinus were radiolabeled by cultunng P marinus presporangia and sporangia in estuanne water (22%c) containing '""C-glycine. The percent of spores phagocytized by hemocytes was detemimed by the uptake of radioactivity by hemocytes. Results from prelimmary experiments to test the efficiency of using an osmotic infiltration method for immunizing oysters are also reported. It was found that oysters can take up both dissolved antigen (radiolabeled bovine serum albumin) and particulate antigen ( '■'C-labeled zoospore homogenate) through osmotic infiltration. The uptake of the antigen was correlated with the concentration of antigen added to the water but was not affected by water temperature. KEY WORDS: Oyster, acquired immunity, techniques. INTRODUCTION Perkinsus marinus (Dermo) and Haplosporidium nelsoni (MSX) are two parasitic pathogens which have been de- structive to estuarine oyster populations in the Middle At- lantic Region since the introduction of P. marinus in the 1950's and MSX in the 1960"s. However, there are some oysters that have survived the invasion of these pathogens (Andrews 1968; Haskin and Ford 1979; Ford and Haskin 1986). Those oysters that survive the epizootics are be- lieved to possess certain genetic, or physiological charac- teristics which make them less susceptible to the pathogens (Maryland Sea Grant 1983, National Fisherman 1983). Two hypotheses have been suggested for the occurrence of this resistance: (1) disease resistant oysters are physiologi- cally or genetically different from non-resistant ones, and (2) disease-resistant oysters acquired immunity through early exposure to the pathogens. These two hypotheses are probably mutually inclusive. Although evidence for the development of acquired im- munity in molluscs is far from satisfactory, there are sev- eral interesting findings. In 1964, Michaelson (1964) re- ported the production of a microacidial immobilizing sub- stance by snails infected with Schistosoma mansoni. Acton and Evans (1968) found that the bacterophage T2 was cleared more rapidly from oyster (C. virginica) hemolymph after secondary injection than after primary injection. Feng and Stauber (1968) suggested that the precipitous reduction in the number of Hexamita sp. in resistant oysters 8 days post-injection might be attributed to the presence of ac- quired immunity. Furthermore, Hardy et al. (1977) demon- strated that exposure of oysters (C. gigas) to bacteria stimu- lated an increase substantially in the titre of bacterial agglu- tinin. Disease problems in oysters and other bivalve species have stimulated interest to determine the feasibility of in- ducing acquired immunity to the pathogen P. marinus in American oysters. Crassostrea virginica. Like other inver- tebrates, bivalve molluscs do not appear to possess immu- noglobulins. Phagocytosis is the principle mechanism by which bivalve molluscs normally defend themselves against invading pathogens and foreign materials (Cheng and Rifkin 1970; Cheng 1981; Cheng 1983). The impor- tance of phagocytosis in determining the outcome of a dis- ease has been established (Metchnikoff 1893; Sindermann 1971). The phagocytic activity of the host to invading pathogen is correlated with the degree of resistance (McKay and Jenkin 1970). Resistance is decreased by a lowering of phagocytic activity (Aarum 1967). In order to test the efficacy of immunization with a possible P. marinus vaccine, a technique was developed to measure the phagocytosis of P. marinus zoospores by oyster hemo- cytes. This paper describes the radiometric technique de- veloped for this purpose. Osmotic infiltration is a practical mass-immunization method which was originally developed by Amend and Fender (1976) for immunizing fishes. This method is less stressful on the animal and less time consuming than indi- vidual inoculation. Antigens are infiltrated into fishes during immersion of the animal in a hyperosmotic solution containing the antigen (Antipa and Amend 1977; Croy and Amend 1977; Bowers and Alexander 1981). Lewis and his associates (The University of the Sea, Vol. 15, No. 1, 1982, Texas A&M University; The University of the Sea, Vol. 14, No. 3, 1981, Texas A&M University) also suc- cessfully immunized shrimp against bacterial diseases by placing shrimp in hyposmotic water containing antigen. Since the osmotic infiltration technique has proven to be very effective and successful for mass-vaccination of small fishes and shrimp, we wanted to evaluate the osmotic infil- tration technique for immunization of oysters. To examine whether oysters can take up antigens by osmotic infiltra- 51 52 Chu tion, experiments were performed to determine the uptake of radiolabeled bovine serum albumin ('''C-BSA, Molec- ular weight = 69,000 daltons) and radiolabeled zoospore homogenate. Preliminary results from these experiments are reported in this paper. MEASUREMENT OF PHAGOCYTOSIS OF P. MARINUS ZOOSPORES BY OYSTER HEMOCYTES A summary of the sequence and procedure developed to measure phagocytosis of P. marimts zoospores by oyster hemocytes is shown in Figure 1 . Immunization and Maintenance of Oysters Oysters (Crassostrea virginica (Gmelin)) were collected from upstream bars of the James River in Virginia. In this area, oysters have been protected by Dermo and MSX dis- eases by low salinity (5- 14%c) (Andrews and Hewatt 1957: Chu, unpublished data). MSX and Dermo are believed to be inactive in salinity below 15%c. There were 2 groups of oysters (15 oysters per group): immunized and sham con- trol. Oysters of the immunized group were injected intra- muscularly with formalin-killed P. marimts zoospores twice (2.0 X 10* zoospores/oyster in 0.1 ml estuarine water) at a one week interval. Each of the sham control oysters was injected with 0. 1 ml estuarine water. Both im- munized and sham control oysters were held in a trough (210 X 60 X 15 cm L X W X H) filled with filtered (10 (xm, 1 |jLm Cuno cotton filters) pasteurized estuarine water. Water in the trough was changed every two or three days. An algal diet (Tetrasehms suecica) was added to the trough daily (500 ml of 1-2 x 10^ cells/ml per day). Twenty-four 1. Prepare hemocyte monolayer by adding known number of hemocytes (1-2 x 10 cells) to a 16 X 75 mm glass test tube- \^ hemolymph hemocyte mo-olaycr 2. Nonadherent cells are removed by 3 gentle washes with culture medium (MEM). 3. Add one ml of medium containing 2.0 x 10 C-zoospores to each test tube. Experimental (+2.0 X 10^ lA C -zoospores) Control . (+ "0" C -zoospores) 4. Incubate at 15°C for 1.5 hrs. 5. Discard culture medium, and wash the monolayer 3 times with clean culture medium. Count in scintillation counter. CPM of hemocyte monolayer CPH 6. % of phagocytosis - <"cubated with '"c-zoospores ' of Control CPM C-2oospores added to hemocyte monolayer Figure 1. Procedure for quantifying phagocytosis of '''C-zoospores of Perkinsus marinus by oyster hemocytes. and 48 hrs after the second immunization, blood samples were taken from oysters (7 oysters/group/time period) for phagocytic activity measurement. Collection of Oyster Hemocytes Hemolymph was collected from oysters. A 27 gauge, 25 mm needle attached to a 1 ml sterile syringe was inserted in the adductor muscle of the oyster. Hemolymph was with- drawn and pooled at 4°C. The number of cells in the hemo- lymph was counted using a hemocytometer. About 1-2 x 10* cells in one ml of hemolymph were used for phagocy- tosis measurement. Preparation of '''C-labeled Zoospores Presporangia and zoospores of P. marimts were cultured by methods described by Perkins and Menzel (1966). '"C- labeled zoospores were obtained by culturing presporangia in 25 ml of 22%c estuarine water containing 10 (jiCi ''*C- glycine (New England Nuclear, U.S.A.) at 27-28°C for 72-96 hrs. '"C-labeled zoospores were harvested, treated with 0.3% formalin, and washed twice with sterile (0.22 |jLm filtered) estuarine water. '"C-zoospores were then con- centrated to a desired level for the phagocytosis study. The '"C-zoospores obtained in this way contained 6.6-13.2 x 10"^ dpm/spore. The percent of radioactivity leached from the zoospores after 24 hrs incubation in '"C-glycine-free sea water was 0-25%. Measurement of Phagocytosis Known numbers of cells ( 1 -2 x 10* cells) in one ml of hemolymph were placed in 16 x 75 mm culture tubes and allowed to adhere at 15°C for 30 minutes. At the end of the time, nonadherent cells were removed by three gentle washes with minimal essential medium (MEM) and counted. About 1.5-2.0 x 10* '"C-zoospores in 1 ml of MEM were added to the hemocytes and incubated at 15°C for 1.5 hours. Phagocytosis was stopped by discarding the supernatant and gently washing 3 times with MEM. Cell pellets were digested with 0.6 ml NCS (tissue solubilizer) at 50°C and the radioactivity in the aliquot was measured in 10 ml Aquasol with a scintillation counter. Percent of phagocytosis of '"C-zoospores by oyster hemocytes was calculated using the following formula: % of phagocytosis = CPM of hemocyte monolayer incubated with '"C-zoospores CPM of control hemocyte monolayer CPM '"C-zoospores added to hemocyte monolayer. Since the number of hemocytes placed in the glass tubes to prepare the hemocyte monolayer and the number of '"C- zoospores added to the hemocyte monolayer are known, percent of phagocytosis can also be expressed in terms of number of '"C-zoospores phagocytized by number of he- mocytes. Counts of the number of nonadhering cells indi- Techniques for Acquired Immunity Study in Oysters 53 cated that about 95-97% of the hemocytes adhered. The phagocytic responses of hemocytes (2 x 10* cells) pooled from 7 immunized oysters and 7 sham control oysters em- ploying this technique are shown in Figure 2. The results showed the uptake of '''C-labeled zoospores by hemocytes sampled 24 and 48 hrs after the second immunization. The uptake of '■*C-labeled zoospores of P. marinus by hemo- cytes from immunized oysters was higher than hemocytes from control (non-immunized) oysters. It was speculated that a cellular response was elicited in oysters at 24 and 48 hrs after the second challenge with formalin-killed zoo- spores, but further study is needed to verify this specula- tion, and the specificity of the response has not been deter- mined. The increased response of sham control at 48 hrs suggests a nonspecific reaction of oyster hemocytes. Phagocytosis is usually measured by enumerating the number of hemocytes which have ingested bacteria (or the pathogen) or by measuring the optical density of abiotic particles (e.g. latex ring) ingested by oyster hemocytes (Anderson and Good 1976; Cheng and Sullivan 1984). The radiometric technique described in this paper is the first re- ported method to radiolabel an oyster pathogen and directly measure the phagocytosis of the radiolabeled pathogen by the oyster hemocytes. Phagocytosis includes processes of recognition, adherence, ingestion, destruction and disposal. Interaction of hemocytes and live or formalin-killed P. marinus zoospores has been examined with light phase contrast microscope, and both adhesion and ingestion were observed. It was assumed that the uptake of radioactivity by the oyster hemocytes was due to either adherence or ingestion. The application of this technique to measure phagocytosis of P. marinus spores by oyster hemocytes will be further evaluated. 2 I O a. I o 480 360 240 - 120 mmunized Control 25 50 75 100 HOURS AFTER SECOND IMMUNIZATION Figure 2. In vitro cellular phagocytic response of oyster hemocytes. UPTAKE OF DISSOLVED AND PARTICULATE ANTIGENS THROUGH OSMOTIC INFILTRATION BY THE OYSTER, C. VIRGINICA The size of oysters used for the osmotic infiltration ex- periments ranged from 2 to 3 cm shell height. Oysters were submerged individually for 1-3 hrs in hyposmotic water containing ['"Cl-methylated bovine serum albumin (soluble antigen '"'C-BSA, molecular weight = 69,000 daltons, Amersham) or '''C-labeled zoospore hemogenate (particu- late antigen). The '''C-labeled zoospore homogenate was prepared by disrupting the zoospores with sonifier cell disruptor (Model WI85, Heat Systems — Ultrasonic Inc.). Hyposmotic water was prepared by lowering the salinity in which the oysters were held from \b%c to 10%c with chlo- rine free tap water. Uptake of '''C-BSA by oysters sub- merged in water of the same salinity (\67cc) was also deter- mined. All the experiments were performed at room tem- perature (=22°C), unless stated otherwise. To examine the effect of temperature on the uptake of antigens, the osmotic infiltration was performed at 18 and 30°C. Results from the preliminary osmotic infiltration studies indicate that through osmotic infiltration the oyster can take up both '"C-BSA and '''C-labeled zoospore homogenate (Fig. 3 and Table 1). The uptake of antigen was correlated with the concentration of antigen added to the water (Figure 3). Elevated water temperature does not appear to increase but to decrease the antigen uptake; it was found that the uptake of C'-BSA in oysters in water of 18°C was 2 times higher than in oysters in water of 30°C. A temperature of 30°C may stress the animals and retard the osmotic activity, 6000-1 5000- cr LU I- > o ^ 4000- Q. Q > o < o < 3000- 2000 1000- y = I02X + 205 7 = 90 — 1 — 10 20 ■ — : — 30 40 50 14 RADIO ACTIVITY OF C-BSA (id* DPM) Figure 3. Uptake of "C-BSA by oysters through osmotic infiltration. 54 Chu TABLE 1. Uptake of radiolabeled zoospore homogenate by oysters* TABLE 2. Uptake of "C-BSA by oysters incubated in hyposmotic (10% 2.6 X 10' 2830 3 3896(2.13 X 10"^ zoospores) 1875 48.3 3 2.6 X 10' 3922 X ± SD 1710.0 ± 419.5 43.9 ± 10.8 4 2.6 X 10= 3297 4 1047 (0.98 X IQO zoospores) 350 33.0 5 2.6 X 105 4380 5 1047 (0.98 X 10*' zoospores) 413 39.0 6 2.6 X 10= 3373 6 1047 (0.98 X 10^ zoospores) 288 27.5 7 2.6 X 105 1598 7 1047 (0.98 X IQO zoospores) 244 23.0 X + SD 4163 + 1469 3162 + 1154 \ -^ SD 328.8 ± 73.7 30.6 ± 6.9 * Oysters were exposed to 2 different concentrations of '''C-labeled zoo- spores from 2 different culture stocks at room temperature (22°C); expo- sure time was two hours. since the oysters were held in ambient water of 18-20°C prior to the experiment. Results also demonstrated that '''C- BSA infiltrated into oysters held in water of the same os- motic concentration (Table 2). The uptake of C'^-BSA was similar to those immersed in hyposmotic water. The effi- ciency and reliability of the osmotic infiltration technique for immunization of oysters will need further investigation. The uptake of antigen in oysters could be the results of the synergy of osmotic infiltration, active transport (pinocy- tosis) and phagocytosis processes. * Oysters were mcubated at room temperature (22°C) for 3 hrs. ACKNOWLEDGMENTS Contribution number 1440 from the Virginia Institute of Marine Science. This work is a result of research sponsored by the NO A A Office of Sea Grant, U.S. Department of Commerce, under grant no. NA86AA-D-SG042 to the Vir- ginia Sea Grant Program. The author wishes to thank Bev- erly Casey for technical assistance and Ms. Shirley Sterling and Ms. Janet Walker for typing and preparing the manu- script. The author thanks Drs. Mary Gibbons, Michael Bender, Kenneth Webb, William Hargis, and Beverly Weeks for critically reviewing the manuscript. REFERENCES Aarum. G. R. 1967. Fagocytose. Nor. Tannlaegeforenings Ted. 77:243- 254. Acton, R. T. & E. E. Evans. 1968. Bacteriophage clearance in the oyster iCrassosrrea virginica). J. Bacieriol. 95:1260-1266. Amend. D. F. & D. C. Fender. 1976. Uptake of bovine serum albumin by rainbow trout from hyperosmotic solutions: a model for vaccinating fish. Science 192:793-794. Anderson, R. S. & R. A. Good. 1976. Opsonic involvement in phagocy- tosis by mollusk hemocytes. J . Imertebr. Pathol. 27:57-64. Andrews, J. D. 1968. Oyster mortality studies in Virginia. VII. Review of epizootiology and origin of Minchinia nelsoni. Proc. Nail. Shellfish. Assoc. 58:23-26. Andrews, J. D. & W. G. Hewatt. 1957, Oyster mortality studies in Vir- ginia. 11. The fungus disease caused by Dermocystidium mariimm in oysters of Chesapeake Bay. Ecological Monographs 27:1-26. Antipa, R. & D. F. Amend. 1977. Immunization of Pacific salmon: com- parison of intraperitoneal injection and hyperosmotic infiltration of Vi- brio anguillarum and Aeromonas salmonicida bacteria. J. Fish. Res. Bd. Can. 34:203-208. Bowers, A. & J. B. Alexander. 1981 . Hyperosmotic infiltration: immuno- logical demonstration of infiltrating bactena in brown trout, Salmo trutta L. J. Fish. Biol. 18:9-13. Cheng, T. C. 1981. Bivalves. In: Invertebrate Blood Cells. N. A, Rat- cliffe and A. F. Rowley (Eds.). Academic Press, London and New York. Vol. 1:233-300. Cheng, T. C. 1983. Internal defense mechanisms of molluscs against in- vading microorganisms: personal reminiscence. Trans. Amer. Mi- crosc. Soc. 102:185-193. Cheng, T. C. & E. Rifkin, 1970. Cellular reactions in marine molluscs in response to helminth parasitism. In: A Symposium on Diseases of Fishes and Shellfishes. S. F. Snieszko (Ed.). Amer. Fish. Soc. Spec Piibl. No. 5;443-496. Cheng, T. C. & I. T. Sullivan. 1984. Effects of heavy metals on phago- cytosis by moUuscan hemocytes. Mar. Environmem. Res. 14:305- 315. Croy, T. R. & R. D. Amend. 1977. Immunization of sockeye salmon {Onchorhynchus nerka) against vibriosis using the hyperosmotic infil- tration technique. Aquaculture 12:317-325. Feng, S. Y. & L. A. Stauber. 1968. Experimental hexamitiasis in the oyster, Crassoslrea virginica. J. Invert. Pathol. 10:94-110. Ford, S. E. & H. H. Haskin. 1987. Infections and mortality patterns in strains of oysters Crassoslrea virginica selected for resistance to the parasite Haplosporidium nelsoni (MSX). J. Parasitol. 73:368-376. Hardy, S. W., T. C. Fletcher. & J. A. Olafsen. 1977. Aspects of cellular and humoral defense mechanisms in the Pacific oyster, Crassostrea gigas. In: Developmental Immunology. Proceedings of the Symposia on Developmental Immunology. J. B. Solomon and J. D. Horton (Eds.), Vol. 5, pp. 59-66. Haskin, H. H. & S. E. Ford. 1979. Development of resistance to Min- chinia nelsoni (MSX) mortality in laboratory-reared and native oyster stocks in Delaware Bay. Mar. Fish. Rev. 4I( l-2):54-63. McKay, D. & C. R, Jenkin. 1970. Immunity in invertebrates: correlation of the phagocytic activity of haemocytes with resistance to infection in the crayfish iParachaerops hicarinatus). Aiisl. J. Exp. Biol. Med. Sci. 48:609-617. Maryland Sea Grant, Vol. 6, No. 1, pp. 4-9, 1983. Techniques for Acquired Immunity Study in Oysters 55 Metchnikoff. E. 1893. Lectures on the comparative pathology of intlam- mation. Delivered al the Pasteur Institute in 1891. Kegan, Paul, Trench, Trubner. and Co.. Ltd.. London. (Republished 1968 by Dover Publications, Inc., New York, 224 p.) Michaelson, E. H. 1964. Microacidia — immobilizing substances in ex- tracts prepared from snails infected with Schistosoma munsoni. J. Amer. J. Trap. Med. Hug. 3:36-42. National Fisherman. May. 1983. p. 75. Perkins. F. O. & R. W. Menzel. 1966. Morphological and cultural studies of a motile stage in the life cycle oi Dermocystidmm mannum. Proc. Natl. Shellfish. Assoc. 56:2-30. Sindermann. C. J. 1971. Internal defenses of Crustacea: a review. Fish. Bull.. Vol. 69. No. 3. 1971. The Umversity of the Sea. Vol. 14(3|. 1981 A&M. Vol. 15(1). 1982, Te.xas Journal of Shellfish Research. Vol. 7. No. I, 57-59. 1988. RAPID DECLINE IN OYSTER CONDITION IN THE PATUXENT RIVER, MARYLAND GEORGE R. ABBE AND JAMES G. SANDERS The Academy of Natural Sciences Benedict Estiiarine Research Laboratory Benedict, Maryland 20612 ABSTRACT The meat condition index of a population of oysters in the Patuxent River, Maryland was determined monthly from February 1986 to February 1988. This period was characterized by dry weather and above average salinity which led to the spread of Haplosporidium nelsoni (MSX) into much of Maryland from Virginia, resulting in widespread high mortalities. The population sampled in this study showed no unusual mortality, but oyster condition declined significantly. It is unknown if this decline is related to sublethal infections of MSX. pollutants, or other factors, but the decline indicates a stress on the system. Routine monitoring of oyster condition could serve as a useful early warning mechanism for threatened populations. KEY WORDS: Condition index, oyster. Crassoslrea virginica INTRODUCTION After several good years of larval recruitment during the early and mid 1980s (Davis et al. 1981; Krantz and Davis 1983; Abbe 1988), the fishery for the oyster Crassoslrea virginica (Gmelin) has reached the lowest level in Mary- land in a century. In 1885 the Maryland harvest was 15 x 10* bu (Kennedy and Breisch 1981). By the early 1900s. production had declined to 4 x 10* bu, and from 1934 to 1984 the annual harvest averaged only 2.5 x 10* bu. In 1984 and 1985, landings were barely 1 x 10* bu. The 1986-87 oyster season produced about 9.8 x 10' bu. and the 1987-88 season may yield only half of that (C. Bonzek, Maryland Department of Natural Resources, pers. comm.). In recent years, low harvests have followed periods of unusually high spat setting which was improved by above average salinity during several dry years. Unfortunately, the high salinity also allowed the invasion of the pathogenic protozoan Haplosporidium nelsoni (Haskin. Stauber. and Mackin) (MSX) from Virginia into Maryland where it gen- erally does not occur (Rosenfield and Sindermann 1966; Andrews and Wood 1967). MSX is most prevalent where salinity is 20-25%f, but it is infectious at 15%f (Andrews 1964; Ford 1985). It becomes less pathogenic as salinity decreases from 15 to \07(c and apparently cannot tolerate salinities below 10%c (Haskin and Ford 1982; Andrews 1983). MSX mortalities were quite high in the middle Chesapeake Bay during 1986 and 1987 as salinity remained above \57cc most of this time. Therefore, although recruit- ment to the population remained high, the high rate of dis- ease-related mortality has kept population densities and fishery yields low. During this period of heavy disease mortality, we were able to observe, as part of an unrelated study, a major de- crease in meat condition of oysters in the upper Patuxent River. Salinity in this area is often 6-8%t, but was 12-157ff during 1986 and 1987. The oysters, however, have shown no unusual mortality. METHODS AND MATERIALS Ten oysters of similar size (means ranged from 79 to 97 mm; overall mean was 87.8 ± 4.8 mm) were collected monthly from February 1986 to February 1988 from the Holland Point oyster bar near Benedict, Maryland (Fig. 1). As part of a study of trace element concentrations in oysters, meat conditions were also determined. Oysters were cleaned of fouling organisms, scrubbed, rinsed in deionized water, and blotted dry. They were then measured for shell length, weighed whole, and shucked. Dry meat weights were determined and the empty valves weighed after 24 hr. Meat condition indices were determined using the methods of Lawrence and Scott ( 1982) according to the formula: dry tissue weight (g) shell cavity volume (ml) 100 where shell cavity volume (ml) is equal to the difference between the weight of the whole oyster (g) and the weight of the empty valves (g). RESULTS AND DISCUSSION The condition index is a unit-free number relating an individual oyster's .soft tissue to its shell cavity volume. Although high values (above 10) generally indicate that oysters are in good physiological condition, low values do not necessarily indicate oysters in poor health because con- dition decreases whenever tissue is lost; spawning therefore results in a short term loss in condition. Long-term loss, however, may indicate stress from other sources such as pollutants, hypoxia, or disease. Mean condition indices decreased from 10.8 in February 1986 to 5.7 in February 1988. The high and low during this time were 12.4 in April 1986 and 4.6 in October 1987, respectively (Fig. 2). Linear regression yielded a signifi- cant decline in condition (p < 0.01; r- = 0.659; df = 24). Oyster size (as shell length) was not related to the decline because size showed no change with time (p > 0.20; r~ = 57 58 Abbe and Sanders Figure 1. Location of Holland Point oyster bar (black dot) near Bene- dict from which oysters were sampled during 1986-88. 0.007; df = 24). The general patterns of 1986 and 1987 were similar and typical for this area of the Patu.xent River (Abbe and Sanders 1986); we expect that these patterns were typical for oysters in much of Chesapeake Bay as well. On a seasonal basis, high values occurred early each year, followed by decreases in late spring or early summer, lows during late summer or early fall, and increases during fall (Fig. 2). These seasonal fluctuations are related to re- productive activity in spring and summer and glycogen storage in fall (Galtsoff 1964). What is unusual about these 2 years is the overall de- cline in condition index, with a mean of 9.4 in 1986 and 6.7 in 1987. This decline occurred during a time of wide- spread disease mortality in Chesapeake Bay. It is unclear, however, whether any relationship exists between mortality and condition, and any attempt to relate them here would be speculative. No unusual mortality was apparent among the oysters on Holland Point bar; it has traditionally been too far upriver to be affected by MSX. Yet something has happened to cause this loss in condition. JFMAMJJASONDJFMAMJJASONOJF 1986 1987 1988 Figure 2. Mean meat condition indices of oysters sampled monthly from February 1986 to February 1988 from Holland Point bar. The regression line and its equation are also shown. (Vertical bars are standard errors of the mean.) We know that MSX has a deleterious effect on the feeding rate of oysters and can result in reduced condition index (Newell 1985), but we also know that contaminants in estuarine water can produce measurable differences in condition (Scott and Lawrence 1982). Were lowered condi- tions of Holland Point oysters typical of other populations in the Chesapeake? If so, reduced condition, possibly indi- cating a poorer physiological state, might help explain the high mortalities observed during 1987, because oysters under stress from any cause would probably be more sus- ceptible to disease. No attempt is made at this time to determine the cause of this declining condition, but the use of a simple index, as discussed by Lawrence and Scott ( 1982), has alerted us to a potentially serious situation. Scott and Lawrence (1982) suggested increased emphasis on the use of condition index over a wider geographical range and for longer periods of time to monitor the wellbeing of oyster populations. We intend to continue to monitor this population in the Pa- tuxent River and urge others to consider the technique for other areas. ACKNOWLEDGMENTS The authors gratefully acknowledge the field and labora- tory assistance of Bill Yates, JoAnn Bianchi, Fritz Riedel, and Debbie Connell. This study was supported by the Bal- timore Gas and Electric Company. LITERATURE CITED Abbe. G. R. 1988. Population structure of the American oyster. Crassos- trea virginica. on an oyster bar in central Chesapeake Bay: changes associated with shell planting and increased recruitment. J. Shellfish Res. In press. Abbe, G. R. and J. G. Sanders. 1986. Condenser replacement m a coastal power plant; copper uptake and incorporation in the American oyster, Crassostrea virginica. Mar. Environ. Res. 19:93-113. Andrews. J. D. 1964. Oyster mortality studies in Virginia. IV. MSX in James River public seed beds. Proc. Natl. Shellfish. Assoc. 53:65-84. Andrews, J. D. 1983. Minchinia nelsoni (MSX) mfections in the James River seed-oyster area and their expulsion in spnng. Estuarine Coastal Shelf Sci. 16:255-269. Andrews. J. D. and J. L, Wood. 1967. Oyster mortality studies in Vir- ginia. VI. History and distribution o( Minchinia nelsoni. a pathogen of oysters, in Virginia. Chesapeake Sci. 8:1-13. Davis, H. E., D. W. Webster and G. E. Krantz. 1981. Maryland oyster spat survey, fall 1980. Maryland Sea Grant Tech. Rept. UM-SG-TS- 81-03. University of Maryland, College Park, MD. 22 pp. Decline in Oyster Condition Index 59 Ford, S. E. 1985. Effects of salinity on survival of the MSX parasite Ha- plosporidiiim nehoni (Haskin, Stauber. and Mackinl in oysters. J. Shellfish Res. 5:85-90. Galtsoff, P. S. 1964. The American oyster Cnissoslrea viri;iiiica Gmelin. US Fish and Wildl. Sen.. Fish. Bull. 64:1-480. Haskin H, H. and S. E. Ford. 1982. Haplosponduun nelsani (MSX) on Delaware Bay seed oyster beds: a host-parasite relationship along a salinity gradient. J. Imertebr. Pathol. 40:388-405. Kennedy, V. S. and L. L. Breisch. 1981. Maryland's oysters: research and management. Maryland Sea Grant No. UM-SG-TS-81-04. Uni- versity of Maryland, College Park, MD. 286 pp. Krantz, G. E. and H. A. Davis. 1983. Maryland oyster spat survey, fall 1982. Maryland Sea Grant Tech. Repl. UM-SG-TS-83-01. University of Maryland, College Park, MD. 14 pp. Lawrence. D. R. and G. I. Scott. 1982. The determination and use of condition index in oysters. Estuaries 5:23-27. Newell. R. 1. 1985. Physiological effects of the MSX parasite Haplospor- idium nelsoiii (Haskin, Stauber, and Mackinl on the American oyster Crassostrea viiginica (Gmelin). J. Shellfish Res. 5:91-95. Scott, G. 1. and D. R. Lawrence. 1982. The American oyster as a coastal zone pollution monitor: a pilot study. Estuaries 5:40-46. Rosenfield, A. and C. Sindermann. 1966. The distribution of "MSX" in middle Chesapeake Bay. Proc. Natl. Shellfish. Assoc. 56:6. Journal of Shellfish Research. Vol. 7, No. 1. 61-71. 1988. ESTIMATING MORTALITY RATES IN THE ICELAND SCALLOP, CHLAMYS ISLANDICA (O. F. MiJLLER) K. S. NAIDU Science Branch Department of Fisheries and Oceans P.O. Box 5667 St. John's, Newfoundland Canada. AlC 5X1 ABSTRACT Annual natural mortality in the Iceland scallop. Chlamys islandica computed from percent occurrence of cluckers was found to be significantly higher (P < 0.001) on exploited beds (0.140 to 0.204) than on unfished grounds (0.024 to 0.084). In a heavily fished population the difference in rates for fully recruited ages (2=9 yrs) between two consecutive years is ascribed to localized fleet movements and provides a first estimate of indirect (incipient) fishing mortality (0.07) for the population. The differ- ence between natural mortality rates for a population heavily fished and one in the virgin state, here estimated to be 0.047 provides a better estimate of gear-induced mortality. Depending on the type of gear used and the intensity of fishing effort, indirect fishing mortality was determined to vary from about 17% in the inshore Digby dredge to about 31% in the heavy, offshore New Bedford dredge, i.e. approximately four (3.9) and eight (7.7) times as many scallops pensh as a result of encounters with fishing gear than through natural causes KEY WORDS: Scallop, mortality. Chlamys isUiiulica INTRODUCTION The Iceland scallop, Chlamys islandica. has its main distiibution within the subarctic zone (Ekman 1953) but fish- able aggregations are found as far south as Nantuckett off the coast of Massachusetts (Serchuk and Wigley 1984). This species, being a filter feeder, is most commonly found in areas characterized by strong currents. Fisheries for the species occur principally in Canada, Iceland, Norway, and to a lesser extent in the United States. More recently, a fishery has developed in western Greenland (S. A. Pe- dersen, Greenland Fisheries Research Institute, Tagensvej 135, 1. DK-2200 Copenhagen N. Denmark, pers. comm.). The mollusc is relatively small, seldom exceeding 1 10 mm in shell height (tangential dorso-ventral axis). In Newfoundland (Canada), populations of the Iceland scallop are normally found in waters deeper than 55 m, usually on hard bottom with variable substrate composi- tion largely consisting of sand, gravel, shell fragments, rocks, and boulders. A pulse fishery for the species occurs in the northeastern Gulf of St. Lawrence, where annual landings have been as high as 239 t meats (Naidu et al. 1982). Recently, a directed fishery for scallops developed offshore on St. Pierre Bank (Fig. 1). This area is unique in that two scallop species (Iceland and sea scallops (Placo- pecten magellanicus)) are found, frequently intermixed. Gear typically used in the Canadian inshore and nearshore fishery consist of Digby buckets (Fig. 2) fished either singly or in a "gang"" of up to eight buckets spaced along a tow bar. In the offshore fishery large New Bedford type dredges (Fig. 2). frequently up to 4.8 m (16 ft) wide are employed. An empty 3.7 m (12 ft) dredge weighs about 0.7 t and upwards of 4 t when full of rocks (Royce 1946; Bourne 1964). Frequently, two such dredges are employed simultaneously, one on each the port and starboard. Fishing vessels, particularly those employing the heavier gear, are necessarily quite powerful. Tow duration is usually to gear saturation, time varying with scallop density. While the heavy, steel dredges customarily employed in harvesting scallops have long been suspected of causing damage to the benthos including targeted species, comparatively few studies have been conducted to assess their impact (Medcof and Bourne 1964; Caddy 1968. 1973; Gruffydd 1972). Caddy (1973). for example, estimated incidental mortali- ties to sea scallops P. magellanicus with an offshore dredge to be at least 13-17% per tow. This investigation provides estimates of non-yield fishing mortality in Iceland scallops. MATERIALS AND METHODS Systematic surveys for the Iceland scallop were con- ducted during 1980 and 1981 in the northeastern Gulf of St. Lawrence where an active commercial fishery had been un- derway (Naidu and Smith 1982). The surveys based on a systematic lattice design (Smith and Naidu 1981). were conducted to determine the distribution of scallops and to assess the suitability of the survey design for biomass esti- mation. Eleven latitudinal transects, each one nautical mile apart, were run in the target area. Stations were assigned at '/> mi. intervals along these lines. One hundred and three stations were occupied in 1980 but operational constraints reduced coverage to 59 in 1981 . All tows were made with a gang of four toothless Digby buckets mounted on a single tow bar. Dredges were equipped with 64 mm (2.5 in.) rings and carried a 38 mm (1.5 in.) polypropylene mesh liner to increase retention of smaller scallops. The liner was fre- quently inspected and repaired or replaced as necessary. Each tow covered Vi nautical mile with a 3: 1 warp to depth ratio. Tow speed varied from 2.5 to 3.0 knots. Dredges were hauled up at the end of each tow and the catch "bushelled"" into baskets and weighed to the nearest 61 62 Naidu 60° 59° 58° 57° 56° 55° 54° 53° 52° 51° 50° 49° 52' 50° 49° 48° 47' 46° 45' 44c T T" T" 52' ICELAND SCALLOPS SEA SCALLOPS ATLANTIC OCEAN 51' 50° 49° - 48' yiERRE BANK/ . •> ;l/\1 r N , 4? "^ \ GRAND BANK \% _L -L I ■ I ^ Lx 45' 60° 59° 58° 57° 56° 55° 54° 53° 52° 51° 50° 49 Figure 1. Map of Newfoundland (Canada) showing principal sampling areas. |0 kg. Individual shell heights were recorded (to the nearest mm) on either the whole catch, or a weighed random sub- sample, depending on the amount caught. All cluckers (persistent paired valves still attached at the hinge line) were counted and measured, again to the nearest mm. A tally was kept of animals so badly damaged that measure- ments were impossible (Fig. 3). Marked cluckers were used to determine the rate of tow- induced disarticulation during typical Va mi. tows. These were completed in the same survey area in the northeastern Gulf of St. Lawrence. Experimental cluckers were mea- sured and individually tagged (Naidu and Cahill 1985) tc distinguish them from those that would be caught during these tows. To better simulate typical dredge tows experi- mental cluckers were interspersed or "seeded"" into each bucket among 4-10 kg of live scallops before each tow. There were five experimental tows, each with 10 tagged cluckers in each of the four buckets. Seventy-five of the Mortality in Chlamys islandica 63 Figure 2. Digby scallop buckets (top) and a New Bedford offshore scallop dredge. tagged cluckers were of the gaping variety and the re- maining 125 were collapsed but still intact at the hingeline (Fig. 4). Detailed data were assembled on various fishery charac- teristics in the Gulf during 1980 and 1981. These included fishing positions, shell height composition of landings, and catch per unit of effort. Additional baseline data were gath- ered during independent research vessel surveys to deter- mine areal distribution and abundance of the mollusc and from selectivity experiments. Data on the percent occur- rence of Iceland scallop cluckers on unexploited beds on the Grand Banks of Newfoundland were available for one year only, but were derived from two separate cruises cov- ering wide areas. Elsewhere, on St. Pierre Bank, frequency of cluckers relative to live captures was available for both the pre- and post-exploited phases. Random samples of shells from the systematic line surveys in the northeastern Gulf of St. Lawrence were re- tained for age determinations. Rings on the left (upper) valve were employed in assigning ages and in back calcula- tions to derive shell heights at age. An age-shell height key was constructed using the complete shell height-at-ring for- mation data (N = 2658; Table 1) and used to generate Figure 3. Typical damage to dredge-caught Iceland scallops. scallop ages at given shell heights. The 50% retention shell height for the 64 mm ring used in the Gulf Chlamys fishery is 70 mm. corresponding to a mean retention age of 8 yrs (Naidu et al. 1982). Nine-year-old scallops were consid- ered fully retained by commercial gear. Natural mortality Natural mortality was computed directly from percent occurrence of cluckers (Dickie 1955) according to the equation: M = 1 - e-('HL)365 where M = annual natural mortality rate, c is the number of cluckers in a sample (adjusted to account for tow-in- duced disarticulation). L is the number of live scallops in the same sample, and t is the average time in days required for the valves of cluckers to separate naturally. Time re- quired for natural clucker disarticulation (210.8 days) was experimentally determined by Mercer (1974). The age- height key was used to determine ages at given shell heights for both live and dead scallops and age-specific nat- ural mortality rates calculated. These were separately com- puted for northeastern Gulf scallops for 1980 and 1981 . For Figure 4. Gaping (left) and collapsed cluckers of the Iceland scallop. 64 Naidu TABLE 1. Summary of back measurements of size-at-ring formation for 284 Iceland scallops taken in 1980 (tangential dorso-ventral distances) from the northeastern Gulf of St. Lawrence. Ring no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Mean size (mm) 6.8 14.5 24.4 36.1 47.8 58.1 66.5 72.7 77.5 81.8 85.0 87.0 89.3 90. ^ No. measurements 256 284 284 284 284 276 266 250 205 125 76 43 18 7 S.D. mean 1.6 2.9 4.3 5.3 5,7 5.6 5.6 5.4 5 5 5.7 5.2 5,2 5.3 5.1 Grand Banks and St. Pierre Bank only overall natural mor- talities were computed. Total mortality rate (Z) for Gulf scallops was calculated from commercial catch and effort data (Naidu et al. 1982). RESULTS Of the 200 tagged cluckers used to determine tow-in- duced disarticulation (75 gaping and 125 collapsed), 171 were retained (64 out of 75 gaping and 107 out of 125 col- lapsed. Tables 2 and 3). There was no difference (P > 0.05) in the mean size of the two categories: 80.8 ± 6.6 mm (SD), and 80.4 ± 7.2 mm for gaping and collapsed respectively. Neither was there a difference (P > 0.05) in the mean size of the two types of cluckers retrieved: 81.4 ± 6.6 mm versus 80.5 ± 7.2 mm. Loss of experimental cluckers during the tows was 15% for both categories. Fre- quency of tow-induced disarticulation was, as expected, higher in gaping than in collapsed cluckers in which hinge resilience had disappeared (29.7 vs 11.2%; Table 3), i.e. 89% of collapsed compared to 70% of gaping cluckers re- mained attached at the hingeline during tows with an overall mean of 82%. Unfortunately, data on the frequency occurrence of each clucker type on scallop grounds are not available. A predictive model was therefore constructed to evaluate the sensitivity of the mortality estimate (M) to varying representations of the two categories of cluckers and to adjustment factors computed therefrom. Disarticula- tion time (t) was allowed to vary from 50 to 300 days. The simulation showed M to decrease exponentially with t (Fig. 5). M values showed progressive convergence with in- creasing t and approached an asymtote towards high values of t. For t values greater than 200 days, varying the propor- tions of gaping to collapsed cluckers from one extreme ( 100% gaping) to another ( 100% collapsed) resulted in only small differences in M. As expected, M was relatively more sensitive to the adjustment factor used in correcting for tow-induced disarticulation than to various representa- tions of the two types of cluckers (Fig. 5). Within the range of adjustment factors obtained empirically, a wide variation in M IS predicted, particularly for small values oft. For the t value used in this study (210.8) a difference of up to 47% in the estimate is possible. Although records were not kept on the frequency of the two types of cluckers it was ob- vious that there were many more collapsed cluckers on the scallop grounds than of the gaping variety. Also, cluckers appear to persist much longer in the collapsed state than in the gaping condition. This results in fewer gaping cluckers in the population. While the higher disarticulation rate among gaping cluckers would lead to a high-biased esti- mate of M, their relative numbers would tend to underesti- mate it. Given the reduced sensitivity of M to t and to various representations of the two categories of cluckers in the context of their persistence over time, it seemed reason- able on empirical grounds to use the weighted average of 1.221 in this first attempt to adjust clucker numbers to ac- count for tow-induced disarticulation (Table 3). Comparisons of the rates of disarticulation within indi- vidual buckets for each tow and between separate tows (Table 2) show greater variability in disarticulation in the latter. Whereas overall rates for individual buckets varied narrowly between 16.7 and 20%, those for the five separate tows ranged from 5 to 39%-. These observations suggest that type of bottom and composition of substrates may have an effect on the fate of cluckers entering the dredge. There TABLE 2. Number of cluckers retained per tow in simultaneous 4-bucket tows and numbers (in parentheses) forcibly disarticulated. Bucket % disarticulating Tow No. A B C D Totals by tow 1 10(1) 8 (1) 9(2) 10(0) 37 (4) 10.8 2 10(1) 9(0) 10(0) 10(1) 39(2) 5.1 3 9(1) 10(1) 10(4) 6(2) 35 (8) 22.9 4 4(2) 8(3) 8(3) 9(3) 28(11) 39.3 5 8(2) 9(3) 8(0) 7(1) 32(6) 18.8 Totals 41(7) 44(8) 45 (9) 42(7) 171 (31) % disarticulating 17.1 18.2 20.0 16.7 18.1 Mortality in Chlamys islandica 65 TABLE 3. Disarticulation based on original condition. Total numbers of each type used, recoveries and numbers forcibly disarticulated (parenthesized). Clucker numbers Initial Recaptured Adjustment factor Tow No. Gaping Collapsed Totals Gaping Collapsed Totals 1 21 19 40 19(3) 18(1) 37 1.121 2 9 31 40 9(11 30(1) 39 1.054 3 12 28 40 11 (51 24(3) 35 1.297 4 18 22 40 13 (7) 15(4) 28 1.647 5 15 25 40 12(31 20(3) 32 1.232 Totals 75 125 200 64 (19) 107(12) 171 (31) % disarticulating 29.7 11.2 18.1 1 221 Adjustment factor 1.422 1.126 is a slight but insignificant positive correlation (r- = 0. 15. P > 0.05) between numbers disarticulating and catch weight (Tables 2 and 4). No correlation between disarticu- lation rate and depth (r-^ = 0.01, P > 0.05) was discemable within the narrow depth range examined (Table 4); indeed none was expected. Overall, ratios of cluckers to live scallops in each of the three geographically discrete populations varied from less than two percent on St. Pierre Bank to 13.27? for the Gulf of St. Lawrence (Table 5). Annual natural mortalities for unexploited stocks of Iceland scallops in the Newfoundland 150 175 200 225 OlSflRTicULflTION TIME IDflTSI Figure 5. (A) Simulated trend with disarticulating time of natural mortality using varying representations of gaping and collapsed cluckers (0 = 100% gaping, 0% collapsed; 1 = 75% gaping, 25% collapsed; 2 = 50% gaping, 50% collapsed; 3 = 25% gaping, 75% collapsed and 4 = 0% gaping, 100% collapsed). (B) Simulated trend with disarticulating time of natural mortality using the range of ad- justment factors obtained experimentally. area varied from 0.025 to 0.084. Computed values were consistently lower than those for exploited populations. The overall mean by averaging values from Grand Banks (1982) and St. Pierre Bank (1976. 1978) was 0.047 (or 5%). Frequency of cluckers in the Gulf population de- creased significantly (P < 0.001) to 8.7% in 1981 from 12.2% the previous year. Although clucker numbers re- mained relatively low on St. Pierre Bank, an increase was evident in 1983 soon after an active commercial fishery had commenced in 1982. albeit for the sea scallop. P. mugel- lunicus. Although mortalities in exploited contagions for both scallop species remained somewhat high through 1985. beds unfished on the Bank continued to register low natural mortalities (Table 5). As expected, the distribution of cluckers showed wide spatial variation, with inordinately large numbers being evident in some areas (Fig. 6). That there is a correspondence between sea scallop clucker numbers and fishing activity is further confirmed by their abundance in relation to total removals by year (Table 6). Mean numbers of sea scallop cluckers per tow from areas for which both research and commercial catch data are available in any year were significantly correlated to the catch in that year (r^ = 0.85, P < 0.05). Although a de- crease in the mean numbers per tow of cluckers of the Ice- land scallop was clearly evident during the same period, the TABLE 4. Scallop catch (whole weight, kg) by individual bucket and mean catch per tow. Depth Range (m) Catch weight by bucket Mean Tow No. A B C D catch/tow 1 75 9.0 8.5 9.0 8.5 8.8 2 77-80 7.0 10.0 8.5 7.0 8.1 3 69-73 12.5 12.5 12.5 9.5 11.8 4 68-73 8.3 12.5 9.5 12.5 10.7 5 69-77 15.5 14.0 16.0 14.0 14.9 Mean/hucket 10.5 11.5 III 10.3 66 Naidu TABLE 5. Estimates of annual natural mortality for three populations of Chlamys islandica from Newfoundland (M,, based on cluckers only; M2, based on cluckers and crushed scallops). Year Location Gear „ Live scallop nos. Tow 1 distance Crushed (nautical mi.) Measured (%) Clucker nos. Total Adjusted Observed (%) M, M, 1980 NE Gulf of St. Lawrence- 1981 NE Gulf of St. Lawrence^ 1982 Grand Banks' 1982 Grand Banks' 1976 St. Pierre Bank' 1976 St. Pierre Bank' 1983 St. Pierre Bank2 1983 St. Pierre Bank^ 1985 St. Pierre Bank^ 1985 St. Pierre Bank' 1985 St. Pierre Bank' 1985 St. Pierre Bank' 1985 St. Pierre Bank' Digby X 4 (lined) Digby X 4 (lined) New Bedford (unlined) New Bedford (unlined) Digby X 4 (lined) Digby X 4 (unlined) New Bedford (lined) New Bedford (unlined) New Bedford (unlined) New Bedford (lined) New Bedford (unlined) New Bedford (lined) New Bedford (unlined) 0.25 0.25 1.0-1.5 1.0 0.25 0.25 0.5 1.0 0.5 0.5 0.5 0.5 0.5 9,950 16.947 4.583 6.069 42,261 3,597 2,410 2,570 1,894 175 (1.7) 396 (2.3) 17,297 3.158(15.4) 11.138 1.279(10.3) 1,235 55 (4.3) 1.442 60(4.0) 114(2.4) 1.143 (15.8) 6,906(14.0) 377 (9.5) 324 (11.9) 255 (9.0) 236 ( 1 1 . 1 ) 10,125 1,095 1,336(13.2) 0.204 0.231 17,343 1,235 1.507(8.7) 0.140 0.17" 20.455 434 529 (2.6) 0.044 0.30S 12,417 340 415 (3.3) 0.056 0.23; 1,290 53 65(5.0) 0.084 0.\5t 1,502 43 52(3.5) 0.058 0.12f 4,697 448 547(11.6) 0.183 0.221 7,212 585 714(9.9) 0.158 0.411 49,167 2,494 3.043(6.2) 0.102 O.iit 3.974 62 76(1,9) 0.033 0.19f 2.734 45 55 (2.0) 0.034 0.23^ 2.825 54 66(2.3) 0.040 0.194 2.130 25 31 (1.5) 0.025 0.21' ' Localized tows associated with selectivity studies (unexploited grounds). ^ Tows restricted to commercially attractive contagions (exploited grounds) ' Exploratory or survey tows covering large areas (unexploited grounds). Figure 6. Spatial distribution of annual natural mortality. M ( x 100) in the Iceland scallop computed from ratio of cluckers to live scallops over a portion of St. Pierre Bank in 1985. Balloons, boxes and columns correspond to areas with catches of 0-49, 50-99, and over 100 live scallops/tow respectively. correlation between their numbers and landings was not statistically significant (r- = 0.68, P > 0.05). The frequency with which animals were crushed during capture shows a correspondence with the type of gear em- TABLE 6. Relationship between mean clucker numbers/tow and removals from St. Pierre Bank. Number of research vessel tows for which data on clucker incidence are available from exploited areas is parenthesized. Mean clucker no./tow Year Landings (MT) Iceland Sea 1982 717 6.271 (59) 1.538(78) 1983 594 5.500 (86) 1.341 (88) 1984 413 5.556(99) 0.856(118) 1985 53 4.543 (79) 0.834(169) 1986 19 1,682 (44) 0,480 (74) Mortality in Chlamys islandica 67 ployed (Table 5). The heavier offshore dredges crushed greater numbers of scallops. The frequency and severity of damage intlicted was greatest with the offshore unlined commercial gear. In independent studies, it has been shown that overall damage to scallops increases with distance towed and that for a given tow distance, frequency of damage is greater in tows completed over rough bottom than those over smooth grounds (Naidu. unpublished). Age determinations using annual rings on the shell were reliable only to about 1.3 yrs. Beyond that there were problems with interpretation. Recaptures of live tagged scallops after nearly a decade (9 yrs.) in the wild showed no shell growth (Naidu, unpublished). Growth rings may merge with one another or become laminated to conceal size increments related to age (Naidu et al. 1982). Age-specific natural mortality in the Gulf during two consecutive years (1980 and 1981) increased with age (Tables 7 and 8). Annual mortality rate computed from the ratio of cluckers to live scallops was substantially lower during the second year of the survey. An analysis of the covariance of the logarithmic transformations of M on age for 1980 and 1981 pointed to no significant differences in slopes (P > 0.05) but elevations were different (P < 0.01). Total mortality (Z) for the 1980-81 period increases with age (Table 9) as a result of increasing natural and fishing mortalities (weighted mean of 0.742 corresponding to an F of 0.533). DISCUSSION In most fisheries literature, particularly stock assess- ments, the rate of natural mortality for a given population is assumed to remain constant throughout the adult history of the animal. This rate is factored into various yield compu- tations either for simplicity in modelling, or more fre- quently, simply because data on age-specific mortality are simply lackii.g. In scallops, we are fortunate in not only being able to age them readily but in distinguishing those that die natum' y from those removed by the fishery where the valves are ^ liberately and forcibly separated at the hinge- line during shucking to remove the adductor muscle or "meat." Senescence alone cannot account for all natural deaths, particularly in an area where an active capture fishery is underway as was the case in the Gulf of St. Lawrence where significant removals had occurred in both 1980 and 1981. This study attempts to address the non-yield mor- tality resulting from fishing. Indirect mortality occurs at various stages during the process of fishing. Not every an- imal in the gear path is caught or retained. Some scallops elude capture gear and recover completely from the "blitz." Others may be crushed in situ as the heavy dredge moves along the sea bottom or may be inadvertently forced into the substrate and become debilitated to face certain death. Shell damage to uncaptured sea scallops (P. magel- lanicus) has been directly observed through diving (Caddy 1968). In addition to physical damage to scallops within the dredge, active muscle adductions associated with escape responses, particularly during saturation tows, lodge grit and shell fragments into the scallop. These contaminants cause various degrees of mantle retraction and prolonged moribundity, sometimes no doubt resulting in death. Dam- aged or weakened scallops are more likely engaged by predators. Caddy (1968) reported numerous scavenging benthic predators to be attracted by the passage of the scallop drag. Further damage is intlicted when scallops are dumped on deck. This involves inverting the scallop rake TABLE 7. Age-specific natural mortalities computed from ratio of cluckers to live scallops in the Gulf of St. Lawrence, 1980. Observed clucker numbers are adjusted upwards by a factor in the range of 1.054 to 1.647 (x = 1.221) to allow for tow-induced disarticulation (C, = adjusted clucker numbers; M„ and M, = observed and adjusted mortalities). No. cluckers observed No Adjustment factor 1.054 1.12 1. 232 1.297 1. 647 X = 1.221 Age live M„ c. M. c. M. C. M. c. M. c. M. c. M. 5 6 127 ,079 6 .083 7 .088 7 .096 8 .101 10 .126 7 095 6 26 288 .145 27 .152 29 .161 32 .175 34 .184 43 .227 32 174 7 68 952 .116 72 .122 76 .129 84 .141 88 .148 112 .184 83 140 8 152 1911 .129 160 .135 170 .143 187 .156 197 .164 250 .203 186 155 9 220 2387 .148 232 .155 247 .164 271 .178 285 .187 362 .231 269 177 10 219 1808 .189 231 .198 245 .210 270 .228 284 .238 361 .292 267 226 11 167 1187 .216 176 .226 187 .239 206 .259 217 .271 275 .330 204 257 12 129 691 .276 136 .289 145 .304 159 .328 167 .342 212 .413 158 326 13 54 300 .268 57 .280 61 .295 67 .319 70 .333 89 .401 66 317 14 39 129 .408 41 .424 44 .444 48 .475 51 .493 64 .578 48 472 L5-14 S9-13 1080 789 9780 6373 .174 .193 1138 832 182 202 1211 885 .193 !I4 1331 973 .210 .232 1401 1023 .220 .243 1778 1299 .270 .297 1320 964 .208 .230 68 Naidu TABLE 8. Age-specific natural mortalities computed from ratio of cluckers to live scallops in the Gulf of St. Lawrence, I98L Observed clucker numbers are adjusted upwards b} a factor in the range of L054 to L647 {x = L221I to allow for tow-induced disarticulation (C, = adjusted clucker numbers; M„ and M^ = observed and adjusted mortalities). No. cluckers No Adjustment factor 1.054 1. 121 1.232 1.297 1. 647 X = 1.221 Age observed live M. c. M. c. M. c. M. c. Ma c. Ma c. M. 5 9 199 .075 9 .079 10 .084 11 .092 12 .097 15 .121 11 .091 6 29 512 .093 31 .098 33 .104 36 .114 38 .119 48 .149 35 .113 7 94 1836 .085 99 .089 105 ,095 116 .103 122 .109 155 .136 115 .103 8 197 3700 .088 208 .093 221 .098 243 .107 256 .113 324 .141 241 .106 9 267 4302 .102 281 .107 299 113 329 .124 346 .130 440 .162 326 .123 10 239 2999 .129 252 .135 268 .143 294 .156 310 .164 394 .203 292 .155 11 174 1768 .157 183 .164 195 .174 214 .189 226 .198 287 .245 212 .188 12 117 948 .192 123 .202 131 .213 144 231 152 ,242 193 .297 143 .230 13 58 368 .239 61 .250 65 .264 71 .286 75 .298 96 .362 71 .283 14 30 134 .321 32 .335 34 .352 37 .380 39 .395 49 .472 37 .377 25-14 1214 16766 .118 1279 .124 1361 .131 1495 .143 1576 .150 2001 .187 1483 .142 S9-13 855 10385 .133 900 .139 958 .148 1052 161 1109 .169 1410 .209 1044 .160 allowing the catch, including rocks, to free fall from heights frequently exceeding 3 m (10 ft.). Dredge-caught scallops nearly always contain a few animals so severely damaged as to provide no yield (Fig. 3). In fact, these may be considered to be "instant" cluckers. It is apparent, however, that the frequency of damage is greater in the unlined offshore dredge than in lined gear normally employed in research vessel tows. Lined gear cushions the severity of impact between scallops and steel, thus minimizing damage. Sometimes, scallops are volun- tarily or inadvertently returned to the sea bed after varying periods of deck exposure. Culling of scallops becomes nec- essary to ensure meat counts are nonviolative (regulated number of scallop meats per unit weight, e.g. 40 meats/500 g). Depending on the severity of damage, these may suffer subsequent mortality. Sorting scallops from trash seldom exceeds '/: hr. Mortality among scallops returned to the sea bed within a reasonable time is low, at least in minimally damaged sea scallops. P. magellaiucus (Naidu. unpub- lished). Medcof and Bourne (1962) estimated mortality in TABLE 9. Estimates of Z and F for Iceland scallops for 1980/81 in the Gulf of St. Lawrence. (Total mortality coefficient, Z is from Naidu et al. 1982.) Age Z M F 9- 10 0.463 150 0.313 10- -11 0.731 U.190 0.541 11- -12 0.964 0.222 0.742 12- -13 1.290 0.278 1.012 13- -14 1.684 0.300 1.384 z„. ,3, = 0.7417 M(198C -81) = 0.195 F(1980 -811 = 0.533 sea scallops resulting from prolonged air exposure to range from 2 to 20%. Iceland scallops maintain tighter shell clo- sure than do sea scallops, perhaps enabling them to tolerate longer periods of air exposure. Scallops that survive turbu- lent encounters with fishing gear register the accompanying stress through the deposition of a shock ring usually re- sulting from mantle damage or temporary mantle retrac- tion. Depending on the severity of impact, this may be a mild shock ring, as is commonly observed in several ex- ploited scallop species or it may be sufficiently violent to require replacement of portions of the shell margin. This results in various changes in the anatomical configuration of the shell and leads to some deformity (Fig. 7). It is evi- dent, however, that types of damage commonly seen may only result from extrinsic man-made causes. The difference between an average weighted M from a population heavily fished and one in the virgin state was used to estimate indirect fishing mortality. While the abso- lute values reported in this paper may be subject to change by a number of factors, the overall difference should re- main unchanged so long as disarticulation time and the factor used to adjust for tow-induced disarticulation remain unchanged. Simply, it is the "added"' non-yield compo- nent that is being estimated. The mean adjustment factor used in this study to compensate for tow-induced disarticu- lation (1.221) must be considered tentative. It recognizes the unequal representation of the two types of cluckers. To have assumed an equinumerical representation would have required adjusting observed numbers by a factor of 1.274 which would result in high-biased mortality estimates. In situ sampling by means more sophisticated than dragging would be required to estimate actual numbers of the two types of cluckers. Also, as already noted, it is likely that Mortality in Chlamys islandica 69 ^^^^^^^^^^^^F w r^^^^^^^^^H ^PP;^^ «ll ^H' f J ^H, hm ^A ff '^'fl 1^ i^l ^^^v^ \ '"V *ii .^^1 r 1 ''-^>^^| ^^^^^BCJw^ -^^ r-^l ■'1 ' '* '^^^ 'V '' ''''^^^1 p '-' ^^^1 ^^P 1 ^^1 ^^^^^^^BC ■ Figure 7. Common shell deformities in Iceland scallops captured from a heavily exploited bed in the northeastern Gulf of St. Lawrence. disarticulation rates will vary with type of bottom and with capture gear. The simple adjustment for clucker numbers by a factor of 1 .221 is based on 'A mi. tows in the Gulf of St. Lawrence. We may expect greater numbers to disar- ticulate during longer tows. Baseline estimates of M for the Grand Banks and St. Pierre Bank must therefore be consid- ered underestimates. Some of the constraints implicit in the calculation of natural mortality rates from clucker to live scallop ratios have been discussed by previous authors (Dickie 1955; Merrill and Posgay 1964; Mercer 1974). Most of them tend to overestimate M. Among these is the assumption that both cluckers and live scallops are equally vulnerable to capture. While precise figures on the effi- ciency of Digby dredges for capturing Chlamys are not available, it is known to vary, depending upon bottom type, between 5 to 15% for the sea scallop. P. magel- 70 Naidu lanicus (Dickie 1955). Iceland scallops are probably more prone to being captured as they are unrecessed and byssally attached to bottom substrates (Gruffydd 1976. 1978). Ex- amining a population of Iceland scallops in Balsfjord in northern Norway, Vahl and Clausen (1980) concluded that nearly all scallops were attached by the byssus and that swimming is a relatively rare occurrence. In the laboratory, 76% (155 of 205) of Iceland scallops (43-93 mm, shell height) were found to be byssally attached, whereas only 9% (8 of 88) of sea scallops (59-70 mm) were smiilarly attached (Naidu and Meron 1986). As active escapement would be minimized, it is probable that the assumption of equal catchability is reasonable. It is likely, however, that long-dead cluckers are more likely to disarticulate than re- cently-collapsed cluckers. This would cause an underesti- mation of their numbers in the catch thereby introducing a low bias in the mortality estimate. On the other hand, the entire population is not equally vulnerable to fishing effort at any time (Caddy 1975). This unequal distribution of fishing effort will inflict a mosaic of M, depending on the duration and intensity of that effort. This, too, will under- estimate M, particularly when it is computed from clucker to live numbers pooled from a large area. Detailed exami- nation of the spatial distribution of M on St. Pierre Bank over three years supports this assumption. Commercially attractive contagions subjected to the most effort produced highest ratios of cluckers to live scallops providing direct confirmation of fishery-induced mortality. This may ex- plain the "graveyards"' containing very high proportions of cluckers observed here and reported elsewhere such as those on Georges Bank (Posgay 1962). He suggested that in a heavily fished area, live sea scallop numbers will be reduced faster than cluckers. thereby creating an artificially high clucker to live scallop ratio. It is difficult to reconcile this with his earlier observation that cluckers may be more catchable than live scallops. The disproportionate numbers of cluckers reported from exploited populations more likely result from mortality inflicted from repeated tows. At the time observations were made, the northeastern Gulf fishery had been expanding and moving away from grounds heavily exploited to virgin beds, but within the same general area (Naidu et al. 1982). Careful monitoring of the fishery in both 1980 and 1981 pointed to fleet move- ments based on localized depletion and declining catch per unit of effort. Nevertheless, the fishery was confined to within a 100 nautical mi"^. Mortality computed from a clucker to live scallop ratio in the second year would in- clude a measure of incipient, fishery-induced mortality in the population. The overall difference in weighted means between the two years of fully-recruited ages (>9-l- yr) provides a minimum estimate of indirect fishing or non- yield mortality in this fishery. For purposes of assessment, particularly for yield-per-recruit considerations, due cogni- zance should be taken of this non-yield mortality, here esti- mated to be at least 0.07. F, then approximates to 0.463 from 0.533 (i.e. 0.533-0.07). Until very recently there was no commercial fishery for scallops on either St. Pierre Bank or the Grand Banks of Newfoundland. Rates of natural mortality computed for unexploited stocks were considerably lower than those in the Gulf of St. Lawrence with an overall mean of 0.047. If we assume that the ratio of cluckers to live scallops in un- fished populations provides a better estimate of intrinsic natural mortality for the species, then indirect fishing mor- tality computed for the exploited Gulf population would be 0.157 (0.204-0.047) and 0.093 (0.140-0.047) in each of 1980 and 1981 respectively (Table 5). Corresponding non- yield mortality (gear-induced) would be 0.184 (0.231- 0.047) and 0. 130 (0. 177-0.047). The absence of a marked increase in the instantaneous mortality coefficient on un- fished beds over a nine-year period on St. Pierre Bank to the order of magnitude observed on exploited beds provides corroborative evidence supporting the hypothesis that fishing itself initiates an additional mortality in the popula- tion. A fishery for sea scallops commenced on St. Pierre Bank in 1982. Concomitantly, but not surprisingly, natural mortality of Iceland scallops computed from clucker to live scallop ratios, particularly in heavily fished contagions be- came elevated and has been reported as high as 0.206 (Naidu and Cahill 1984). Meanwhile, the Iceland scallop fishery in the northeastern Gulf of St. Lawrence has ex- panded into hitherto unexploited grounds off Labrador. As expected, pre-exploitation natural mortality computed from the ratio of dead to live scallops was considerably lower than those separately determined by Mercer (1974) and Naidu et al. (1982) and has been reported to be as low as 0.066 (Lanteigne et al. 1986). Previous estimates no doubt included the mortality component ascribable to fishing. More recently, Lanteigne and Davidson (1987) noted a 74% difference in the frequency of cluckers in this fishery over a two-year period during which significant removals had occurred. Repetitive tows over productive grounds may explain the observed disparity between the two years. There was probably considerable overlap between the areas surveyed in the second year and where an intense fishery had occurred one year previously. The absence of statistical correlation between frequency of cluckers and spatial removals of Iceland scallops from St. Pierre Bank (Table 6) is not surprising. The directed fishery on St. Pierre Bank was aimed at sea scallops only. Iceland scallops are frequently intermixed with sea scallops and when caught in large numbers were considered to be a nuisance bycatch and discarded. There was a directed fishery for Iceland scallops in 1984 (Naidu and Cahill 1985), but this was spatially well removed from the areas of sea scallop exploitation. The majority of crushed scallops are discarded and pro- vide little yield to the fishery. The ones categorized in this study as being damaged most certainly perish when re- turned to the sea bed. These may be considered as contrib- uting to the net clucker population. When these are added to the mortality computations, it becomes apparent that Mortality in Chlamys islandica 71 total non-yield mortality (natural and indirect fishing mor- tality) could be as high as 20% in lined Digby dredges and possibly as high as 347c in the unlined New Bedford dredge. Clearly, the ratio of dead to live scallops in un- fished populations provides a more realistic measure of in- trinsic natural mortality for the species. This being so, an- nual indirect fishing mortality must be estimated to be at least as high as 0.184 (0.231-0.047) or 17% and possibly as high as 0.364 (0.41 1 -0.047) or 31% for the Digby and New Bedford dredges respectively, i.e. approximately four (3.9) and eight (7.7) times as many scallops die from en- counters with fishing gear than through natural causes. In- cidental mortality to Iceland scallops with the offshore dredge is approximately twice the value (13-17%) reported for sea scallops. P. magellanicus (Caddy 1973). The higher rate may be due to the more sedentary nature of the species as well as the greater propensity to being byssally attached, sometimes at exceedingly high densities. Annual natural mortality reported here for unexploited stocks of Iceland scallops (5%^) is approximately one-half the previous estimate of 11% (Mercer 1974). His estimate for the exploited Gulf of St. Lawrence population probably includes the moiety of non-yield mortality intlicted by fishing. The lower rate reported in this paper appears to be compatible with the expected longevity of this arctic-boreal species. ACKNOWLEDGMENTS Principal research support for this study was provided by Messers F. M. Cahill and D. B. Lewis, both from the Science Branch. Department of Fisheries and Oceans. St. John's. Newfoundland. Canada. Data processing assistance by Mr. D. Stansbury is gratefully acknowledged. All pho- tographs are by Mr. G. King. Drs. G. P. Ennis and G. H. Winters provided comments on an earlier version of this paper. REFERENCES Boume. N. 1964. Scallops and the offshore fishery of the Maritlmes. Bull. Fish. Res. Board Canada. No. 145, 60 p. Caddy. J, F. 1968. Underwater ob.servations on scallop {Placopeclen ma- gellanicus) behaviour and drag efficiency. J . Fish Res. Board Canada 25(10);2123-2141. Caddy. J. F. 1973. Underwater observations on tracks of dredges and trawls and some effects of dredging on a scallop ground. J. Fish. Res Board Canada 30(21:173-180. Caddy. J. F. 1975. Spatial model for an exploited shellfish population, and its application to the Georges Bank scallop fishery. J. Fish. Res. Board Canada 32(8):1305- 1328. Dickie, L. M. 1955. Fluctuations in abundance of the giant scallop, Pla- copeclen magellanicus (Gmelin) in the Digby area of the Bay of Fundy. J. Fish Res. Board Canada 12(61:797-857. Ekman, S. 1953. Zoogeography of the Sea. Sigwick and Jackson, London, 417 p. Gruffydd, LI. D. 1972. Mortality of scallops on a Manx scallop bed due to fishing. J. Mar. Biol. Ass. U.K. Vol. 52(2):449-455. Gruffydd, LI. D. 1976. Swimming in Chlamys islandica in relation to current speed and an investigation of hydrodynamics lift in this and other scallops. Nor. J. Zool. 24(4):365-378. Gruffydd, LI. D. 1978. The byssus and byssus glands in Chlamys islan- dica and other scallops (Lamellibranchia). Zool. Scnpia. 7(4):277- 285. Lanteigne, M., L.-A. Davidson & J. Worms. 1986. Status of the Iceland scallop. Chlamys islandica in the northeastern Gulf of St. Lawrence. 1985. Can. All. Fish. Sci. Adv. Committee Res. Doc. 86176. 20 p. Lanteigne. M. & L.-A. Davidson. 1987. Status of the northeastern Gulf of St. Lawrence Iceland scallop (Chlamys islandica) stock- 1986. Can. Atl. Fish. Sci. Adv. Committee Res. Doc. 87/83. 21 p. Medcof. J. C. & N. Boume. 1964. Causes of mortality of the sea scallop. Placopeclen magellanicus. Proc. Natl. Shellfish As.wc. 53:33-50. Mercer. M. C. 1974. Natural mortality of the Iceland scallop (Chlamys islandicus) in the Gulf of St. Lawrence. ICES CM. 1974/K:7. 1 1 p. Merrill. A. S. & J. A. Posgay. 1964. Estimating the natural mortality rate of the sea scallop (Placopeclen magellanicus). Intern. Comm. Norlhw. Atl. Fish. Res. Bull. 1:88-106. Naidu, K. S.. F. M. Cahill & D. B. Lewis. 1982. Status and assessment of the Iceland scallop, Chlamys islandica. m the northeastern Gulf of St. Lawrence. Can. Atl. Fish. Sci. Adv. Committee Res. Doc. 82/2. 66 p. Naidu, K. S. & S. J. Smith. 1982. a two-dimensional systematic survey of the Iceland scallop. Chlamys islandica. in the Strait of Belle Isle. Can. All. Fish. Sci. Adv. Committee Res. Doc. 82/4. 24 p. Naidu. K. S. & F. M, Cahill. 1984. Status and assessment of St. Pien-e Bank scallop stocks. 1982-83. Can. All. Fish. Sci. Adv. Committee Res. Doc. 84/69. 56 p. Naidu. K. S. & F. M. Cahill. 1985. Offshore fleet directs fishing effort on the Iceland scallop. Chlamys islandica. Can. Atl. Fish. Sci. Adv. Committee Res. Doc. No. 85/20. 13 p. Naidu. K. S. & F. M. Cahill. 1985. Mortality associated with tagging in the sea scallop. Placopeclen magellanicus (Gmelinl. Can. Atl. Fish. Sci. Adv. Committee Res. Doc. 85/21. 10 p. Naidu, K. S. & S. Meron. 1986. Predation of scallops by Amencan plaice and yellowtail flounder. Can. Atl. Fish. Sci. Adv. Committee Res. Doc. 86/62. 25 p. Posgay, J. A. 1962. Maximum yield per recruit of sea scallops. ICNAF Res. Doc 62/73. 20 p. Royce, W. F. 1946. Gear used in the sea scallop fishery. Commer. Fish Rev. 8(12):7-ll. Serchuk. F. M. & S. E. Wigley. 1984. Results of the 1984 sea scallop research vessel survey: status of sea scallop resources in the Georges Bank. Mid-Atlantic, and Gulf of Maine Regions and abundance and distributions of Iceland scallops off the southeastern coast of Cape Cod. Nat. Mar. Fish. Ser.. Northeast Fishenes Center. Woods Hole Laboratory Ref Doc. No. 84-34. 74 p. Smith. S. J. Si K. S. Naidu. 1981. Estimating the variance of the mean from a systematic sample in two dimensions — a simulation study. Can. Atl. Fish. Sci. Adv. Committee Res. Doc. 81/74. 21 p. Vahl. O. & B. Clausen. 1980. Frequency of swimming and energy cost of byssus production in Chlamys islandica (O. F Mullen. J. Cons int. E.xplor. Mer. 39(I):I0I- 103. Journal of Shtllfish Research. Vol. 7, No. I, 73-76, 1988. DAILY GROWTH RINGS IN JUVENILE SAUCER SCALLOPS, AMVSWM BALLOTI (BERNARDI) L. M. JOLL Western Australian Marine Research Laboratories P.O. Bo.x 20 North Beach 6020, We.stern Australia .ABSTRACT The number of pignienled rings occurring in the growth increment of tagged juvenile saucer scallops Amusium halloli (Bernardo in Shark Bay. Western Australia was closely related (r = 0.9%) to the number of days at liberty. The number of rings was always slightly less than the number of days, because of the loss of small amounts of shell from around the lip during the recapture process. The nature of the pigment forming the rings, which are laid down in the outer, calcific layer of the shell. i.s unknown. The rings become very closely packed when animals mature and enter their first spawning season, which provides a means of distin- guishing recruits from older year classes. Total ring counts may provide a means of direct ageing of recruits while growth per ring gives a permanent record of the daily rate of growth. KEY WORDS: Scallop. Pectinidae. Amusium. daily growth, growth rings INTRODUCTION The shells of a variety of recent and fossil bivalves have been shown to have marks which delineate increments of growth claimed to relate to various time periods. The most widely recognized is the annual growth ring, but daily and sub-daily marks have also been reported (Barker 1964, Broom and Mason 1978, Deith 1985, Lutz and Rhoads 1980 (review)). These marks may be structural changes on the surface of the shell (e.g. as in Argopecten irradians, (Lam.) (Wheeler, et al. 1973), Pecten diegensis (Dall) (Clark, 1968)) or internal marks which are only visible in thin sections or acetate peels (e.g. as in Mercenaria mer- cenaria (Linn.) (Panella and McClintock, 1968), Cerasto- derma edule (Linn.) (Richardson et al. 1981)). Supporting evidence for the validity of some of the claims for the time periods represented by the increment between the marks has, however, not always been pro- duced. Gruffydd (1981) examined the production of sup- posed daily ridges by Pecten maximus (Linn.) under exper- imental conditions and found that the rate of production of ridges was not always daily and that it was particularly af- fected by temperature and shell size. The outer surface of scallops of the genus Amusium are smooth and polished, with the left valve being brown to deep red and the right valve white (Habe 1964). The colour of the left valve of A. halloti (Bemardi) from Western Aus- tralia comes from concentrically arranged, fine, pigmented lines on a pale background. In the juvenile zone of shell growth the rings are readily resolved under a dissecting mi- croscope, but in the more distal parts of the shells of mature animals the pigmented rings blend together. Specimens from the east coast of Australia also show fine, pigmented rings in the juvenile areas of the left valve, although the overall effect is of a darker shell as the rings are darker and there is some pigmentation between the rings. This paper reports the observation that the number of these pigmented rings in the growth increment of tagged juvenile A. halloli from Shark Bay is very closely related to the number of days at liberty and provides evidence that they are laid down daily. METHODS (i) Tag-recapture Three groups of individually tagged scallops were re- leased on the commercial scallop trawling grounds in Shark Bay. Western Australia (Lat. 25°00'S; Long 113°30'E) in the course of studies on growth and mortality. Orange PVC tags C'Dymo"" tape. Esselte Dymo) were applied to the shell using a cyano-acrylate glue ("Supa-Glue". Selleys). Scallops were captured by trawling using the research vessel "Flinders", measured, tagged and returned to the sea within 1-4 hours. Except for brief periods during the sorting of the trawl and the measuring and tagging, scallops were stored in running sea water in a deck tank. The three groups of animals were at liberty for differing amounts of time (Table I). Recaptures of Group 1 animals were made by commerical vessels, while recaptures of the other two groups were made by R. V. "Flinders". Exact dates were available for all recaptures by "Flinders", but only 67 of the 140 commercial vessel recaptures had known recapture dates. (ii) Ring Counting As a result of the loss of a small amount of the shell around the lip at the time of capture, most recaptured scallops show a small step in the otherwise smewth surface of the shell, corresponding to the size at which they were tagged. Many also show a reddish flaring of pigment for 1 to 2 mm over the surface of the shell, but the more darkly pigmented rings are still visible through this flaring. The number of rings on the left valve of recaptured juveniles was counted under a dissecting microscope from the tag- ging "step" to the ventral edge of the shell along a line drawn from the umbo to the most distant part of the lip (shell height axis) (Figure 1). All counts were done by the 73 74 JOLL ^N ♦ TABLE 1. Release and recapture data for tagged Amusium balloti. GROUP 1 2 3 Date released November 1984 September 1985 November 1985 Date recaptured March 1985 November 1985 November 1985 Days at liberty 107-140 49-65 4-12 Number released 1000 157 500 Size range of releases (mm) 32-114 33-104 60-115 Number recaptured 67' 12 451 Size range of recaptures (mm) 83-108 58-100 60-115 Number suitable for ring analysis 36 8 6 Number for which exact recapture date known. Figure 1. Recaptured A. balloti with ink marlis highlighting rings along shell height axis. Note flaring of pigment at the tagging "step". Tagged: 7 November. 1984 — Size: 50 mm; Recaptured March. 1985 —Size 94 mm. (Mag xO.76) author. The rings, which are present as a reddish-brown pigmented hne approximately 0.15 to 0.25 mm wide on a pale background, were marked with a fine ink line to pre- vent mis-counting. RESULTS The pigmented rings of juvenile scallops are separated by up to 0.4 mm of unpigmented or weakly pigmented shell and are readily separated at low power under a dissecting microscope. However the outer band of rings of sexually mature scallops (i.e. scallops greater than about 90 to 95 mm shell height) are usually too closely packed to separate reliably. Also, animals tagged as immature scallops may reach maturity and have an outer band of rings which are too compacted to count if they are recaptured after long periods. Of the 67 recaptures of Group 1 animals with known numbers of days liberty (Table 1). only 36 had rings which were sufficiently separated to permit counting over the whole growth increment. Of these 36 animals, 35 were ju- veniles at the time of tagging, with initial sizes of 36-78 mm. One animal (initial size 93 mm) was mature at the time of tagging, but the rings in the growth increment of 8 mm could just be resolved. Most of the animals in which ring counts could not be done were mature animals, in which all rings in the growth increment were compacted. However, a few were juveniles which had matured during the time at liberty and had rings near the shell margin which were too compacted to resolve. With the 12 recaptures of group 2 animals, four scallops (initial sizes 88-96 mm) were mature animals and had rings which were too tightly packed to resolve and data were available only from eight juveniles (initial sizes 37-51 mm). Although there were 45 1 recaptures of group 3 animals (Table 1). all the larger animals (>85 mm initial size) showed no recognisable growth increment for the brief pe- riod (4 to 12 days) for which this group was at liberty. Most scallops of smaller initial size showed the fragmentary remnants of newly-formed shell around the shell margin, but only six individuals (initial sizes 60-73 mm) retained a sufficiently undamaged rim of this delicate new shell for analysis. The relationship between the number of pigmented rings from the tagging "step"" to the ventral margin of the shell and the number of days at liberty was very close to 1 : 1 (Figure 2). The correlation coefficient between the number no 160 NUMBER OF DAYS AT LIBERTY Figure 2. Number of rings in the growth increment of tagged A. bal- loti. Regression line is: Number of Rings Days). 1.0-1- 0.939 (Number of Daily Growth Rings in Amusium ballot/ 75 of pigmented rings and the number of days at liberty (0.996) was significant at the 0.1% level. However, in all cases there were marginally fewer rings than days and the discrepancy was greater and more variable for the Group 1 animals, which had long periods at liberty. This shortfall in the number of rings is considered to occur primarily as a result of the loss of shell from around the lip during the recapture of scallops by trawling. The thin shell of A. balloti is reinforced by internal radial ribs, but these do not reach right to the margins of the valves. As a consequence there is between 1 and 3 mm of un-rein- forced lip, depending on the size of the shell, which is only approximately 0.15 mm thick (in scallops of 40-50 mm shell height) to approximately 0.35 mm thick (in scallops of 90- 100 mm). The un-reinforced lip usually breaks off in a horseshoe shaped strip when scallops are spilled from the cod-end of the trawl. This was a particular problem with the recaptures of Group 1 animals from commercial vessels, where the cod-end contents can be very heavy. The fitted regression line — No. of Rings = -1.0 + 0.939 (No. of Days) (Figure 2) explained 99.1% of the sum of squares. There was, however, a significant partial correlation between the size at tagging and the number of rings after the number of days had been taken into account, with animals of large initial size having a greater difference between the number of days at liberty and the number of rings. This source of bias may result from (i) a greater loss of rings by large scallops nearing maturity in which the loss of the same amount of shell causes a greater loss of rings than for im- mature scallops with less compacted rings, or (ii) blurred or mis-counted rings in large scallops entering maturity. The rings visible on the left valve of A. balloti result from a band of pigmented laid down in the outermost shell layer. This layer (Figure 3) is composed of micro-crystal- line calcite approximately 0.05 mm thick (at about 45 mm from the umbo) to approximately 0.13 mm thick (at about 90 mm from the umbo), overlaying a crossed-lamellar middle layer (terminology of Taylor et al. 1969) of calcite. Banding is visible in this middle layer and results from the differential retlection of incident light, presumably as a re- sult of alternation in the direction of orientation of the crystal bundles. This banding is at a higher frequency than the pigmented rings and there was not any regular relation- ship between the banding in the middle layer and the pig- mented rings in the outer layer. The nature of the pigment forming the rings in the outer layer is unknown, but is cur- rently under investigation. DISCUSSION Clark (1974) noted from experiments on a variety of pectinids that ridges were formed nearly, but not always, daily and that the daily periodicity of ridge formations was ■ ^SP^-'^L 0.1 11111124.2 kU 6.55E2 2431/93 SE Figure 3. Electron micrograph of fracture along the shell height axis of .4. balloti A = microcrvstalline, calcitic outer layer; B = crossed- lamellar, calcitic middle layer. preserved only under the best conditions. Gruffydd (1981) examined the production of possible daily ridges by Pecten maximits under four light regimens and at two tempera- tures. He found that ridge production was not always daily and that it was particularly affected by temperature. There was, however, no consistent effect of the various photo- periods. He also found that the production of ridges around the margins of the shell varied with the size of the shell and that larger scallops produced more ridges over a given time period. The observation on the relationship between the number of rings in the growth increment of tagged A. balloti and the number of days at liberty comes from animals of a lim- ited initial size range growing under normal environmental conditions between September and March at one point in the species' range. These limitations are, however, dictated by the nature of the annual cycle of A. balloti in Shark Bay. New recruits (0+ age group, 30+ mm shell height), arising from a spawning season beginning in April, are first catchable in trawls around September of each year. Juve- niles grow rapidly over the period September to March, by which time the bulk of the recruits reach a size of 90 to 95 mm and begin to enter sexual maturity (Joll, unpub. data). Consequently the time of year when juveniles are available is limited. Even late recruits, which only achieve a size of around 70-75 mm by March, begin maturing and show closely packed rings from March onwards. In only 1 ma- ture animal was it possible to count the rings laid down in the growth increment. This animals was a late recruit and the rings, although distinguishable, were still very closely packed. The distribution of A. balloti in Western Australia is from Shark Bay to Esperance (Lat. 34°'S Long. 122°'E) and specimens from throughout this range all show a sim- ilar pattern of markings of the left valve, although there are local variations in the intensity of colour. It is not known if 76 JOLL the rings observed in shells from these other locations also reflect daily growth or whether local environmental condi- tions cause a variation in their frequency of deposition. Specimens from Hervey Bay, Queensland (25°00'S: Long 152°30"E) of the species reported as A. japonicum balloti (Bemardi) by Dredge 1981 and as A. japonicum balloti Habe by Williams and Dredge 1981 also show distinct rings in the juvenile parts of the shell. However, the rings are much darker and, in addition, there is some pigmenta- tion of the inter-ring spaces so that the overall effect is of a much more darkly coloured left valve. The reason for there being fewer rings than days is con- sidered to be most likely an artefact of post-capture han- dling, rather than a failure of the animals to lay down a ring on some days or a steady production of rings at a lower than daily frequency. The failure of the larger animals (>85 mm) from group 3 to grow any new shell over the short period that these animals were at liberty indicates that tagging may sometimes result in a brief cessation of growth in larger animals. However, nearly all the smaller anmials showed the fragmentary remains of new shell deposition indicating that, for the size range of scallops considered, tagging shock was probably not a major factor reducing the number of rings produced. The data indicate that the rings produced by juvenile A. balloti growing in the Shark Bay environment are of daily origin. Back- counting of rings from the outer margin of juveniles may, therefore, be used to determine the date of deposition of a ring. There will be some error in dating caused by the loss of shell from the lip if scallops are re- captured by trawling, but for rings within the range of the present data (i.e. within about 140 rings of the edge), a correction based on the regression could be applied. The distance between rings will give a record of the daily rate of growth on that date. Although there is some flaring and patchiness in the pigmentation in the first 5-10 mm of the shell, it is possible to count the number of rings on the left valve almost back to the umbo. On the assumption that the daily rhythm of ring formation is maintained in very small animals it would be possible to directly age juveniles by total ring counts. The close-packing of rings which occurs when animals mature and enter their reproductive season would prevent the direct ageing of older shells. However, the presence of this zone of closely-packed rings provides a means of distinguishing the immature recruits from the older year classes. ACKNOWLEDGMENTS The possibility that the rings represented daily growth was first suggested by S. Garcia. X-ray diffraction and thin sections to identify the composition of the various layers were done by B. Logan. J. Kuo assisted with the electron microscopy. Critical comment on the manuscript was given by J. Penn and N. Caputi. REFERENCES Barker, R. M., 1964. Microtextural variation in pelecypod shells. Muta- cologia 2:69-86. Broome, M. J. and Mason. J., 1978. Growth and spawning in the pectinid Chlamys opercularis in relation to temperature and phytoplankton concentration. Mar. Biol. 47:277-285. Clark, G. R. II, 1968. Mollusk shell: Daily growth lines Science 161:800-802. Clark, G. R. II, 1974. Periodic growth and biological rhythms in experi- mentally grown bivalves pp 103-117. In: Growth rhythms and the history of the earth's rotation, (eds. G. D. Rosenberg and S. K. Run- corn). John Wiley and Sons. London. Deith. M. R.. 1985. The composition of tidally deposited growth lines in the shell of the edible cockle, Cerasloderma ediile. J . Mar. Biol. Ass. U.K. 65:573-581. Dredge, M. C. L. 1981. Reproductive biology of the saucer scallop /Am»- slum japonicum balloti (Bemardi) in central Queensland waters. Aiisl. J. Mar. Freshwal. Res. 32:755-787. Gruffydd, LL. D., 1981. Observations on the rate of production of ex- ternal ridges on the shell of Pecten mcLximus in the laboratory. J. Mar. Biol. Ass. U.K. 61:401-411. Hahc. T., 1964. Notes on the species of the genus Amusiiim (Mollusca). Bull. Nat. Sci. Mus. Tokyo 7:1-5. Lutz. R. A. and Rhoads. D. C. 1980. Growth patterns within the mol- luscan shell. An overview, pp. 203-254 In: "Skeletal growth of aquatic organisms. Biological records of environmental change" Plenum. N.Y. Panella. G. and MacClintock. C, 1968. Biological and environmental rhythms reflected in molluscan shell growth. J. Paleont. 42(Mem. 2):64-80. Richardson. C. A., Crisp. D. J. and Runham, N. W., 1981. Factors in- Huencing shell deposition dunng a tidal cycle in the intenidal bivalve Cerastoderma edule. J. Mar. Biol. Ass. U.K. 61:465-476. Taylor, J. D. Kennedy, W. J. and Hall, A., 1969. The shell structure and mineralogy of the bivalvia. Introduction. Nuculacae-Trigonacae. Bull. Brit. Mus. Nat. Hist. (Zoot.) Suppl. 3:125 pp. Wheeler. A. P.. Blackwelder. P. H. and Wilbur. K. M.. 1975. Shell growth in the scallop Argopecten irradians. I. Isotope incorporation with reference to dirumal growth. Biol. Bull. 148:472-482. Williams. M. J. and Dredge, M. C. L., 1981. Growth of the saucer scallop Amusium japonicum balloti Habe in central eastern Queens- land. Aust. J . Mar. Freshwat. Res. 32:657-666. Journal of Shellfish Research. Vol, 7, No. 1, 77-82, 1988. SEASONAL CHANGES IN OXYGEN CONSUMPTION OF THE GIANT SCALLOP, PLACOPECTEN MAGELLANICVS (GMELIN) SANDRA E. SHUMWAY, JANEEN BARTER, AND JAMES STAHLNECKER State of Maine Department of Marine Resources West Booihbay Harbor. Maine 04575 ABSTRACT Rates of oxygen consumption (VO,) by the giant scallop. Placopecten magellanicus were measured monthly over a penod of fifteen months. In addition, scallops were acclimated to a senes of temperatures (Tj in the laboratory and the rates of oxygen consumption monitored. In acclimated animals, VO, increased with experimental temperatures with a concomitant decrease in Q,o value. Although the VO, of scallops from the field was consistently higher than values obtained from acclimated scallops at similar temperatures, the general trend was in keeping with rates which varied with the environmental temperature. It was shown that the seasonal changes in respiration rate are intimately related to changes in the gametogenic cycle with the highest rates exhibited during the summer months (npening of the gonads! and the lowest rates during the winter months. While the observed changes in metabolic rate generally follow the changes in environmental temperature, it is suggested that seasonal changes in food availability and reproductive stage have a greater affect on VO, than temperature per se. KEY WORDS: Scallop. Placopecten magellanicus. oxygen consumption INTRODUCTION The giant scallop, Placopecten magellanicus. represents one of the major fisheries in the Gulf of Maine. The species supports a large commercial fishery throughout its range and is currently considered as a prime species for aquacul- ture efforts. In spite of its economic importance, a recent review of the existing literature on this species (Shumway et al.. in preparation) and the development of a preliminary model for the giant scallop ecosystem by Campbell ( 1985) revealed a number of aspects of the biology of this species that are still in need of investigation. These information gaps severely limited our ability to accurately model the system and included such basic data as seasonal changes in fecundity, growth rate, food availability, respiration rates and the effects of temperature on these parameters. A number of studies were undertaken to fill these gaps and the results have been reported elsewhere (Langton et al.. 1987; Shumway et al., 1987; Schick et al., 1987; Barber et al., 1988; Schick etal., 1988a,b). There are few data available on respiration rates of scallops in general, or P. magellanicus in particular. Vahl (1978), Shafee (1982). Shafee and Lucas (1982). Barber and Blake (1985), MacDonald and Thompson (1986a.b) and Bricelj et al. (1987) have all reported on seasonal changes in the metabolic rate in various scallop species and their results are summarized in the Discussion. In a recent series of papers, MacDonald and his co- workers ( 1985a, b; 1986a, b; 1987) reported on the influ- ences of temperature and food availability on the ecological energetics of P. magellanicus from Newfoundland. In his studies, water depth was used as a model for variable food supply and temperature. Water depth per se was not of par- ticular interest, but rather the conditions at those depths. Although his station depths were only separated by approx- imately 20m, he was able to demonstrate marked differ- ences in growth rates, gamete production, reproductive ef- fort and other parameters as a result of differences in food and temperature. Our studies extend the depth factor (and associated environmental factors) considerably in that our stations range from approximately 20 to 180 m. Since the peculiarities of any given environment affect the fishery locally, it is important to establish a comprehen- sive data base for individual areas if fishery management is to be efficient. Further, before any major aquaculture ef- forts can be undertaken, it is essential to have a firm under- standing of the species' biology and the effects of varia- tions in environmental factors on their performance. The majority of energy losses, or 'costs of living" (Sibley and Calow, 1986) can be measured as heat losses or respiration rates. In the present paper, we report on the seasonal changes in respiration rate for P. magellanicus in coastal Maine waters. The study is part of an ongoing research pro- gram designed to establish such a data base for P. magel- lanicus in Maine waters and the subsequent production of a model to describe growth and spawning in this species. MATERIALS AND METHODS Specimens of the sea scallop, Placopecten magellanicus Gmelin, were collected by divers at a depth of approxi- mately 20m, from the lower Damariscotta River on a monthly basis between October 1984 and January 1986. Immediately after capture, the animals were transported to the laboratory, scrubbed free of epiphytes and maintained in running seawater from Boothbay Harbor prior to use in experiments. Vahl (1978) found that oxygen consumption in Chlamys islandica decreased during the first 20 days in the laboratory. Preliminary experiments indicated that no differences in rates of oxygen consumption (VO2) were ap- parent between the day of capture and up to 4 weeks after capture, as long as the temperature remained constant. 77 78 Shumway et al. Since it was our intention to monitor, as closely as pos- sible, the changes in VO2 under ambient conditions, mea- surements were made within 1-2 days of capture. Sea- water temperatures at the collection site were within 2°C of the seawater in the laboratory and all experiments were run at ambient temperatures. Each month, VO2 was determined on scallops of a wide size range (0.01- 18g dry tissue weight; approximately 10-130 mm shell height). The number of individuals measured varied (see Table 1). Dry tissue weights were obtained by oven drying to constant weight at 60°C for 24-48h. Rates of oxygen consumption were determined for indi- vidual scallops using a Radiometer oxygen electrode in a closed system (Taylor and Brand. 1975; Shumway, 1983). Preliminary experiments indicated that VOj was indepen- dent of oxygen tension (POj) only to approximately 70% saturation. Similar results have been shown for other species of scallops (van Dam, 1954; Vahl, 1978; MacKay and Shumway. 1980). Therefore, ambient PO2 was not al- lowed to drop below approximately 80% saturation. Sexes were not separated. Since the animals were freshly col- lected and VO2 measurements taken immediately, the rates reported for the seasonal study are assumed to represent a "routine' rate of oxygen consumption (see Bayne, 1976; Bayne and Newell. 1983). Results are expressed as least squares regression ac- cording to the formula: Y = aWb where Y is the predicted rate of oxygen consumption in ml oxygen hour"'. W is the dry tissue weight in g, a is the intercept and b is a constant. All regression and statistical TABLE 1. Parameters of the regression equations relating oxygen consumption (VO2; ml O2 • h"') to tissue dry weiglit (W; g) for Placopecten magellanicus. Data were fitted to the equation: VOj = aW''. Values preceded by an * are signincantly different (P < 0.05) from the previous value. Values of b are given ± s.e. DATE b a r2 n T(°C) 20 Oct 1984 0.838 + 0.029 0.363 0.992 19 10 27 Nov 1984 0.740 ± 0.039 *0.304 0.969 12 9 28 Dec 1984 0.714 ± 0.052 *0.220 0.949 12 6 29 Jan 1985 0.761 + 0.049 *0.069 0.964 11 1 26 Feb 1985 0.755 -»- 0.057 *0.196 0.951 11 4 29 Mar 1985 0.752 + 0.037 *0.283 0.976 12 5 25 Apr 1985 0.848 + 0.026 *0.259 0.990 13 8 25 May 1985 0.862 ± 0.062 *0.344 0.960 10 11 3 Jul 1985 0.837 + 0.053 *0.386 0.968 10 17 31 Jul 1985 0.837 -H 0.037 0.399 0.985 10 19 5 Sep 1985 0.820 ±_ 0.030 0.428 0.984 14 16 1 Oct 1985 0.831 ■+■ 0.104 *0.361 0.875 11 15 31 Oct 1985 0.814 -+- 0.039 *0.382 0.973 14 11 15 Nov 1985 0.736 ±^ 0.039 *0.281 0.962 16 9 10 Jan 1986 0.740 -t- 0.081 *0.162 0.903 9 3 analyses were carried out on an IBM 370 computer using SAS programs (SAS, 1985). For experiments to determine the effects of acclimation temperature on the scallops, animals were collected as above and maintained in ambient seawater and constant temperatures for 3 weeks prior to use in experiments. No food other than that which was available in the seawater supply was provided. For temperature acclimation which involved large changes in temperature, a step- wise series of acclimations was carried out whereby animals were main- tained at an intermediate temperature for at least a week prior to being subjected to the final temperature of acclima- tion. This procedure eliminated mortalities due to tempera- ture shock. Metabolic rates (routine) were measured as de- scribed above. RESULTS Figure 1 shows the weight-specific rates of oxygen con- sumption for scallops acclimated to a series of experimental temperatures. There was a steady increase in the rate of oxygen consumption with increasing temperature coupled with a decrease in the calculated Qiq value. These data have been used to calculate the 'expected' VOj at the various environmental temperatures (shown in Figure 2). With the exception of January, the observed rates of ox- ygen consumption were higher than the predicted values; however, the general trend was in keeping with rates that varied with the environmental temperature. Similar Qiq values for routine metabolic rate were reported for C. varia by Shafee (1982). The monthly rates of oxygen consumption by P. magel- lanicus are summarized in Table 1 . While the slopes of the lines are not significantly different as a group or individu- ally, significant differences (p < 0.01) between levels of VO2 were found. The seasonal changes in VO, are summa- rized in Figure 2 where it can be seen that the oxygen con- sumption rates generally followed the changes in environ- mental temperature. There were two major exceptions to this trend: 1 ) during the late winter and early spring where oxygen consumption rates increased at a much more rapid rate than would be predicted from the temperature changes alone and 2) during the late spring and early summer when rates of oxygen consumption remained fairly constant de- spite a fairly steep rise in environmental temperature. In addition, two significant decreases in VOj occurred be- tween March and April (when the temperature actually in- creased by 3°C) and between August and September (when the temperature only decreased by TC). Both of theses de- creases are significant (p < 0.05) and are indicated in Figures 2 and 3 by arrows. DISCUSSION Metabolic rate in scallops has been studied by few workers. Vahl (1978) monitored the changes in metabolic rate of the Iceland scallop, Chlamys islandica (O. F. Oxygen Consumption in Placopecten magellanicus 79 0.4 - .1 r- -'■■f _^ -- • '^^ I -'"' 1 y 1 ^cvj 0.2 _ QO2 = 0.0543 T°^^' r2=o.938 O • y y y i Qio l-IO 4.87 — y 5-15 2.38 CJ • O 10-20 1.58 O / 0.1 - y /■ T- ^ •? 1 1 1 1 10 15 ACCLIMATION TEMPERATURE (°C ) 20 Figure I. VOj/T, curve for Placopecten magellanicus over the range of normally experienced temperatures. Data are plotted for a standard scallop of 1 g dry tissue weight. Muller) throughout the year. In addition, he determined seasonal changes in the relationship between body size and respiration rate. His results showed that the rate of oxygen uptake increased rapidly in the beginning of April, reached maximum levels in late April and May and thereafter de- creased. This corresponds to the later part of the growth period of both shell and gonad in this species. He con- cluded that the seasonal variations in oxygen consumption exhibited by C . islandica were not explained by the sea- sonal variations in temperature but more likely were due to seasonal variability of the food supply. In addition, he showed significant seasonal variation in the dependence of oxygen consumption upon body weight. The exponent of the allometric equation relating VO2 to body weight for the period May-February was 0.78 while the exponent for the period March-April was approximately 0.90. Shafee (1982) reported a common slope of 0.72 for all seasons in the Black scallop. Chlamys varia. He demon- strated a marked seasonal fluctuation in respiration rate with highest values in August/September and lowest values during February/March corresponding to periods of high temperature/peak gonad development and low temperature and little or no gonad activity respectively. Similarly, Bri- celj et al. (1987) found that the oxygen consumption of the bay scallop, Argopecten irradicms irradians (Lamarck) closely paralleled seasonal changes in water temperature. They showed that temperature explained 93% of the sea- sonal variation in metabolic rate with minimum values re- corded in January/February and maximum values during June/July. In the only published account on P. magellanicus. Mac- Donald and Thompson (1986) measured seasonal changes in metabolic rate and found no significant differences in weight exponents (common slope = 0.89), but significant differences in metabolic rates between seasons. Measured rates were lowest during January-May and much higher during summer, June-September. Further, they found dif- ferences with depth in that rates for scallops from shallow water were highly correlated with water temperature whereas those from deeper water (31 m) were not. 80 Shumway et al. 05 - \ ^ 04 - ^^_^ '^ o 1 03 o •> 02 X- _ _ V ^\ \ / X Temperature ^^ X 1 I \ \ \ w / X 1 > II I 1 1 s X 111 II 20 -15 u cr 3 1- < CL 10:^ UJ -5 O'NDJ FMAM J J AS ONDJ (1984) (1985) (1986) Figure 2. Seasonal changes in oxygen consumption of Placopecten magellanicus. (• •) represents measured VOj (• •) expected rates of VO2 based on acclimation data (see Figure 1) and the environmental temperatures (x x). Data presented are for a standard animal of Ig dry tissue weight. Seasonal changes in metabolic rate reflect the various interactions between food availability, temperature, growth and reproductive activities. P. magellanicus. like other scallops (Mason. 1958: Ansell. 1974; Comely. 1974; Barber and Blake, 1981, 1983), exhibits a distinct annual reproductive cycle and as shown in Figure 3, the gameto- genic cycle and energy utilization are intimately related. This in turn affects metabolic rate (see Barber and Blake. 1985). Previous authors have studied the pattern of gonad de- velopment in P. magellanicus from various areas with con- trasting results. Thompson (1977) showed that reproductive development in Newfoundland begins in spring and the gonads mature in the summer. He further showed that en- ergy reserves from the previous year played no part in go- nadal development. In contrast, Robinson et al. (1981) found that gametogenesis began in December/January in animals from Boothbay Harbor, Maine and that gonad de- Figure 3. Seasonal changes in oxygen consumption of Placopecten magellanicus (from Figure 2) and approximate periods of the gametogenic cycle in the Gulf of Maine. Oxygen Consumption in Placopecten magellanicus 81 velopment takes place during January-March concurrently with somatic tissue growth. The energy reserves in the so- matic tissues were lost in late spring-summer after the mat- uration of the gametes, i.e. energy for gametogenesis comes from both the stored reserves and from the ingested ration unlike the more northern populations in Newfound- land where energy reserves from the previous year play no part in gonadal development. The gametogenic cycle for P . nuigeUanicus from Maine can be summarized as follows: During January, gameto- genesis has already reached the early developmental stage; energy reserves are at their lowest level (Robinson et al.. 1981), and energy must be mobilized from the accumulated reserves. During the spring, gametogenesis is underway, gonad size increases, feeding begins with coincidental spring phytoplankton blooms and energy reserves begin to accumulate. During the summer (June-August), food is plentiful, gametes are ripening and energy is derived from spring storage and from food intake (Robinson et al., 1981). Spawning takes place in September/October and the animals enter a reproductively quiescent or 'rest" period. Barber et al. ( 1988) found that primary oogenesis was initi- ated in February, secondary oogenesis in March and vitel- logenesis after June in P. magellanicus from Boothbay Harbor. Maine. Spawning and resorption of mature ova was evident in September and to a greater extent in October after which the animals were in a period of recovery (De- cember/January). In the present study, we have shown that respiration rates of P . magellanicus exhibited pronounced seasonal fluctuations which generally followed the changes in envi- ronmental temperature. While seasonal changes in meta- bolic activity are probably more closely related to food supply or reproductive activity than to temperature per se, in P. magellanicus there is a strong relationship between environmental temperature and seasonal changes in meta- bolic rate. These results are in general agreement with those of MacDonald and Thompson ( 1986a, b). Highest rates are exhibited during the summer months (ripening of the gonads) and lowest rates are exhibited during the winter months when gametogenesis is initiated. It is demonstrated here that seasonal variations in meta- bolic rate are intimately linked with the gametogenetic cycle as has been demonstrated for several other species of molluscs (see Bayne and Newell 1983 for review). This cycle is not strictly related to temperature. During the spring and summer months, glycogen is stored and during the autumn and winter months this energy store is utilized for metabolism (including gametogenesis, see Gabbott, 1975). In scallops, glycogen stored in the adductor muscle is the major energy substrate (Ansell, 1974; Barber and Blake, 1981; Robinson et al., 1981). The greatest discrepancy between the observed rates of respiration and those expected based on Q,o values oc- curred during February and March when there was a sudden increase in VO, coupled with only a slight increase in environmental temperature. This increase is most likely associated with the increased energy requirements of the scallop to fuel the sudden increase in gonad growth. This further suports the suggestion of Vahl and Sundet (1985) that the attainment of sexual maturity has an energetic cost. The observed VO2 from the end of January and the end of March is greater than would be expected based on Qjg values. Gonad development is an energy demanding pro- cess usually requiring mobilization of nutrients from in- gested food or the storage and subsequent utilization of re- serves from the body tissues. This period of time may cor- respond to the period of proliferation of the gonad/differentiation of gametes which is then followed, between April and August, by the less (?) energetically de- manding process of gamete ripening. The two 'unexpected' decreases in VOj observed during March/ April and August/September are of particular in- terest (see Figures 2,3). Similar decreases have been seen previously for this and other species. Ehinger (1978) con- cluded that reproductive stage had no effect on the rate of respiration. Closer examination of her data, however, re- veals that the trends in her study were similar to those re- ported here and in fact, the same decrease after spawning is reported in her thesis. Bricelj et al. (1987) noted a similar decrease in VO2 after spawning in A. irradians. Barber and Blake (1985) reported a similar decrease in VOt between mid- and late June (temperatures virtually the same) at about the time cytoplasmic growth of oocytes was initiated in A. irradians from Florida. He also noted a concomitant increase in RQ and O/N ratio both indicative of a shift to- ward greater carbohydrate utilization. March/April is the period of secondary oogenesis in P. magellanicus (Barber et al., 1988). It is possible that the initial decrease in VO2 observed in early spring by Barber and seen in this study is the result of the invocation of this metabolic machinery. The second decrease, corresponding to spawning, is ac- companied in bay scallops by a shift from carbohydrate ca- tabolism to protein catabolism (Barber and Blake 1985) and similar mechanisms are probably in effect in P. magel- lanicus It is difficult to separate the effect(s) of food, tempera- ture and reproductive stage on metabolic rate because they all vary simultaneously. Only when we have a clear knowl- edge of the seasonal changes in food availability and feeding strategies in this species can the allocation of en- ergy between somatic and gametogenic growth be clearly understood. ACKNOWLEDGEMENTS We are indebted to the Department of Marine Resources dive team for supplying scallops. Drs. B. Barber, B. Mac- Donald and D. Campbell provided helpful discussions and comments on an earlier version of the manuscript. 82 Shumway et al. REFERENCES CITED Ansell, A. D. 1974. Seasonal changes in biochemical composition of the bivalve Chlamys septemradiala from the Clyde Sea area. Mar. Biol. 25:85-99. Barber, B. J. & N. J. Blake 1981. Energy storage and utilization in rela- tion to gametogenesis in Argopecien irradians concemricus (Say). J . E.xp. Mar. Biol. Ecol. 52:121-134, Barber. B. J. & N. J. Blake 1983. Growth and reproduction of the bay scallop. Argopecten irradians (Lamarck), at its southern distributional limit. J. Exp. Mar. Biol. Ecol. 66:247-256. Barber. B. J. & N. J. Blake 1985. Substrate catabolism related to repro- duction in the bay scallop. Argopecten irradians concenlriciis. as de- termined by 0/N and R.Q. physiological indices. Mar. Biol. 87:13- 18. Barber. B. J.. R. Getchell. S. E. Shumway & D. Schick 1988. Reduced fecundity in a deep-water population of the giant scallop, Placopecten magellaniciis (Gmelm), in the Gulf of Maine, U.S.A. Mar. Ecol. Prog. Ser. (in press). Bayne. B. L. (ed) 1976. Marine Mussels: Their ecology and physiology. Cambridge University Press, London. 506 p. Bayne, B. L. & R. C. Newell 1983. Physiological energetics of marine molluscs. In: Saleuddin, A. S. M., Wilbur. K. M. (ed.) Academic Press, New York, 407-515 p. The Mollusca 4:407-515. Bricelj. V. M., J. Epp & R. E. Malouf 1987. Comparative physiology of young and old cohorts of the bay scallop, Argopecien irradians irra- dians (Lamarck): Mortality, growth and oxygen consumption. J. Exp. Mar. Biol. Ecol. 112:73-91. Campbell, D. E. 1985. Modelling the sea scallop ecosystem of the Gulf of Maine. State of Maine Dept. of Marine Resources. Res. Ref. Doc. 85/16. 50 p. Comely. C. A. 1974. Seasonal variations in the tlesh weights and bio- chemical content of the scallop Pecten maximus L. in the Clyde Sea area. J. Cons. Int. E.xplor. Mer. 35:281-295. Ehinger, R. E. 1978. Seasonal energy balance of the sea scallop. Placo- pecten magellanicus from Narragansett Bay. M. S. Thesis University of Rhode Island. 86 p. Gabbott. P. A. 1975. Storage cycles in marine bivalve molluscs: a hy- pothesis concerning the relationship between glycogen metabolism and gametogensis. Proceedings of the ninth European marine biology symposium. H. Barnes (ed.) Aberdeen University Press, Aberdeen. 191-211 p. Langton, R, W., W. E. Robinson & D. Schick 1987. Fecundity and re- productive effort of sea scallops Placopecten magellanicus from the Gulf of Maine. Mar. Ecol. Prog. Ser. 37:19-25. MacDonald, B. A. & R. J. Thompson 1985a. Influence of temperature and food availability on the ecological energetics of the giant scallop Placopecten magellanicus. I Growth rates of shell and somatic tissue. Mar. Ecol. Prog. Ser. 25:279-294. MacDonald, B. A. & R. J. Thompson 1985b. Influence of temperature and food availability on the ecological energetics of the giant scallop Placopecten magellanicus. II Reproductive output and total produc- tion. Mar. Ecol. Prog. Ser. 25:295-303. MacDonald, B. A. & R. J. Thompson 1986a. Influence of temperature and food availability on the energetics of the giant scallop Placopecten magellanicus. Ill Physiological ecology, the gametogenic cycle and scope of growth. Mar. Biol. 93:37-48. MacDonald, B. A. & R. J. Thompson 1986b. Production, dynamics and energy partitioning in two populations of the giant scallop Placopecten magellanicus (Gmelin). J. E.xp. Mar. Biol. 101:285-299. MacDonald, B. A., R. J. Thompson & B. L. Bayne 1987. Influence of food, temperature and food availability on the ecological energetics of the giant scallop Placopecten tnagellanicus. IV Reproductive effort, value and cost. Oecologia (Berlin):550-556. Mackay, J. & S. E. Shumway 1980. Factors affecting oxygen consump- tion in the scallop Chlamys delicatuta (Huttonl. Ophelia 19(0:19-26. Mason, J. 1958. The breeding of the scallop Pecten maximus (L.) in Manx waters. J. Mar. Biol. Assoc. U.K. 37:653-671. Robinson, W. E., W. E. Wehling. M. P. Morse & G. C. McCleod 1981. Seasonal changes in soft body component indices and energy reserves in the Atlantic deep-sea scallop, Placopecten magellanicus. Fish. Bull. 79(31:449-458. SAS Institute Inc. 1985. SAS User's Guide: Statistics. Version 5 Edition. Cary. N.C.: SAS Institute Inc. 956 p. Schick. D. F.. S. E. Shumway & M. Hunter 1987. A comparison of growth rate between shallow water and deep water populations of the scallops Placopecten magellanicus (Gmelin, 1791) in the Gulf of Maine. American Malacological Bulletin 6:1-8. Schick. D. F.. S. E. Shumway & M. Hunter 1988a. Allometric relation- ships and growth in Placopecten magellanicus: The effects of season and growth. Unitas Malacologia (In press). Schick. D. F., S. E. Shumway & M. Hunter 1988b. Seasonal changes of allometric relationships in two populations of the scallop, Placopecten magellanicus (Gmelin) in the Gulf of Maine. Submitted. Sibley, R, M. & P. Calow 1986. Physiological Ecology. An Evolutionary Approach. Blackwell Scientific Publications, Oxford 179 p. Shaffee, M. S. 1982. Variations saisonnieres de la consommation d'oxy- gene chez petoncle noir Chlamys varia (L.) de Lanveoc (rade de Brest). Oceanologica acta 5:189-197. Shafee. M. S. & A. Lucas 1982. Variations saisonnieres du bilan energe- tique chez les individus d'une population de Chlamys varia (L.): Bi- valvia. Pectinidae. Oceanologica acta 5:331-338. Shumway. S. E. 1983. Factors affecting oxygen consumption in the coot clam Mulinia lateralis (Say). Ophelia 22:143- 171. Shumway, S. E., R. Selvin & D. F. Schick 1987. Food resources related to habitat in the scallop Placopecten magellanicus (Gmelin. 1791). A qualitative study. J. Shellfish Res. 6:89-95. Shumway, S. E., S. K. Naidu & D. F. Schick 1988. A synopsis of avail- able data on the giant scallop, Placopecten magellanicus. (in prep). Taylor, A.C. & A. R, Brand 1975. Effects of hypoxia and body size on oxygen consumption of the bivalve Artica islaiulica (L). J. Exp. Mar. Biol. Ecol. 19:187-196. Thompson, R. J, 1977. Blood chemistry, biochemical composition and the annual reproductive cycle in the giant scallop, Placopecten magel- lanicus. from southeast Newfoundland. J. Fish. Res. Board Can. 34:2104-2116. Vahl. O. 1978. Seasonal changes in oxygen consumption of the Icelandic scallop Chlamys islandica (O. F. Mullcr) from 70°N. Ophelia 17(11:143-154. Vahl. O. & J. H. Sundet 1985. Is sperm really so cheap? Marine Biology of Polar Regions and Effects of Stress on Marine Organisms. Gray, J. S. and M. E. Christiansen (ed.) John Wiley and Sons. N.Y. 281-285 p. van Dam, L. 1954. On the respiration in scallops (Lamellibranchia). Biol. Bull: 107:194-202. Journal of Shellfish Reseunh. Vol. 7, No. I, 83-88, 1988. BIOLOGICAL FEASIBILITY OF GROWING THE NORTHERN BAY SCALLOP, ARGOPECTEN IRRADIANS IRRADIANS (LAMARCK, 1819), IN COASTAL WATERS OF GEORGIA PETER B. HEFFERNAN, RAnDAL L. WALKER AND DAVID M. GILLESPIE Marine Extension Service University of Georgia P.O. Bo.x 13687 Savannah, Georgia 31416-0687 ABSTRACT Two studies were carried out to evaluate ihe potential of pearl net cultivation of non-native (northern) Bay Scallops, Argopeclen irrudians irradians (Lamarck, 1819). in the coastal waters of Georgia. In October 1984. scallops (x = 9.8 mm ± 0.21 S.E.) were placed in pearl nets suspended from a floating raft in House Creek. Little Tybee Island. Georgia, at a density of 70/net. By May 1985. scallops averaged 43.2 mm ± 2.3 S.E. in shell length with 27% larger than 50 mm (commercial size) and 47% survival. In a second study (October 1985-June 1986). density dependent and site specific effects on scallop (initial size = 6.5 mm ± O.I S.E.) growth were evaluated between an "exposed" site (Priest Landing) and a "sheltered" site (House Creek). Growth at the "sheltered" site was significantly greater at similar stocking densities (lOO/net; 200/net) and also at higher stocking densities than at the "exposed" site. By June 1986. the House Creek lOO/net treatment showed mean shell length of 37.6 mm with 68.2% survival. At 200/net mean shell length was 37.9 mm with 48.5% survival. At the exposed site, growth was significantly greater in nets placed <0.3 m below the surface (x = 36. U mm at lOO/net) as opposed to nets at greater depths (3 ml (x = 33.3 mm at lOO/net and 32.2 mm at 200/net). Survival rates at Pnest Landing ranged from 36.5 to 64.2% in June 1986. None of the scallops in the 1985-86 experi- ments had reached market size by June. These results show that non-native Bay Scallops can grow and survive in the coastal waters of Georgia with acceptable survival between October and May -early June, Although smaller northern bay scallop seed {ca. 10 mm) can grow to a minimal commercial size with acceptable survival when grown from fall to spring in the coastal waters of Georgia, it is doubtful that this subspecies has a maricultural future for Georgia fishermen. To ensure adequate growth and survival, nets must be cleaned monthly to control fouling organisms. This constant cleaning process and the fact that scallops only grow to a minima! marketable size will prevent the farming of this subspecies in coastal Georgia. KEY WORDS: Scallop mariculture, Argopetten irradians. peari net cultivation INTRODUCTION protected bays where seagrasses or seaweeds provide a nat- ural refuge for the animal (Dreyer and Castle 1941; Thayer The bay scallop, Argopecien irradians irradians (La- and Stuart 1974; Orth 1977; Stoner 1980; Eckman 1987). marck 1819). is an important commercial species of scallop Argopecten irradians (Lamarck 1819) is divided into two along the Atlantic and Gulf coasts of the United States, subspecies: Argopecten irradians irradians (Lamarck with the fishery centered in the New England states and on 1819) which occurs from Cape Cod to New Jersey and Ar- the Gulf coast of Florida. In 1985, 0.735 million pounds of gopecten irradians concentricus (Say 1822) which occurs scallop meat were landed in the United States (National from Maryland to Georgia and Louisiana to Tampa, Florida Marine Fisheries Service 1986). During peak season in (Abbott 1974). Although Georgia has approximately '/a of 1985, bay scallops from northern areas (Long Island and the salt marsh acreage along the Atlantic coast, bay Massachusetts) fetched $6.66/lb. while those from North scallops do not occur there, probably due to the lack of Carolina commanded $5.69/lb.. according to the green submerged vegetation (personal observations). The absence sheet (New York market reports). Dockside values of of the bay scallop is regrettable since it is an ideal species $3.50/lb. were common during peak season while a 35Vc which commands high prices (Castagna and Duggan price reduction occurred during the marketing of the calico 1971a. 1971b). scallop harvest (anonymous reviewer). The life span of the bay scallop is 1 .5 to slightly over 2 Due to its commercial and ecological significance, the years (Marshall 1963; Taylor and Capuzzo 1983). The time bay scallop has traditionally received a lot of attention from required for bay scallops to reach marketable size (50 mm) researchers. While only those works with direct relevance decreases with decrease in latitudes (Castagna and Duggan to the present study are cited herein, the readers are referred 1971b). Hard clams, Mercenaria mercenaria (L.), from to the excellent reviews of bay scallop literature contained natural populations in coastal Georgia obtain marketable in the works of Broom (1976). Robert (1978), and Fay et size in 2 to 3 years (Walker 1987); whereas, 4 to 5 years of al. (1983). growth are required in the Long Island Sound, New York The distribution of the bay scallop ranges from New En- area (Greene 1978). Northern hard clam seed (x = 6 mm gland to the Gulf of Mexico (Abbott 1974), primarily in in shell length) obtained marketable size in 18 months 83 84 Heffernan et al. when grown in the coastal waters of Georgia (Walker 1984). Thus, it may be possible that growth to a marketable size for the northern bay scallops might take less time in the warmer Georgia waters than in their native northern waters. As part of an ongoing Mariculture Development Pro- gram at the University of Georgia, a wide variety of poten- tial mariculture species are being studied. The initial inves- tigation involves testing the biological feasibility of the species in coastal Georgia. If this is deemed acceptable, an economic feasibility study is then conducted. The purpose of the research reported here was to deter- mine if the northern bay scallop, a non-native species, can grow and survive in the coastal waters of Georgia, and to evaluate the potential fishery that might be developed for this species in Georgia. Ease of access to seed of northern bay scallops dictated that they be investigated at this stage in preference to the southern bay scallop whose seed is not readily available. MATERIAL AND METHODS 1984-1985 Experiment at House Creek: Bay scallops were obtained from Bristol Shellfish Farms. Round Point. Maine. They were set up at a mean shell length of 9.8 mm ± 0.2 S.E. in 1 1 pearl nets (0.3 x 0.3 m) suspended (0.3 m depth) from a floating raft at a density of 70/net on 21 October 1984. The raft was an- chored within a small tidal creek at Little Tybee Island (Figure 1). The nets were checked monthly, cleaned ot fouling organisms and the shell length of at least 200 scallops (except for October and November) was measured using vernier calipers. There were a considerable number Figure 1. Map of the study area indicating the exposed (Priest Landing) and sheltered (House Creek) sites in which pearl net cultivation of Argopecten irradians irradians was carried out. Bay Scallop Cultivation in Georgia 85 of deformed seed present from the onset of the study (see Results). 1985-1986 Experiment at House Creek and Priest Landing: Scallops were shipped from the Aquaculture Research Corporation. Dennis, Massachusetts. They were set up in density replicates (100/net and 200/net) in pearl nets (0.3 X 0.3 m) suspended (0.3 m depth) from floating rafts at Priest Landing and House Creek on 9 October 1985 at a mean shell length of 6.5 mm ± 0.1 S.E. The Priest Landing experiment also contained a trial designed to eval- uate the effect of depth on growth, with a suite of nets (n = 6; 3 at 100/net. 3 at 200/net) suspended ca. 0.3 m below the surface and another at ca. 3 m depth (n = 6; 3 at 100/ net and 3 at 200/net). House Creek replicates (n = 12; 5 at 100/net and 7 at 200/net) were all at a depth ca. 0.3 m. Nets were examined, with total counts and measure- ments of all scallops carried out on January 14- 15. March 30-April 1 and June 4-5. 1986. Fouling organisms were removed and/or nets were replaced during each sampling season. Results were compared statistically using student t-tests (at 95% confidence level). RESULTS 1984-1985 Experiment at House Creek: From 21 October 1984 to 13 June 1985 scallops grew from a mean shell length of 9.8 ± 0.2 mm to 49.0 ± 3.4 mm in pearl nets suspended from a floating raft (Table 1 and Figure 2). Although scallop growth slowed during Jan- uary and February, it was continuous throughout the study period. Large sample errors were due to difference in indi- vidual growth rates and shell deformities. For instance, in April scallops without deformities ranged from 12.9 mm to 61.6 mm in shell length (x = 43.3 ± 0.6 mm). Deformed scallops, accounting for 21% of the animals sampled, ranged from 12.9 to 54.6 mm in shell length (x = 23.8 ± 0.9 mm). By June 13. 1985, only 3% of the scallops remained alive. Water temperatures rose from 26°C (June 1) to 29.7°C during a heat wave in the first week of June. By June 13, water temperatures dropped to 25°C (as recorded at the Marine Extension Service Skidaway Island dock). However, on May 16, survival was at 47%. with a mean shell length value at 43.3 mm (±2.3) (Figure 2). By this time, 27% of the scallops sampled were marketable size (Table 1). The daily growth rate from October 21 to May 26 (206 days) was 0. 162 mm/day. 1985-1986 Experiments at House Creek and Priest Landing: By January 1986. site specific significant differences were evident between treatments at Priest Landing and House Creek. Scallop growth was significantly greater in the House Creek 100/net (HlOO) treatment (x = 24.9 mm) than in the "surface" (<0.3 m depth) treatment with 100/ net at Priest Landing (PSIOO) (x = 20.9 mm) (see Tables 2 and 3). There were no significant differences in growth be- tween other House Creek and Priest Landing treatments by January. There were no density dependent significant dif- ferences in growth between House Creek treatments. Priest Landing 200/net "bottom" {ca. 3 m depth) treatment PB200 (X = 23.2 mm) and the 100/net "bottom" (PBIOO) (X = 24.8 mm) treatment both displayed significantly greater growth than the surface treatment (PSIOO, x = 20.9 mm) (Table 2). Mean survival rates were all greater than 90% (90.8-98.8%) during January (see Figure 3). By March 31 -April 1 there were site specific and den- sity dependent significant differences in growth rates ob- served between the various treatments. The House Creek treatments were significantly greater in mean size than their counterparts at Priest Landing, e.g., HIOO (x = 36.1 mm) > PBIOO (X = 29.1 mm) > PSIOO (x = 28.9 mm) and H200 (X = 32.9 mm) > PB200 (x = 26.7 mm) (Tables 2 and 3. Figure 2). The H200 treatment was also significantly greater in mean size than the PBIOO and PSIOO treatments. TABLE 1. Growth of Bay Scallops, Argopecten irradians irradians, in the coastal waters of Georgia. Number of Scallops Shell Length Percentage of measured for in Scallops Percentage Date shell length mm ± S.E. >50 mm Survival 21 October 1984 57 9.8 ± 0.2 — 17 November 1984 59 14.0 ± 0.7 — 9 December 1984 272 24.7 ± 0.5 — 18 January 1985 368 34.4 ± 1.8 0.3 — 18 February 1985 257 38.6 ± 2.4 7.8 — 18 March 1985 477* 37.7 ± 1.7 15.9 62,0 18 Apnl 1985 462* 39.0 ± 1.8 21.4 60.0 13 May 1985 361* 43.3 ± 2.3 26.6 47.0 1 3 June 1985 24* 49,0 ± 3,4 56,0 3,1 Total count and measurement of scallops from all pearl nets. 86 Heffernan et al. 70/nel 1984-85 PS 100 1985-86 PB100 1985-86 ■-PB200 1985 86 i-H100 1985-86 '- H2G0 1985-86 Figure 2. Growth of Argopeclen irradians irradians in various density treatments during October-June 1984-85 and 1985-86. PS = Priest Landing Surface (0.3 m), PB = Priest Landing Bottom [ca. 3 m), H = House Creek, e.g. HlOO = House Creek 100/net etc. Density dependent significant differences in growth were observed at House Creek, with HlOO greater than H200 (see Table 3). Similarly PBIOO was significantly greater than PB200 in observed growth while the differences be- tween PSIOO and PBIOO were not significant. Survival rates ranged from 69.3% (PB200) to 87% (PBIOO) on March 31 -April 1 (see Figure 3). At the termination of the study on June 4-5, 1986 the mean growth of the House Creek treatments was signifi- cantly greater than any of the Priest Landing treatments (see Tables 2 and 3). Mean sizes at House Creek varied from 34.4 mm -40.1 mm (HlOO) and 35-40.9 mm (H200) while those at Priest Landing ranged from 32.2 mm (PB200) to 33.8 mm (PBIOO) and 34-37.7 mm (PSIOO) (Tables 2 and 3). There were no density dependent signifi- cant differences in scallop growth among treatments at House Creek or on the bottom at Priest Landing (i.e.. PBIOO and PB200). The surface treatment at Priest Landing (PSIOO, x = 36.0 mm) showed significantly greater growth than the bottom treatment (PBIOO, x = 33.3 mm) (see Figure 2). Survival rates by 4-5 June varied from 36.5% (PBIOO) to 68.2% (HlOO). Daily growth rates over the study period (238 and 239 days at Priest Landing and House Creek, respectively) are presented in Table 4. The H200 treatment exhibited the fastest growth rate with 0.132 mm being added to shell length daily (Table 4). The observed bay scallop growth rates are not sufficient to support this species as a good mariculture candidate for Georgia. Furthermore, the labor intense nature of pearl net cultivation for bay scallops will also rule out this grow-out system for mariculturists in Georgia. DISCUSSION The results presented here show that bay scallops can survive and be grown to a minimum commercial size in the coastal waters of Georgia. The growth rates reported here are comparable to those published from other areas of the eastern U.S. In Georgia scallops grew from 9.8 mm to 49 mm (1984/5) and from 6.5 mm to 33.3-37.9 mm (1985/6) in ca. 8 months. In Massachusetts, 12 to 17 months are required for scallops to reach 50 mm (Belding 1910), whereas, in North Carolina, 10 months are required (Gut- sell 1928). In Virginia seed scallops grew from 12.7 mm to 57 mm in 6 to 7 months (Castagna and Duggan 1971a) and from 14.4 mm to 47 mm in 4 months (Duggan 1973). However, growth patterns are quite different in Georgia than in northern states. Growth in Georgia was continuous throughout the winter months (Figure 2), whereas growth frequently ceases during this period in northern states, e.g., scallop growth stops below 7°C in Massachusetts (Belding 1910). Since 1958 Georgia coastal water temperatures have dropped below 7°C during only 10 winters, with the TABLE 2. Argopeclen irradians irradians pearl net growth experiments at Priest Landing October 1985 — June 1986. 9 October 15 January 86 1 April 86 5 June 86 Initial Size (mm) Size Size Size Density (± S.E.) (±S.E.) % Surv. (±S.E.) % Surv. (±S.E.) % Surv. Surface (<0.3 m) too 6.5 ± 0.1 21.1 ± 0.9 100 31.3 ± 1.0 87 37.7 ± 0.9 70 100 6.5 -t- 0.1 19.7 ± 0.8 59 29.3 ± 1.0 55 35.5 ±1.1 49 100 6.5 -H 0.1 19.4 ± 1.4 99 29.1 ± 0.9 87 35.0 ± 0.8 70 100 6.5 + 0.1 22.7 ± 1.2 100 30.2 ± 0.8 97 37.7 ± 0.9 72 100 6.5 ± 0.1 21.7 ± 1.2 96 21.4 ± 0.6 66 34.0 ± 0.8 60 100 6.5 ± 0.1 Lost — — — — — Subtidal 100 6.5 ± 0.1 24.8 ± 1.0 93 29.2 ± 0.7 100 32.3 ± 0.9 27 100 6.5 -h 0.1 24.1 ± 1.2 97 28.0 ± 0.7 77 33.8 ± 0.7 46 100 6.5 ^ 0.1 25.7 ± 1.4 89 29.3 ± 0.7 84 Lost 200 6.5 ± 0.1 23.7 ±1.1 98.5 26.2 ± 0.5 62.5 32.2 ± 0.5 58 200 6.5 -t- 0.1 22.7 ± 1.0 98 26.9 ± 0.5 84.5 Lost 200 6.5 ± 0.1 23.1 ± 0.8 100 26.6 ± 0.5 61 Lost Bay Scallop Cultivation in Georgia 87 TABLE 3. Argopecten irradians irradians pearl net growth experiments at House Creek October 1985-June 1986. (All subtidal). 9 October 85 14 .lanuarv 86 I April 86 5 June 86 Initial Size (mm) Size (mm) Size (mm) Size (mm) Density (± S.E.) (± S.E.) % Surv. (±S.E.) % Surv. (±S.E.) % Surv. 100 6.5 + 0.1 26.4 ± 0.8 97 .^5.7 ± 1.0 90 40.1 ± 0.5 78 100 6.5 ± 0.1 22.9 + 0.9 100 35.3 ± 0.9 97 38.7 ± 0.7 73 100 6.5 -*- 0.1 24.4 ± 0.9 100 38.6 ± 0.9 40 38.4 ± 1.0 51 100 6.5 ± 0.1 26.8 ■+ 0.8 93 36.1 ± 0.8 85 36.9 ± 0.6 68 100 6.5 ± 0.1 24.0 -t- 1.0 96 35.2 ± 1.0 ') 34.4 ± 0.7 71 200 6.5 -(- 0.1 23.5 ± 1.0 86 34.4 ± 1.0 65.5 40.9 ± 0.6 43 200 6.5 + 0.1 26.4 ± 1.0 97.5 34.6 ± 1.0 77 37.8 ± 0.6 45.5 200 6.5 ± 0.1 25.1 + 0.8 97 31.2 ± 0.9 78.5 39.0 ± 0.6 43.5 200 6.5 + 0.1 22.4 ± 0.8 99 32.7 ± 1.0 71.5 38.9 ± 0.7 43 200 6.5 -t- 0.1 24.5 m 0.8 97.5 Lost — — — 200 6.5 ± 0.1 22.9 ± 0.7 90 5 .■^1,3 ± 1.1 82.5 35.0 ± 0.5 67.5 200 6.5 ± 0.1 24.7 ± 0.7 97.5 Lost — — — '"drop" in temperatures usually of less than l-week"s dura- tion. The two experiments reported here indicate commer- cially acceptable survival rates (i.e., 47-68%) using pearl net cultivation as long as northern bay scallop harvesting is completed before lethal water temperatures (ca. 26°C) are reached in June. As demonstrated in June 1985, very high mortalities occur at water temperatures in excess of 26°C. Site characteristics have been shown to have a profound effect on the growth of scallops during this study. The sheltered creek site (House Creek) exhibited significantly greater scallop growth rates than the exposed site (Priest Landing). The creek location obviously afforded a lot of shelter from natural elements as well as commercial and recreational boat traffic disturbances, with consequent ben- eficial effects on observed srowth rates. This is in agree- lUU — ,^,^ •-PS100 " ^^^~~-q O-PB100 80- ^^v^\ ■ -PB200 ^^^i^C^^^^ * -H200 CO > > 60-^ 40 — \ ss 20 — 1 1 1 January 13^14 March-April 31-1 June 4-5 Figure 3. Mean survival rates for Argopecten irradians irradians pearl net cultivation experiments at Priest Landing and House Creek Oc- tober 1985-June 1986. PS = Priest Landing Surface, PB = Priest Landing Bottom, H = House Creek, e.g. HlOO = House Creek 100/ net. ment with the findings of Kirby-Smith (1972). where a fast water-tlow rate was shown to adversely affect bay scallop growth. The raft with scallops at House Creek sits in a pool of water (0.6 to 1 m deep) at low tide. Most of the creek is drained at low tide while scallops at Priest Landing are an- chored in a major river with 6 meters of water below the raft at low tide. In the absence of water temperature and current velocity data pertaining to the two sites, one can only speculate that temperatures fluctuated greater in the shallower sheltered creek and water movement was higher at the exposed site on Wilmington River. The significantly superior growth of scallops in the exposed sites "surface" treatment as opposed to the "bottom" treatment contrasts with the findings of Duggan (1973), where growth was equal throughout the water column. The higher mortalities suffered in the "bottom" nets agrees with the results of Duggan ( 1973). The lower growth and survival of scallops reared in "bottom" pearl nets may be related to varying pressures exerted by fouling organisms. The "surface" nets were fouled mainly by Molgula (sea squirt), oyster spat, barnacles and branching bryozoans while the "bottom" nets were predominantly fouled by encrusting bryozoans. There was no clear relationship between sur- vival and site selection and/or density treatments. table 4. Mean daily growth rates for Argopecten irradians irradians reaed in pearl nets in two sites in coastal Georgia October 1985-June 1986. Site/Densitv Growth Rate mm/day Priest Landing (Exposed) 100/net (Surface) 100/net (Bottom) 200/net (Bottom) House Creek (Sheltered) 100/net (Bottom) 200/net (Bottom! 0.124 0.112 0.108 0.130 0.132 88 Heffernan et al. Growth rates of northern bay scallops cultivated in pearl nets during periods of amicable water temperatures in coastal Georgia do not suggest a bright future for this grow-out system for the southern mariculturist. The ob- served growth rates (0.162 mm/day) at a stocking density of 70/net (9.8 mm seed) yielded barely minimum commer- cial size scallops within a 206 day growing season. Use of larger seed (15-25 mm) and lower initial stocking densities (30-70/net) could yield bigger scallops, but the extra cost of larger seed and additional nets may be offsetting. Fur- thermore, the labor input necessary to maintain relatively unfouled nets in open seawater systems may prove prohibi- tive for any commercial undertaking in Georgia. Consider- able attention is currently being devoted to the interstate movement of shellfish by state and federal legislators. Fu- ture legislation prohibiting and/or limiting the movement of shellfish resources due to the threat of disease, is a distinct possibility. Such prohibitive legislation could rule out such overwintering programs as those discussed here. ACKNOWLEDGMENTS This work was funded by the Georgia Sea Grant Pro- gram under grant number NA84AA-D00072. Dr. J. Cren- shaw, Jr. is thanked for his many contributions. Mr. J. Carr, G. Paulk and D. Jacobi are thanked for their assis- tance in field work. Mrs. J. Haley is thanked for typing the manuscript. REFERENCES CITED Abbott. R. T. 1974. American Seashells. Second Edition. Van Nostrand Reinhold. New York. 663 pp. Belding, D. L. 1910. A report upon the scallop fishery of Massachusetts, including the habits, life history of Argopeclen irradians. its rate of growth and other factors of economic value. Massachusetts Comm. Fish. Game. Spec. Repl. 150 pp. Broom, M. J. 1976. Synopsis of biological data on scallops. FAO Fish Synopsis No. 114 (FIRS/5 114). 43 pp. Castagna, M. A. and W. P. Duggan. 1971a. Spawning and rearing the bay scallop VIMS Laboratory Method. Virginia Institute of Marine Sciences. Sea Grant Advisory Service Project No. 5. 3 pp. Castagna, M. A. and W. P. Duggan. 1971b. Reanng the bay scallop. Argopeclen irradians. Proc. Natl. Shellfish. Assoc. 61:80-85. Dreyer, W. A. and W. A. Castle. 1941. Occurrence of the bay scallop, Argopecten irradians. Ecology 22:425-427. Duggan, W. P. 1973. Growth and survival of the bay scallop, Argopecten irradians. at various locations in the water column and at various den- sities. Proc. Natl. Shellfish. Assoc. 63:68-71. Eckman, J. E. 1987. The role of hydrodynamics in recruitment, growth and survival of Argopeclen irradians (L.I and Anomia simplex (D'Or- bigny) within eelgrass meadows. J. Exp. Mar. Biol. Ecol. 106:165- 191. Fay, C. W., R. J. Neves, and G. B. Pardue. 1983. Species profiles: life histories and environmental requirements of coastal fishes and inverte- brates (Mid-Atlantic)-bay scallop. U.S. Fish and Wildlife Service, Division of Biological Services, FWS/OBS-82/1 1.12. U.S. Arniy Corps of Engineers, TR EL:-82-4. 17 pp. Greene, G. T. 1978. Population structure, growth, and mortality of hard clams at selected locations in Great South Bay. New York. Masters Thesis, State University of New York at Stony Brook. 199 pp. Outsell, J. S. 1928. Scallop industry of North Carolina. Rep. U.S. Comm. Fish, for 1928. Append. 5, pp. 173-197. Kirby-Smith, M. W. 1972. Growth of the bay scallop: the influence of experimental currents. J . E.xp. Mar. Biol. Ecol. 8:7- 18. Marshall. N. 1963. Mortality rates and the life span of the bay scallop, Argopecten irradians. Proc. Nail. Shellfish. Assoc. 54:87-92. National Marine Fisheries Service. 1986. Fisheries of the United States, 1986. U.S. Department of Commerce, NOAA. National Manne Fish- eries Service, Washington, DC. 119 pp. Orth. R. J. 1977. The importance of sediment stability in eelgrass com- munities. In: Ecology of marine benthos edited by B. C. Coull, Uni- versity of South Carolina Press. Columbia, pp. 281 -300. Robert, G. 1978. Biological assessment of the bay scallop [Argopecten irradians) for maritime waters. Can. Fish. Mar. Sen'. Tech. Rep. No. 778. 13 pp. Stoner, A. W. 1980. The role of seagrass biomass in the organization of benthic macrofaunal assemblages. Bull. Mar. Sci. 30:537-551. Taylor, R. E. and J. M. Capuzzo. 1983. The reproduction cycle of the bay scallop, Argopecten irradians (Lamarck), in a small coastal em- bayment on Cape Cod, Massachusetts. Estuaries 6:431-435. Thayer, G. W. and H. H. Stuart. 1974. The bay scallop makes its bed of seagrass. Mar. Fish. Rev. 36:27-30. Walker, R. L. 1984. Effects of density and sampling time on the growth of the hard clam, Mercenaria mercenaria. planted in predator-free cages in coastal Georgia. The Nautilus 98:1 14- 1 19. Walker, R. L. 1987. Hard clam, Mercenaria mercenaria (Linne), popula- tions of coastal Georgia. Georgia Marine Science Center Technical Rept. Ser.. No. 87-1. 73 pp. Journal of Shellfish Research. Vol. 7, No. 1. 89-94, 1988. BACTERIAL DEPURATION BY THE HARD CLAM, MERCENARIA MERCENARIA LISA JOHNSON AND STEVEN HAYASAKA Department of Microbiology Clemson University Clemson. SC 29634-1909 ABSTRACT Bacterial loads varied in hard clams obtained from a South Carolina estuary with lower counts dunng cooler months and higher counts during warmer months; the water source from which clams were taken was considered unpolluted as low levels of fecal conforms and no Salmonella or Shigella spp. were detected. Vibrio parahaeinolylicus was detected only in low numbers in hard clams. Two important parameters considered m depuration were salinity and temperature. Clams raised in South Carolina coastal waters provided good depuration results within 24 h at a temperature close to 24°C and salinities of 24-31%c. No apparent "thermal shock" was observed when clams were depurated at higher than ambient temperatures; "cold shock" was observed when clams acclimated at higher temperatures were depurated at lower temperatures. Clams preloaded with bacteria at 30°C (salinity of 26.5%(j) required 24-48 h of depuration at 20°C. KEY WORDS: Bactenal depuration, hard clam. Mercenaria mercenaria INTRODUCTION The hard clam. Mercenaria mercenaria, is a natural re- source along the South Carolina coast. However, the avail- ability of edible shellfish throughout the year is dependent upon the bacterial content of their environment and of the mollusks themselves. Along the coast of South Carolina, sewage outfalls are major sources of pollution; an increase in summer visitors also presents a public health problem as waste disposal needs increase. High coliform counts, indic- ative of sewage pollution, may be attributed to stomi water run-off from residential areas and shopping mall parking lots, livestock, domestic fowl, migratory water fowl and/or heavy rains which contribute to overloaded sewage treat- ment facilities. As the temperature increases during warmer months, total bacterial content and fecal coliform counts increase. Once fecal coliform levels rise above the guide- lines set forth by the National Shellfish Sanitation Program (NSSP), the shellfish are no longer safe for human con- sumption. At this time, a major resource and food source is unavailable. Shellfish purification has been studied for clams (Can- telmo and Carter 1984. Hartland and Timoney 1979. Mac- Millan and Redman 1971. Ritchie 1976, and Timoney and Abston 1984). oysters (Eyles and Davey 1984, Huntley and Hammerstrom 1971. Janssen 1974. and Rowse and Fleet 1984) and mussels (Houser 1964 and Ledo et al. 1983) and is a viable alternative for using mollusks that have been exposed to polluted waters. These reports suggest that site- specific studies would be required to determine optimal conditions for depuration of bacteria by shellfish. The present study was undertaken to examine specific param- eters (salinity and temperature) necessary for depuration of bacteria from clams growing in coastal waters of South Carolina, and to determine levels of sewage pollution and indigenous bacteria in clams from an unpolluted estuary in South Carolina. MATERIALS AND METHODS Collection and Maintenance of Shellfish Hard clams (Mercenaria mercenaria) were obtained from Trident Sea Farms of Charleston. South Carolina and transported to the Waddell Mariculture Center at Bluffton, South Carolina. The basic grade (size) of clams used was the little neck with some cherrystones. Clams averaged 50.0 mm in length with a size range of 38-74 mm. Clams were maintained in a fiberglass tank supplied with a flow- through water system of sand-filtered seawater. Water was obtained from the Colleton River Estuary. Clams used for depuration studies at Clemson University were shipped in styrofoam containers with ice, packed so as not to di- rectly contact the clams. After arriving at Clemson Univer- sity, the clams were placed in tanks (22 1) containing arti- ficial seawater (ASW; Rila Marine Mix, Rila Product Company, Teaneck. New Jersey) at 26.57cc and 25°C. Water was aerated and circulated using a commercial aquarium pump. Clams were fed a variable amount of algal suspension of Isochnsis galhana and Chaetoceros gracilis (cell density, 10^/ml) at 3-day intervals. Three days prior to each experimental study, clams were placed in a tank (22 1 ) containing fresh artificial seawater at 26.5%c at 25°C and allowed to pump without feeding. Bacterium A chloramphenicol resistant strain of Escherichia coli was obtained by mutating a parental strain (ATCC 25922) with ultraviolet light. The mutant was cuUivated at 37°C for 18 h in Brain Heart Infusion (BHl) broth (Difco Lab) or at 37°C on BHI agar medium (Difco Lab) to which filter ster- ilized (0.45 |jLm; Gelman Acrodisc) chloramphenicol (Sigma Chemical Company) was added to a final concen- tration of 100 |jLg/ml. 89 90 Johnson and Hayasaka Loading Clams were initially scrubbed with a stiff bristled brush under running potable water to remove debris from their shells. They were then placed in a plastic container, cov- ered with damp paper towels, and placed in a 4°C room for 24 h and then at 25°C for 12 h. This procedure facilitated uptake (personal communication, John Manzi) of bacteria by clams. A pre-loading sample of three to five clams was removed for processing to determine the initial bacterial concentration. Groups of 36 clams were then evenly sepa- rated in tanks (22 1) containing ASW. at 26.5%c, and a temperature of 25°C. The water was aerated and circulated using a commercial aquarium pump. Clams were exposed for 4 h to chloramphenicol resistant E. coli (approximately 5 X lO"* per ml), removed from the contaminated water, placed in wire baskets and scrubbed with a stiff-bristled brush under potable running water. Only those clams that were observed to have extended siphons during loading were measured and used for depuration studies. The per- centage of clams that siphoned was recorded. Depuration Studies Clams preloaded with bacteria were placed in tanks (22 1) containing clean seawater as previously described. Zero h was recorded at the moment shellfish were placed into experimental tanks. Clam samples were removed at 0, 4, 8, 12 and 24 h intervals. Seawater salinity and temperature were monitored for their effect on shellfish depuration. Temperature was controlled by placing experimental tanks into temperature controlled chambers (Scientific Systems). A synthetic seawater mix (Rila Marine Salts) was used to prepare water at different salinities. Processing of Shellfish Samples of three or more clams were washed and scrubbed with a stiff-bristled brush under running potable water, air dried and measured before being aseptically shucked into a sterile tared beaker. Clam meat and liquor were weighed to the nearest gram and transferred to a sterile blender (Waring). To each sample, sterile 0.2 M phosphate buffer, pH 7.6, was added to dilute the sample 1 :4 ( w/v) and blended for 30 s at low speed, then for 60 s at high speed. Four milliliters of homogenate contained ap- proximately 1 g of clam meat and liquor. Tenfold serial dilutions of homogenate were made to the 10"" dilution, using sterile 0.2 M phosphate buffer (pH 7.6). Samples were spread-plated, in duplicate, with 0. 1 ml of serially diluted clam homogenate on BHI agar medium supplemented with chloramphenical (100 |xg/ml). Plates were incubated at 37°C and colony forming units were counted after 24 h. Temperature Shift Studies Collection of shellfish, bacterium used, and loading procedures were as previously described. Maintenance of shellfish was altered; after arrival at Clemson University, clams were divided evenly into three groups and placed in three tanks (22 1 each) containing ASW at 26.5%c and water temperatures of 20° or 30°C. Clams preloaded with bacteria were placed in tanks as follows: one-half of the clams loaded at 20°C were placed in a tank of 26.5%f ASW at 30°C, the other half in a tank of 26.5%c ASW at 20°C; the clams loaded at 30°C were placed in a tank of 26.5%o ASW at 20°C. Zero time for each experiment was the mo- ment shellfish were placed into tanks. Clam samples were removed at 24 and 48 h intervals for depuration measure- ments. Temperature was controlled by placing experi- mental tanks into temperature controlled chambers (Scien- tific Systems). Processing of shellfish was as described previously. In situ Bacterial Levels in Clams Clams (Mercenaria menenaria) were obtained at ap- proximately 2-month mtervals from an unpolluted source in the Colleton River Estuary near the Waddell Mariculture Center at Bluffton. South Carolina. Clams were processed within 2 h after arrival, as previously described. At least 25 g of clam meat and liquor were used for each sampling. The clams were homogenized and ten-fold serially diluted to the 10 ■* dilution. Total plate counts were obtained by mixing I ml of di- luted clam homogenate into a pour plate containing melted standard methods plate count agar (American Public Health Association 1984). Duplicate plates for each dilution were prepared and incubated at 35°C for 48 h before colony- forming units were counted. The most probable number (MPN) of fecal coliform bac- teria was obtained using the presumptive and complete tests recommended by the American Public Health Association ( 1984). Four milliliter aliquots of serially diluted clam ho- mogenate were used to inoculate sets of either three or five MPN tubes in the presumptive test. Salmonella. Shigella and Vibrio spp. were enumerated using procedures recommended by the American Public Health Association (1984). A 25 g sample of clam meat and liquor was used for estimating each bacterial popula- tion. RESULTS A survey was taken of the natural bacterial loads in clams that were in the Colleton River Estuary next to the Waddell Mariculture Center (Table 1). Bacterial loads in clams were highest in the September sample (1.5 x 10^ cfu/g shucked clanij and lowest in the March sample (5.3 X 10- cfu/g shucked clam). Few fecal coliforms and no Salmonella or Shigella spp. were detected. Vibro parahae- molyticiis was detected in low numbers (3-23 bacteria/g clam meat) throughout the year. The effect of salinity on hard clam depuration was ex- amined for 24 h at 25°C. Results showed that a salinity of Bacterial Depuration by the Hard Clam 91 TABLE I. Bacterial Survey of Hard Clams in the Colleton River Estuary Sample Standard Plate Count Fecal Coliforms v. parahaemolyticus" Salmonella' Shigella' Month/Year (CFU/g Shucked Clam) (MPN/g) (MPN/g) Detected Detected March 1985 5.3 X 10- 8» 23 Apnl 1985 8.9 X 10' O" 4 June 1985 7.4 X 10^ 4" 11 September 1985 1.5 X 10' O" 9 Niivcmhcr 1985 1.9 X 10' <3'' 5 January 1986 .\1 / 10' <3" 3 ' Three tube MPN method. '' Five tube MPN method. ' Enrichment culture. 25%f or 30%<- was more favorable than 20%f for elimination of preloaded chloramphenicol resistant E. coli (Figure 1). The salinities showed no statistical difference at 4 and 8 h; at 12 h. depuration at 30%f was significantly faster than at 207(c: and at 24 h, both ISVcc and 307rc showed significantly more depuration than that at 2(Wcc. Clams depurated for 24 h at 25°C and salinities of 25%o and 30%c had coliform levels well below standard levels (230 coliforms/100 g clam meat) that are acceptable for marketing (NSSP); clams depurated at a salinity of 20%c were in excess of NSSP guidelines. The effect of temperature on hard clam depuration during a 24 h period at a salinity of 257rc was examined (Figure 2). In general, results showed better depuration at 25°C and 30°C than at 20°C and 35°C (Figure 2). During the initial 4- 12 h depuration, temperatures of 30° and 35°C showed, for the most part, significantly better depuration; after 12 h, temperatures of 25° and 30°C gave significantly lower bacterial levels. Depuration was slowest at 20°C. The interaction of temperature and salinity over a 24 h depuration period is shown in Figure 3. The data suggest that depuration best occurs at a temperature between 24-28°C and a salinity between 24-3\7cc A temperature-shift study was run to determine the ef- fect of a shift in temperature on depuration (Figure 4). Clams that were initially acclimated to a temperature of 20°C showed similar depuration for the first 24 h whether the temperature was kept at 20°C or shifted-up to 30°C. An :^ 4 • S 2 ■ v> • 20 0/00 O 25 o/oo D 30 o/oo \\.\ ■ ^ V o/b ^^ N^< ^s. a • \b \ \ ■ N^b 6 ' 16 20 24 4 8 12 TIME :H) Figure I. The Effect of SaHnity on the Time Course of Depuration of Bacteria (£. coli) from Clams (A/, mercenaria) Incubated at 25°C. Clams were preloaded with E. coli at 25°C and 26.5'/ff ASW. Data points with different or the same letters at corresponding time in- tervals are significantly or not significantly different, respectively (analysis of variance, LSD = 0.98 at 0.05 confidence level). 20''C O 25''C D 30° C 5 v"^ a A 35°C £ V\ — — — ; w 4 \^\ Q ^3 \ ^ix^ °xi\ - 3 b — -D~-~. \,^b^ b ^ ^^~A °2 N\ 1 • \ 16 20 24 4 8 12 TIME IH) Figure 2. The Effect of Temperature on the Time Course of Depura- tion of Bacteria (£. coli) from Clams (M. mercenaria) Suspended in 25'/,c ASW. Data points with different or the same letters at corre- sponding time intervals are significantly or not significantly different, respectively (analysis of variance, LSD = 0.98 at 0.05 confidence level). 92 Johnson and Hayasaka 10 lo' lold decrease m'' W^ lold decreose 20 25 30 35 IfMPEIIAIUilf °C Figure 3. The Effect of Salinity and Temperature on Depuration of Bacteria (£. colt) from Clams (A/, mercenaria). The data represent 10-fold decreases in the number of bacteria in clams after 24 h of depuration. Clams were preloaded with £. coli (final concentration: 5 X 10-' CFU/g shucked clam I at 25°C and 26.5%^ ASW. Experimental data points (•) and extrapolated data points (C) from Figures 1 and 2. increase in bacterial load in clams depurated for longer than 24 h at 30°C may be attributed to regrowth of £. coli and/or resiphoning of previously depurated bacteria. On the other hand, clams acclimated to 30°C and shifted-down to 20°C showed decreased depuration over 24 h. However, their bacterial load was reduced over a 48 h period to a level that was comparable to clams that were not subjected to a tem- perature-shift (loaded, 20°C/depurated, 20T). 6 1 "C LOADING /°C DEPURATION \ • 20°/ 20° \ O 20°/ 30° 5 A ■ 30°/ 20° 2 4 ^"~^^-.\\ a ^V\ =3 5 3 \ \^^^-^'' =3 \ \ b ■ \ \^ 1 \ V? — ^ ^\ ° \a / / \: TABLE 2. Percent Clams Siphoning at Various Salinities (Temperature 25°C) Salinitv i'i,) 9c Siphoning (# clams) 24 48 TIME (H) Figure 4. The Effect of a Temperature-Shift on Depuration. Clams were cither preloaded with E. coli at 20°C and depurated at 20°C or 30°C, or preloaded with E. coli at 30°C and depurated at 20''C. The salinity of the experiments was 26.5%c Data points with different or the same letters at corresponding time intervals arc significantly or not significantly different, respectively (analysis of variance, LSD = 0.98 at 0.05 confidence level). 15 20 25 30 4.2 83.3 89.6 (28) (47) (48) (48) The percentage of clams observed siphoning at different salinities and temperatures was recorded. At a temperature of 25°C, salinities of 25%c and 30%c showed the highest percentage of siphoning clams (Table 2). Few siphoning clams were observed at 207cc At a salinity of 26.5%c. tem- peratures of 20° and 25°C showed the highest percent si- phoning clams (Table 3). A low percentage of siphoning clams was observed at 30°C and none were seen at 16°C. DISCUSSION Bacterial loads in clams showed seasonal variation with relatively low counts in colder months and higher counts in warmer months. Clams accumulated up to 1.5 X 10^ cfu/ gram shucked clam (September, 1985). Accumulation oc- curred even though water from the Colleton River Estuary contained low fecal coliform counts and undetectable levels of Salmonella and Shigella spp. in clams and would be con- sidered unpolluted from sewage. Shellfish accumulate mi- croorganisms through filter feeding, especially during warmer seasons when temperatures promote siphoning (Van Winkle et al. 1976) and bacterial counts are highest in natural waters (Zobell 1946). Vibrio parahaemolyticus. an mdigenous bacterium in estuaries, was accumulated only in low numbers in clams, indicating low population levels in this area. Nevertheless, its presence indicates a potential public health problem if appropriate handling and pro- cessing of clams are not stringently followed. A number of food-borne illnesses have resulted from improper handling or cooking of seafoods contaminated with Vibrio parahae- molyticus (Center for Disease Control 1981, Idler 1970, and Wood 1970). Using clams that were preloaded with an indicator bac- terium (chloramphenicol-resistant E. coli). at concentra- tions comparable to in situ levels, allowed for observation of depuration. Temperature and salinity are two important parameters that affect depuration. Effects of temperature TABLE 3. Percent Clams Siphoning at Various Temperatures (Salinity 26.5 %c) Temperature (°C) % Siphoning (# clams) 16 20 25 30 83.7 78.2 33. S (97) (257) (284) (227) Bacterial Depuration by the Hard Clam 93 and salinity on depuration are known to vary with the kind of shellfish, bacterial load and turbidity of the water (Can- telmo and Carter 1984, Eyles and Davey 1984, Hartland and Timoney 1979, Heffernan and Cabelli 1970, Janssen 1974, Ledo et iil. 1983, MacMillan and Redman 1971, Perkins et al. 1980, Ritchie 1976, Rowse and Fleet 1984, Timoney and Abston, 1984 and Van Winkle et al. 1976). Other studies have indicated that a temperature range of 25°C to as low as 10°C (Hartland and Timoney 1979, Mac- Millan and Redman 1971, Perkins et al. 1980, and Van Winkle et al. 1976) and a salinity range of 24%p to as low as 20%o (Hartland and Timoney 1979, Heffernan and Ca- belli 1970, MacMillan and Redman 1971, Timoney and Abston 1984, and Van Winkle et al. 1976) would give good depuration results for hard clams in New York (Can- telmo and Carter 1984 and MacMillan and Redman 1971) and England (Houser 1964). Our results show that a higher temperature (24-28°C) and salinity (24-3 l%c) range would be necessary for optimal hard clam depuration in South Carolina. These variations in environmental require- ments for depuration may reflect adaptation of hard clams to regional differences in water temperature and salinity. In order to depurate microorganisms from clams, it is likely that the water temperature will need to be changed form the //; .situ temperatures of clam beds. In the summer, the temperature of South Carolina coastal waters may ex- ceed 30°C. Higher temperatures do not promote the shelf- life of fresh clams during processing and marketing (Furfari 1966). Therefore, it would be beneficial if depuration tem- peratures were lowered. Similarly, in the winter, when temperatures are low (approximately 16°C), decreased si- phoning by clams is expected. Although bacterial loads in clams will also be reduced due to decreased siphoning, a low temperature will inhibit further depuration. Tempera- tures will need to be shifted upward for significant depura- tion to occur. Heffernan and Cabelli (1970) showed a marked decrease in siphoning when clams were removed from 10°C and placed in 20°C water. However, our results indicated that clams acclimated to 20°C could be shifted to a higher optimal temperature for better depuration without apparent '"thermal shock."" The difference in the range of temperatures used in these experiments may account for the difference in results. In contrast, our findings show that clams acclimated at a higher temperature and shifted to a lower temperature had a lower rate of depuration compared to clams that were both acclimated and depurated at the same lower temperature. Longer depuration times would be required to overcome this "cold shock"" on depuration. In our depuration studies, clams were preloaded with in- dicator bacteria prior to monitoring depuration. However, only clams that had siphons visibly extended at the end of the loading period were used in depuration experiments. It is not known whether clams siphon intermittently or throughout a loading or depuration period. Nevertheless, at different salinities and temperatures there was a change in the number of clams that were observed siphoning. Our results showed that for siphoning clams, depuration would be best at approximately 24-3 1%( salinity and at tempera- tures of approximately 24-28°C. However, the highest percentage of clams that were observed siphoning occurred at similar salinities but lower temperatures (20-25°C). Small numbers of "unsiphoning" clams would prolong re- quired depuration times since they pose potential public health problems if consumed. This study has shown that forM. mercenaria cultured in South Carolina coastal waters, a temperature and salinity close to 24°C and 24-31%c, respectively, would favor a high percentage of depurating clams and faster depuration. "Temperature shock" would also be more apparent when shifting from a high temperature to a lower temperature for depuration. Lowering the temperature should help promote the shelf-life of clams and maintain lower bacterial concen- trations, but would also extend the period of the time re- quired for effective depuration. ACKNOWLEDGMENTS We thank J. S. Hopkins, J. H. Hoats and J. D. Hol- loway of the Waddell Mariculture Center in Bluffton, SC and N. H. Hadley of the Wildlife and Marine Resources Department in Charleston, SC for obtaining and main- taining clams. We especially thank J. J. Manzi of the Wildlife and Marine Resources Department in Charleston, SC for his technical advice. This research was supported by the South Carolina Sea Grant Consortium, a Cooperative Mississippi-Alabama Sea Grant Consortium/Southeast Fisheries Center Fellowship Program (project no. E/0-17(6)) and the South Carolina Wildlife and Marine Resources Department (grant no. 3-30-XXX-I909-49-2599). REFERENCES American Public Health Association. 1984, Compendium of methods for Furfari the microbiological examination of foods. American Public Health As- sociation, Washington, DC. Cantelmo, F, R. and T. H. Carter, 1984, Commercial depuration of the hard clam. Am. Zoo. 24:84A- Center for Disease Control: Foodbome Disease Outbreaks. Annual Sum- mary 1979. Issued April 1981. Eyles, M. J. and G. R. Davey. 1984. Microbiology of commercial depura- tion of the Sydney rock oyster. Crassostrea commerciali . J . Food Prot. 47:703-706. S. A. 1966, Depuration Plant Design, HEW USPHS. Division of Environmental Engineering and Food Protection, Washington, DC. Public Health Service Publication No. 999-FP-7. Hartland. J. H. and J. F. Timoney. 1979. In vivo clearance of enteric bacteria from the hemolymph of the hard clam and the American oyster. Appl. Environ. Microbiol. 37:517-520. Heffernan, W. P. and V. J. Cabelli. 1970. Elimination of bactena by the northern quahog (mercenaria mercenaria): Environmental parameters significant to the process. J. Fish. Res. Bd. Canada. 27:1569-1577. 94 Johnson and Hayasaka Houser, L. 1964. Depuration of shellfish. J. Environ. Health. 27:477- 481. Huntley. B. E. and R. J. Hammerstrom. 1971. An experimental depura- tion plant; Operation and evaluation. Chesapeake Science. 12:231- 239. Idler, D. R. 1970. Effects of pollutants on quality of marine products and effects on fishing, pp. 535-541. In M. Ruivo (ed.). Marine Pollution and Sea Life. Fishing News (Books) Ltd., London, England. Janssen, W. A. 1974. Oysters: Retention and excretion of three types of human waterbome disease bacteria. Health Lab Sci. 1 1:20-24. Ledo A., E. Gonzalez, J. L. Barja and A. E. Toranzo. 1983. Effect of depuration systems on the reduction of bacteriological indicators in cultured mussels {Mytiliis edulis Linnaeus). J. Shellfish. Res. 3:59-64. MacMillan, R. B. and J. H. Redman. 1971. Hard clam cleansing in New York. Commer. Fish. Rev 33:25-33. Perkins, F. O., D. S. Haven, R. Morales-Alamo and M. W. Rhodes. 1980. Uptake and elimination of bacteria in shellfish. J. Food. Prot. 43:124-126. Ritchie, T. D. 1976. A comprehensive review of the commercial clam industries in the United States. U.S. Dept. Commerce, NOAA, Nat'l Mar. Fish. Serv. (Delaware Sea Grant Program, Coll. Mar. Stud., Univ. Del., Newark and Lewes, Del. DEL-SG-26-76), 106 pp. Rowse, A. J. and G. H. Fleet. 1984. Effects of water temperature and salinity on elimination of Salmonella charity- and Escherichia coli from Sydney rock oysters (Crassostrea commercialis) . Appl. Environ. Mi- crobiol. 48:1061-1063. Timoney, J. F. and A, Abston. 1984. Accumulation and elimination of Escherichia coli and Salmonella ryphimurium in an In vitro system. Appl. Environ. Microbiol. 47:986-988. Van Winkle, W., S. Y. Feng and H. H. Haskin. 1976. Effect of tempera- ture and salinity on extension of siphons by mercenaria mercenaria. J. Fish. Res. Bd. Canada. 33:1540-1546. Wood, P. C. 1970. The principles and methods employed for the sanitary control of molluscan shellfish, pp. 560-565. In M. Ruivo (ed). Ma- rine Pollution and Sea Life. Fishing News (Books) Ltd., London. En- gland. ZoBell, C. E. 1946. Marine microbiology. Chronica Botanica Publishing Company, Waltham, MA. Journal of Shellfish Research. Vol. 7, No. I, 95-100, 1988. A THEORETICAL EVALUATION OF SHELLFISH RESOURCE MANAGEMENT" STEVEN MALINOWSKP AND ROBERT B. WHITLATCH Department of Marine Sciences The University of Connecticut Avery Point. Groton. CT 06340 ABSTRACT Resource management allematives for three commercially important bivalve species (Mya arenana Linne Mercenaria mercenana Lmne and Crassoslrea virgnuca Gmelm) are examined by applying population projection models and sensitivity analyses to age(stage)-specific life history information. All species showed positive correlations between size and fecundity and size and survivorship. Population growth rates were 2-3 orders of magnitude more sensitive to changes in survivorship in larval and juvenile stages of the life cycles than proportional changes in either survivorship or fecundity in adult size classes. The greatest return will be realized in shellfish production if management efforts are directed to increasing juvenile survivorship and the quality and/or quantitv of the juvenile habitat. KEY WORDS: Bivalve resource management, theoretical analysis INTRODUCTION Presently there exists a variety of shellfish resource management practices ranging from harvesting strategies which limit minimum size and quantity of adults that may be taken, to re-seeding programs which supplement local populations with artificially planted juveniles. Other poli- cies include predator control, enhancing the suitability of potential larval settlement areas, and planting adults for spawning purposes. Conceptually, these resource manage- ment alternatives are dramatically different since each con- centrates on a different portion of the organism's life his- tory. To date, no quantitative attempt has been made to assess the relative benefits of the various programs. While their relative success can be assessed after implementation, realistically, it may be years before this is possible. An in- terim approach is to theoretically evaluate the various man- agement policies using mathematical models. It is obvious that the success of any management pro- gram is dependent upon the biology of the exploited species. For example, Adams (1980) illustrated the theoret- ical relationship between the harvestability and life history strategies of various fish species. He found that those species with more "r-selected" traits (e.g., fast growth, early maturity, production of large number of offspring) could withstand more intense harvesting than more "K-se- lected" (e.g.. slow growth, delayed maturity) species. Only after understanding how a natural population main- tains itself through time, is it possible to predict its re- sponse to human intervention. To simplify the problem of comprehending how a popu- lation projects itself through time, it is often convenient to use demographic data as the parameters in mathematical ' Contnbution Number 180 of The University of Connecticut, Manne Sciences Institute - Mailing Address: The University of Connecticut. Marine Research Lab- oratory Noank, CT 06340 models that depict the growth (or decline) of a population. One advantage of this type of analysis is that it computes a population growth rate which may then be used as a cur- rency to assess the relative fitness of different suites of de- mographic parameters. After establishing baseline demo- graphic data (e.g., life table), it is then possible to represent management policies demographically by identifying their impacts on the baseline data. Population growth rate re- sulting from the altered life table can then be used as a relative measure of the long-term impacts of various man- agement alternatives on target species. Matrix population models are commonly used for this purpose. These models have been used extensively for ana- lyzing life history tactics (e.g.. Hartshorn 1975; Longstaff 1977; Caswell and Werner 1978; Enright and Ogden 1979; Pinero et al. 1984; Levin et al. 1987) and optimum har- vesting problems (e.g.. Usher 1966; Beddington and Taylor 1973; Doubleday 1975; Rorres 1976; Gopalsamy 1976; Harley and Manson 1981). Since this is an age- or stage-classified model, it is possible to investigate pro- cesses which impact the population at specific isolated por- tions of the life history. For instance, Lefkovitch (1967) classified a population of insects by life stages and simu- lated differential stage-specific mortality rates, and Jensen (1971) used the Leslie matrix to determine the effect of increased juvenile mortality on trout population yield. Bivalve management policies can be translated into changes in the demography of local populations. For ex- ample, planting spawner stocks may increase the fecundity of the population, re-seeding programs may increase survi- vorship of juveniles, and harvest strategies affect the survi- vorship of adults and fecundity of the population. In the present study, the Leslie matrix is used to analyze the life history tactics of three species of commercially important bivalves: Mya arenaria (Linne) (soft shell clam); Merce- naria mercenaria (Linne) (hard shell clam); and Crassos- lrea virginica (Gmelin) (American oyster). The relative benefits of commonly used resource management policies 95 96 Maunowski and Whitlatch are then assessed in light of their differential impacts on various stages of the species' life cycles. METHODS The Leslie Matrix The Leslie matrix is a linear, discrete, time invariant model, based on an age-structured population, that may be used to describe direction and magnitude of population growth, as well as the stable age distribution and reproduc- tive value of each age class. A detailed description of the model appears in Pielou (1977). The dominant eigenvalue, X.^ is equal to e' where r is the intrinsic rate of increase (Lotka 1925) and is used as a relative measure of evolu- tionary fitness (Fisher 1930). An analysis of the relative importance of each of the parameters in the matrix, with respect to population growth rate, can be accomplished by sequentially changing single parameters in the matrix and observing the relative effect on X^. An alternative and more straightforward procedure, known as the sensitivity analysis, has been derived by Cas- well (1978) and uses stable age distributions and reproduc- tive values (right and left eigenvectors of the matrix, re- spectively) to assess the sensitivity of \^ to proportional changes in fecundity and survivorship. The Use of Leslie Matrices for Resource Management Problems While the Leslie matrix has been frequently used for the analysis of life history strategies and optimum harvest problems, Mendelssohn (1976) has criticized the approach because it neglects density effects. Indeed, density depen- dence is intimately linked to the harvesting problem since if population growth is limited by intraspecific interactions then harvesting can change the demography of the har- vested species. Sustained harvesting is most easily tolerated by those species that experience increased productivity after individuals have been removed. Simulated exploita- tion studies in the laboratory have documented enhanced productivity following removal of individuals (Nicholson 1954; Watt 1955; Slobodkin and Richman 1956; Sillman and Outsell 1958; and Usher et al. 1971). Enhanced pro- ductivity may frequently be the result of diminished rates of mortality following decreases in population density (DeAn- gelis et al. 1977). Changes in population productivity result in changes in the matrix parameter values thereby violating the time invariant assumption of the model. Beddington (1974) mentioned that unless a population were in the middle of a colonizing episode or at a density below lim- iting resource levels, parameters of the matrix would be valid for only a few generations. Analytical solutions to the density dependence problem have been incorporated into the matrix model (e.g., Leslie 1948; Pennycuick et al. 1968; Usher 1972; Van Winkle et al. 1978); however there is conflicting evidence that intra- specific interactions limit the population density of large bivalves. For example, while Peterson (1982) concluded that growth, fecundity, and recruitment of Protothaca stamina were all significantly reduced by increases in adult population density, Malinowski and Whitlatch (1988) found that adult growth, fecundity, survivorship and re- cruitment were not significantly affected by population density in M. mercenaria. In general, studies that have documented density-mediated growth effects in M. mer- cenaria (e.g., Eldridge et al. 1979; Walker 1984; Hadley and Manzi 1984) have tested densities characteristic of aquacultural operations which may be an order of magni- tude or higher than population densities normally encoun- tered in natural populations. In addition, several studies have concluded that predation (particularly among the juve- nile age classes) ultimately determines the distribution and abundance of M. mercenaria and M. arenaria (MacKenzie 1977; Blundon and Kennedy 1982; Malinowski and Whit- latch 1988). Therefore, the omission of density dependence from the model may not necessarily sacrifice realism and many objections to the use of the model may not be rele- vant. Other biological processes unique to organisms with a planktonic life stage do, however, violate the assumptions of the model. For example, populations of large bivalves experience intense annual variability with respect to suc- cessful recruitment (e.g., Carriker 1961; Saila et al. 1967) and since these bivalves have a one-to-three week larval stage, it is unlikely that larvae settling into a population originated solely from that population. Furthermore, the matrix parameters may be extremely site-specific and vary over relatively small spatial scales (Flagg and Malouf 1983; and Malinowski and Whitlatch 1988). Solutions to the optimal yield problem require a precise prediction of the actual number and/or biomass of or- ganisms in a population through time. Since these bivalve populations violate the assumptions of the model, it is clear that the model cannot be used to accurately predict the size of these populations through time. It is, however, possible to use this model to address a more general problem. Our purpose is to use the Leslie matrix in its simplest form in order to translate induced demographic changes into a meaningful population parameter such as population growth rate which can then be used as a measure to assess the relative importance of processes which affect the or- ganism at specific stages of the life cycle. Available Data and Analytical Procedures To use the Leslie matrix model, only age- or stage-spe- cific fecundity and survivorship data are needed. Life table data for each species were derived from the literature (Table 1). While there may be considerable variation in size among individuals of the same age class (e.g., Pe- terson et al. 1983) and both fecundity and survivorship are more closely correlated to size than age in these bivalves, age classes have been assigned to size class intervals for convenience in interspecific comparisons and presentation Shellhsh Resource Management 97 TABLE 1. Survivorship (1,) and fecundity (m,l data used in Leslie matrices Tor the three bivalve species. SURVIVORSHIP Age C. virginica^ M. mercenaria' M. arenaria' Post-set 0.091 169 0.007 1 0.453 0.169 0.833 2 0.350 0.910 0.790 3 0.342 0.910 0.902 4 0.032 0.910 0.899 5 0.910 0.941 6 0.910 0.969 7 0.910 0.824 8 0.910 FECUNDITY (LOG) Age C. virginica' M. mercenaria^ M. arenaria^ Post-set 1 6.24 2 6.94 4.38 3 7.51 6.76 4.79 4 7.72 6.96 4.88 5 7.89 7.16 4.98 6 7.25 5.08 7 7.33 5.18 8 7.44 5.26 ' Dame ( 1976) for 0- 1 year old; Mackin ( 1961 ) for 1 + age-classes. - Connell et al . (1 98 1 ) for juveniles; Carriker ( 1 96 1 ) for adults , ^ Modified from Brousseau (1978a) where 1, = I - q, and then col- lapsing the life table so it represents yearly age mtervals. " Estimated from Davis and Chanley (1956) for a 40 ml (interval shell volume) individual and then extrapolating for other ages using regression equation of Dame ( 1976, p. 248). ' Estimated from Davis and Chanley (1956) for a 65 ml (internal shell volume) individual and then extrapolating for other ages using regression equation of Hibbert (1977). ^ Modified from Brousseau ( 1977b). of results. Of the three species in question, a single stuiJy summary of a complete life table is only available for M. arenaria (Brousseau 1978a). Larval survivorship has not been documented for any of the bivalves. Vaughn and Saila (1976) derived a formula for indirectly computing survi- vorship of an age-class if all other life table parameters are known. This equation was used to determine the minimum larval survivorship necessary to generate a stable popula- tion. The compiled life tables should be viewed only as generalized descriptions of the demography of each species. These species share the characteristics of high fe- cundity which increases with size/age. low larval and juve- nile survivorship and high adult survivorship. These char- acteristics, rather than the exact life table values, are under scrutiny in the present analyses. For each species the Leslie matrix model was run with larval settlement rates varying from 100% to I x 10"*%. The model was also used to evaluate the effect on popula- tion growth rate of three harvesting strategies, each re- moving the same number of individuals but concentrated harvesting on different age classes within the populations. Sensitivity analyses, as derived by Caswell (1978). were used to determine the relative effects of altering age-spe- cific fecundity and survivorship on the population growth rate. RESULTS AND DISCUSSION The life history tactics (sensu Stearns 1976) of each of these bivalves examined are characterized by high fecun- dity, iteroparity, large size, and high larval and juvenile mortality (relative to adults). Consequently, the analyses of the different species yielded approximately similar results and the species will be discussed collectively. While it was possible to estimate most parameters of the life tables, larval and early post set mortality was deduced indirectly from the other parameters of the life table. As- suming 100% survivorship during this stage of the life his- tory resulted in population growth rates (k^ ranging from 20.69 for C. virginica to 3.30 for A/, arenaria. Larval sur- vivorship values that yielded a stable population (e.g.. X^, = 1) range from about 0. 1% to 0.0001% (Table 2). These values were used for all subsequent analyses except simu- lated harvests (see Figures 3-4). Current commercial shellfish management policies have taken many forms and while the ultimate goal of each is to increase yield of the fishery, there have been no attempts to quantitatively determine the relative effectiveness of each policy. The establishment of minimum harvest sizes serves at least two purposes. First, by allowing the harvest of only large individuals, potential yield (biomass) of a single co- hort is increased (provided adult survivorship is high). A second purpose, and one more relevent to the present anal- ysis, is that larger individuals realize increased fecundity and the reproductive contribution of a cohort to future gen- erations will be increased as the minimum harvestable size increases. For example, Bricelj and Malouf (1981) and McHugh (1981) suggest that current hard shell minimum size limits and harvest strategies which concentrate on small individuals may not effectively protect adequate breeding stocks. MacArthur ( 1960) reasoned that harvesting should be re- stricted to those older age classes with low reproductive values because the reproductive values of most organisms TABLE 2. Population growth rates (X„) assuming 100% larval survivorship and the larval survivorship that will yield a stable (X„ = I) population (critical larval survivorship) for each species. Species Critical Larval Survivorship' M. mercenaria 9.59 M. arenaria 3.30 C. virginica 20.69 1.1 X 10-* 6.7 X 10-" 8.0 X 10-' Computed from equation given by Vaughn and Saila ( 1976). 98 Malinowski and Whitlatch peak early in life and then decline throughout adulthood. The three bivalves examined here deviate from this general pattern. The reproductive values peak early in life and, since fecundity is size related, remain high throughout adulthood (Figure 1). Coupled with stable age distribu- tions, reproductive values are used to compute the survi- vorship sensitivity analyses (Caswell 1978) which apply di- rectly to the determination of optimal harvest sizes. Sensi- tivity of \^ to changes in survival describes the effect on population growth rate of proportional changes in age-spe- cific survival or age-specific harvesting. In theory, har- vesting should be concentrated on those age classes that contribute least to X^i- Identification of a major change in slope of the survivorship sensitivity function (Figure 2) would suggest a logical minimum harvest size. Results, however, indicate that there are only slight differences in the sensitivity of \^ to changes in survivorship among the adult (>2 years old) age classes. Therefore, removing a certain proportion of an old age class is essentially equiva- lent to removmg that same proportion of three or four year olds. Any age class may be harvested with approximately equivalent effects on the population growth rate provided minimum size limits allow the organism to reach reproduc- tive maturity and spawn. Since predation is generally the cause for extremely high rates of juvenile mortality (MacKenzie 1977: Blundon and Kennedy 1982; Malinowski and Whitlatch 1988), both re- seeding beds and predator control are synonymous with in- creasing juvenile survival. The sensitivity of \^ to changes in survivorship (Figure 2) indicates that population growth rate is most affected by alterations in juvenile survivorship. In fact, there are at least two orders of magnitude difference between adult and juvenile survivorship. Reductions in ju- venile mortality will have 100 times the impact on the fu- ture number of individuals in the population than a propor- Figure 1 .^gc-specific reproductive values given by the left eigen- vector 01 the Leslie matrix. The shaded area encompasses all values of all three species. Figure 2. The relative sensitivity of the population growth rate 1\„) to single matrix parameter changes in survivorship and fecundity [equa- tions derived by Caswell (1978)]. The shaded area encompasses all values of all three species of bivalves. tional reduction in adult mortality, suggesting that pro- tecting the juvenile stages of the life history rather than establishing size-specific harvest strategies would be a more beneficial management policy for the three species examined. To assess the relative benefits of these two alter- natives, three different harvest strategies were simulated on M. arenaria and compared with reductions in juvenile sur- vivorship. Each harvest strategy removed 97% of the adult population and differed with respect to the intensity of the age-specific removal. Despite drastic differences between harvest strategies, there was little difference with respect to effects on the population growth rate (Figure 3) and, when compared to simulated reductions in juvenile survivorship (Figure 4), indicate that removing 97% of the adults is equivalent to decreasing juvenile survivorship from 0.7 to 0.3%. Sensitivity analyses can also be used to evaluate the rel- ative effectiveness of planting adult shellfish for spawning purposes as compared to the alternative strategy of in- creasing the survivorship of juveniles. The consequences of increasing reproductive potential of the population can be examined by determining the magnitude of the response of \„. A comparison of the sensitivity of X^, to changes in fecundity with respect to changes in survivorship reveals a dramatic result. In all cases, population growth rate is at least four orders of magnitude more sensitive to changes in juvenile survivorship than it is to deviations in fecundity of adults (Figure 2). Since larval and early post set settlement mortality rates are extremely high, very large increases in fecundity are necessary to duplicate the consequence of Shellrsh Resource Management 99 10' 9 ,— 8 IS ^ H . A = 1.27 Ug AM. 23 B A -1.34 AGE AGE AGE Figure 3. Simulated harvesting strategies of M. arenaria. Each strategy removed 97% of the adults in the population but concen- trated on different adult age classes (Strategy I: equal intensity of harvesting on all adult age classes; Strategy 11: concentrated har- vesting on older adult age classes). The shaded areas represent the age-specific decreases in survivorship resulting from the simulated harvesting. The relative effect of each strategy on the population is assessed by comparing the resultant values of \„ = (\„ 1.43 before harvesting). A larval survivorship of 0.01 was arbitrarily chosen for these simulations. only slight increases in juvenile survivorship. Furthermore, Kasner and Malouf (1982) have demonstrated that a basic assumption of the spawner transplant concept is incorrect (transplanted clams spawned at the same time as native clams) and estimate that recruitment resulting from typical spawner transplants will be insignificant compared to con- tributions from native standing stocks. These results suggest that far greater return may be gained from management efforts aimed at increasing juve- nile survivorship than from other alternatives. In a compar- ison of several natural populations of M . mercenaria. MacKenzie (1977) arrived at a similar conclusion. He found differences in densities of adult clams between local populations were correlated to the abundance of juvenile clam predators and observed a 7-8 fold increase in clam density after juvenile clam predators were poisoned. MacKenzie (1979) further suggested practical techniques of increasing clam abundance which focused on the juvenile stage of the life history. Since juvenile bivalves (1-10 mm shell length) may experience intense predator-mediated density dependent mortality (Boulding and Hay 1984; Ma- LAS ■ ^ 1 40 1 35' « 130- -i 1 25> 1 20 A=1.28 075 0065 0055 0045 0035 JUVENILE SURVIVORSHIP ■*- 0025 0015 Figure 4. The effect of simulated reductions in juvenile survivorship on the population growth rate (A„) of A/, arenaria. A larval survivor- ship value of 0.01 was arbitrarily chosen for these simulations (Figure 3). \„ = 1.28 corresponds to the average value of m resulting from the harvest simulations shown in Figure 3. linowski and Whitlatch 1988). it is possible that increased recruitment (resulting from spawner transplants or in- creased average size of adults) will be followed by in- creased rates of predation. By similar reasoning, the return on seed plantings is not likely to be greater than the harvest of the individuals planted (there will be little if any contri- bution to future generations). It appears, therefore, that persistent juvenile clam predator control has the greatest potential to significantly increase the maximum sustainable harvest of these species. ACKNOWLEDGMENTS V. Starczak, J. Weinberg, R. Zajac and anonymous re- viewers provided insightful comment on earlier versions of the manuscript. H. Caswell provided the computer program used to perfomi the analyses. J. Rodriguez typed the final version of the manuscript. This work was supported by a grant from NOAA, National Sea Grant Program, Depart- ment of Commerce, Grant No. NA82AA-D-0018, R/LR-1. LITERATURE CITED Adams, P. B. 1980. Life history patterns in marine fishes and their conse- quences for fisheries management. Fish. Bull. 78:1-12. Beddington, J. R. 1974. Age structure, sex ratio and populalion density in the harvesting of natural animal populations. J Appl. Ecol. 11:915- 924. Beddington. J. R. and D. B. Taylor. 1973. Optimum age-specific har- vesting of a population. Biometrics 29:801-809. Blundon, J. A. and V. S. Kennedy. 1982. Refuge for infaunal bivalves from blue crab. Callinecies sapidus. predation in Chesapeake Bay. J . Exp. Mar. Biol. Ecol. 65:67-82. Boulding, E. G. and T. K. Hay. 1984. Crab response to prey density can result in density-dependent mortality of clams. Can. J. Fish. Aquai. Sci. 41:521-525. Bricelj. V. M. and R. E. Malouf. 1980. Aspects of reproduction of hard clams [Mercenaria mercenaria) in Great South Bay, New York. Proc. Nail. Shellfish. Assoc. 70:216-229. Brousseau, D. J. 1978a. Population dynamics of the soft-shell clam. Mya arenaria. Mar. Bio. 50:63-71. Brousseau, D. J. 1978b. Spawning cycle, fecundity, and recruitment in a population of soft-shell clams. Mya arenaria, from Cape Ann. Mass. Fish. Bull. 76:155-166. Carriker. M. R. 1961. Interrelation of functional morphology, behavior, and autecology in early stages of the bivalve Mercenaria mercenaria. J. Elisha Mitchel Sci. Sac. 77:168-241. Caswell. H. 1978. A general formula for the sensitivity of population 100 Malinowskj and Whitlatch growth rate to changes in Hfe history parameters. Theor. Popiil. Biol. 14:215-230. Caswell, H. and P. Werner. 1978. Transient behavior and life history analysis of teasel, Dipsacus sylvestris . Ecology 59:53-66. Connell. R., R. E. Loveland and W. Cokeley. 1981. Factors of mortality and growth in an intertidal population of juvenile M. mercenaria from Shark River, New Jersey, over a two-year period, J. Shellfish Res. 2:92(abstr.). Dame, R. F. 1976. Energy flow in an intertidal oyster population. Estiiar. Coasl. Mar. Sci. 4:243-253, Davis, H. C. and P. E. Chanley. 1956. Spawning and egg production of oysters and clams. Biol. Bull. 110:117-128. DeAngehs, D. L., S. W. Christensen and A. G. Clark. 1977. Responses of a fish population model to young of the year mortality. J. Fish. Res. Bd. Can. 34:2124-2132. Doubleday, W. G. 1975. Harvesting in matrix population models. Bio- mcfWci 31:189-200. Eldridge, R. J., A. G. Eversole and J. M. Whetstone. 1979. Comparative survival and growth rates of hard clams, Mercenaria mercenaria planted subtidally and intertidally at varymg densities in a South Caro- lina estuary. Proc. Nail. Shellfish. Assoc. 69:30-39. Ennght, N. and J. Ogden. 1979. Applications of transition matri.x model in forest dynamics. Araucaria in Papau New Guinea and Nolhofagus in New Zealand. Austral. J. Ecol. 4:3-23. Fisher, R. A. 1930. The Genetical Theory of Natural Selection. Dover Publ. Co., New York. 272 pp. Flagg. P. J. and R. E. Malouf. 1983. Experimental plantings of juveniles of the hard clam Mercenaria mercenaria (Linne) in the waters of Long Island, New York. J. Shellfish Res. 3:19-27. Gopalsamy, K. 1976. Optimal control of age-dependent populations. Math. Biosci. 32:155-163. Hadley, N. H. and J. J. Manzi. 1984. Growth of seed clams (Mercenaria mercenaria) at various densities in a commercial scale nursery system. Aquaculture 36:369-378. Harley, P. J. and G. A. Manson. 1981. Harvesting strategies for age- stable populations. J. Appl. Ecol. 18:141-147. Hartshorn. G. 1975. A matrix model for tree population dynamics. In: P. G. Golley and E. Medina (Eds.), Tropical Ecology Systems: 45-51. Springer-Veriag. NY. Hibbert, C. J, 1977. Growth and survivorship in a tidal tlat population of the bivalve, Mercenaria mercenaria. from South Hampton waters. Mar. Biol. 44:71-76. Jensen, A. L. 1971. The effect of increased mortality on the young in a population of brook trout, a theoretical analysis. Trans. Amer. Fish. Sac. 100:456-459. Kasner, J. and R. E. Malouf. 1982. An evaluation of "spawner trans- plants" as a management tool in Long Island's hard clam fishery. J. Shellfish Res. 2:165-172. Lefkovitch. L. P. 1967. A theoretical evaluation of population growth after removing individuals from some age groups. Bull. Entoin. Res. 57:437-445. Leslie, P. H. 1948. Some further notes on the use of matrices in popula- tion mathematics. Biomeirika 35:213-235. Levin, L. A., H. Caswell. K. D. DePatra and E. L. Crego. 1987. Demo- graphic consequences of larval development mode: planktotrophy vs. lecithotrophy in Strehlospio benedicti. Ecology 68:1877-1886. Longstaff, B. C. 1977. The dynamics of collembolan populations: A ma- trix model of single-species population growth. Can. J. Zool. 55:314- 324. Lotka, A. J. 1925. Elements of Physical Biology. Williams and Wilkins, Baltimore, Maryland. MacArthur, R. H. 1960. On the relation between reproductive value and optimal predation. Proc. Natl. Acad. Sci. 46:443-445. MacKenzie, C. L. 1977. Predation on hard clam (Mercenaria merce- naria) populations. Trans. Amer. Fish. Soc. 106:530-537. MacKenzie, C. L. 1979. Management for increasing clam abundance. Mar. Fish. Rev. 22:10-22. Mackin, J. G. 1961 . A method of estimation of mortality rates in oysters. Proc. Natl. Shellfish. Assoc. 50:41-51. Malinowski, S. and R. B. Whitlatch. 1988. Adult hard clam (Mercenaria mercenaria) population dynamics: Management implications. Manu- script in preparation. Malinowski. S. and R. B. Whitlatch. 1988. Survivorship of juvenile hard clams {Mercenaria mercenaria): The importance of population den- sity, clam size, site and time. Submitted manuscript. McHugh, J. L. 1981. Decline has NY. clammers worried. Nat. Fish- erman Yearbook. Journal Publications, Inc.. Maine and Washington. Mendelssohn. R. 1976. Optimization problems associated with a Leslie matnx. Amer. Nalur. 1 10:339-349. Nicholson, A. J. 1954. Compensatory reactions of populations to stresses and their evolutionary significance. Aust. J. Zool. 2:1-8. Pennycuick. C. J., R. M. Compton and L. Bechinham. 1968. A computer model for simulating the growth of a population of two interacting species. J. Theoret. Biol. 22:381-400. Peterson, C. H. 1982. The importance of predation and intra- and inter- specific competition in the population biology of two infaunal suspen- sion feeding bivalves, Protothaca stamineu and Chione undatella. Ecol. Monogr. 52:437-475. Peterson, C. H., P. B. Duncan. H. C Summerson and G W. Safrit. Jr. 1983 A mark-recapture test of annual periodicity of internal growth band deposition in shells of hard clams, Mercenaria mercenaria. from a population along the southeastern United States. Fish. Bull. 81:765- 779. Pielou. E. C. 1977. Mathematical Ecology. J. Wiley and Sons, New York. Pinero, D., M. Martinez-Ramos and J. Sarukhan. 1984. A population model for Astrocar\um mexicanum and a sensitivity analysis of its finite rate of increase. J. Ecol. 72:977-991. Rorres, C. 1976. Optimal sustainable yields of a renewable resource. Bio- metrics 32:945-948. Saila, S. B., J. M. Flowers and M. R, Cannario, 1967, Factors affecting the relative abundance of Mercenaria mercenaria in the Providence River, Rhode Island, Proc. Natl. Shellfish. Assoc. 57:83-89. Silliman, R. P. and J. S. Gutsell. 1958. Expenmental exploitation offish populations. U.S. Fish. Wild. Ser. Fish. Bull. 58:215-252. Slobodkin, L. B. and S. Richman. 1956, The effect of removal of fixed percentages of newborn on size and variability in populations of Daphnia pulicaria (Forbes). Limnol. Oceanogr. 1:209-237. Steams. S. C. 1976. Life history tactics: A review of the ideas. Quart. Rev. of Biol. 51:3-47, Usher, M, B, 1966, A matrix approach to the management of renewable resources with special reference to selection forests. J. Appl. Ecol. 3:355-367. Usher, M. B., B. C. Longstaff and D. R. Southall. 1971. Studies on pop- ulations of Folsomia Candida (Insecta: Collembola). Oecologia 7:68-79. Van Winkle, W., D. L. DeAngelis and S. R. Blum. 1978. A density-de- pendent function for fishing mortality rate and a method for deter- mining elements of a Leslie matrix with density-dependent parameters. Trans. Amer. Fish. Soc. 107:395-401. Vaughan, D. S. and S. B. Saila. 1976. A method for determining mor- tality rates using the Leslie matrix. Trans. Amer. Fish. Soc. 105:380- 383. Walker, R. L. 1984. Effects of density and sampling time on the growth of the hard clam, Mercenaria mercenaria, planted in predator-free cages in coastal Georgia. Nautilus 98:1 14- 1 19. Watt, K. E. F. 1955. Studies on population productivity. 1. Three ap- proaches to the optimum yield problem in populations of Triholium confusum. Ecol. Monogr. 25:269-290. Journal of Shellfish Research. Vol. 7, No 1. 101-138, 1988. ABSTRACTS OF TECHNICAL PAPERS Presented at 1986 Annual Meeting NATIONAL SHELLFISHERIES ASSOCIATION Seattle, Washington June 22 — 26, 1986 National Shellfisheries Association, Seattle. Washington Abstracts. 1986 Annual Meeting, June 22 — 26, 1986 103 CONTENTS George Abbe, James G. Sanders and JoAnn M. Bianchi Pathways of silver accumulation by the American oyster {Crassoslrea virginica Gmelin) in Chesapeake Bay 107 S. K. Allen Cytology of gametogenesis in triploid Pacific oyster, Crassoslrea gigas 107 Tissa Amaratunga and Terence W. Rowell Age and meat yield of Stimpson's surf clam, Spisiila polynyimi. a recently found commercial bivalve resource in Eastern Canada 107 Richard S. Appledoorn Assessment of mortality in an offshore population of Queen conch, and comparative natural mortality estimation in mollusks 108 David Armstrong and Donald Gunderson Interannual variability in recruitment of juvenile Dungeness crab: Is an estuary important after all? 108 Peter J. Auster Response of megafaunal predators to synchronous settlement of sessile prey 108 Malin M. Babcock and John F. Karinen Reproductive success in Tanner (Chionoecetes bairdi) and Dungeness (Cancer magisler) crabs held on oiled sediments 109 Bruce J. Barber, Susan E. Ford and Harold H. Haskin Relationships among condition index, glycogen level, reproductive effort and intensity of MSX (Haplosporidium nelsoni) infection in oysters, Crassostrea virginica 109 Hal Bealtie, R. Elslon, C. Friedman and R. Hedriels Geographically widespread bacterial infection and mortality in Pacific oysters, Crassostrea gigas 109 Ronald E. Becker Summary of the Louisiana oyster industry depuration conference 110 Mark Berrigan and John W. Schneider Status of controlled purification of shellfish in the Southeastern United States 110 Robert Bisker and Michael Castagna Predation on single spat oysters Crassostrea virginica (Gmelin) by blue crabs, Catlinectes sapidus and mud crabs Panopeus herbstii Milne-Edwards Ill Pierre Bocquillon Oyster breeding cages [OBC] — The "Above ground" 'growing system Ill Louis W. Botsford and Jonathan M. Shenker Possible influence of wind on Cancer magister settlement Ill A'. Bourne Scallop breeding studies Ill V. Monica Bricelj, J. Epp and R. E. Malouf Comparative physiology of two cohorts of the Northern bay scallop. Argopeclen irradians irradians 112 James R. Brown A habitat suitability index model for the aquaculture of the Pacific oyster. Crassoslrea gigas 112 Brenda J. Burd and G. Jamieson Biology and commercial potential of Galatheid crabs in British Columbia 112 Fu-Lin Chu Preliminary results from the study of acquired immunity in the oyster, Crassostrea virginica 113 Christine A . Cooke Larval development of the spiny scallop, Chlamys hastata (Sowerby) 113 Ken Cooper The potential for direct application of university-developed research findings to the commercial oyster industry 113 M. Alison Craig and Eric N. Powell A survey of Pcrkinsus marinus infection in the Gulf of Mexico 114 Lee R. Crockett and R. W . Whitlatch Growth rate and age structure comparisons of geographically isolated hard clam, Mercenaria mercenaria. populations 114 104 Abstracts. 1986 Annual Meeting, June 22 — 26, 1986 National Shelifisheries Association, Seattle, Washington CONTENTS (Continued) Jonathan P. Davis Energetics of sterile triploid oysters uncouple the reproductive and somatic effort of diploids 114 Adolphe O. Debrot Comparative coastal ecology of the tropical rocky-intertidal snail Ciitarium pica in the Exuma Islands, Bahamas .... 114 Sandra L. Downing Optimal induction of triploidy in Crassosotrea gigas depends on temperature 115 Brett R. Dumbauld, David A. Armstrong, Donald R. Gunderson and A. Ross Black The importance of intertidal shell as nursery habitat for young-of-the-year Dungeness crab in Grays Harbor, Washington 115 Christopher F. Dungan and Ralph A. Elston Destruction of bivalve mollusc hinge ligament by cytophaga-like bactena: Association with mortality in hatchery-reared juvenile Pacific oysters, Crassostrea gigas 115 Albert F. Eble Depuration of heavy metals by hard clams, Mercenaria mercenaria 116 Ralph A. Elston Bonamia osirea disease of the European flat oyster (Ostrea edulis) in North America: Occurrence, environmental effects and host range 116 B. Emmett Transplant of abalone in Barkley Sound, British Columbia 117 Jennifer A . Epp Energy storage and utilization in the Bay scallop, Argopecten irradians 117 Marilyn C. Erickson and D. P. Selivonchick Egg yolk vesicles as a potential food system for juvenile Pacific oysters 117 Arnold G. Eversole Reproductive biology of clam populations in North America; A review 117 John W. Ewart. Melbourne R. Carriker, Janzel R. Villalaz, Juan A. Gomez and Luis D'Croz Gametogenic development of the venerid clam Prololhaca aspernma in the Bay of Panama 118 Sung Y. Feng Host response to Procloeces inacidatiis infection in the blue mussel, Mytilus edulis L 118 Antonio J. Figueras, Sheila A. Kanaley, Susan E. Ford and Eugene M. Burreson Development of enzyme-linked immunosorbent assays for detection of molluscan parasites 118 Susan E. Ford and Antonio J . Figueras Effects of MSX (Haplosporidniin nelsoni) parasitism on reproduction of the oyster, Crassostrea virginica 119 S. Cynthia Fuller Comparative analyses of larval and early post-larval shell morphology in seven Mytilid species 119 Santo A. Furfari Status of commercial shellfish depuration in the Northeast- 1 986 120 Raymond Grizzle Preliminary studies on the effects of tidal currents, food concentration, and sediment characteristics on ontogenetic growth of Mercenaria mercenaria 120 F. Brandt Gutermuth and David Armstrong A bioenergetic model of juvenile Dungeness crab {Cancer magister) population dynamics in Grays Harbor, Washington 1 20 Harold H. Haskin and Eric Wagner Assessment of mortalities in surf clams (Spisula soUdissima) due to dredging, sorting and discard 120 Herbert Hidu and Greg Podniesinski The distribution of larvae of the blue mussel Mytilus edulis Linne in three Maine estuaries 121 Glen S. Jamieson and A. C. Phillips The spatial distribution of Dungeness crab (Cancer magister Dana) megalopae off the West coast of Vancouver Island. Canada 121 Francis Juanes The foraging behaviour of Cancer magister feeding on Protnthaca slaminea: Size selection and risk 121 Jeffrey Kassner Public fishery and private mariculature conflict in Long Island, N.Y.'s shellfish industry 122 National Shellfisheries Association, Seattle, Washington Abstracts, 1986 Annual Meeting, June 22 — 26, 1986 105 CONTENTS (Continued) Shannon Kelly Aspects of the life history of the pea crab. Pinnotheres macidatus 122 Richard S. Knaub and Arnold G. Eversole Reproductive development in three Mercenaria mercenarta stocks grown in South Carolina waters 122 Daniel A. Kreeger, Christopher J. Langdon and Roger 1. E. Newell Utilization of refractory carbon by the ribbed mussel, Guekensia demissa (DillwynJ 123 Chris J. Langdon Use of downwelling chambers in studies with bivalve molluscs 123 Jack L. Lilja Depuration: Policy and practice on the west coast 123 Bruce A. MacDonald Energy partitioning patterns in cultured and wild populations of the giant scallop, Placopecten magellanicus 124 Steve Malinowski and Scott E. Siddall Recirculation of seawater through upwelling silos in a hard clam nursery system 124 Roger Mann, Robert J. Byrne and Bernardita M. Campos Dispersal of bivalve larvae at a front in the James River Estuary, Virginia 124 John Manzi, Nancy H. Hadley and R. T. Dillon Improved stocks of hard clams (Mercenaria spp.) through genetic manipulation 125 John J. Manzi, A. G. Eversole, J. Hilbish and R. T. Dillon Genetic improvement of hard clam, Mercenaria spp., populations for commercial mariculture stock development in South Carolina 125 Robert C. Maris and John R. McConaugha Diurnal vertical distribution and dispersal-recruitment mechanisms of decapod crustacean larvae and postlarvae in the Chesapeake Bay, Virginia and adjacent offshore water 125 John R. McConaugha and Robert C. Maris Spacial and temporal variability of decapod larval distributions as regulating factors in estuarine decapod populations dynamics 126 R. O. McMillan, D. A. Armstrong and P. A. Dinnel Intertidal distribution and abundance of young-of-the-year Dungeness crab Cancer magister in Northern inland water of Washington 1 26 Edgar Miller Responses of Geukensia demissa to dissolved copper at various salinities 127 Robert E. Miller and W. F. van Heukelem Threshold levels of crab chemoreception for amino acids and results of field tests using these amino acids as attractants in artificial bait 1 27 J. Frank Morado and Albert K. Sparks A review of infectious diseases of the Dungeness crab. Cancer magister 1 27 Roger I. E. Newell and Christopher J. Langdon Digestion and absorption of refractory carbon by the oyster, Crassostrea virginica (Gmelin) 128 Eugene J. Olmi, III and Papal A. Sandifer Recruitment of Blue crab, Callinecies sapidus, in open and impounded marsh systems in South Carolina 128 William P. Osborne and William N. Shaw The setting patterns of the purple-hinge rock scallop, Hinnites multiruqosus in Humboldt County, California 128 Mark Page Temporal variation in growth rate, body and gonad weight in a population of Mytilus edulis in the Santa Barbara Channel 129 Bruce Pease and K. Cooper A relationship between selective larval settlement and adult distribution patterns of Geoduck clams and the presence of Chaetopterid polychaete tub mats in Puget Sound, Washington 129 Greg Podniesinski Short-term and long-term settlement of larval and juvenile Mytilus edulis L 129 E. N. Powell, M. E. While and E. A. Wilson Small-scale spatial distribution of oysters {Crassostrea virginica) on oyster reefs 130 106 Abstmcts. 1986 Annual Meeting, June 22 — 26, 1986 National Shellfisheries Association, Seattle, Washington CONTENTS (Continued) R. D. Rheinhardt and Roger Mann Development of epibenthic fouling communities on shells depositied on a natural oyster bed in the James River of Virginia 130 Edwin W. Rhodes and John J. Manzi Interstate shipment of larval and juvenile bivalves: Effects of shipping duration and method on survival 130 Raymond J. Rhodes Economics of shoreside depuration 131 Neil A. Richard and Robert A. Newman Development of technology for harvesting and transplanting subtidal juvenile Pacific razor clams, Siliqua panda Dixon, along the coast of Washington State 131 Neil A. Richard, Alan D. Rammer and Donald Simons Aspects of the early subtidal life history of the Pacific razor clam, Siliqua patida Dixon, off the coast of Washington State 131 G. E. Rodrich Bacterial and viral elimination in commercial plants 132 John Scarpa and E. T. Bollon Experimental production of gynogenetic and parthenogenetic Mulinia lateralis (Say) 132 William N. Shaw A proposed standardization of the stages in the gametogenesis cycles of bivalves 132 Sandra E. Shumway, Terry L. Cucci, Clarice M. Yentsch, Richard C. Newell and Louis Gainey The effects of the toxic dinoflagellate, Proiogonyaulax tamarensis. on the physiology and behavior of marine molluscs 132 Scott E. Siddall, Robert E. Malouf, Mario E. C. Vieira and Eugenia Gomez-Reyes Use of dispersion models for prediction of bivalve larval recruitment 133 Thomas C. Siewichi Overview of NMFS shellfish depuration research 133 Barry D. Smith and Glen S. Jamieson Spatial and temporal variation in the abundance of male and female Dungeness crabs (Cancer magister) near Tofino, B.C., with implications for the commercial fishery 134 T. M. Soniat and M. S. Brody A field test of the American oyster habitat suitability (HSI) model 134 Albert K. Sparhs and J. Frank Morado Diseases of Alaskan king crabs 134 G. Sumner and C. Sumner The Tasmanian shellfish control program — a growers perspective 134 David M. Taylor and Paul G. O'Keefe Recovery period of newly-molted snow crab. Chionoectes opilio. to a hard-shelled condition 135 Stephen T. Tettelbach Crabs vs bay scallops; Effects of predator and prey size on feeding rates and predatory behavior 135 George A. Trevelyan On the use of hatchery producted mussel {Mytilus edulis L.) spat in mussel aquaculture 135 R. H. Watson, G. G. Jones and B. L. Jones Using centrifuged algae for feeding oyster larvae 136 Jach M. Whetstone, Eugene J. Olmi, III and Paul A. Sandifer Extensive culture of Panacid shrimp in coastal impoundments in South Carolina 136 Marie E. White, Eric N. Powell, Elizabeth Wilson and Sammy M. Ray A model of the energy budget of healthy and parasitized oysters, with validation by growth experiments 136 John N. C. Whyte Metabolic reserves and caloric content of six species of phytoplankton cultured as food for bivalve larvae 137 Daniel E. Wichham Interaction between Dungeness crab abundance and infestation intensity with nemertean crab-egg predators 137 Elizabeth A. Wilson, Eric N. Powell, Marie E. White and Sammy M. Ray The effect of the ectoparasitic snail, Boonea impressa on oyster growth and health in the field with comments on patch formation in snail populations 137 National Shellfisheries Association. Seattle, Washington Abslrucls. 1986 Annual Meeting, June 22 — 26, 1986 107 PATHWAYS OF SILVER ACCUMULATION BY THE AMERICAN OYSTER {CRASSOSTREA VIRGINICA GMELIN) IN CHESAPEAKE BAY GEORGE R. ABBE, JAMES G. SANDERS, AND JOANN M. BIANCHI The Academy of Natural Sciences, Benedict Estuarine Research Laboratory . Benedict. Maryland 20612 Because of low-level aquatic releases of radioactive """"Ag (silver) from some nuclear power plants, and because of the lim- ited information on the effects of these releases, a series of studies was conducted to investigate processes and pathways of Ag accu- mulation by oysters (Crassostrea virginica). During 1984 and 1985 three pathways were examined at three temperatures. Groups of 50- to 70-mm hatchery-reared oysters (0.97 to 1.13 g initial dry weight) were exposed to stable Ag from several sources including enriched water (5 p,g • 1^'), enriched sediment (1.76 (j-g • g~'). enriched algae (77.6 p,g • g"'). and a combination of these three at 15°, 20° and 25°. Ag was rapidly accumulated in soft tissues over a 3- to 4-week period until body burdens were nearly five times controls (4.02 vs. 0.83 |JLg Ag • oyster"'), ex- cept for oysters exposed to enriched sediment which probably re- jected the sediment as pseudofeces. Ag body burdens rapidly de- creased during subsequent depuration periods, but after 9 weeks the oysters from the enriched water, algae and combination tanks still averaged 1 .7 times the Ag found in controls ( 1 .46 vs. 0.85 |j.g Ag • oyster"'). It is clear that oysters accumulate Ag from water, but not from sediment. The degree to which Ag is accumulated from algae, however, is still in question as experiments yielded conflicting results. Uptake rates increased with temperature, but differences among rates at 15°, 20° and 25° were small. These studies indicate that the primary source of Ag accumulated by oysters is dissolved in water and not in particulate form. tissue confirmed ploidy; this same analysis on gonadal tissue pro- vided information on distribution of DNA content in component cell types. Relative maturity was compared by quantifying cross sectional area of gonad relative to total area. Diploid oysters matured normally, relative maturity was equal between sexes, and all spawned at the end of July. Glycogen con- tent decreased by 72% prior to spawning and increased afterward. Triploids matured abnormally. Females matured less than males, producing few oocytes and mobilizing less glycogen during maturation. Males matured about half as much as diploids but twice as much as triploid females. Glycogen levels in triploids decreased only 8% prior to spawning but continued to decline for eight more weeks. Greater maturity in males resulted from mitotic proliferation during maturation. Cytofluorometric results indicate that some meiosis does occur in triploid males, resulting in aneuploid ga- metes. Cell cycle analysis on dividing tissues suggests the rate, in addition to extent, of gametogenesis is retarded. Twenty percent of triploids but no diploids were hermaphro- ditic. Excluding hermaphrodites, sex ratios in diploids and trip- loids were the same, although males were significantly more fre- quent in earlier sampling periods, i.e. sexes were heterogeneously dispersed within the gametogenic period. Surprisingly triploids also spawned. Aspects of reproductive biology in diploids, inferred from aberrant maturity in triploids, are discussed. AGE AND MEAT YIELD OF STIMPSONS SURF CLAM, SPISULA POLYNYMA, A RECENTLY FOUND COMMERCIAL BIVALVE RESOURCE IN EASTERN CANADA TISSA AMARATUNGA AND TERENCE W. ROWELL Fisheries Researcli Branch. Department of Fisheries and Oceans, Halifax. Nova Scotia B3J 2S7 CYTOLOGY OF GAMETOGENESIS IN TRIPLOID PACIFIC OYSTER, CRASSOSTREA GIGAS S. K. ALLEN, JR. School of Fisheries. WH-10, Universirs' of Washington. Seattle, Washington 98195 Diploid and triploid cohorts were reared together in Eureka, CA and sampled histologically at two to four week intervals during their first reproductive season. Flow cytometry of somatic Exploratory surveys conducted between 1980 and 1983 by the Invertebrates and Marine Plants Division, Department of Fisheries and Oceans, Halifax, described commercially harvestable concen trations of Stimpson's surf clam, Spisula polynyma, on the Sco- tian Shelf. Preliminary estimates of virgin biomass on Banquereau Bank were upward of 750,000 t. This species is thought to have commercial potential similar to the economically important At- lantic surf clam, Spisula solidissima. being harvested in the eastern United States. A commercial test fishery is in progress, and this study is intended to provide biological information per- taining to age and meat yield of 5. polynyma essential for resource management. 108 Abstracts, 1986 Annual Meeting. June 22 — 26, 1986 National Shellfisheries Association, Seattle, Washington Random samples representative of the size range were col- lected during the 1980-83 research surveys and the test fishery. Gear selectivity resulted in the minimum observed shell length at 24 mm, while the largest was 157 mm. A total of 355 shells ranging from 24 mm to 143 mm were aged by taking thin sections (approximately 0.25 mm thick) across the chondrophores and counting age lines using a light microscope, as described by Ropes (1984). Length/age relationships were determined for se- lected areas of Banquereau Bank and were compared with 5. po- lynyma in Alaskan waters (Hughes and Brown 1981). More than 4,000 fresh-frozen clams from the research surveys were analysed to determine meat yield and were cross checked with fresh clams obtained from the test fishery. Meat yields ranging from 35. 57? to 44.7% were compared with the clams in Alaskan waters and with S. solidissima being harvested in the U.S. fishery. ASSESSMENT OF MORTALITY IN AN OFFSHORE POPULATION OF QUEEN CONCH, AND COMPARATIVE NATURAL MORTALITY ESTIMATION IN MOLLUSKS RICHARD S. APPELDOORN Depurtment of Marine Sciences, University of Puerto Rico, Maygiie:, Puerto Rico 00708 A 2-year Jolly-Seber multiple tag-recapture experiment was conducted on a queen conch, Strombus gigas L., population off- shore of La Parguera, PR in order to estimate mortality. Over 2000 individuals were tagged in 9 sampling periods spaced at 3- month intervals from August 1983 to August 1985. The occur- rence of fishing during half the intervals allowed estimates to be made of both fishing and natural mortality. Fishing mortality averaged F = 1.14 over the study period. An upper limit of nat- ural mortality M = 1.53, including effects of emigration, was estimated. Assuming random diffusion, emigration was estimated and subtracted yielding M = 1.05. This and other reported estimates of M for queen conch are considerably higher than those reported for temperate mollusks. A preliminary relationship predicting M from von Bertalanffy growth parameters P = Log k ■ W'^ was developed. The regres- sion of Log M vs P accounted for over 90% of the variability of M. making it useful for corroborating and comparing indepen- dently derived estimates. For queen conch, M was consistent with the relationship; other large tropical mollusks should also be ex- pected to have high natural mortality rates. INTERANNUAL VARIABILITY IN RECRUITMENT OF JUVENILLE DUNGENESS CRAB: IS AN ESTUARY IMPORTANT AFTER ALL? DAVID ARMSTRONG AND DONALD GUNDERSON School of Fisheries. University of Washington, Seattle, Washington 98195 A working hypothesis during an ongoing study of juvenile Cancer magister population dynamics has been that major es- tuaries of the southern Washington coast (Grays Harbor and Wil- lapa Bay) provide critical nursery habitat to a significant propor- tion of newly settled + crab in comparison to those that settle directly nearshore. The estuarine population estimate of O-H, based only on subtidal surveys, supported this hypothesis in one year (1983) of three (1983-85). Otherwise, 0-1- estimates for a nearshore coastal area (albeit much greater than the subtidal es- tuarine area) were as much as two orders of magnitude higher. Three provisos to this apparent refutation of our hypothesis are in order: I ) Estuarine + grow much faster than nearshore siblings and are 2 X larger and 7 x heavier by September of the first year. This presumably imparts a survival advantage to the larger, es- tuarine siblings. 2) Intertidal population abundance of O-l- may be 2-3 orders of magnitude greater than subtidal estimates. Addi- tional work to better quantify intertidal estuarine populations is needed to provide a better comparison between nearshore and es- tuarine 0-1- abundance. 3) !n all three years, 1 -I- population esti- mates for the estuaries have generally equalled or greatly ex- ceeded estimates for the nearshore. Apparently the species oc- cupies estuaries en masse for a second summer as 1 -I- juveniles. As the study proceeds, the general hypothesis has been retained that estuaries are of critical importance to C . magister. but our perspective as to location and habitat type used by -I- and impor- tance of the system to older juveniles has changed. RESPONSE OF MEGAFAUNAL PREDATORS TO SYNCHRONOUS SETTLEMENT OF SESSILE PREY PETER J. AUSTER NOAA' s National Undersea Research Program. The University of Connecticut at Avery Point. Groton. Connecticut 06340 Direct underwater observations by biologist-divers have re- vealed small-scale aggregations of high density prey greatly influ- ence nearfield predator populations. Synchronous and aggre- gative settlement of benthic prey (i.e., Mytilus edulis. Modiolus modiolus. Balanus spp., Mercenaria mercenaria) provide a short term prey pool which requires little or no search time once the patch is located by predators, thus facilitating prey capture. National Shellfisheries Association. Seattle, Washington Abstracts, 1986 Annual Meeting, June 22 — 26, 1986 109 Grasping/crushing predators (i.e.. Tautogolabrus adspersus. Cancer spp., Carcinus maenas. Pagurus spp.) often forage on aggregative prey laterally along established fronts (the interface between a prey and non-prey area) which presents easier access to individual prey items. Non-grasping predators (i.e. Asteroidea) are not apparently selective to individual prey position within prey aggregations. These types of predators help establish fronts in high density prey patches for other species to cue on. Motile pred- ators cue on other individuals which have located areas of easy prey access and also forage at these locations (social facilitation). Predators aggregate in prey patches, rapidly depleting prey den- sities to nearfield or lower densities. These predation events are highly localized (patchy), have short temporal scales, and have profound effects on the distribution and subsequent growth of prey species. In order to understand such events, high resolution (short-term) sampling is required. REPRODUCTIVE SUCCESS IN TANNER (CHIONOECETES BAIRDI) AND DUNGENESS i^CANCER MAGISTER) CRABS HELD ON OILED SEDIMENTS MALIN M. BABCOCK AND JOHN F. KARINEN Northwest and Alaska Fisheries Center Aulie Bay Laboratory, National Marine Fisheries Sen-ice, NOAA, P.O. Bo.x 210155, Alike Bay. Alaska 99821 Gravid female Tanner (Chionoecetes bairdi) and Dungeness {Cancer magister) crabs were held on one of several concentra- tions (0-8.9 \i.\ Cook Inlet crude oil/gm sediment) of oiled sedi- ments through one complete reproductive cycle. These two species of commercially important crabs, although found in sim- ilar habitats but at different depths, exhibit somewhat different behavioral characteristics while carrying eggs. Brooding Dunge- ness crab females consistently bury in the sediments while gravid Tanner crab females rarely bury. Because of differences in be- havior, we predicted dissimilar responses to oiled sediments. Dungeness crabs held on all doses of oiled sediments produced significantly fewer numbers of larvae than did control crabs. Larvae from crabs in the high-dose tanks survived significantly shorter periods than did larvae from the control, low- and mid- dose tanks. Eggs from crabs in the high-dose tanks had signifi- cantly elevated levels of aromatic and aliphatic hydrocarbons, compared with eggs from control crabs. Production of larvae and viability of larvae of Tanner crabs held on oiled sediments showed no differences from that of con- trol crabs. Likewise, there was no significant uptake of hydro- carbons in eggs over levels found in controls. We feel that the intimate contact of gravid Dungeness crabs with sediment-absorbed oil and oil present in interstitial waters in a polluted habitat can lead to reduced reproductive success in this species. RELATIONSHIPS AMONG CONDITION INDEX, GLYCOGEN LEVEL, REPRODUCTIVE EFFORT, AND INTENSITY OF MSX (HAPLOSPORIDIUM NELSONI) INFECTION IN OYSTERS, CRASSOSTREA VIRGINICA BRUCE J. BARBER, SUSAN E. FORD AND HAROLD H. HASKIN Rutgers Shellfish Research Laboratory, NJAES, Rutgers University. Port Norris. NJ 08349 As an initial attempt at describing the energetics of the MSX/ oyster relationship, breeding stocks maintained in Delaware Bay, New Jersey were examined in May and June, 1985 for condition, glycogen level, and reproductive effort as well as intensity of in- fection by the haplosporidan parasite, MSX. Although slight dif- ferences occurred between the two months, oysters parasitized by MSX had less glycogen (% dry wt.), a lower condition index, and a reduced relative reproductive effort compared to non-infected oysters. These relationships were related to the intensity of infec- tion, as oysters with systemic infections were affected to a greater extent than oysters with epithelial infections. The inter-relation- ship between condition, glycogen level, and gamete production was demonstrated by significant correlations between condition index and glycogen level, glycogen level and gonad index, and gonad index and condition index. It is presently unclear whether MSX is competing for glycogen reserves which in turn reduces reproductive effort or whether MSX interferes directly with normal gametogenesis. thus limiting the amount of glycogen being stored in the tissues. GEOGRAPHICALLY WIDESPREAD BACTERIAL INFECTION AND MORTALITY IN PACIFIC OYSTERS, CRASSOSTREA GIGAS HAL BEATTIE University of Washington. Seattle. WA 98195 R. ELSTON Battelle Marine Research Laboratory , Seqiiim. WA 98382 no Abstracts. 1986 Annual Meeting. June 22 — 26, 1986 National Shellfisheries Association, Seattle. Washington C. FRIEDMAN AND R. HEDRIELS University of California Davis, Davis, CA 95616 For the past ten years the University of Washington School of Fisheries has been studying mortalities of Pacific oysters in the Puget Sound basin. This study has focused on developing stocks of oysters through selective breeding which are resistant against summer mortality. Since 1981 oysters have been observed dying past the summer mortality period and on through December. These fall mortalities have been recorded in two bays in south Puget Sound. Similar occurrences have been reported from Wil- lapa Bay and from several bays in British Columbia. Mortality is variable among the array of the selected experimental oyster stocks, ranging from to 20%, suggesting a possibility of breeding for improved survival. Over the last two years, the dying animals have been studied at the Center for Marine Disease Control at Battelle. Associated with the fall mortality pattern is the presence of a systemic bacterial infection of the oysters. In severely affected animals, green or yellow pustules occur on the mantle surface, on the gill, and in the adductor muscle and pericardial cavity. Histological examination of infected animals indicates the presence of a highly inflamma- tory, gram-variable, systematically distributed bacterium. Infec- tions can be observed microscopically in clinically normal an- imals. The grossly visible pustules represent a terminal inflamma- tory response to the disease. Pathological observations and mortality data suggest that disease can be lethal in individuals and significant as a cause of mortalities in the populations of oysters. Identity of the organism and the relationship of the disease to previously reported oyster diseases and mortalities will be dis- cussed. SUMMARY OF THE LOUISIANA OYSTER INDUSTRY DEPURATION CONFERENCE RONALD E. BECKER Office of Sea Grant Development. Louisiana State Universin-. Baton Rouge, Louisiana 70803 Louisiana's oyster industry produces about 12 million pounds of meat annually and provides more than 5,000 jobs. With 17 percent of the nation's classified shellfish waters, Louisiana has great potential for expanding production to offset declining em- ployment in the mining sector. Yet all the state's shellfish waters are "conditionally approved" and subject to seasonal closures that have severely curtailed oyster production in recent years. De- puration affords a means to overcome environmental problems through technology, but its use for treatment of oysters from re- stricted waters to achieve public health standards is new to the state. Louisiana producers are experimenting with several ap- proaches such as a batch-processing unit involving one-time use and disposal of seawater, a closed, recirculating system with bio- logical filtration and ultraviolet light for bacterial reduction, and a flow-through system utilizing ozone for disinfection. Experience to date indicates that added costs of depuration can be offset by improved quality in terms of a cleaner shellstock product, the control of salty flavor, and longer shelf life, as well as the ability to meet mandated health standards. The conference was held to provide current information about depuration to members of the oyster industry, regulatory agencies, financial institutions, governing bodies, academia, and the news media. A secondary purpose was to identify socioeconomic, regu- latory, and technological factors that may inhibit the widespread practice of depuration in order to plan future research. The one- day conference dealt with the necessity for depuration and its ef- fectiveness, state and federal regulations, economic aspects, the physiology of bacterial and viral accumulation and elimination, and the characteristics of depuration systems. STATUS OF CONTROLLED PURIFICATION OF SHELLFISH IN THE SOUTHEASTERN UNITED STATES MARK E. BERRIGAN AND JOHN W. SCHNEIDER Florida Department of Natural Resources, Tallahassee, Florida 32303 Increased competition for resources in the coastal zone along both the Southeastern Atlantic and Gulf Coasts has resulted in diminished quality and quantity of shellfish growing waters. In many coastal regions where water quality and shellfish resources are threatened by contamination from microbial pollutants, alter- native technologies and processes have been implemented to pro- mote responsible utilization of these resources. Controlled purifi- cation provides a practical method for cleansing potentially con- taminated shellfish, insuring product quality, and protecting public health. Commercial purification operations and facilities vary between producing states and are contingent upon numerous controlling factors including species of shellfish, plant design specifications, water quality, microbiological criteria, and pro- cessing procedures. Each producing state is responsible for effec- tively regulating scheduled control purification processes (SCPP) according to recommended NSSP standards. The status of con- trolled purification processes, commercial operations, production levels, as well a a summary of guidelines regulating shellfish puri- fication in the Southeastern States are presented. National Shellfisheries Association, Seattle, Washington Abilracls. 1986 Annual Meeting. June 22 — 26. 1986 111 PREDATION ON SINGLE SPAT OYSTERS CRASSOSTREA VIRGINICA (GMELIN) BY BLUE CRABS CALLINECTES SAPIDUS AND MUD CRABS PANOPEUS HERBST/I MILNE-EDWARDS ROBERT BISKER AND MICHAEL CASTAGNA Virginia Institute of Marine Science, School of Marine Science, College of William and Mary, Wachapreague, Virginia 23480 POSSIBLE INFLUENCE OF WIND ON CANCER MAGISTER SETTLEMENT LOUIS W. BOTSFORD Department of Wildlife and Fisheries Biology , University of California, Davis, California 95616 JONATHAN M. SHENKER Bodega Marine Laboratorv, P.O. Box 247, Bodega Bay. California 94937 Single spat oysters Crassostrea virginica of four size classes (3.4-24.6 mm mean shell heights (SH)) were exposed to six size classes of blue crabs Callinectes sapidus (9.3-85.5 mm mean carapace width (CW)) and five size classes of mud crabs Pano- peus herbstii (7.1-34.4 mm mean CW) for two days. Crab pre- dation, recorded as the number of dead oyster spat/crab/day. was directly proportional to crab size and inversely proportional to oyster size. Mud crabs of 34.4 mm CW and blue crabs of 85.5 mm CW had predation rates of 22.5 and 16.7 spat/crab/day on oyster spat of 24.6 and 24.4 mm SH. respectively. Larger sized spat could be more readily preyed upon by mud crabs than by blue crabs of similar size. Mud crabs of 7. 1 and 25.2 mm CW caused significant mortalities to oyster spat of 8. 1 and 24.6 mm SH. re- spectively. Blue crabs of 9.3. 24.5 and 85,5 mm CW caused sig- nificant mortalities to oyster spat of 3.4, 13.9 and 24.6 mm SH, respectively. OYSTER BREEDING CAGES (OBCJ— THE "ABOVE GROUND" GROWING SYSTEM PIERRE BOCQUILLON Nortene, Inc., Norcross.Ga. 30091 STEVEN TALIS ADPI Enterprises, Inc., Philadelphia. Pa. 19134 The use of above ground cages to breed and farm oysters is a technique that has increased European oyster production by more than 30%. The concept — based on specially designed polyeth- ylene cages — has dramatically improved the economics of oyster farming. The cage system increases the productivity of farming labor, enhances oyster growth and reduces the risks of oyster loss. Three different style cages are used in the growing cycle. They are installed either on "breeding racks" or floated. Photos of typ- ical European oyster farms are shown. Catch of Cancer magisler along California. Oregon and Wash- ington fluctuates cyclically. One proposed cause is an influence of oceanographic conditions on the larval stages. Analysis of avail- able environmental data (upwelling index, surface temperature, sea level, and wind stress! showed a statistical relationship be- tween catch and southward wmd stress during the late larval pe- riod. The effect of wind on larvae depends on their vertical distri- bution. Preliminary sampling has indicated megalopae are neus- tonic at night and at low light levels, but distributed to at least 60 m in daylight. Implications of these results for a wind forced mechanism of larval recruitment are discussed. SCALLOP BREEDING STUDIES N. BOURNE Fisheries and Oceans Canada. Pacific Biological Station, Nanaimo. British Columbia, V9R 5K6 Scallop breeding studies were undertaken to determine the fea- sibility of raising juveniles in a hatchery for aquaculture purposes. Three species of native scallops: weathervane. Patinopecten caurinus: rock. Chlamys gigantea: and spiny. Chlamys hastata; and two exotics, Japanese. Patinopecten yessoensis: and sea. Pla- copecten magellanicus , have been investigated. Most of the work has been with Japanese scallops and results for this species are described in detail. In 1985 about 25% of fertilized eggs of Japa- nese scallops developed to the veliger stage; 60 million veliger larvae were produced. Survival from veliger to metamorphosis stages was about 18.5%; 11.2 million mature larvae were pro- duced . Larvae were fed single species or a mixture of five spwcies of algae; Isochrysis galbana, Tahitian Isochrysis: Chaetoceros calcitrans, Chaetoceros sp. A.R.C. variety, and Thalassiosira pseudonana. Larvae developed from fertilized egg to metamor- phosis in about 28 days when raised at 15°C. Several materials were used as cultch and five methods were used to settle meta- morphosing larvae and raise spat; upwellers, downwellers, static water, flowing water in tanks and raceways. The best cultch was 112 Abstracts. 1986 Annual Meeting. June 22 — 26. 1986 National Shellfisheries Association. Seattle, Washington "kinran"" and the best method to raise spat was in flowing water or in raceways. About 50% of mature larvae metamorphosed but heavy unexplained mortalities were experienced in spat when about 0.4-0.6 mm shell height. About 1 .500 juvenile Japanese scallops are being held in the natural environment to assess growth and mortality, the largest measure over 4 cm shell height. Results of studies with other species are described briefly. COMPARATIVE PHYSIOLOGY OF TWO COHORTS OF THE NORTHERN BAY SCALLOP, ARGOPECTEN IRRADIANS IRRADIANS V. MONICA BRICELJ, J. EPP AND R. E. MALOUF Marine Sciences Research Center, State University of New York at Stony- Brook. N.y. I J 794 Argopecten irradians undergoes rapid population decline in its second year of life. Mortality, growth and metabolic rates, and net growth efficiencies were compared for two cohorts of bay scallops held in cages between September 1984 and July 1985 at two sites in Long Island, N.Y. Mass natural mortality of second year scallops occurred in mid-winter (January through March), before the onset of a second reproductive period, and coincided with a period of minimal temperatures. At the site where milder environ- mental conditions prevailed, both cohorts maintained a positive energy balance throughout the fall, and experienced comparable tissue weight losses (9 and 11%) during the winter. The life span of Long Island bay scallops is thus estimated at 20 months. The main period of gametogenesis and gonadal buildup (April-May) did not coincide with the winter peak in phytoplankton abundance (Feb. -March). Over the temperature range rC-23°C, metabolic rates of both year classes closely paralleled seasonal changes in water tempera- ture. The latter explained a highly significant proportion (93%) of the seasonal variation in weight-specific oxygen consumption rate. An increase in oxygen uptake of first year olds was observed in conjunction with increased gametogenic activity in May. Meta- bolic rate during this period was about 50% higher than that pre- dicted based on temperature alone, providing an estimate of the metabolic cost of reproduction in this species. A HABITAT SUITABILITY INDEX MODEL FOR THE AQUACULTURE OF THE PACIFIC OYSTER, CRASSOSTREA GIGAS JAMES R. BROWN Department of Biological Sciences, Simon Eraser University, Burnaby. British Columbia. Canada, V5A IS6 A Habitat Suitability Index (HSI) model was developed in order to evaluate the suitability of coastal areas in British Co- lumbia for the aquaculture of Crassostrea gigas. In the model, the effects of abiotic and biotic factors upon oyster growth and mor- tality are quantified through the use of a relative index. Funda- mental to the model is the comparison of existing habitat condi- tions to the optimum conditions of the habitat variables for the oyster as described in the literature. The performance of the model was evaluated utilizing environ- mental and oyster production data collected from 10 field sites over a 14 month period. Site-specific HSI values derived from the environmental data were found to be significantly correlated with the increase in shell length of two groups of oysters which, over the course of the study, aged from 0- 14 months (r- = 0.82, p < 0.001) and 14-28 months (r^ = 0.88, p < 0.001). The use of the HSI model in the selection of sites for aquacul- ture operations and in the management of coastal areas will be discussed. BIOLOGY AND COMMERCIAL POTENTIAL OF GALATHEID CRABS IN BRITISH COLUMBIA BRENDA J. BURD Galatea Research, Box 202, Brentwood Bay, B.C. VOS lAO G. JAMIESON Shellfish Division. Pacific Biological Station. Nanaimo. B.C. V9R 5K6 The biology and commercial potential of the crab Munida qua- drispina were examined by surveying the prawn fishermen and processors and by collecting data during prawn fishing expedi- tions in B.C. Photographic data from a previous study was ana- lysed statistically to determine substrate preference. Pigment, lipid and protein contents were analysed to examine the potential for use of this spcies in fish food. Fishermen believed that M. quadrispina is increasing in abun- dance as prawn stocks decline. A representative cruise and the surveys indicated that catches of 600 or more pounds of the crabs per day might be expected in some areas, in addition to similar catches of prawn. Seasonal size frequency data indicate that growth rates of M. quadrispina are similar to other Pacific galatheid crab species. Females have a slower growth rate than males, and decline in abundance dramatically following reproduction in spring. Para- sitism by isopods varied between sample locations. Substrate pref- erence analyses indicated that M. quadrispina consistently prefer heterogeneous substrates (such as wood-fibre beds) to homoge- neous ones, which may be related to hiding behaviour. National Shellfisheries Association, Seattle, Washington Abslnuls. 1986 Annual Meeting, June 22 — 26, 1986 113 Proximate analysis indicated that carotenoid content is 2-3 times higher in M. quadrispina than in prawn or shrimp. Precent protein, lipid and pigment were higher in the waste portion than in the meat portion, suggesting a possible dual commercial use of tails for meat and waste portion for fish food. PRELIMINARY RESULTS FROM THE STUDY OF ACQUIRED IMMUNITY IN THE OYSTER, CRASSOSTREA VIRGINICA FU-LIN CHU Department of Estuarine and Coastal Ecology. Virginia Institute of Marine Science. School of Marine Science. The College of William and Mary. VA 23062 larvae at different developmental stages were prepared for scan- ning electron microscopy and histological examination to compli- ment live observations. Morphology of several key organs, valves, velum, gut, foot, and gill rudiment were examined. Struc- tures described for the first time include the interlocking crown- and-groove feature on denticles of the larval hinge region, loca- tion of secretory cells on the outer margin of the velum, and spe- cialized compound cilia at the mouth region. Growth and survival of the larval stage was examined when reared at 12, 16, 19, and 24°C, Larvae reared at 12°C had a 42% survival, reached maximum valve length of 238.9 ± 0.93 (xm and were capable of metamorphosis by 42 days, but were able to sur- vive as larvae for as long as 130 days. Larvae reared at 16°C had a 33% survival, were able to metamorphose by 34 days and sur- vived as larvae for as long as 115 days. Larvae reared at 19 and 24°C had slower growth and lower survival rates. Throughout de- velopment, valve length corresponded to valve height by a linear correlation (r- = 0.87) with a ratio of 1.1 ;1 for length to height. For the determination of the concentration of antigen which will elicit a cellular response in oysters, preliminary experiments have been performed by challenging oysters with various concen- trations of formalin-killed zoospores of Perkinsus mariniis. Re- sults indicate that a concentration of 0.07-2 x 10* formalin- killed zoospores could induce a cellular response (phagocytosis by hemocytes). The uptake of '''C-labeled zoospores of Perkinsus marinus. in vivo and in vitro, by hemocytes from immunized oysters was shown to be higher than by hemocytes from control (non-immunized) oysters. Short-term exposure of oysters to the living pathogen also elicit a similar cellular response. Initial studies reveal that the composition of hemolymph in immunized oysters is different from that of control oysters. LARVAL DEVELOPMENT OF THE SPINY SCALLOP, CHLAMYS HASTATA (SOWERBY) CHRISTINE A. COOKE Department of Biology, University of Victoria, Victoria. British Columbia, Canada V8W 2Y2 The early life history of the spiny scallop, Chlamys hastata. from gamete release through metamorphosis to a benthic juvenile, was observed and described. Ripe adults were induced to spawn by using a combination of UV-irradiated seawater and thermal stress. Newly released oocytes had a mean diameter of 71 jjim and were surounded by a thick jelly coat. Larvae were planktotrophic and capable of metamorphosing about five weeks after fertiliza- tion when reared at 16°C (240 (j-m in valve length). Embryos and THE POTENTIAL FOR DIRECT APPLICATION OF UNIVERSITY-DEVELOPED RESEARCH FINDINGS TO THE COMMERCIAL OYSTER INDUSTRY KEN COOPER Coast Oyster Company, P.O. Box 327. Quilcene. Washington 98376 Recent research findings appear to have potential for direct application to the oyster culture industry. The most promising re- search involves the production of polyploid oysters using cytocha- lasin B; the use of chemicals to trigger metamorphosis and/or at- tachment of competent oyster lavae; and microencapsulation to potentially enable the formulation of complete diets to supplant microalgae, but to more realistically enable the easy addition of supplements and/or specific agents such as antibiotics and hor- mones. Other research with potential for application includes the development of techniques for gynogenesis to enable the accelera- tion of genetic selection of oysters and genetic engineering. The direct application to commercial production-scale of findings developed at universities in small-scale laboratories is often not simple nor straight forward. Application to industry re- quires an intimate relationship between the university researchers who developed a concept and industry researchers who understand the requirements that must be met and limits to the application of the initial concept. In this presentation I shall use the above ex- amples in attempting to explain the complexities and potential pit- falls in developing a working relationship between universities and private industry. 114 Abstracts, 1986 Annual Meeting, June 22 — 26. 1986 National Shellfisheries Association. Seattle. Washington A SURVEY OF PERKINSUS MARINUS INFECTION IN THE GULF OF MEXICO M. ALISON CRAIG AND N. ERIC POWELL Department of Oceanography . Texas A&M University. College Station. Texas 77843 Perkinsus marinus is an important cause of mortality in oyster populations in the Gulf of Mexico. Incidence of the disease has been related to salinity and temperature, however local variations in disease incidence among neighboring reefs are also well de- scribed. Infection by Perkinsus is being monitored in oysters in connection with NCAA's Status and Trends ("mussel watch") program. Fifty locations along the Gulf coast from southern Texas to southern Florida were sampled between January and April. 1986. To assess within site variability, twenty oysters were col- lected from each of three stations at each of the fifty sites. Mantle tissue from each oyster was cultured in thioglycollate medium. Preliminary results of Perkinsus incidence and intensity at each site are reported. GROWTH RATE AND AGE STRUCTURE COMPARISONS OF GEOGRAPHICALLY ISOLATED HARD CLAM, MERCENARIA MERCENARIA, POPULATIONS LEE R. CROCKETT AND R. W. WHITLATCH Marine Sciences Institute. The University of Connecticut, Avery Point. Groton. Connecticut 06340 Samples were collected from five geographically isolated hard clam populations in eastern Long Island Sound. Age determina- tions were made by counting the internal annuli of transversally sectioned valves. The von Bertalanffy growth model was used to generate growth parameters for each population. The parameter w (the growth rate at to) was used for statistical comparisons between populations. A total of 896 clams were collected from the five sites and aged. Recruitment into each population occurred at generally low levels with strong year classes occurring aperiodically. Strong year classes were not found simultaneously at more than one site. This suggests that site specific factors, such as, differential preda- tion or settlement or both, may be governing recruitment. Strong year classes were rare and as few as 4 year classes dominated a single population. Growth rates, using the w parameter, were sig- nificantly different between all populations. Differences in water depth, and therefore temperature and food availability, are pro- posed to explain growth differences. The implications of these results are discussed with regard to management schemes. ENERGETICS OF STERILE TRIPLOID OYSTERS UNCOUPLE THE REPRODUCTIVE AND SOMATIC EFFORT OF DIPLOIDS JONATHAN P. DAVIS School of Fisheries. Utuversiry of Washington. Seattle, WA 98195 Discrete energy budget analyses of diploid and triploid oysters, Crassostrea gigas. were made under ambient conditions of tem- perature, salinity and seston levels during the period corre- sponding to peak reproductive condition in diploids. Results indi- cate that ripe yearling diploid oysters are in negative energy bal- ance while triploid siblings remain in a state of positive energy balance. Both reduced metabolic costs (measured as VO,) and nitrogen excretion (VNH,-N) in triploids specifically contribute to significant differences in the energy available for tissue produc- tion. Lower 0/N ratios in diploids suggest that germinal tissue production coupled with relatively warm water temperatures may contribute to a stress condition and negative energy balance at this time of year. Rates of consumption and efficiency of absorption were similar for diploids and triploids. Histological analyses of cross-sectional areas in diploid and triploid oysters demonstrate the virtual exclusion of gametes in female triploids and reduction in gametes in male triploids com- pared to the normal proliferation of gametes in sibling diploids. Sterile triploids provide a means of assessing the significant impact that seasonal reproductive cycles have on the physiology of bivalve molluscs and may be estimated in terms of reduced metabolic costs and increased somatic growth in sterile triploids serving as synchronous controls. Recent models of reproductive effort in invertebrates are discussd with reference to metabolic costs associated with reproduction. COMPARATIVE COASTAL ECOLOGY OF THE TROPICAL ROCKY-INTERTIDAL SNAIL CITTARIUM PICA IN THE EXUMA ISLANDS, BAHAMAS ADOLPHE O. DEBROT Rosenstiel School of Marine and Atmospheric Science. University of Miami. 4600 Rickenbacker Causeway. Miami, Florida 33149 National Shellfisheries Association, Seattle, Washington Absinicis. 1986 Annual Meeting, June 22— 26, 1986 115 The population ecology of Citlarium pica was studied on shores of low, intermediate and high wave exposure, using both population sampling (114 sites) and transplant experiments (14 sites). When compared to quiet sites, more exposed sites had higher population densities and highe densities of predators. At more ex- posed sites the snails showed higher rates of dispersal and mor- tality, and lower rates of growth. Dead shells from more exposed sites showed a higher proportion of lethal shell damage due to drilling predators. The decrease in mortality with increase in size was most pronounced at the more exposed sites. Size of matura- tion and relative fecundity were least at the more exposed sites. Coastal differences in population structure were consistent with coastal differences in growth and mortality, and were not ascribed to differences in recruitment pattern. The results suggest that low population densities at quiet sites are due to poor recruitment. OPTIMAL INDUCTION OF TRIPLOIDY IN CRASSOSTREA GIGAS DEPENDS ON TEMPERATURE SANDRA L. DOWNING School of Fisheries. WH-IO. University of Washington, Seattle. WA 98195 Egg lots from six mass spawns were treated using cytochalasin B (CB) from fertilization to past first cleavage at three different temperatures, 18, 20 and 25°C. Treatments of 1 mg CB/1 were applied at 0, 15, 30 and 120 min after fertilization for the lowest temperature, 18°C. After 15 minutes the eggs were then filtered and resuspended in a 0.1% DMSO bath for another 15 minutes. Replicates were run for most treatments. Control eggs were exposed to 0.1% DMSO for 15 minutes at the appropriate temperature and time. Triploid percentages, larval growth and survival rates were measured to determine the optimal treatment at each temperature. Large differences in survival to straight hinge were found among mass spawns. Some treated groups outperformed controls, but on average CB reduced larval survival during the first 48 hrs in all treated groups. This was most apparent during critical pe- riods of zygotic development (e.g., fertilization). After 48 hrs, survival rates were not significantly different among control and treatment groups. No significant differences in growth were found between treated and control groups from the same spawn and having sim- ilar densities. Replicates yielded similar percentages of triploids with stan- dard errors below 10%. Induction curves were derived for each temperature. These curves illustrate that lower temperatures pro- duced fewer triploids in their best treatments; highest percentages attained at 18, 20 and 25°C were 62, 74 and 88%, respectively. In addition, lowering the temperature delayed these maximum peaks; maxima at 18, 20 and 25°C are approximately 50. 45 and 30 min post-fertilization, respectively. Overall, the optimal treatment for inducing triploidy in the Pacific osyter (C. gigas) appears to be 30-45 min at 25°C which yielded 88-1-/- 9% (SE) triploidy over four replicates. THE IMPORTANCE OF INTERTIDAL SHELL AS NURSERY HABITAT FOR YOUNG-OF-THE-YEAR DUNGENESS CRAB IN GRAYS HARBOR, WASHINGTON BRETT R. DUMBAULD, DAVID A. ARMSTRONG, DONALD R. GUNDERSON AND A. ROSS BLACK School of Fisheries, University of Washington. Seattle, Washington 98195 Studies of juvenile Dungeness crab ecology and population dy- namics in Grays Harbor ( 1983-85) show that intertidal areas, par- ticularly those with an overlying shell substrate, play a critical role in the survival of newly recruited 0+ crab. Optimal habitat in Grays Harbor consists of shell depoits of the eastern softshell clam M>(7 arenana and live commercial oyster beds (Crassostrea gigas). Recruitment to the intertidal was monitored for 3 summers with an intensive study initiated in May 1985. Although initial settlement densities in May of 1983 were as high as 362 crabs/m^, numbers fell to much lower but relatively stable levels of 15-20 crabs/m- in June and 5-10 crabs/m- in July and August of all 3 years. Even at these densities, population estimates were much higher for 0-1- crabs in the intertidal subtidal areas where I -I- juveniles are prevalent. Crabs greater than 40 mm carapace width were rarely found m the intertidal indicating that they 1 ) physically outgrow the shell habitat; 2) leave due to agonistic behavior and displacement and/ or; 3) can no longer find suitable prey. This exodus from the in- tertidal to the subtidal in late summer may be the source of distinct increases in subtidal -I- populations in September and October. Movement from shell refuge may also indicate attainment of size refuge since crabs of 30-40 mm CW at this time are not nearly so vulnerable to predation as are small, early summer instars. DESTRUCTION OF BIVALVE MOLLUSC HINGE LIGAMENT BY CYTOPHAGA-LIKE BACTERIA: ASSOCIATION WITH MORTALITY IN HATCHERY-REARED JUVENILE PACIFIC OSYTERS, CRASSOSTREA GIGAS 116 Abstracts, 1986 Annual Meeting, June 22 — 26, 1986 National Shellfisheries Association, Seattle, Washington CHRISTOPHER F. DUNGAN AND RALPH A. ELSTON Center for Marine Disease Control. Marine Research Laboratory, Sequim, Washington 98382 Histopathoiogical examination of individuals from captive populations of juvenile Pacific oysters. Crassostrea gigas, experi- encing high mortality levels revealed the presence of severe de- generative bacterial lesions in the hinge ligament. The ligament lesions were associated with bacterial infections in mantle and connective tissues. Ultrastructure of infected hinge ligament dem- onstrated a morphologicaly distinct and homogeneous bacterial population at the eroding ligament surface. Bacterial isolations from hinge ligaments of juvenile oysters in high-mortality popula- tions yielded cytophaga-likc bacteria as the dominant flora. These isolates are morphologically identical to bacteria associated ultra- structurally with ligament destruction. Among the bacterial taxa isolated from hinge ligament, the cytophaga-likc isolates demon- strate a unique capability for in vitro proliferation using hinge lig- ament as the sole source of organic carbon and nitrogen. Coloni- zation of oyster resilium by these isolates results in liquifaction or loss of mechanical resiliency. Serological and biochemical tests suggest that these bacterial isolates belong in the genus Cytophaga and are previously undescribed. Fitness consequences of hinge ligament loss are discussed in light of ligament structure and func- tion. (69%), mantle (51%), foot (407f ). gill (31%) and adductor muscle (7%). Depuration of "Cr at average temperatures of 15°C for 35 days showed the digestive gland again to have the greatest percent de- puration (91%) followed by the mantle (77%). gill (76%r), foot (72%). adductor muscle (65%) and kidney (49%). In the process of depuration of heavy metals by clams several important factors are noteworthy; (1) depuration is a slow process even at optimal temperatures, the time period being months in- stead of days in the case of bacterial depuration. (2) different organs depurate at different rates depending on the metal in- volved, (3) clams shift the metal burden during depuration usually from the digestive gland and gills to the kidney for final elimina- tion, (4) the adductor muscle consistently showed the lowest levels of activity of all organs studied both during uptake as well as throughout the depuration process. BONAMIA OSTREA DISEASE OF THE EUROPEAN FLAT OYSTER {OSTREA EDULIS) IN NORTH AMERICA: OCCURRENCE, ENVIRONMENTAL EFFECTS AND HOST RANGE RALPH A. ELSTON Center for Marine Disease Control. Marine Research Laboratory. 439 West Sequim Bay Road. Sequim. Washington 98382 DEPURATION OF HEAVY METALS BY HARD CLAMS, MERCENARIA MERCENARIA ALBERT F. EBLE Department of Biology. Trenton Stale College. Trenton, N.J. 08625 Hard clams. Mercenaria mercenaria. were exposed separately to three isotopes, ""Cd. "Cr and ""'Zn for a period of 15 days after which the label was removed and depuration followed for one to several months. Depuration is a function of temperature and time but even at optimal temperatures (20°C) removal of metals by clams is slow compared to bacterial depuration. After 14 days of depurating '''Zinc, the digestive gland showed the highest per- centage of removing the metal (56%) followed by the gills (36%), foot (31%) and mantle (27%); several organs however, increased their concentration of metal; the adductor muscle ( 102%). gonad (114%) and the kidney, as expected, concentrated the metal the greatest (157%)). If 'o*Cd depuration is allowed to proceed for 45 days at optimal temperatures all organs show a decrease in metal content with the digestive gland having a 74% depuration followed by kidney The European flat oyster. Oslrea ediilis. is cultured in rela- tively small numbers in western North America. During 1985 and 1986. flat oysters from 12 locations in western North America were examined in the laboratory using a minimum 60-day ele- vated temperature regime, clinical, and histological methods. Oysters from three sites in Washington State, U.S.A., were dis- covered to be infected with a haplosporidian parasite of the ame- bocytes, identical in ultrastructure and disease manifestation to Bonamia ostreae. which is known to cause substantial oyster mor- talities in Europe. Laboratory studies showed that infected stocks of animals would manifest the terminal signs of the disease and die at I6°C but animals from the same stocks exhibited no clinicals signs of the disease or mortalities when held for up to six months at 8°C. In oysters which were sampled in June, prior to exposure to warming summer temperatures, the disease could not be detected by histo- logical methods. However, individuals from the same groups ex- hibited the disease when held for 45 days at 15°C. During a 5- month period of exposure of Crassostrea gigas and Ostrea lurida to Ostrea edulis known to be infected with Bonamia. the disease was not detected in these two species. The localization of the par- asite in amebocytes suggests that the animal's ability to dispose of invading infectious agents is severely impaired. Transmission of National Shellfisheries Association, Seattle, Washington Abstracts. 1986 Annual Meeting, June 22 — 26, 1986 117 the disease appears to be limited by environmental factors in the areas where it has been discovered. TRANSPLANT OF ABALONE IN BARKLEY SOUND, BRITISH COLUMBIA B. EMMETT Archipelago Marine Research, 4-1140 Fort St.. Victoria. B.C.. Canada V8V 3K8 G. S. JAMIESON Department of Fisheries and Oceans. Pacific Biological Station. Nanaimo. B.C.. Canada V9R 5K6 Biological and economic feasibilities of transplanting sub- legal-sized abalone iHaliotis kamtschatkana) from exposed beds to more sheltered, productive abalone habitat were investigated. After nine months, recovery and growth of transplanted, tagged abalone were significantly better than nontransplantcd. tagged controls. Recovery rates were 38% and 71% at the two replicate transplant sites. This difference was attributed to variation in both habitat composition and topography, which affected relative survey ease and success, and predator presence. There was little evidence of extensive emigration of transplanted abalone from the transplant sites. The study demonstrates that it is biologically feasible to trans- plant abalone 50 to 100 mm in length. Economic feasibility is dependent on recovery rates attained, which is quite site specific. The population dynamics of abalone in exposed beds and the long-term potential for enhancing juvenile abalone settlement in abalone-depleted areas by transplanting adult broodstock into them are two remaining elements which need investigation to es- tablish the overall biological merit of abalone transplants. ENERGY STORAGE AND UTILIZATION IN THE BAY SCALLOP, ARGOPECTEN IRRADIANS JENNIFER A. EPP Marine Sciences Research Center. State University of New York at Stony- Brook. N.Y. 11794 Energy storage/utilization and reproductive condition were monitored in first and second year Long Island scallops held in cages between September 1984 and July 1985. Condition indices and proximate composition were determined for each tissue com- ponent. Gonadal growth of first year scallops occurred in early spring at the expense of adductor muscle protein and lipid reserves. This contrasts with reported utilization of digestive gland reserves for reproduction in Massachusetts populations of the northern bay scallop. Carbohydrate utilization, commonly observed in other bi- valves, was not apparent. Histological analysis revealed that no residual, ripe oocytes remained in the gonads of older scallops by November. In March, first year and 80% of second year scallops were undergoing early gametogenesis. Twenty percent of the surviving, older cohort showed anomalous gonadal development, with the presence of ripe and resorbing oocytes. Mass senescent mortality of older scallops occurred before the period of gonadal buildup in early spring. This phenomenon does not appear to be associated with post-spawning energy depletion, nor with increased energy de- mand for a second reproductive event. A three week starvation experiment at 15°C in January resulted in 4 and 38% mortality of first and second year scallops respec- tively. The two age classes showed different responses to starva- tion stress. Second year individuals exhibited significantly greater depletion of gonadal and mantle reserves than first year scallops. EGG YOLK VESICLES AS A POTENTIAL FOOD SYSTEM FOR JUVENILE PACIFIC OYSTERS MARILYN C. ERICKSON and D. P. SELIVONCHICK Department of Food Science and Technology. Oregon State University, Corvallis. Oregon 97331 Vesicles prepared from egg yolk were shown to encapsulate protein and to be in a size range that would be filtered by the oyster. A radiotracer study involving addition of radiolabeled phosphatidylcholine to egg yolk demonstrated that the egg yolk vesicles were taken up and metabolized by juvenile Crassostrea gigas. Catabolism of the radiolabeled lipid and subsequent re- synthesis into non-lipid components occurred to a slight extent. The main factor responsible for the distribution of radioactivity amongst the lipids in the stomach tissue was believed to be trans- acylation. The use of aspartate transcarbamylase as a potential indicator of growth will also be discussed. REPRODUCTIVE BIOLOGY OF CLAM POPULATIONS IN NORTH AMERICA: A REVIEW ARNOLD G. EVERSOLE Department of Aquacidture . Fisheries and Wildlife. Clemson University. Clemson. South Carolina 29634-0362 118 Abstracts, 1986 Annual Meeting. June 22— 26, 1986 National Shellfisheries Association, Seattle, Washington Several generalizations can be made about those clam species which are cultured or have been recommended as candidates for aquaculture in North America. Most of the clam species are dioe- cious with Mercenaria menenaha. a consecutive hermaphrodite, being the most notable exception. Dioecious species produce her- maphrodites approximately 0.1% of the time and exhibit a 1:1 sex ratio in most populations. Greater percentages of hermaphrodites and unequal sex ratios are more frequently encountered in stressful environments. Male clams tend to mature at a smaller size and younger age than females, one reason the sex ratio favors males in young uniformly-aged populations. Most species of clams mature by 3 years of age and before they reach 25% of their maximum size. Spawning cycles vary with latitude and ambient water temperature. Clams spawn on an annual or semiannual cycle during the warmer months; only Tresus capax and T. nuttalU spawn at the seasonal minimum water temperatures. Clams must achieve a certain degree of ripeness before they can respond to a spawning stimulus, and the key factor appears to be the spring water temperatures when gametogenesis occurs. Gamctogenesis and spawning is orchestrated for maximum reproductive success. Fecundity is high and a costly expenditure of energy by clams. GAMETOGENIC DEVELOPMENT OF THE VENERID CLAM PROTOTHACA ASPERRIMA IN THE BAY OF PANAMA JOHN W. EWART AND MELBOURNE R. CARRIKER College of Marine Studies, University of Delaware . Lewes. Delaware 19958 JANZEL R. VILLALAZ, JUAN A. GOMEZ AND LUIS D'CROZ Centra de Ciencias del Mar y Limnologia. Universidad de Panama. Republica de Panama Little is known about reproductive cycles of bivalve molluscs in the Bay of Panama. Although local coastal upwelling during the dry season (January- April) appears to significantly influence ga- metogenesis. the effect of the rainy season (April -December) on annual reproductive cycles is not known. Gametogenic develop- ment of the venerid clam Protothaca asperrima was studied as part of an ongoing hatchery development project for the produc- tion of commercially important bivalves. Adult clams used as broodstock were suspended in Japanese lantern nets in the Bay of Panama and were sampled biweekly for a period of one year. Go- nadal development was determined histologically. Preliminary observations of tissue sections from specimens collected during the rainy season indicate that spenmaries remainin a ripened state and are continually replenished. Follicles were consistently observed to be either full or partially spent. Observa- tions of simultaneous spawning and germinal activity were char- acterized by the presence of prominent bands or zones of sperma- tocytes and spermatids on the periphery of partially depeleted fol- licles. Observations of ovaries during the rainy season indicate that female spawning activity is also continuous. In contrast to the male gonadal material studied, no evidence of fully ripened ovaries was found. Rather, follicles were observed to be partially, or in some cases, fully spent. Although gonadal replenishment or regeneration appeared to occur at a slower rate in females, there was extensive germinal cell activity with many oogonia observed forming on follicular membranes. HOST RESPONSE TO PROCTOECES MACULATUS INFECTION IN THE BLUE MUSSEL, MYTILUS EDULIS L. SUNG L. FENG Marine Sciences Institute. University of Connecticut. Groton. Connecticut 06340 Myiilus edulis from Ram Island Reef in Fishers Island Sound. Connecticut were sampled over a one year period. Prevalence of Proctoeces maculatus infection in and gonadal development of the host were determined histologically. The results showed that the development of P. maculatus was synchronized with the host re- productive cycle, that the mean prevalence of infection was 32.9 ± 3.50%' and that in moderate to heavy infections, normal game- togenesis was either impaired or totally absent in 6 to 11% of the infected mussels. In addition, subtle effects of the infection mani- fested as delaying the early stage of gametogenesis and sup- pressing the number of mussels reaching the mature stage were also revealed. Discharged sporosysts, free cercariae and adults elicited in- tense hemocytic infiltrations which were effective in destroying some of the parasite. Relevance of previously reported changes in the hemolymph biochemical constituents: carbohydrates, proteins and major free amino acids, will be discussed in terms of the infection and the gonadal development. Furthermore, ecological and genetic implications of the rediscovery of at least two distinc- tive reproductive patterns in Long Island Sound mussel popula- tions will also be presented. DEVELOPMENT OF ENZYME-LINKED IMMUNOSORBENT ASSAYS FOR DETECTION OF MOLLUSCAN PARASITES National Shellfisheries Association, Seattle. Washington Abstracts, 1986 Annual Meeting, June 22—26, 1986 119 ANTONIO J. FIGUERAS, SHEILA A. KANALEY AND SUSAN E. FORD Rutgers Shellfish Research Laboratory. NJAES. Rutgers University, Port Morris, NJ 08349 EUGENE M. BURRESON Virginia Institute of Marine Science, Gloucester Point. VA 23062 The detection of protozoan and metazoan parasites of commer- cially important molluscs is routinely done by expensive, time- consuming histological methods. A number of modern immuno- logic techniques are available, however, that might significantly reduce the time and expense of detecting parasites and, at the same time, provide valuable tools for basic research, including elucidation of life cycles. We report here preliminary results of a project to develop an enzyme-linked immunosorbent assay (ELISA) for detection of the oyster parasite Haplosporidium nelsoni (MSX). We have had moderate success in isolating parasites from the hemolymph of oysters with advanced infections using velocity sedimentation with Percoll gradients. This method takes advantage of the fact that, on the average, plasmodial stages of MSX are larger (mean diameter 20 (xm) than hemocytes (diameter 10 |j.m). Another source of (crude) antigen has been the blood and pericardial fluid of very heavily infected oysters (>10*' plasmodia/ml) with high parasite-to-hemocyte ratios, used without further enrichment of MSX. Anti-sera to crude antigen have been raised in rabbits and mice using standard methods and tested with both fluorescent and enzyme conjugates. MSX gives an intense reaction with the fluo- rescent conjugate, indicating that it is highly antigenic and fore- casting success with the use of more highly purified antigen to produce polyclonal antibodies or with the production of mono- clonal antibodies. EFFECTS OF MSX {HAPLOSPORIDIUM NELSONI) PARASITISM ON REPRODUCTION OF THE OYSTER. CRASSOSTREA VIRGINICA SUSAN E. FORD AND ANTONIO J. FIGUERAS Rutgers Shellfish Research Laboratory, NJAES. Rutgers University. Part N orris. NJ 08349 Besides causing heavy mortalities, the parasite Haplospori- dium nelsoni (MSX) may inflict considerable sublethal damage to its host, the oyster Crassostrea virginica. A potentially very se- rious sublethal effect is that on reproductive success of popula- tions in enzootic waters. We have analyzed tissue sections of nearly 2000 oysters collected during the oyster reproductive pe- riod in Delaware Bay, where MSX pressure is heavy. The object was to determine 1 . the extent to which MSX interferes with ga- metogenesis, 2. whether it differentially affects males and females or alters sex ratios, and 3. whether there is a difference between the effect of MSX on reproduction of mortality-resistant oysters compared to mortality-susceptible oysters. Results show that MSX parasitism significantly depresses ga- metogenesis in systeniically infected oysters, but not in those with infections confined to the gills. Inhibition is proportional to infec- tion severity, but affects males and females equally. There was evidence that infected oysters had a higher proportion of females than did uninfected oysters. Gametogenesis was significantly de- pressed in susceptible oysters with no patent infections and with advanced infections compared to mortality-resistant oysters in the same infection categories. There was no correlation between year- to-year fluctuations in MSX levels and spawning/setting patterns in Delaware Bay. COMPARATIVE ANALYSES OF LARVAL AND EARLY POST-LARVAL SHELL MORPHOLOGY IN SEVEN MYTILID SPECIES S. CYNTHIA FULLER Department of Zoology. Rutgers University, Piscatawax, New Jersey 08854 The shell morphology of larval and early post-larval specimens is compared in seven mytilid species indigenous to the Atlantic Coast of North America. Mussels were spawned in the laboratory, and the resulting larvae were cultured through the early juvenile stage. Ontogenetic changes in shell morphology were documented with scanning electron photomicrographs of disarticulated valves. Relationships of four quantitative features including the length and height of the shell, the number of provincular teeth, and the length of the provinculum vary among lar\'al valves of the seven species. The practical value of these features for the discrimina- tion of sympatric species is discussed. Examination of the lateral hinge system facilitates distinction among juvenile mussels. Post-larval mytilids may have zero. one. two, or three types of marginal teeth. Geukensia demissa and Amygdalum papyrium lack marginal teeth. Only dysodont teeth are found in Ischadium recurvum. and only primary lateral teeth are found in Modiolus modiolus. Mytilus edulis has secondary lat- eral and dysodont teeth, and Brachidontes exustus has all three types of postlarval, marginal teeth. 120 Absiracis. 1986 Annual Meeting, June 22—26, 1986 National Shellfisheries Association, Seattle, Washington STATUS OF COMMERCIAL SHELLFISH DEPURATION IN THE NORTHEAST-1986 SANTO A. FURFARI Department of Health and Human Services, North Kingstown. Rhode Island 02852 Currently, controlled purification (depuration) plants exist in three northeast states: Three soft clam plants and one hard clam plant in Maine; one soft clam plant in Massachusetts; and two soft clam and one hard clam plant in New Jersey. The plants range in age from four years to 58 years with six of the eight plants more than 10 years old. The plants use surface sea water, shallow salt water wells, and deep salt water wells. Some of the plants use flow through systems and some use recirculating systems. Each plant operates under authority given by state regulations. These plants were constructed under older regulations and concepts for the process, however few of the older concepts have changed since the plants were built. Because of the successful operation of the older commercial plants, there does not appear to be any great economic hardship. fifth site affected by strong tidal currents that at times exceed crit- ical entrainment velocities. These results support Peterson et al.'s hypothesis, but they also suggest that at some point current velocity becomes inhibitory, and (in support of earlier reports) that sediment characteristics may be a factor. Additional studies including manipulative exper- iments are ongoing. It is suggested that in future work on growth of suspension feeders the effects of food concentration and water flow (or velocity) be determined using calculations of FPR, which may estimate the net effect of concurrent changes in both flow and food concentration. A BIOENERGETIC MODEL OF JUVENILE DUNGENESS CRAB {CANCER MAGISTER) POPULATION DYNAMICS IN GRAYS HARBOR, WASHINGTON F. BRANDT GUTERMUTH AND DAVID ARMSTRONG School of Fisheries. University of Washington, Seattle. Washington 98195 PRELIMINARY STUDIES ON THE EFFECTS OF TIDAL CURRENTS, FOOD CONCENTRATION, AND SEDIMENT CHARACTERISTICS ON ONTOGENETIC GROWTH OF MERCENARIA MERCENARIA RAYMOND E. GRIZZLE Center for Coastal and Environmental Studies. Rutgers University. New Brunswick. New Jersey 08903 Water currents, food, and sediments are major environmental factors that affect growth rates of hard clams [Mercenaria mer- cenaria), but their relative effects are not well understood. Pe- terson et al. (J. Mar. Res. 42, 123-138, 1984) hypothesized that current velocity and food concentration interact so that the net effect depends on their relative strengths. The relation of sediment characteristics to these factors is not known. Preliminary results from students at four sites in a coastal la- goon in southern New Jersey show that average growth rates of wild clams determined by examining annual bands in sectioned shells were (1) not correlated with near-bottom food concentra- tions estimated in 1985 by chlorophyll a and particulate organic matter; (2) positively correlated with near-bottom tidal current ve- locities, and "food provision rates" [FPR = current velocity converted to flow (e.g. 1/s) x food concentration (e.g. mg/1) = biomass per unit time (e.g. mg/s)]; and (3) negatively correlated with sediment organic content. The slowest growth rates were at a A bioenergetic model of juvenile Dungeness crab population dynamics has been developed to estimate assimilated energy re- quired by a resident spring/summer population in Grays Harbor, a major west coast estuary. Principle parameters of the model are respiration at several temperatures (laboratory), population abun- dance and growth (both determined in the field). Population respiration values were computed from crab weight/temperature relationships (0.03 to 30g; 6° to 18°C). The mean respiration for 10 mm carapace width (CW) intervals was determined based on average bottom temperature between two week sampling periods, and multiplied by the biomass calculated for that size interval throughout the estuary. Growth estimates were determined for four strata (areas) of Grays Harbor and for three moving age/size classes: 0-1- (6-59 mm CW), 1 -I- (30- 1 15 mmCW), and >H- (77-160 mmCW). Juvenile population abundance ranged from less than 1 million to over 20 million crab in constantly changing proportions of age class (size) and, therefore, biomass. The growth calculation in- cludes energy as somatic growth and also energy lost as exuvia over each time interval. Population respiration and growth, as Kcal/ha, were summed over the estuarine strata and through the spring/summer in a cumulative estimate of energy assimilated by the juvenile crabs of Grays Harbor. ASSESSMENT OF MORTALITIES IN SURF CLAMS (SPISULA SOLIDISSIMA) DUE TO DREDGING, SORTING AND DISCARD National Shellfisheries Association, Seattle. Washington Ahsimcts. 1986 Annual Meeting. June 22—26, 1986 121 HAROLD H. HASKIN AND ERIC WAGNER Rutgers Shellfish Research Laboratory'. NJAES. Rutgers University. Port Norris. NJ 08349 A significant poilion of tlie clams available in the Mid Atlantic region are less than the current minimum legal size limit. Har- vesters must either sort their catch and discard undersize clams, or modify their catch and discard undersize clams, or modify their gear so undersize clams will not be caught. Current harvesting practices cause significant mortalities to clams dredged, sorted, and returned to the sea. To assess mortality associated with sorting and discard, clams were dredged and run through a mechanical sorter. Post sort "catch" (larger clams) and "discards" (smaller clams) were transplanted to marked plots at nearby areas. Plots were sampled with a hydraulic dredge and SCUBA divers 1, 24, 48, 72, and 144 hours after planting. Samples were sorted to determine percent mortality. To evaluate efficiency and effects of "bottom sorting" ie. increasing bar spacings in the dredge to allow smaller clams to pass through, divers sampled both inside and outside the path of a bottom sorting dredge fished through a dense population of under- size clams. Results of this study show that 1) with careful han- dling, minimal mortality to clams captured in a hydraulic clam dredge will average about 17-18% 2) sorting the dredged catch by steel rollers (current practice) adds another 18- 197c kill 3) additional stress e.g. holding on deck, shovelling overboard etc. can add another 17-18% mortality. Predators increased in abun- dance and diversity in planting areas. A single tow evaluation of bottom sorting in this study confirmed high mortality rates re- ported by others in clams left behind in the dredge path (62% this study). This is a NJAES Publication No. K-32503-1-86, sup- ported by federal funds. THE DISTRIBUTION OF LARVAE OF THE BLUE MUSSEL MYTILVS EDULIS LINNE IN THREE MAINE ESTUARIES HERBERT HIDU Department of Animal ami Veterinary Science. Ira Darling Center, University of Maine. Walpole. Maine 04573 GREG PODNIESINSKI Department of Zoology . Ira Darling Center. University of Maine, Walpole, Maine 04573 Water column abundance of blue mussel larvae as it related to tidal stage was investigated in three widely differing estuarine systems in Maine. A relatively wide coastal embayment, Webb Cove on Deer Isle, showed negligible differential position of larvae with tidal stage. The Damariscotta River, a 19-mile long, narrow drowned-river-mouth estuary, consistently produced en- hancement of Mytilus (and Mya arenaria Linne) larvae by a factor of 3 to 5 on the flood tide. Minimum salinity (less than \7fc) and temperature (less than 1°C) changes occurred over the tidal cycle. A 3-mile long, narrow inlet, the Jordan River, in contrast, pro- duced enhancement of larval and pelagic juvenile mussels by a factor of 35 on the flood tide. The effects of basin morphometry as it alters hydrography, particularly current velocity with resultant effects on larval distribution, is discussed. THE SPATIAL DISTRIBUTION OF DUNGENESS CRAB (CANCER MAGISTER DANA) MEGALOPAE OFF THE WEST COAST OF VANCOUVER ISLAND, CANADA GLEN S. JAMIESON AND A. C. PHILLIPS Department of Fisheries and Oceans. Pacific Biological Station. Ncmaimo. British Columbia, Canada V9R 5K6 Relative abundance and spatial distribution of dungeness crab (Cancer magister) megalopae was monitored between March and August, 1985, along a transect line extending 180 km seaward off the west coast of Vancouver Island. Sampling for megalopae and newly settled crabs was also carried out in bays and estuaries around Tofino, B.C., from May through September, 1985. Mega- lopae was first collected offshore in April, reached peak abun- dance in June, and remained present to the end of sampling in late August. Neuston samples provided the best estimate of megalopae abundance. Inshore sampling indicated little crab settlement during 1985. Surface currents in the area of the transect line were northwest within 35 km offshore, but were predominantly south- west over the remainder of the continental shelf and slope. The boundary of these two currents shows properties of a convergence zone. Relatively large numbers of megalopae were found seaward of this boundary. This oceanographic regime is known to break down under certain meteorological conditions resulting in trans- port of surface water inshore. Hypotheses involving both large- scale and local oceanographic and meteorological events are pro- posed to explain the distribution of crab larvae and their settlement magnitude in inshore waters. THE FORAGING BEHAVIOUR OF CANCER MAGISTER FEEDING ON PROTOTHACA STAMINEA: SIZE SELECTION AND RISK 122 Abstracts, 1986 Annual Meeting, June 22—26, 1986 National Shellfisheries Association, Seattle, Washington FRANCIS JUANES Department of Biological Sciences Simon Fraser Universir\\ Burnaby, British Columbia. Canada V5A IS6 The foraging behaviour of the Dungeness crab. Cancer ma- gister. was studied using optimal foraging theory. A model was constructed to predict optimal sizes of prey (.Prototliaca staminea) from values of energy content of the prey, energetic cost of feeding, and handling time. The model predicted that the larger sizes of prey were the most profitable. The predators preferred the smaller sizes of clams, considered unprofitable. These results can be explained with the use of energetic efficiency ratios (benefits/ costs) and as a trade-off to minimize claw wear and the risk of claw breakage. PUBLIC FISHERY AND PRIVATE MARICULTURE CONFLICT IN LONG ISLAND, N.Y.'s SHELLFISH INDUSTRY JEFFREY KASSNER Environmental Protection , Town of Brookhaven. Patchogue. N.Y. 11772 Long Island, N.Y.'s coastal waters support several commercial shellfish resources and are believed to have a high but unrealized potential for private shellfish mariculture. While many factors constrain mariculture development, the primary obstacle is the in- ability of private ventures to obtain exclusive use of suitable un- derwater levels: baymen who harvest natural shellfish stocks strongly oppose private mariculture and have successfully blocked every request made to or by government agencies for allocations of public lands for private mariculture use. Until this impasse is resolved, private mariculture will be severely limited. The baymen have considerable political power and public sup- port so that private mariculture will require at least their passive support. The benefits of private mariculture are considerable — it could be an important fishery management option while providing baymen with alternative employment and supplemental income. However, simply projecting benefits is insufficient to win baymen support. It is therefore first necessary to understand the basis of their opposition and then develop a private mariculture policy that addresses their concerns. Baymen oppose private mariculture fearing loss of traditional freedoms, competitive disadvantages, and displacement by big business as well as a distrust of government regulators. Past and some current practices reveal these to be legitimate concerns. An acceptable mariculture program would therefore need the fol- lowing attributes: accessibility by baymen and individuals, pro- tection of the natural fishery limited scale, and strict oversight. These are not insurmountable and there is no reason why private mariculture cannot exist with commercial shellfishermen. ASPECTS OF THE LIFE HISTORY OF THE PEA CRAB, PINNOTHERES MACULATUS SHANNON KELLY Marine Sciences Institute, The University of Connecticut, Groton, Connecticut 06340 In samples of the mussel. Mytilus edulis, collected over a fif- teen month period from Ram Island Reef in eastern Long Island Sound. 98% contained the pea crab. Pinnotheres maculatus. Males and females were frequently found together in the same mussel except during summer months when females are ovig- erous. A higher percentage of females than males was always ob- served. To study the effect of tidal height on pea crabs two stocks of mussels of different sizes were used. The larger group, ranging between 6-8 cm in length, had a 98% prevalence of crabs while the other, ranging between 4-6 cm had a 10% prevalence. From September 1985 through January 1986, mussels were placed in cages and hung one foot off the bottom, one foot from the surface of mean high water (12 ft), and half-way between the two at the end of a dock. The results indicate an apparent attrition of crabs from mussels held at the high intertidal level. The larger mussels in the two deeper cages maintained the high prevalence of crabs. However, the smaller mussels experienced an increase in crab prevalence at the deeper depths in October, November and De- cember and a sharp decrease in January. The apparent attrition of the crab is probably due to the fact that the smaller mussels were either physically or nutritionally unable to accomodate and sup- port the growing crab when environmental conditions were unfa- vorable. REPRODUCTIVE DEVELOPMENT IN THREE MERCENARIA MERCENARIA STOCKS GROWN IN SOUTH CAROLINA WATERS RICHARD S. KNAUB AND ARNOLD G. EVERSOLE Department of Aquaculture, Fisheries and Wildlife, Clemson University, Clemson, South Carolina 29634-0362 JOHN J. MANZI Marine Resources Research Institute. Charleston, South Carolina 29412 National Shellfisheries Association, Seattle, Wasliington Abslracls, 1986 Annual Meeting, June 22 — 26, 1986 123 Juvenile Menenaria mercenaria from Aquaculture Research Corporation in Dennis, Massachusetts (ARC), South Carolina wildstock (SCW) and a cross of the two were planted from July to October 1983 at a mean shell length of 8 mm. Beginning in July 1984, clams from each stock were collected monthly for a 14- month period. Clams were measured and gonads sectioned for histological examination. Binary coding based on developmental categories permitted analysis of variance by general linear model. Seventy-five percent of the clams in all three stocks had reached the differentiated stage by December 1984. Significant differences (p < .05) among spawning peaks were detected for the stocks and the interaction of month and stock. The ARC stock showed dis- tinct spawning peaks in August and April, with a smaller peak in December while the SCW stock spawned only in July and March. Offspring of the ARC a SCW cross exhibited at intermediate spawning pattern. UTILIZATION OF REFRACTORY CARBON BY THE RIBBED MUSSEL, GEUKENSIA DEMISSA (DILLWYN) DANIEL A. KREEGER College of Marine Studies, University of Delaware. Lewes. Delaware 19958 CHRISTOPHER J. LANGDON Hatfield Marine Science Center. Oregon State Universin-, Newport, Oregon 97365 ROGER I. E. NEWELL Horn Point Environmental Laboratories. University of Maiyland. Cambridge, Maiyland 21613 The ribbed mussel, Geukensia demissa, is a common inhabi- tant of the intertidal region of east coast salt marshes. The ability of G. demissa to incorporate cellulosic carbon prepared from Spartina alterniflora was investigated with '"C radiotracer tech- niques. Spartina alterniflora was grown in an atmosphere enriched with '"COt. The refractory cellulosic component was then chemi- cally extracted from the harvested plants. G. demissa were main- tained in a 6 hour immersed; 6 hour emersed simulated tidal cycle for a 30 hour period. While immersed the mussels received marsh water, together with its natural complement of particulate mate- rial. The '''C-cellulose was added to the marsh water only during the first 6 hours. Filtration, fecal deposition, pseudofecal deposi- tion, respiration, and final body burden of '"C were measured to assess how readily the maierial was digested and incorporated. The significance of cellulosic carbon in the marsh environment will be discussed. USE OF DOWNWELLING CHAMBERS IN STUDIES WITH BIVALVE MOLLUSCS CHRIS J. LANGDON Department of Fisheries and Wildlife, Oregon Stale University-. Corvallis, Oregon 97331 J. EVAN WARD College of Marine Studies, Universin- of Delaware, Lewes, Delaware 19958 Flow-through chambers have been used by many researchers to study the feeding behaviour of bivalve mollusc species. The approach has many advantages over static systems. Resuspen- sion of feces is avoided, particle concentrations can be main- tained, low bacterial concentrations in the chamber can be achieved using filtered seawater, the animals can be rapidly ex- posed to various stimuli. We have designed and constructed a downwelling flow-through system for use in feeding studies with the oyster [Crassostrea virginica) and the mussel [Mytilus edulis). Experiments indicated that the flow-through apparatus was especially useful in determining the effects of the ectoparasite Boonea impressa on the filtration rate of C. virginica because the experiments could be run for several days without the need to disturb the animals. Experiments have also been carried out to test the effects of dissolved substances from cultures of various algal species on the filtration rate of M. edulis. Downwelling chambers have also been used in growth experi- ments with C. virginica fed on microencapsulated, artificial diets. Concentrations of bacteria were lower but growth of oysters poorer in downwelling chambers compared with those of animals in tlask cultures. DEPURATION: POLICY AND PRACTICE ON THE WEST COAST JACK L. LILJA Department of Social and Health Services, State of Washington, Olympia, Washington 98504 Recent closures of productive shellfish growing areas in some west coast states have generated an increased dialogue between the shellfish industry and state officials regarding depuration. Al- though there has been an increased interest in depuration, a major focus has been on the problems associated with the depuration alternative. There arc currently no active depuration operations on the west coast and only California and Hawaii have had operations in the past. 124 Abstracts. 1986 Annual Meeting, June 22—26. 1986 National Shellfisheries Association, Seattle. Washington The absence of depuration plants on the west coast is due more to industry resistance and the lack of need rather than state policy. Industry resistance is b