MBL/WHOI JOURNAL OF SHELLFISH RESEARCH VOLUME 12, NUMBER 1 JUNE 1993 Wonrts Hoi. Oceanoflr«phlc Inttitulion Library P 2 6 1994 Hole, MA 02543 The Journal of Shellfish Research (formerly Proceedings of the National Shellfisheries Association) is the ofTicial 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 Dr. Standish K. Allen, Jr. (1993) Rutgers University Haskin Laboratory for Shellfish Research P.O. Box 687 Port Norris, New Jersey 08349 Dr. Neil Bourne (1994) Fisheries and Oceans Pacific Biological Station Nanaimo, British Columbia Canada V9R 5K6 Dr. Andrew Brand (1994) University of Liverpool Marine Biological Station Port Erin, Isle of Man Dr. Monica Bricelj (1994) Marine Sciences Research Center State University of New York Stony Brook, New York 1 1794-5000 Dr. Alan Campbell (1994) Fisheries and Oceans Pacific Biological Station Nanaimo, British Columbia Canada V9R 5K6 Dr. Peter Cook (1994) Department of Zoology University of Cape Town Rondebosch 7700 Cape Town, South Africa EDITORIAL BOARD Dr. Robert Elner (1994) Canadian Wildlife Service Pacific and Yukon Region 5421 Robertson Road P.O. Box 340 Delta, British Columbia Canada V4K 3Y3 Dr. Ralph Elston (1993) Battelle Northwest Marine Sciences Laboratory 439 West Sequim Bay Road Sequim, Washington 98382 Dr. Susan Ford (1993) Rutgers University Haskin Laboratory for Shellfish Research P.O. Box 687 Port Norris, New Jersey 08349 Dr. Jonathan Grant (1994) Department of Oceanography Dalhousie University Halifax. Nova Scotia Canada B3H4J1 Dr. Paul A. Haefner, Jr. (1994) Rochester Institute of Technology Rochester, New York 14623 Dr. Robert E. Hillman (1994) Battelle Ocean Sciences New England Marine Research Laboratory Duxbury, Massachusetts 02332 Dr. Lew Incze (1994) Bigelow Laboratory for Ocean Science McKown Point West Boothbay Harbor, Maine 04575 Dr. Roger Mann (1994) Virginia Institute of Marine Science Gloucester Point, Virginia 23062 Dr. Islay D. Marsden (1994) Department of Zoology Canterbury University Christchurch, New Zealand Dr. Roger Newell (1994) Horn Point Environmental Laboratories University of Maryland Cambridge, Maryland 21613 Dr. A. J. Paul (1994) Institute of Marine Science University of Alaska Seward Marine Center P.O. Box 730 Seward, Alaska 99664 Journal of Shellfish Research Volume 12, Number 1 ISSN: 00775711 June 1993 >^M 5Y^y Journal of Shellfish Reseiirch. Vol 12. No. 1, 1-7. I99.V EFFECT OF SALINITY ON INFECTION PROGRESSION AND PATHOGENICITY OF PERKINSUS MARINUS IN THE EASTERN OYSTER, CRASSOSTREA VIRGIN IC A (GMELIN) LISA M. RAGONE AND EUGENE M Virginia Institute of Marine Science School of Marine Science The College of William and Mary- Gloucester Point. Virginia 23062 BURRESON^.;arineBfoloflical Laboratory/ ^oods Hole Oceanoflraphic InstiMon Library SEP 2 6 1994 ABSTRACT The effect of salinity on Perkinsus mannus, a protozoan pathogen of the eastern oyster, Crassostrea virginica (Gmelin .MA 02543 1791) was investigated. Oysters parasitized by P. marinus were exposed in the laboratory to 6, 9, 12, and 20 ppt at a temperature ranging from 20-25°C. for an eight week period. Infection prevalence and intensity were assessed in samples (n = 25) from each treatment following 2, 4. 6, and 8 weeks of exposure and oyster mortality was determined daily. The pathogen persisted, at high prevalences, throughout the course of the experiment at all treatment salinities; however. P marinus infection development was retarded at 12 ppt and did not progress at 6 and 9 ppt. Cumulative oyster mortalities progressively increased with increasing salmity and at the termination of the expenment were 9.1, 11. 6, 21.1, and 27.8 percent at 6. 9. 12, and 20 ppt, respectively. A cntical range for parasite pathogenicity apparently exists between 9 and 12 ppt. Although P. marinus is able to tolerate salinities as low as 6 ppt it is less virulent at salinities below 9 ppt. KEY WORDS: Perkinsus. oyster, salinity INTRODUCTION During the past three decades commercial oyster landings in Virginia have declined from an average of 3.5 million bushels per year prior to 1960. to a record low 0.1 million bushels in 1990- 1991 (Virginia Marine Resource Commission landings data). This decline has been attributed to over fishing, declining water quality and disease (Hargis and Haven 1988). Factors that lead to disease epizootics in marine organisms are extremely complex and include biotic and abiotic parameters (Thorson 1969, Rohde 1982). For osmoconformers. such as the eastern oyster. Crassostrea virgi- nica. salinity plays a major role in modulating its association with disease organisms (Hepper 1955, Bayne et al. 1978, Gauthier et al. 1990). Generally, oyster parasites have a narrower salinity tolerance than their host and are more common in high salinity areas (Hopkins 1956, Wells 1961, Andrews 1964, Farley 1975, Ford and Haskin 1982, Andrews 1983, Gauthier et al. 1990). Low salinity exposure (< 10-15 ppt) often reduces the occurrence and the virulence of disease organisms. In the last decade salinity increases in Virginia's upper estuaries, resulting from four con- secutive drought years (1985-1988), have caused an intensifica- tion of Perkinsus marinus (commonly known as Dermo), one of the Chesapeake Bay's most problematic oyster pathogens (Burre- son and Andrews 1988, Burreson 1989). In response to increasing salinities in upper bay waters the parasite has spread to previously disease free seed areas and has had a severe impact on the oyster resource and industry. A more thorough understanding of the in- fluence of salinity on the relationship between P . marinus and the eastern oyster will help elucidate the annual variability in the dis- tribution and pathogenicity of this parasite and allow resource managers and oyster growers to forecast and perhaps avoid disease epizootics. The influence of salinity on the activity of P. marinus has been the focus of numerous studies. Several investigators have docu- mented a positive correlation between salinity and P. marinus infection intensity through field surveys (Mackin 1951, Mackin 1956, Andrews and Hewatt 1957. Somat 1985, Craig et al. 1989, Gauthier et al. 1990, Crosby and Roberts 1990. Powell et al. 1992). Oysters grown in high salinity areas (15-30 ppt) experi- enced higher disease prevalence than those grown at lower salin- ities {<15 ppt). It has been suggested that the correlation between disease level and salinity is not a result of a limiting physiological effect on host or parasite but rather is due to the dilution of infec- tive elements by freshwater inflow into the estuary (Mackin 1956. Ray 1954. Andrews and Hewatt 1957); however, disease devel- opment was retarded and oyster mortality was suppressed in in- fected oysters that were transplanted to a low salinity site (1-13 ppt) in the James River, Virginia (Andrews and Hewatt 1957) suggesting that salinity may have some physiological effect on the parasite. Few laboratory studies investigating the effect of salinity on P. marinus have been conducted. Ray (1954) investigated the com- parative development time of P . mannus in artificially infected oysters maintained at high (26-28 ppt) and low (10-13.5 ppt) salinity in closed aquana. The parasite tolerated the low salinity treatment; however, development of infection and subsequent mortalities of oysters were delayed relative to the high salinity group. Similarly. Scott et al. (1985) found lower mortality in oysters held at 8-10 ppt than in oysters held at 21-25 ppt. Inhi- bition of P. marinus zoosporulation by low salinity (5-10 ppt) has been documented in vitro studies conducted by Perkins ( 1966) and Chu and Greene (1989). Studies to date have greatly enhanced our understanding of the influence of salinity on the relationship between C. virginica and P. marinus: however, further investigations under controlled con- ditions are needed to substantiate and elaborate current knowl- edge. The experiment reported here investigated the effect of low salinity exposure (6. 9. and 12 ppt) on established infections of P. marinus. MATERIALS AND METHODS Approximately 900 oysters (60-1 10 mm) were collected 5 May 1989 from Deep Water Shoal, the uppermost natural oyster reef in the James River. Virginia. Immediately following collection the oysters were placed in two trays and suspended from a pier at the Virginia Institute of Marine Science in the lower York River, Virginia. The oysters remained at this location until mid Septem- Ragone and Burreson ber 1989, during which time they acquired P . marinus infections. The mean daily salinity at the site during the exposure period ranged from XA-ll ppt and the mean daily water temperature ranged from 19-27°C. On 14 September 1989 the oysters were transferred to the laboratory and cleaned of fouling organisms. Three replicate samples (n = 25) were analyzed for P . marinus intensity and prevalence. Oysters serving as uninfected controls were collected 8 Sep- tember 1989 from Ross" Rock located in the upper Rappahannock River, Virginia. At the time of collection a sample of 25 oysters was examined for P. marinus prevalence and intensity. The laboratory portion of the experiment was conducted at the Virginia Institute of Marine Science Eastern Shore Laboratory, Wachapreague, Virginia. Oysters were randomly assigned to one of four salinity treatments; a high salinity (20 ppt) control treat- ment and three low salinity treatments, 12, 9, and 6 ppt. Five replicate, 50 liter, polypropylene tanks, each containing 30 oysters were established at each salinity treatment. One tank containing 30 uninfected control oysters was also established at each salinity treatment. All oysters were conditioned to salinity treatments so that no greater than a 5 ppt change in salinity was experienced in a 24 hour period. Water of the desired treatment salinity was prepared daily by diluting filtered sea water (pumped from Finnery Creek) with fresh well water in 44 gallon plastic containers. The sea water was filtered through a series of filters including a 25 micron bag filter, two sand filters containing sand and activated carbon, and a 1 micron bag filter. Filtration removed seston, ensuring that food availability did not vary between treatments, and reduced the pos- sibility of exposure to P . marinus and other oyster parasites which may have been present in influent water. Aquaria water was aer- ated and changed daily. Mean daily water temperature was 23.3°C (±1.9 s.d.). Oysters were fed a commercial algal diet (Diet A, Coast Oyster Co., WA) daily. An aliquot of the algal mix (2.5 ml) was diluted with 250 mis of filtered sea water and added to each aquaria. The food source was adequate on the basis of feces and pseudofeces production by most individuals and by the overall condition of uninfected oysters and oysters with low level infections (i.e. firm and opaquely colored tissue and well developed gonads). The experiment was conducted for a period of eight weeks. Oyster mortality was recorded daily. All gaping oysters were re- moved from aquaria and examined for disease organisms. Random samples of live oysters, five from each replicate tank, were taken from each treatment group on day 14, 28, 42, and 56. The oysters were shucked and P. marinus prevalence and intensity were de- termined using thioglycollate culture of rectal, gill, and mantle tissue; infection intensities were rated as negative, light, moderate, and heavy (Ray 1952). Diagnosis of other oyster parasites [Hap- losporidium nelsoni (Haskin, Stauber, and Mackin), Bucephalus cuculus McGrady and Nematopsis ostrearum Prytherch] was by routine paraffin histology of tissue fixed in Davidson's AFA. Par- asite intensity and prevalence in the control groups were evaluated only at the termination of the experiment. On day 28 of the experiment, 25 oysters from each low salinity (6, 9. and 12 ppt) treatment group were transferred to 20 ppt in order to determine if infections that may have become subpatent would reappear upon exposure to high salinity. Mortality of the transferred oysters was followed daily for the remainder of the experiment and at the termination of the experiment all live oysters were analyzed for P . marinus and other parasites. Control oysters were not treated similarly; hence, we did not have an appropriate control to assess the solitary effect of the salinity change on the survival of oysters transferred from the low to high salinity con- ditions. Cumulative mortality was determined for each treatment rep- licate on day 14, 28, 42. and 56. In order to adjust for samples removed from each replicate, mortality was calculated as follows. Interval mortality, mortality occurring between sample dates (day 1-14. 15-28, 29-^1. 43-56), was determined for each replicate group by dividing the number of oysters dying during an interval by the number of oysters that were alive at the beginning of the interval. Interval mortality was then multiplied by the proportion of survivors of the previous interval (1 -cumulative mortality of preceding interval) to yield the adjusted interval mortality. Suc- cessive cumulative mortalities were then determined by summing adjusted interval mortalities and preceding cumulative mortalities. Differences in mean cumulative mortality and mean prevalence between treatment groups and through time were determined by a two factor analysis of variance (ANOVA). Differences in mean cumulative mortality and mean prevalence between treatment groups on each sample date and on data collapsed across time were determined by a one factor ANOVA. When significant differences were found a Student-Newman-Keuls (SNK) test for multiple comparison among means was performed (Zar 1984). Prior to analysis the dependent variable was arcsine transformed and eval- uated for compliance to the test assumptions. Normality was ex- amined using a Komogorov-Smimov goodness of fit test and ho- moscedasticity was evaluated with a Cochrans C test (Sokal and Rohlf 1981, Zar 1984). A hierarchical log-linear test (log-likelihood ratio test) was uti- lized to detect differences between salinity treatments and through time in the distribution of oysters within the four P. marinus intensity categories (Sokal and Rohlf 1981). All tests were judged significant at an alpha level of 0.05. Computations were made on a Prime computer using a SPSSX statistical package. RESULTS The mean prevalence of P. marinus in oysters sampled at the initiation of the experiment was 80% (±8% s.d.) and infection intensity did not vary greatly between replicates (Figs. 1 and 2). Prevalence of P. marinus in oysters sampled from treatment groups on day 14, 28, 42, and 56 ranged from 76% to 100% (Fig. 1). A two factor analysis of variance indicated that the effect of salinity on prevalence was significant (P = 0.031), while the effect of time and the interaction of salinity and time were not significant (P = 0.285 and P = 0.915 respectively). Prevalence, however, did not significantly differ among treatment groups on any sample date (day 14 P = 0.3910. day 28 P = 0.9446, day 42 P = 0.1752, day 56 P = 0.1538). A significant difference in mean prevalence between treatment groups was observed when data was collapsed across time (P = 0.0235). A SNK test revealed a significant difference only between 9 ppt and 12 ppt treatments. Perkinsus marinus was not detected in control oysters sampled at the initiation of the experiment; however, the parasite was present at low prevalences in live control oysters sampled at the termina- tion of the experiment (0% at 20 ppt, 4% at 12 ppt. 12% at 9 ppt, and 0% at 6 ppt). The effect of salinity on infection intensity was significant (P = 0.0338). Oysters maintained at 6 and 9 ppt had a higher total number of negative and light infections and a lower total number Effect of Salinity on Phrkinsus 120 100- CD O c Q > c CO CD □ 6ppt n 9ppt B 12ppt ■ 20ppt Figure 1. Mean prevalence (±1 standard deviation) of P. marinus in oysters sampled from each treatment group at the initiation of the experiment (day 0) and after 14, 28. 42, and 56 days of exposure to treatment salinities. Day mean prevalence is based on three samples of 25 oysters, all other means are based on five replicate samples of 5 oysters. of moderate and heavy infections than oysters held at 12 ppt and 20 ppt (Fig. 3). On day 14 there were relatively large differences between treatments in the number of light and heavy infections (Fig. 2). Differences in infection intensity between treatment groups were not as great as the experiment progressed and the number of oysters within each infection category (negative, light, moderate, and heavy) did not significantly differ through time (P = 0.0624). The interactive effect due to salinity and time was not significant (P = 0.7087). Despite the high prevalence of P. marinus at all four salinity treatments a marked difference in mortality was observed. Mean cumulative mortality progressively increased with increasing sa- linity (Fig. 4). Mean cumulative mortalities at 6. 9, 12. and 20 ppt were respectively: 0.7%, 2.0%, 2.0%. and 7. 3% on day 14; 1.5%. 6.8%, 10.1%, and 17.7% on day 28: 2.5%, 8.8%. 16.4%, 21.1%' on day 42: and 9.1%, 11.6%, 21.1%, 27.8% on day 56. The effects of salinity and time on cumulative mortality were highly significant (P < 0.0001 ) while the interactive effect of salinity and time was not significant (P = 0.8907). Treatment means signifi- cantly differed on days 14 and 28 (P < 0.0281 and P < 0.0037, respectively) but did not significantly differ on days 14 and 56 (P < 0.0956 and P < 0.0607) (Fig. 4). Oysters transferred to high salinity, 20 ppt, following a 28 day low salinity treatment experienced a much higher mortality rate than those remaining continuously at the original treatment salin- ity. The mortality began soon after the transfer and continued until the termination of the experiment (Fig. 5). Mortality of the uninfected control oysters was as follows: 4% at 20 ppt, 12% at 12 ppt, 4% at 9 ppt, and 0% at 6 ppt. Three of the five dead control oysters were infected by P. marinus. All three infections were light. Histological analysis revealed the presence of H. nelsoni, B. cuculus. and N. osirearum m 3%, 4%, and 20%, respectively, of the total number of live oysters sampled. In general, as the exper- iment progressed the prevalence of all three parasites declined (Table 1). Prevalence of H. nelsoni at 6 and 9 ppt declined from an initial mean prevalence of 12% to 0% within the first 14 days of the investigation and remained below 4% for the remainder of the experiment. Haplosporidium nelsoni was present in only 3 of 73 gaping oysters that were examined histologically. In agreement with thioglycoUate cultures, P. marinus was present in 100% of the dead oysters examined histologically. Ninety percent of the dead oysters had moderate to heavy P. marinus infections. DISCUSSION Previous investigations have indicated that low salinity sup- presses oyster mortality caused by P. marinus (Andrews and Hewatt 1957. Ray 1954. Scott et al. 1985). This investigation substantiates their results and further extends our understanding of this relationship by defining 9-12 ppt as a critical range for P. marinus activity. Oyster mortality at 6 and 9 ppt was reduced by more than 50% compared to oysters maintained at 12 and 20 ppt. At the end of the experiment mean cumulative mortality of oysters at 6 ppt was 67% lower than at 20 ppt. Additionally, oyster mor- tality was delayed at 6. 9. and 12 ppt relative to 20 ppt. Oysters exposed to 20 ppt began dying soon after the initiation of the experiment and continued to die through the duration of the ex- periment. An abundance of advanced infections in the dead oysters at 20 ppt indicates that infections were progressing during the course of the study. The pattern at 12 ppt was similar although the onset of mortality was slightly delayed relative to the 20 ppt group. Mortality of oysters at 6 and 9 ppt primarily occurred during the final two weeks of the experiment, presumably as a result of advanced infections which were present at the start of the exper- iment. Ragone and Burreson H M ■ H ABC 6 9 12 20 14 6 9 12 20 6 9 12 20 6 9 12 20 Treatment Salinity (ppt) 28 42 56 Day Figure 2. Perkinsus marinus infection intensity (H = heavy, M = moderate, and L = light) in oysters sampled at the initiation of the experiment (day Ol and after 14, 28, 42, and 56 days exposure to treatment salinities (6, 9, 12, and 20 ppt). Day replicates are designated as A, B, and C. Sample size for the 12 ppt treatment group on day 56 was 20, all other samples consisted of 25 oysters. Enhanced survival was not a permanent attribute of oysters exposed to low salinity. When transferred to high salinity, the oysters died at a relatively high rate compared to those continu- ously held at their original salinity. The sharp increase in mortality most likely reflects increased multiplication of the parasite in re- sponse to more favorable conditions. It is also possible that the change in salinity may have created additional stress thereby in- creasing mortality of oysters which had already been weakened by disease. Although exposure of infected oysters to low salinity reduced oyster mortality, a concomitant decrease in P . marinus prevalence was not observed. Unlike H. nelsoni. which is readily eliminated from the oyster after a two week exposure to salinities less than 10 ppt (Ford 1985), P. marinus, once established in the eastern oys- ter, can tolerate salinities as low as 6 ppt for a period of at least 56 days at temperatures exceeding 20°C. Infection intensities were also indicative of a lack of P . mari- nus expulsion. Had low salinity induced expulsion, a coincident decline in parasite intensity would have been observed in sampled oysters. A striking decrease in parasite intensity was not observed at any treatment; however, low salinity did prevent, or at least delay development of infections to pathogenic levels. Infections at 6 and 9 ppt did not significantly change during the experiment while infections at 12 and 20 ppt progressed and caused mortality within the first few weeks. Advanced infections were more nu- merous in oysters maintained at 12 and 20 ppt than in oysters held at 6 and 9 ppt. Statistical analysis suggest that infection intensity did not significantly change through time at any treatment. How- ever, it is important to note that as the experiment progressed the number of oysters sampled from the high salinity groups having advanced infections is obscured by the high mortality of oysters having advanced infections. Many oysters from the 12 and 20 ppt groups perished early in the experiment, as a result of moderate to heavy infections, and were not included in subsequent samples. Hence, the actual number of advanced infections at 12 and 20 ppt is not reflected in the statistical analysis. Development of P. mari- nus in the Ross' Rock "uninfected" control oysters may be at- tributed to infections which were present but undetectable at the initiation of the experiment. Perl' {Crassostrea virginica) is primaiily a leased-based industry (Keithly et al. 1992). More than 300 thou- sand acres are currently under lease for oyster production, and poundage taken from these leases averaged about 9.2 million pounds annually during the 1980s (National Marine Fisheries Ser- vice, unpublished data). In addition to the private grounds. Louisiana also maintains considerable acreage devoted to public seed grounds and oyster seed ground reservations. These public grounds include the most productive natural reef area east of the Mississippi River (see Fig. 1), and encompass some 896 thousand acres in total (Perret et al. 1991). Vermillion Bay. west of the Mississippi River, is also a major public seed ground. The state periodically seeds these public grounds and the fishermen, in the fall season, will transport and bed the seed on their individual leases. Leases used for the purpose of producing oysters are rented from the state on an annual basis for a duration of 15 years. The annual rental rate was one dollar per acre prior to 1980, and in subsequent years, two dollars per acre. Procedures for a lease cancellation in the event of default of rental payments are de- scribed in La.R.S. 56:429; "The failure of the tenant to pay the rent punctually on or before the first of each January, or within thirty days thereafter, ipso facto and without demand or putting in default, terminates and cancels the lease and forfeits to the depart- ment all the works, improvements, betterments, and oysters on the leased water bottoms. The department may at once enter on the water bottoms and take possession thereof. Such water bottoms shall then be open for lease to the highest bidder." While provisions of La.R.S. 56:429 have been law for more than 30 years, the non-payment situation was minimal prior to the early to mid 1980s and. hence, no auctions were held prior to this period. By the mid 1980s, however, the delinquent payment sit- uation had escalated to the point of mandating an auction as pro- vided by statute. Some of the reasons speculated for the increased delinquent payment situation were: ( 1 ) the annual rental fee dou- bled from one dollar to two dollars per acre in 1980, (2) there was a shift away from the once traditional oyster producing areas, making leases in these areas no longer productive (see Van Sickle et al. 1976). (3) the decline in the oil-and-gas related activities in Louisiana's coastal area during the early to mid 1980s resulted in reducing oyster damage compensation payments, and (4) the sharp decline in Louisiana's coastal economy during the early-to-mid 1980s, tied to the decline in oil-and-gas related activities, resulted in reduced income which could be devoted to rental payments for nonproductive oyster acreage. Interestingly, the lease default sit- uation escalated during a period of above average production from private leases in Louisiana and record real value of production. Annual production from private leases during the 1980s, for in- stance, was 675 thousand pounds above that reported in the 1970s, or about 8%. and because of a substantial increase in the real dockside price of Louisiana produced oysters, the real value of production during the 1980s was almost 40% above that reported during the 1970s (Keithly et al. 1992). The first public auction in reference to La.R.S. 56:429 was held in April 1987, and an auction was held in March or April each year thereafter. Since the first auction and through the most recent auction held in March 1992, a total of 764 leases were offered. This paper provides an analysis of the auction data from 1987- 1992. RESULTS Physical Characteristics Number of Leases Offered and Taken As noted. 764 leases were offered at auction between 1987 and 1992. or about 10% of the total number of leases under ownership 10 Keithly et al. Figure 1. Louisiana Public Seed Grounds. as of mid- 1992.' As indicated in Figure 2, with the exception of the first auction in 1987, the number of leases offered for auction increased during each year in which the auction was held. The number of leases included in this first auction, however, is mis- leading to the extent that it included cancellations dating back to 1975. Overall, 136 of the 164 leases put up for auction in 1987 were canceled during the 1980s, with 21 of the cancellations oc- curring in 1986 and another 57 in 1987." In all but the 1987 auction, the number of leases auctioned reflected the number can- celed in that year. A particularly sharp increase in leases offered for auction was observed in the latest two years of analysis with the 242 leases offered in 1992 representing three times the number reported as recently as 1990. The extremely sharp rise in 1992 may be related to the $19.2 million 1991 dockside value of Louisiana's oyster harvest which was the lowest since 1983 when evaluated in current dollars and the lowest since 1981 when adjusted for inflation. This decline in value came despite a sharp increase in leased acreage during the 1980s (see Keithly et al. 1992). Leases opened for auction had minimum bid requirements of two dollars per acre per year in default plus interest. Thus, for example, a ten acre lease defaulted upon in 1985 and opened for auction in 1987 required a minimum bid of sixty dollars plus accrued interest.^ Annual auctions were held since 1988. Leases 'The number of leases and total acreage devoted to oyster production in Louisiana changes daily as new leases are Issued or combined with ex- isting leases. The number of leases and acreage lease size discussed in this paper are approximations as of June 1992. "Prior to 1986, no more than ten leases were canceled in any one year, except for 1980. In 1980, Ihe year In which rental feeds increased from $1.00 to $2.00 per year, there were .^0 cancellations. 'An exception to the minimum bid requirement Involved those leases de- faulted upon prior to 1980. Among this small group of leases, minimum bid requirement was one dollar per acre per year through 1979 and two dollars per year thereafter, plus accrued interest. auctioned since 1988 required minimum bids of $2.20 per acre ($2.00 rental plus $0.20 interest). Almost 70% of the 764 leases offered for auction during the study period, or 524 in total, were taken. The remaining 240 leases did not receive bids. On a yearly basis, the percentage of leases taken ranged from less than 45% in 1989, when 33 of the 78 leases offered were taken, to more than 80% in 1988, when 53 of the 66 offered leases were taken (Fig. 2). With the exception of 1989, the percentage of leases taken at auction has declined in each year since 1988. In 1988, for instance, 80.3% of the leases offered at auction were taken. The percentage fell to 76.5% in 1990, 73.7% in 1991, and 66.5% in 1992. The abnormally low proportion of leases taken at the 1989 auction (42%) reflects, as discussed be- low, the abnormally large size of leases canceled that year. 1987 1988 19S9 1990 1991 1992 Year of Auction Figure 2. Number of Leases Offered and Taken at Annual LDWF Auctions. 1987-92. Louisiana's Oyster Lease Auctions 11 120 1987 1988 1989 1990 1991 1992 Year of Auction Figure i. Average Number of Acres Among Leases Offered and Taken at Annual LDWF Auctions, 1987-92. Size of Leases Offered and Taken Average size of leases offered at auction during the study pe- riod was 59 acres compared to about 33 acres (as of mid 1992) for the industry in total. As indicated in Figure 3. the average size ranged from a low of 40 acres in 1992 to a high of 1 12 acres in 1989. Total acres offered at auction during the study period equalled almost 45 thousand, or about 13% of total acreage leased. ■* On average, 7,470 acres were offered each year at auc- tion, with a range from 3,454 to 10,972 (Fig. 4). The average size of leases taken at auction was consistently smaller than among those offered (Fig. 3). In 1987, for instance, size of leases taken averaged 5 1 acres compared to 67 acres among leases offered. The average size of leases taken compared to leases offered ranged from 50% ( 1991 ) to 85% ( 1989) and averaged 73% during the six year period. The proportion of acreage taken at auction ranged from 36.0% in 1989, when 3,148 of the 8,755 acres offered at auction were taken, to a high of 55.8% in 1990, when 2,087 of the 3,737 acres offered were taken (Fig. 4). The lease size, as documented in Table 1 , is an important determinant as to whether a particular lease is taken at auction. During 1987-92, for example, 564 leases, or 73.8% of the total number of leases offered at auction, were =£50 acres. Among these 564 leases, 427 or 75.7% received at least the minimum bid. Of the 94 leases ranging from 51 to 100 acres in size, 57.5% were taken. Only 41 .9% of the leases 101-500 acres in size were taken while less than a third of the leases >500 acres offered at auction were taken. The negative relationship between the size of lease and its probability of being taken at auction is, in all likelihood, a function of ( 1 ) the amount of unproductive acreage associated with larger leases and (2) the risk of large monetary losses that would be incurred if a large lease proves to be unproductive. With respect to the first point, larger leases may have relatively few productive areas. Hence, potential investors may be unwilling to pay the required delinquency fees for relatively few productive acres. With respect to the second point, there is a certain amount of risk, or uncertainty, related to bidding on acreage of which little or nothing is known. As such, there is likely to be reluctance in bidding on the larger leases, vis-a-vis smaller leases. The transfer of leases between willing parties also exhibited the small lease phenomena. The average size of leases transferred during 1980-89 was 46.5 acres (Keithly et al. 1992). Area Of the 45 thousand acres offered for auction during 1987-92, approximately 23% (10,467) was located in Plaquemines Parish (see Fig. 1). By comparison, Plaquemines Parish accounts for about 40% of the state's total leased acreage (128 thousand acres) and 4.8 million pounds of the state's 9.2 million pound annual production from private leases during the 1980s. Of the 10,467 acres offered at auction which were in Plaquemines Parish, 7,232 (69%) were subsequently taken. This proportion is considerably higher than that reported for the state in total (48%). St. Bernard Parish based leases represent less than a quarter of Louisiana's total leased acreage. However, this parish accounted for 24.9 thousand of the 44.8 thousand acres (56%) offered for auction during 1987-92. Furthermore, only 8.7 thousand of these 24.9 thousand acres, or 35.1%, were taken at auction, i.e., re- ceived at least minimum bid. This is well below the state average. It is evident from these figures that St. Bernard leases face a challenging production environment. Terrebonne Parish, with 15%' of the state's leased acreage leases (51.8 thousand acres), had only three percent of the acreage offered at auction during 1987-92. The remaining acreage opened at auction included almost three thousand acres in both Iberia and Vermillion Parishes, 2.3 thousand acres in Lafourche Parish, and a lesser number of acres in Jefferson , St . Tammany , St . Mary , and Orleans parishes. The relatively large number of leases canceled in St. Bernard parish is consistent with observed activities in Louisiana's oyster industry. For example, while there has been a large decline in the productivity per leased acre throughout Louisiana's oyster industry during the past three decades, this decline has been especially apparent in St. Bernard Parish. Estimated production per leased 12.0 0.0 ■'Some of the acreage may have been auctioned more than once. Due to the change in the number assigned to each lease, it is impossible to determine the extent of this. 1987 1988 1989 1990 1991 1992 Year of Auction Figure 4. Total Number of Acres Offered and Taken at Annual LDWF Auctions, 1987-92. 12 Keithly et al. TABLE 1. Selected information pertaining to Louisiana Department of Wildlife and Fisheries water bottom lease auctions by size of lease, 1987-92 average. Percent of Sales Price of Sales Price Size of Leases Leases Leases Taken Leases Offered of Leases Lease (acres) Offered Taken ($/acre) ($/acre) Taken «50 564 427 75.7 10.63 15.23 51-100 94 54 57.5 6.01 10.22 101-500 93 39 41.9 1.56 4.08 >500 13 4 30.8 1.03 3.01 Source: Compiled form unpublished data mamtamed by Louisiana Department of Wildlife and Fisheries, Oyster Division. acre at the state level, for instance, fell from 108 pounds annually during 1960-64 to 30 pounds during 1985-89, a decline of almost 75%. In St. Bernard Parish, however, the decline was almost 90%, from 70 pounds per acre to only eight pounds per acre. In Plaquemines Parish the decline was only about 60% (96 pounds per acre to 37 pounds per acre) which approximated that observed in Terrebonne Parish (97 pounds to 40 pounds). St. Bernard's close proximity to the state's largest public seed grounds was evidently of no advantage (see Fig. 1). As noted by Perret et al. (1991), the effectiveness of the public seed grounds in the St. Bernard area has declined by some 60 to 65 percent through time as a result of salt water intrusion, accelerated by the Mississippi River Gulf Outlet (MRGO). This channel, a man-made deep-water structure built in the 1960s, is approximately 75 miles in length and connects the open Gulf of Mexico waters to the Port of New Orleans. It traverses St. Bernard Parish and as noted by Dugas (1979). resulted in pronounced salinity changes in the MRGO surrounding areas upon its completion. These salinity changes destroyed many productive oyster beds in the area and. according to Dugas. led to an inland shift of the oyster growing area. Among other things, this inland shift resulted in the growing areas being closer to domestic sources of pollution which has resulted in pe- riodic and permanent closures by health officials. Auction Prices The average sales price of leases taken at auction varied sub- stantially on an annual basis, as indicated in Figure 5. To some extent, this variation is commensurate with differences in average lease sizes from year to year. In 1989, for instance, when leases taken averaged a record 95 acres, the average bid per lease taken was a record $577. Since 1989, however, leases taken have con- sistently averaged from 32 to 34 acres each, yet the sales price has ranged from $157 to $378. When evaluated on a per acre basis, the sales price of leases taken at auction varied from a low of $4.66 per acre to a high of $1 1 .92 per acre (Fig. 6). Overall, the average sales price of leases taken at auction declined steadily from 1987 through 1990 but increased considerably in the subsequent two years. When leases not taken at auction are included in the analysis, the average sales price per acre declined considerably due to the large proportion of leases not receiving at least the minimum bid (Fig. 7). The average price per acre, however, has increased stead- ily since 1989 with the 1992 sales price of $6.32 per acre being more than 75%' above that reported in 1991 and almost three times that reported in 1989. As indicated in Table 1 . there exists a strong negative relation- ship between the size of a lease and its per acre sales price. This relationship was also apparent among leases transferred (see Keithly et al. 1992). Leases taken at auction which were «50 acres, for example, received $15.23 per acre compared to $10.22 among leases of 51-100 acres, $4.08 among leases 101-500 acres, and $3.01 among leases in excess of 500 acres. In other words, leases of s50 acres taken at auction went for about five times as much as those >500 acres when evaluated on a per acre basis. $600.00 -F so.oo 1987 1988 1989 1990 1991 1992 Y«ar of Auction Figure 5. Average Bid Per Lease Taken at LDWF Auctions. 1987-92. LDWF Auctions, 1987-92. 1987 1988 1989 1990 1991 1992 Y«ar of Auction Figure 6. Average Bid Per Acre Among Leases Taken at Annual Louisiana's Oyster Lease Auctions 13 $7.00 $6.00 $5.00 e u < t $4.00 a> a a a $3.00 o Q $2.00 $1.00 $0.00 Avarago Bid Donerad $6.32 »S.09 1967 1988 1989 1990 1991 1992 Year of Auction Figure 7. Average Bid Price Per Acre Among Leases Offered at An- nual LDVVF Auctions. 1987-92. When leases not receiving the minimum bid are included in the analysis, the average sales price of leases «50 acres was more than ten times that of leases >500 acres. This increase reflects the greater proportion of larger leases not receiving minimum bid requirements. Among leases taken at auction, those in Terrebonne Parish received the highest per acre bid ($23.89). This was followed by Lafourche Parish ($14.83), Plaquemines Parish ($11 .93), and Jef- ferson Parish ($10.85). Lowest per acre bids among leases taken were received in St. Bernard Parish ($4.43), Iberia Parish ($3.34), and St. Mary Parish ($2.99). The relatively low price observed for auctioned leases in St. Bernard Parish is consistent with deterio- ration of leases in that area. DISCUSSION Several salient features were highlighted by the analysis of Louisiana's oyster lease auction data. One of these features is that a considerable proportion of Louisiana's leased acreage, about 10%- 1 5% of the total, was valued at less than two dollars per acre among those individuals and companies who voluntarily relin- quished their claim to the property by failure to pay annual rental fees. Of the almost 45 thousand acres that reverted back to the state for nonpayment of rental fees, however, about 45% was subsequently taken at auction. The cancellation and subsequent purchase of oyster producing grounds at auction raises the issue of why this occurs. There are at least three plausible answers to this question. First, speculation likely plays a major role in any decision to relinquish property rights and/or to rent additional property. As noted by Perret and Chatry (1988) "Itlishermen not only lease areas which are cur- rently productive, but they also hold leases in areas which may become productive as salinity conditions change." As such, what may be considered an exceedingly high risk by one individual (company), i.e., he relinquishes his rights to a given lease, may be considered an acceptable risk by another individual (company). The lease holder, however, evidently can not identify prospective buyers of the lease, since the cost of searching for prospective buyers is exceedingly high relative to potential benefits. A transfer via sale not being possible, the lease holder cancels the lease. The lease is then subsequently offered at auction. A second explanation to the issue of cancellation and subse- quent purchase of a given lease at auction relates to location . There are multiple lease holders in Louisiana. They are only subject to a one-thousand acre maximum. In some instances individuals (com- panies) may possess marginally productive leases far away from their major producing leases. It may not be profitable for these individuals (companies) to manage and harvest these leases due to the long distance which would need to be travelled. Other indi- viduals (companies) who control leases in close proximity to the lease may, however, find such activities profitable. As such, it may be relinquished by the original owner only to be purchased at auction by an individual who maintains other leases in its general vicinity. A direct transfer may not occur because of lack of knowl- edge about availability or failure to agree on terms. A final explanation for the cancellation and subsequent reissu- ing of oyster leases at auction relates to issue of absentee owner- ship in Louisiana's oyster industry. It is generally recognized that some of Louisiana's oyster lease holders are not active participants in the oyster industry and, as such, do not have current information on the productivity of all the leases under their ownership. Thus, they may relinquish certain leases they believe to be no longer productive. Active participants, recognizing the productivity in the area, may then purchase these leases at auction. A second feature highlighted by the analysis reflects the ob- served increase in the number of leases being canceled through time and the related decline in the proportion of these leases sub- sequently being taken at auction. Both of these situations suggest continued deterioration in Louisiana's oyster lease-based busi- nesses. Another feature gleaned from the analysis reflects the low de- mand for the larger leases, vis-a-vis smaller leases, when evalu- ated on a per acre basis. As noted, this lower demand likely reflects increased monetary risk associated with the purchase of larger leases. Finally, the analysis indicated that leases canceled, at least to some extent, were related to areas of declining productivity. This was found to be particularly the case in St. Bernard Parish. ACKNOWLEDGMENTS Partial support of this research was provided by the National Marine Fisheries Service, United States Department of Com- merce, through MARFIN Contract # NA90AA-H-MF092. LITERATURE CITED Dugas, R.J. 1979. Some observations on the post-construction effects of the Mississippi River Gulf Outlet on Louisiana oyster production. La. Dept. Wild, and Fish. Tech. Bull. No. 28:1-15. Keithly, W. R. Jr., K. J. Roberts & D. Brannan. 1992, Oyster Lease Transfers and Lending; Roles in Rehabilitation of Louisiana's Oyster Industry. J. Shellfish Res. 11(1): 125-13 1. Perrett. W S . R, J. Dugas & M. F, Chatry 1991, Louisiana Oyster: enhancing the Resource Through Shell Planting, World Aquacuhure 22(4):42^5. Van Sickle. V,, B, Barrett, T, Ford & L, Gulick, 1976, Barataria Basin: Salinity Changes and Oyster Distribution, Louisiana Sea Grant Publi- cation No. LSU-T-76-02. Journal of Shellfish Resi'iirch. Vol. 12. No. 1. 15-19. IW.V ESTIMATION OF OYSTER SHELL SURFACE AREA USING REGRESSION EQUATIONS DERIVED FROM ALUMINUM FOIL MOLDS' REINALDO MORALES-ALAMO The ColU'i;e of William and Mary Virginia Institute of Marine Science School of Marine Science Gloucester Point. Virginia 23062 ABSTRACT A method is described for estimation of surface area of shells of the American oyster. Crassoslrea virginica (Gmelin IV^l). as an alternative to direct measurement of surface area with aluminum foil molds. It is based on computation, from a small sample of shells, of the equation for regression of area of aluminum foil molds of shells on area enclosed within tracings of the shell outline. Area of other shells is then predicted from their shell outlme area using the equation. Accuracy of the regression method in spatfall studies was established using data from shellstring collectors suspended in the Piankatank River. Virginia. For the most part, differences between foil mold area of individual shellstring shells and the area predicted from regression equations were small, and spat densities on individual shells, as computed from foil mold area and from regression-predicted area, were almost identical. KEY WORDS: Crassoslrea virginica. larval settlement, spatfall, oyster shells, surface area, aluminum foil INTRODUCTION Quantitative field studies of settlement of oyster lai^ae (spat- fall) on shell cultch of the same species is hampered by difficulty in measurement of shell suif ace area ( Butler 1 954 ). For that reason settlement data have been presented most frequently as number of spat per shell or per oyster (e.g., Singarajah 1980, Haven and Fritz 1985, Morales-Alamo and Mann 1990, Adams et al. 1991); those data, however, cannot be compared with each other or with other data because they lack shell dimensions. Some investigators have estimated shell surface area from the dimensions of the longer and shorter axes of the shell (Lunz 1954, Carreon 1973), from the weight of paper cutouts of shells (Mc- Nulty 1953) or from shell height (Galtsoff 1964. Marcus et al. 1989). Those methods, however, failed to account for shell shape and texture. Other investigators avoided the problem by using alternate materials with flat surfaces and square comers (e.g., Kennedy 1980. Osman et al. 1989. Kenny et al. 1990). Healy ( 1 99 1 ) made direct surface area measurements of oysters using aluminum foil molds that accounted for shell shape and surface texture. Foil had been previously used to measure surface area of corals (Marsh 1970), stones (Shelly 1979). and the bivalve mollusc DoncLx serra (Donn 1990). Whereas Donn (1990) and Healy ( 1991 ) prepared foil molds of each animal in their studies, a technique is presented here for estimation of the surface area of shells of Crassoslrea virginica (Gmelin 1791) that reduces time and tediousness because it does not require a foil mold of every shell examined. Shell surface area is predicted from the equation for regression of actual (foil mold) area on the area enclosed by a tracing of the shell perimeter outline; Marcus et al. (1989) mea- sured the area within the shell perimeter outline to validate their area estimates but apparently did not consider shell shape and texture. MATERIALS AND METHODS Source of Oyster Shells Area measurements using aluminum foil were made on random samples of C. virginica shells from a natural oyster reef in the 'Contribution No. 1788. Virginia Institute of Manne Science. School of Marine Science. The College of William and Mary. James River, Virginia (referred to as reef shells), and from refuse piles at local oyster-packing houses (house shells). Regression equations were computed for three samples; a 1983 sample of 48 mixed reef and house shells, a 1983 sample of 68 reef shells, and a 1990 sample of 80 house shells. Attached organisms were scraped off reef shell surfaces before foil molds were made. The 1990 sample of house shells came from stock used to construct shellstrings deployed in the Piankatank River, Virginia, as part of a spatfall monitoring program (Barber 1990), and the equation derived from those shells was used to predict surface area of shellstring shells. Shellstring collectors were described by Ha- ven and Fritz (1985). Foil Mold Preparation and Area Measurements Molds were made by pressing aluminum foil over the shell surface and molding it over mounds and ridges and into depres- sions and crevices. The mold of the inner surface included the ligament furrow in the left valve and the buttress and umbonal cavity in the right valve. The foil was smoothed out continuously during the molding process to avoid pleating. Excess foil extend- ing over the shell edge was trimmed and the mold removed from the shell without distorting mold shape. Slits were cut into the mold from the margin inward and carefully flattened out, concave surface down. The outline of the flattened mold was traced on paper and area of the tracing measured with an electronic digitiz- ing planimeter; this area will be referred to as the foil mold area (FMA). Shell outline area (SOA) was also measured with the planimeter from a tracing on paper of the perimeter outline of each shell. Accuracy of FMA Measurements The accuracy of FMA measurements was evaluated by com- parison with another measure of true surface area based on divi- sion of the shell surface into 1 -cm segments across the long axis of the shell and addition of the segment areas. Length of the lines between segments was measured with a cotton string following shell contours and surface area computed using the equation for the Trapezoidal Rule (Britton et al. 1965). Lohse (1990) also measured the area of Mytilus edulis valves directly by adding segmental areas. 15 16 Morales- Alamo REGRESSION OF FMA ON SOA OYSTER SHELL SURFACES 1983 MIXED SHELLS 150- r2 = 0.91 ^ n = 25 / 100- / 50- /^. n- ^-"-'"''^ LEFT OUTER 1983 MIXED SHELLS rJ = 0.99 n = 23 S^*^ RIGHT INNER IT < n 1 o 50 ^ —I o LI. 200 1983 REEF SHELLS r2 = 0.91 n.29 ^ ./^ '<^^^^^^^ LEFT OUTER 1983 REEF SHELLS r^ = 096 n = 39 / ^-"""^^ RIGHT INNER ' 1990 HOUSE SHELLS 150 r! = 0.86 n = 40 100 ■1^' 50 ■■'^ ^l'-"'^ LEFT OUTER n 1990 HOUSE SHELLS r2 =0 98 n = 40 -"^ RIGHT INNER SHELL OUTLINE AREA (cm^ ) Figure 1. Line and 95% prediction interval for tlie regression of FMA (foil mold area) on SOA (shell outline area) in tiirec different samples of oyster shells. Lack of symmetry of prediction intervals is due to conversion of computed values from logarithms to original form. Mixed shells were a mixture of reef and house shells. Left: outer = outer surface of left valve; Right: inner = inner surface of right valve. Reproducibility of FMA Measurements Reproducibility of FMA measurements was tested by replicat- ing the process 10 times for the outer surface of each of two shells and computing the coefficient of variation (CV). The outer surface was selected for this test because it is more uneven and complex than that of the inner surface, thus providing a more rigorous test. One of the shells was a very convex left valve with outer surface deformations originating from another oyster previously attached to it; the other shell was a relatively flat right valve. Regression Equations Equations for regression of FMA on SOA were computed by the least-squares method after logarithmic transformation of the data to correct for heterogeneity of variance. Shell surface area was then predicted from those equations for a multiple number of SOA measurements. Use of the same regression equation to make multiple predictions precludes application of the usual prediction interval (Tiede and Pagano 1979, Snedecor and Cochran 1980). In its place, a prediction interval given by Snedecor and Cochran (1980) was computed. Accuracy of Predicted Surface Areas Accuracy of surface area predictions was tested by comparing FMA of shcllstring shells with the area predicted from the regres- sion of FMA on SOA (the regression-predicted area, or RPA). Spat densities on the shellstring shells as derived from FMA and as obtained from RPA were also compared. RESULTS Accuracy of the Aluminum Foil Mold Measurement There was a high correlation between FMA and the area ob- tained from the sum of the segmental areas of the shell; the coef- ficients of determination (r") for the outer and inner surfaces were 0.99 and 0.98 in a mixed sample of 20 reef and 20 house shells which ranged from 10.27 to 70.64 cm" in SOA. The absolute percent difference between the two types of measurements for individual shells ranged between 0.1 and 9.6 (mean = 3.0; stan- dard deviation (SD) = 2.4) for the outer surface and between 0. 1 and 13.6 (mean = 4.8; SD = 3.8) for the inner surface. Mean surface areas obtained by the two methods were almost identical; for the outer surface, 45.2 cm" (SD = 20.8) by the foil mold method and 45.5 cm" (SD = 21 .2) by the sum of segmental areas; for the inner surface they were 36.9 cm" (SD = 16.9) and 36.6 cm" (SD = 16.8), respectively. The coefficient of variation for ten FMA replications of the outer surface of each of two individual house shells was very low (1.2 and 1 .4), indicating that this technique is highly reproducible. Regression of FMA on SOA There was a strong correlation between FMA and SOA in each of three shell samples analyzed (Fig. 1 , Table 1 ). All coefficients TABLE I. Equations for the regression of foil mold area (Y) on shell outline area ( X ) in shells of Crassostrea virginica from three sources. Source of Shells Regression Equation Valve and Surface (logt = a -t-logX) r-" 1983 Mixed Reef & House Shells: Left Valve (n = 25) Outer Surface: logt = 0.249 -1- 0.954 logX 0.91 Inner Surface: logt = 0.171 -1- 0.949 logX 0.96 Right Valve (n = 23) Outer Surface: logt = 0.072 + 1.039 logX 0.97 Inner Surface: logt = 0.115 + 0.964 logX 0.99 1983 Reef Shells: Left Valve (n = 29) Outer Surface: logt = 0.057 -1- 1.094 logX 0.91 Inner Surface: logt = 0.038 -1- 1.047 logX 0.96 Right Valve (n = 39) Outer Surface: logV = 0.051 -1- 1.047 logX 0.91 Inner Surface: logt = 0.006 + 1.038 logX 0.96 1990 House Shells: Left Valve (n = 40) Outer Surface: logt = 0.153 -1- 1.004 logX 0.86 Inner Surface: logt = 0.085 -1- 0.994 logX 0.95 Right Valve (n = 40) Outer Surface: logt = -0.012 + 1.086 logX 0.95 Inner Surface: logt = 0.093 -1- 0.967 logX 0.98 Reef shells collected from Wreck Shoal in the James River Virginia; house shells, ongin unknown, were obtained from shucking-house refuse piles in Virginia. Logarithms to the base 10. V = fitted Y, i.e.. estimated Y (RPA in text). Estimation of Surface Area of Oyster Shells 17 TABLE 2. Cumulative percent frequency distribution of the difference (in percentages, sign ignored) between foil mold area (FMA) and regression-predicted area (RPA) for individual oyster shells from shellstrings suspended in the Piankatank River in 1990 (predictions based on 1990 house shells). Left Valves Right Valves Pet Outer Surf. Inner Surf. Outer Surf. Inner Surf. Difference n Pet. n Pet. n Pet. n Pet. <5.00 15 53.6 18 64.3 4 25.0 12 75,0 5.01-10.00 6 75.0 9 96.4 5 56.3 3 93.8 10.01-15.00 5 92.9 1 100.0 5 87.5 93.8 15.01-20.00 1 96.4 2 100.0 1 100.0 20.01-25.00 1 100.0 n 28 28 16 16 Mean 6.5 4.2 9.1 4.1 SD 5.-^ 2.8 5 1 4.7 SD = standard deviation. of determination were higher than 0.86. Prediction intervals for the regression lines were very wide because a value of 500 was used for the number of future predictions in the equation from Snedecor and Cochran ( 1980). Figure 1 only includes the data for inner surface of right valves and outer surface of left valves be- cause they represent extremes of shell flatness and concaveness (or convexity), respectively; regression data for the other two valve- surface combinations were intermediate in prediction interval width. Comparison of Area and Spat Density Estimates Differences between FMA and RPA were either all or mostly all under \59c for both surfaces of left and right valves in indi- vidual shells from Piankatank River shellstrings (Table 2). The same was true for shellstring shells used in the James River in 1983 (R. Morales-Alamo and D. S. Haven, unpublished data). Means for FMA and RPA were very close in each of the four groups of Piankatank River shellstring shells (Table 3). Spat densities computed for individual shellstring shells using the two area estimates were identical or nearly identical in most shells (Table 3). Mean spat densities for each shellstring were identical in 6 of the 8 surface comparisons and very similar in the other two. The large variation around these means is associated with variations in larval settlement between shells in the same shellstring and not with variation in area estimates. DISCUSSION Surface area measurements of oyster shells using aluminum foil molds provide the closest approximation to true surface area of any technique proposed to date because they are the only ones that account for variations in shell shape and texture among individual shells. Their accuracy was demonstrated here by comparison with the sum of shell segmental areas. Direct foil measurement of every shell examined (as done by Donn 1990 and Healy 1991) is the most desirable option. How- ever, in studies that involve large numbers of shells, as in exten- sive spatfall monitoring programs, that method would require an inordinate amount of time and effort. The same would be true in studies involving natural reef shells because direct measurement with foil molds would require preliminary removal of fouling or- ganisms. The regression method presented here is a suitable alter- native that would substantially reduce time and effort investment because few actual foil mold measurements are required. A max- imum of 40 each of the right and left valves would be satisfactory to derive a regression equation; tracing shell outline and measure- ment of the area enclosed for all other shells is done relatively quickly. Use of direct foil mold measurements or predictions made from regression of FMA on SOA solves some of the problems associ- ated with substrate suitability in larval settlement studies with oysters: (1) they provide dimensional measures of spat density, unlike data presented in terms of spat per shell, (2) they make it unnecessary to use, just for dimensional purposes, alternate ma- terials that may be potentially less attractive than oyster shells to oyster larvae (Kennedy 1980), and (3) they permit comparison of settlement on oyster shells with settlement on alternate materials when such comparisons are required. They also offer the option of making counts on several measured small areas of the shell sur- face, instead of on the whole shell, when the number of spat is extremely large. The mean of those counts would be comparable to those made on whole shells. Regression of FMA on SOA may be characterized by wide predictive intervals, depending on the valve and surface being analyzed, which would ordinarily handicap use of the regression for prediction purposes. In actual practice, however, percent dif- ferences between individual FMA and RPA were for the most part small and when tested in spatfall studies, their effect was incon- sequential: spat density values for Piankatank River shellstring shells were almost identical regardless of whether the area mea- sured with foil mold or the area predicted from regression was used. In that context, therefore, it is acceptable to ignore the wide predictive regression intervals. A drawback of methods based on foil molds is that a foil mold of the outer surface of an oyster shell cannot account for surface areas inside very small depressions, crevices and pits on the shell surface. They may, thus, underestimate the total area available to settling oyster larvae in heavily-pitted shells. That, however, is not a serious problem when house shells are used because their outer surfaces are relatively unblemished. Old shells from natural reefs are usually heavily pitted and the problem created by that condi- tion must be acknowledged when surface area estimates are made using foil molds. Investigators using shellstrings in spatfall studies have often stated that they used shells of similar size (Lutz et al. 1970, Kennedy 1980, Singarajah 1980). Although those data may present an adequate picture of relative spatfall at different stations and in different years, absence of dimensional units considerably reduces confidence in comparisons with other data. Adoption of the technique presented here, as an alternative to direct foil mold measurements of all shells, would be advisable in spatfall studies that use whole oyster shells as collection substrate. Refinement of the method for improved accuracy is possible by using very flat right valves and examination of only the inner surface. Differences in size and shape of shells from different geographic locations and environments require computation of separate regression equations in subsamples from each of those populations to ensure the highest accuracy of predictions based on the equations. 18 Morales-Alamo TABLE 3. Surface area of oyster shells and density of spat in shellstrings suspended in the Piankatank River, VA. Outhne Area Shell Surface Area (cm^) No. Spat and Density (No./cm^) Outside Surface Inside Surface Outside Surface Inside Surface No. Dens. Dens. No. Dens. Dens. V (cm^) FMA RPA FMA RPA Spat (FMA) (RPA) Spat (FMA) (RPA) Exposure ; Period: 16-23 Aug 1990 Palace Bar (n = 12) L 36.33 54.62 52.42 43.03 43.24 8 0.15 0.15 8 0.19 0.19 L 41.99 67.88 60.62 46.10 49.94 14 0.21 0.23 7 0.15 0.14 R 45.57 64.74 61.56 49.02 49.77 2 0.03 0.03 9 0.18 0.18 R 30.67 40.73 40.05 32.91 33.94 7 0.17 0.17 10 0.30 0.29 L 31.22 50.95 45.02 37.51 37.19 12 0.24 0.27 10 0.27 0.27 R 48.86 61.67 66.41 52.59 53.24 16 0.26 0.24 13 0.25 0.24 R 60.15 79.67 83.22 64.77 65.09 3 0.04 0.04 6 0.09 0.09 R 47.50 57.94 64.40 51.31 51.80 9 0.16 0.14 6 0.12 0.12 R 40.27 58.78 53.83 42.88 44.16 5 0.09 0.09 11 0.26 0.25 L 55.41 82.18 80.09 66.63 65.78 13 0.16 0.16 3 0.05 0.05 L 61.06 85.66 88.29 71.26 72.45 7 0.08 0.08 4 0.06 0.06 L 52.77 78.93 76.26 65.41 62.67 10 0.13 0.13 9 0.14 0.14 Mean 45.98 65.31 64.35 51.95 52.44 8.8 0.14 0.14 8.0 0.17 0.17 SD 10.30 13.92 15.26 12.47 12.01 4.4 0.07 0.08 2.9 0.09 0.08 Burton Point (n = 12) Mean 39.17 59.22 56.20 47.04 46.38 5.0 0.09 0.09 3.0 0.07 0.07 SD 7.90 14.31 11.93 10.46 Exposure 9.63 : Penod: 4.2 23-30 Aug 0.07 1990 0.07 2.3 0.05 0.05 Palace Bar (n = 10) L 49.97 75.38 72.19 57.90 59.36 22 0,29 0.30 32 0.55 0.54 L 69.60 100.38 100.69 89.92 82.52 30 0.30 0.30 31 0.34 0.38 L 60.56 100.97 87.56 74.98 71.86 7 0.07 0.08 52 0.69 0.72 R 45.01 55.49 60.74 53.76 49.18 41 0.74 0.67 29 0.54 0.59 R 39.13 58.86 52.18 44.22 42.95 9 0.15 0.17 25 0.57 0.58 L 48.93 74.50 70.69 62.62 58.14 10 0.13 0.14 30 0.48 0.52 L 35.92 55.62 51.83 43.58 42.76 5 0.09 0.10 25 0.57 0.58 L 53.26 70.33 76.97 59.36 63.25 23 0.33 0.30 55 0.93 0.87 R 39.83 66.45 53.19 53.01 43.69 13 0.20 0.24 22 0.42 0.50 L 37.42 62.98 54.00 46.51 44.53 1 0.02 0.02 9 0.19 0.20 Mean 47.96 72.10 68.00 58.59 55.82 16.1 0.23 0.23 31.0 0.53 0.55 SD 10.90 16.63 16.79 14.53 13.74 12.6 0.21 0.19 13.6 0.20 0.19 Ginney Point (n = 10) Mean 47.89 66.09 66.73 56.01 54.03 3.5 0.06 0.06 18.8 0.34 0.35 SD 10.90 15.66 14.43 10.00 10.91 3.3 0.06 0.06 7.6 0.12 0.13 Key to abbreviations: n = Number, V = Valve, L = Left. R = Right. SD = Standard Deviation. Individual data for Palace Bar strings and means only for two other stations. Areas given as measured from aluminum foil molds (FMA I and as obtained from the regression equation of foil mold area on shell outline area for a sample of the house shells used to construct the shellstnngs (RPA). Spat density computed using both surface area values. ACKNOWLEDGMENTS The author is indebted to Dexter S. Haven and Roger Mann for their support, manuscript reviews, and many valuable suggestions. Comments by Patrick K. Baker, Bruce J. Barber. David P. Lohsc and other anonymous reviewers contributed significantly to the final form of the manuscript. Responsibility for its shortcomings, however, rest exclusively with the author. The technical assistance of Catherine A. Lawrence and Mary Y. Munro is gratefully ac- knowledged. LITERATURE CITED Adams, M. P . R L. Walker, P B Hcffeman & R E. Reinert. 1991. Eliminating spat settlement on oysters cultured in coastal Georgia: a feasibility study. J. Shelfish Res. 10:207-213. Barber, B.J. 1990. Oyster Spatfall in Virginia Waters: 1990 Annual Sum- mary. Marine Resources Special Report, Virginia Institute of Marine Science. College of William and Mary. Gloucester Point. VA. 7 pp. + 2 tables. (Available from the author). Britton, J. R., R. Ben Kriegh & L. W. Rutland. 1965. University Mathematics. I. W. H. Freeman & Co.. San Francisco, CA. 658 pp. Estimation of Surface Area of Oyster Shells 19 Butler. P. A. \'-)5-i- Selective setting of oyster larvae on artificial cultch Proc. NalL Shellfish. Assoc. 45:95-105. Carreon, J. A. 1973. Ecomorphism and soft animal growth of CraiiOiJrco (>C(/a/f; (Faustino). Proc. Null. Shellfish. Assoc. 6.^:12-19. Donn. T. E.. Jr. 1990. Morphometries oi Doiulx scrru Roding (Bivalvia: Donacidae) populations with contrasting zonation patterns. J. Coaslal Res. 6:893-901. Galtsoff, P. S. 1964. The American Oysler: Crassostrea virgmica GmeYin Fish. Bull. U.S. Fish and Wildlife Service, 64:1-480. Haven. D. S. & L. W. Fritz. 1985. Setting of the American oyster Cras- soslrea virginica in the James River. Virginia. USA: temporal and spatial distnbution. Mar. Biol. 86:271-282. Healy. D. K. 1991. The effect of physical disturbance on settlement and recruitment of the American oyster. Crassoslrea virginica (Gmclm). Thesis. University of South Carolina. 60 pp. Kennedy. V. S. 1980. Comparison of recent and past patterns of oyster settlement and seasonal fouling in Broad Creek and Tred Avon River. Maryland. Proc. Nail. Shellfish. Assoc. 70:36-46. Kenny, P. D., W. K. Michener & D. M. Allen. 1990. Spatial and tem- poral patterns of oyster settlement in a high salinity estuary. J . Shellfish Res. 9:329-339. Lohse. D. P. 1990. Interactions between a manne mussel and its epibionts: influence on population dynamics and community structure. Ph.D Dissertation. University of California. Santa Barbara. CA. 174 pp. Lunz, G. R. 1954. The general pattern of oyster setting in South Carohna. Proc. Natl. Shellfish. Assoc. 45:47-51. Lutz, R. A.. H. Hidu & K. G. Drobeck. 1970. Acute temperature increase as a stimulus to setting in the American oyster, Crassoslrea virginica (Gmelin). Proc. Nail. Shellfish. Assoc. 60:68-71. Marcus, J. M., G. I. Scott & D. D. Heizer. 1989. The use of oyster shell thickness and condition index measurements as physiological indica- tors of no heavy metal pollution around three coastal marinas. J . Shell- fish Res. 8:87-94. Marsh, J A., Jr. 1970. Primary productivity of reef-building calcareous red algae. Ecology 51:255-263. Morales-Alamo, R. & R. Mann. 1990. Recruitment and growth of oysters on shell planted at four monthly intervals in the lower Potomac River, Maryland. J. Shellfish Res. 9:165-172. McNulty, J. K. 1953. Seasonal and vertical patterns of oyster setting off Wadmalaw Island. S.C. Conlrih Bears Bluff Labs. No. 15. 17 pp. Osman. R. W., R. B. Whitlach & R. N, Zajac. 1989. Effects of resident species on recruitment into a community: larval settlement versus post- settlement mortality in the oyster Crassoslrea virginica. Mar. Ecol. Progr. Ser. 54:61-73. Shelly. T. E. 1979. The effect of rock size upon the distribution of species of Orthocladiinae (Chironomidae: Diplera) and Baetis inlercalaris Mc- Dunnough (Baetidae: Ephemeroptera). Ecol. Eniomol. 4:95-100. Singarajah. K. V. 1980. Some observations on spat settlement, growth rate, gonad development and spawning of a large Brazilian oyster. Proc. Nail. Shellfish. Assoc. 70:190-200. Snedecor. G. W. & W. G. Cochran. 1980. Statistical Methods. The Iowa State University Press. Ames. lA. 507 pp. Tiede, J. J. & M. Pagano. 1979. The application of robust calibration to radioimmunoassay. Biometrics 35:567-574. Journal of Shellfish Research. Vol. 12, No. 1. 21-27, 1993. DEVELOPMENT OF DISEASE CAUSED BY THE PARASITE, PERKINSVS MARINUS AND DEFENSE-RELATED HEMOLYMPH FACTORS IN THREE POPULATIONS OF OYSTERS FROM THE CHESAPEAKE BAY, USA FU-LIN E. CHU AND J. F. LA PEYRE Viri^inia Institute of Marine Science School of Marine Science College of William and Mary Gloucester Point. Virginia 23062 .ABSTRACT The development of infection cau,sed by the protozoan parasite. Perkinsus marinus (Dermo) and some specific potential defense-related cellular and humoral components in oysters collected from three geographic areas, Deepwater Shoal of James River (DW), Wachapreague (WPl, and Mobjack Bay (MJ) were examined over time. Oysters were maintained in estuarine water (salinity = 20 ppt) or in water at a salinity similar to the ambient salinity of the collection sites. Oysters were sampled at the initiation of the experiment (day 01, day 35, and day 100 to determine defense-related parameters and disease prevalence and intensity. All populations experienced a significant increase in P. marinus infection prevalence and intensity from the initiation of the experiment to the termination of the study. Oyster mortality differed between oyster populations. None of the DW oysters perished while cumulative mortalities for WP at 32 ppt and 20 ppt and MJ oysters were respectively, 23, 25, and 35%. The experimental oyster populations demonstrated significant differences with respect to cellular and humoral defense-related variables. As the study progressed, the mean number of total hemocytes declined in the WP and MJ populations and increased in the DW population . The percentage of granulocytes in DW oysters was consistently higher than other populations. DW oysters also had the highest concentrations of protein and lysozyme. This pattern persisted throughout the expenmental pcnod. Oyster condition index significantly decreased during the course of the study in all populations except the DW oysters at 10 ppt. Results suggest that the increase of hemocyte number and higher percentage of granulocytes, and lysozyme concentration in DW oysters may have contributed to the high (100%) survival rate of this population. Salinity may have affected disease development. Disease prevalence and intensity tended to be lower in the WP oysters maintained at low salinity than those maintained at high salinity. In the DW population, unexpectedly, oysters maintained at 20 ppt had lower infection prevalence and intensity than oysters maintained at 10 ppt. Salinity induced, to some extent, changes in certain hemolymph components: lysozyme concentration tended to be higher in oysters maintained at low salinity than those maintained at high salinity. Increase in percentage of granulocytes was also observed in WP oysters after transferring to a salinity lower than ambient salinity. KEY WORDS: oyster disease, hemolymph factors, Perkinsus marinus INTRODUCTION Disease-induced mortality in eastern oysters (Crassotrea vir- ginica) caused by two parasites, Perkinsus marinus (Dermo) and Haplosporidium nelsoni (MSX) is one of the factors contributing to the decline in oyster harvest in the Chesapeake Bay, U.S.A. Previously, disease pressure from H. nelsoni has been more in- tense on oysters than that from P. marinus. Because of its current expanded distribution and increase in abundance in waters of the Chesapeake Bay. P. marinus is now considered more significant than H. nelsoni as an oyster pathogen (Andrew 1988, Burreson 1989). It has been well documented that prevalence and intensity of P . marinus infections in oysters are related to milieu salinity (e.g., Soniat 1985, Soniat and Gauthier 1989, Crosby and Roberts 1990. Gauthier et al. 1990, Paynter and Burreson 1991). Signif icant growth reduction due to P. marinus infection in oysters raised in habitats of different salinity in the Chesapeake Bay has been reported by Paynter and Burreson (1991). Hemocyte activities and lysozyme concentrations of eastern oysters have been reported to change seasonally (Fisher et al. 1989, Feng and Canzonier 1970, Chu and La Peyre 1989) and to be affected by salinity (Fisher 1988, Chu and La Peyre 1989, Chu et al. In review). Increased salinity suppressed hemocyte spread- ing and locomotion. Hemolymph lysozyme concentration in oys- ters was negatively correlated with salinity in oysters (Chu et al. In review). The purpose of this study was to compare the development of disease caused by P marinus in oysters collected from three dif- ferent salinity habitats of the lower Chesapeake Bay and to deter- mine if any changes occurred in some measurable cellular and humoral components in these oysters during the course of disease development. METHOD AND MATERIALS Experiment To encompass the natural salinity range of oysters in the lower Chesapeake Bay, oysters were collected from 3 locations; a low salinity site. Deep Water Shoal of James River (DW, ambient temperature = 22.5°C, salinity = 10 ppt), a high salinity site, Burtons Bay, Wachapreague (WP, ambient temperature = 19.5°C, salinity = 32 ppt), and a moderate salinity site, Mobjack Bay (MJ, ambient temperature = 20.0°C, salinity = 20 ppt), in early October 1990. Oysters were cleaned of fouling organisms and a hemolymph sample was withdrawn from 30 oysters from each population to measure initial total hemocyte count (TC), percent of granulocytes (PG) and protein and lysozyme concen- trations. Oysters were then sacrificed to determine initial condition index (CI = dry meat weight/dry shell weight x 100. Lucas and Beninger 1985) and to examine for P. marinus infection (Ray 1952, 1966). Sixty oysters from each population were maintained in 250 1 static fiber-glass tanks at 22 ± TC and at conditions indicated below. Oysters from MJ (N = 60) were maintained in filtered ( 1 \i. filter) estuarine water (York River water, YRW, salinity = 20 ppt). Oysters from DW and WP were each divided into 2 groups 21 22 Chu and La Peyre (60 oysters/group/tank); one group of the oysters was maintained in filtered YRW; the other group was maintained in water adjusted to ambient salinity (i.e. 10 ppt for DW oysters. 32 ppt for WP oysters). Oysters were fed daily with an algal diet (a mixture of Pavlova lutheri, Isochrysis galbana and Tahitian Isochrysis gal- bana). Water was changed every other day. The experiment was terminated in the middle of January. 1991 (100 days after exper- iment initiation). Thirty five days after the initiation of the exper- iment and at the end of the experiment, subsamples of oysters (N = 20 oysters, 35 days after initiation, N = 30 oysters at the end of the experiment) from each group were sampled for TC, PG, protein, lysozyme and CI measurement and P. marinus diagnosis. Total and Differential Counts and Preparation of Sera Hemolymph from individual oysters was withdrawn with a syringe from the adductor muscle sinus through notches in the shell and placed in micro test tubes in an ice bath. Total and differential (number of granulocytes and agranulocytes) hemocyte counts were obtained on each hemolymph sample using a hemo- cytometer. Differential counts were expressed as percentage of granulocytes (PG = 100 x number of granulocytes/total hemo- cytes). To determine protein and lysozyme concentrations in oys- ter serum (cell-free hemolymph). serum of each hemolymph sam- ple was separated from hemocytes through centrifugation (400 x g at 4°C for 10 min). Serum was withdrawn and stored in a freezer ( — 20°C) for subsequent protein and lysozyme measurement. Protein and Lysozyme Measurements Lysozyme concentration was determined spectrophotometri- cally according to the method of Shugar (1952) and Chu and La Peyre ( 1989). Cell-free oyster serum (0. 1 ml) was added to 1.4 ml of the bacterial {Micrococcus lysodeiklicus) suspension and the decrease m the absorbance was recorded at 450 nm on a Schi- madzu UV 600 spectrophotometer for 2 minutes. All measure- ments were duplicated and were taken at room temperature (21 ± 1°C). Recorded lysozyme activities were converted to lysozyme concentration using egg white lysozyme as a standard. Standard curves at different salinities were constructed by dissolving egg white lysozyme in a balanced salt solution of appropriate salinities (i.e. 10, 20, and 32 ppt). assuming that reactivity of oyster lysozyme and egg white lysozyme were similar if assayed in buffer of the same salinity. Serum protein was measured by the method of Lowry et al. (1951) using bovine albumin as a standard. Ten |jil of a cell-free hemolymph sample from individual oysters was used for the serum protein measurement. Perkins us Assay The thioglycollate assay described by Ray (1952. 1966) was used for P. marinus diagnosis. Rectal tissue was removed from each oyster and incubated in thioglycollate medium for 4—5 days. Intensity of infection was ranked from (negative) to 5 (heavily infected) based on the relative number of stained P. marinus hyp- nospores contained in the tissue smear. Statistical Analysis One factor analysis of variance (ANOVA) and Student- Newman-Keuls test were used to compare total hemocyte counts (TC) and percentage of granulocytes (PG), protein (P) and lysozyme concentrations, condition index, and prevalence and in- tensity of P. marinus infection between population groups and between different salinity treatments of the same population (i.e. DW and WP oysters). Data were Log.o or Arcsine transformed whenever data showed a large variance. Differences were consid- ered statistically significant at P « 0.05. Linear correlation (Pear- son correlation analysis) was calculated between the measured variables, condition index, serum protein and lysozyme concen- trations, and P. marinus infection intensity. RESULTS The infection prevalence and intensity of oysters sampled from DW, WP, and MJ populations on day 0, day 35, and day 100 are shown in Figure 1. At the beginning of the experiment, preva- lences in DW, WP, and MJ oyster samples (N = 30/population) PERKINSUS PREVALENCE Q UJ O O cr DW10 DWYRW WP32 WPYRW MJYRW POPULATION DAY ^ DAY 35 ^^ ^^^ 100 WEIGHTED INCIDENCE OF INFECTION DW10 DWYRW WP32 WPYRW MJYRW POPULATION DAYO DAY 35 ^ DAY 100 Figure 1. Perkinsus marinus prevalence and weighted incidence in DW (Deep Water Shoal, James River), WP (Burtons Bay. Wachapre- ague) and MJ (Mobjack Bay) oysters at day (N = 30), day 35 (N = 20) and day 100 (N = 30). DWIO = DW oysters at 10 ppt water, DWYRW = DW oysters in York River Water. WPYRW = WP oys- ters in York River Water, WP32 = WP oysters at 32 ppt water, MJYRW = MJ oysters in York River Water. Disease Development in Three Populations of Oysters 23 were 63, 80. and 70%, respectively (Fig. 1). Infection intensities expressed as weighted incidence (WI = sum of disease code num- bers/number of oysters) in WP and MJ oysters were significantly higher than in DW oysters (Fig. 1). P. mahnus prevalence in oysters sampled on day 35 (N = 20/group) were 50% in DW at 10 ppt (DWIO), 55% in DW at YRW (20 ppt. DW20), 70% in MJ, 100% in WP at 32 ppt (WP32) and at YRW (20 ppt, WP20). Weighted incidence increased in both WP and DW populations and decreased in the MJ oysters at day 35. At the termination of the experiment, prevalence in oyster samples (N = 30/group) were 93, 83, 96, 93 and 100% in DWIO, DW20, WP32, WP20 and MJ populations respectively. All population groups experi- enced a significant increase in P. marinus infection prevalence and intensity from the initiation to the termination of the experiment, a period of 100 days. Generally, as in the beginning of the exper- iment, at the end of the experiment, DW oysters maintained rel- atively lower P. marinus weighted incidence than WP and MJ oysters. At all sampling dates, DW20 oysters had significantly lower (P < 0.05) weighed incidence than all other groups of oysters. The DW oysters maintained in YRW (20 ppt) had lower prevalence and weighted incidence than those maintained at 10 ppt. Only four DW20 oysters developed moderate to advanced (level 3-5) infections. Disease prevalence did not appear to differ in WP20 and WP32 oysters, but disease intensity was lower in the fomier than the latter at both day 35 and day 100. Oyster mortality differed among populations (Fig. 2). During the course of the study, none of the DW oysters perished. Cumu- lative mortalities in WP at 32 ppt, WP at 20 ppt, and MJ groups were 23, 25, and 35%, respectively. At the initiation of the experiment, mean TC was significantly higher in WP and MJ oysters than in DW oysters (Fig. 3). How- ever, mean PG was much higher (P < 0.05) in the DW oysters than in the other oyster populations. As the study progressed, the mean TC declined in the WP and MJ groups and increased in the DW20 group. In the DW20, WP20 and MJ groups, the final mean TC differed significantly from the mean TC at day 0. No signif- TOTAL HEMOCYTE COUNT w o o o UJ m 3 DW10 DWYRW WP32 WPYRW MJYRW POPULATION DAYO DAY 35 ^ DAY 100 DIFFERENTIAL HEMOCYTE COUNT DW10 DWYRW WP32 WPYRW MJYRW POPULATION CUMULATIVE MORTAUTY 1 DAYS 35 DAYS ^ 100 DAYS DWIO DWYRW WP32 WPYRW MJYRW POPULATION Figure 2, Cumulative mortality of DW (Deep Water Shoal, James River), WP (Burtons Bay, Wachapreague) and MJ (Mobjack Bay) oysters at the termination of the experiment. DWIO = DW oysters at 10 ppt water, DWYRW = DW oysters in York River Water, WPYRW = WP oysters in York River Water, WP32 = WP oysters at 32 ppt water, MJYRW = MJ oysters at 20 ppt in York River water. Figure 3. Mean total hemocyte counts and percentage of granulocytes (±SE) in DW (Deep Water Shoal, James River), WP (Burtons Bay, Wachapreague) and MJ (Mobjack Bay) oysters at day (N = 30). day 35 (N = 201 and day 100 (N = 30). DWIO = DW oysters at 10 ppt water, DWYRW = DW oysters in York River Water, WPYRW = WP oysters in York River water, WP32 = WP oysters at 32 ppt water, MJYRW = MJ oysters at 20 ppt in York River water. icant change in mean TC was observed over time in the DWIO and WP32 oysters. A trend of increasing TC with time was noted in the DW groups, although differences were not statistically significant. Generally, DW oysters had the highest PG over the course of the study. Generally, in both WP and DW oysters, no significant difference was observed in both TC and PG between salinity treat- ments. Serum protein and lysozyme concentrations differed among the three oyster populations (Fig. 4). Concentrations of lysozyme and protein were significantly lower (P < 0.05) in the WP and MJ than in DW populations on day (Fig. 4). No significant difference in lysozyme or protein concentration was observed between MJ and WP oysters. This pattern persisted throughout the experimental period; DW oysters had the highest (P < 0.05) concentration 24 Chu and La Peyre SERUM PROTEIN CONCENTRATION C5 2 (5 DW10 DWYRW WP32 WPYRW MJYRW POPULATION DAYO DAY 35 SS DAY 100 SERUM LYSOZYME CONCENTRATION DW10 DWYRW WP32 WPYRW MJYRW POPULATION protein concentrations than WP oysters. Although insignificant statistically, lysozyme concentrations tended to increase in WP20 oysters and to decrease in DW20 oysters. The mean protein con- centrations in DW20 oysters was higher (P < 0.05) than DWIO oysters. The lysozyme concentrations in DWIO, MJ, and WP32 oysters sampled at the termination of the experiment were nega- tively correlated with infection intensity. Oyster condition, as indicated by condition index (Fig. 5), was significantly lower at the end than the beginning of the experiment in all population groups except the DWIO group. When Pearson correlation analysis was performed on data pooled from each group, it revealed that the condition index of DWIO, WP32, WP20 and MJ oysters was negatively correlated with P. marinus infection intensity; condition index of DW20, WP32. WP20, and MJ oysters were positively correlated with serum protein concen- trations. DISCUSSION Oysters from the upper James River, in areas such as Horse- head and Deep Water Shoal, are quite vulnerable to both P. mari- nus and H. nehoni (Andrews 1984, Ford and Haskin 1987. An- drews 1988, Barber et a! . 1991, Burreson 1992) but have remained relatively disease free because of prevailing low salinity (Andrews 1988, Burreson 1989, 1990, 1991). Mobjack Bay of the lower York river is an endemic area for both P. marinus and MSX. Progeny from survivors of the 1960 MSX epizootics in Mobjack Bay were shown to be less susceptible to MSX than seed oysters from the James River (Andrews 1971, Andrews 1984). Until 1990, oysters from Wachapreague had a low incidence of P. mari- nus and low mortality caused by P. marinus (Burreson 1990, 1991). The three oyster populations under investigation may be genetically different. However, they displayed a similar response to P marinus. Almost all oysters (83 to 100%) from each popu- CONDITION INDEX DAYO DAY 35 ^ DAY 100 Figure 4. Mean hemolymph protein and lysozyme concentrations (±SE) in DW (Deep Water Shoal. James River), WP (Burtons Bay, Wachapreague) and MJ (Mobjack Bay) oysters at day (N = 30), day 35 (N = 20) and day 100 (N = 30). DWIO = DW oysters at 10 ppl water, DWYRW = DW oysters in York River Water, WPYRW = WP oysters in York River water, WP32 = WP oysters at 32 ppt water, MJYRW = MJ oysters at 20 ppt in York River water. of protein and lysozyme on all sample dates. The concentrations of these two serum components fluctuated between sample dates. Within the DW populations, oysters sampled at day 35 had a significantly higher protein concentration than those sampled at day and day 100; but lysozyme concentration in DW oysters did not change significantly through time. Protein concentration in MJ oysters also peaked at day 35 and differed significantly from both initial and final sample concentrations. Both lysozyme and protein concentrations in MJ oysters declined from day 35 to day 100; MJ oysters sampled at day 100 had the lowest protein and lysozyme concentrations. Protein and lysozyme concentration in the WP population did not differ significantly over time. WP oysters did not differ from MJ oysters in protein and lysozyme concentration except at day 35. At day 35, MJ oysters had significantly higher 2 o z z o Q z O o DWIO DWYRW WP32 WPYRW MJYRW POPULATION DAYO DAY 35 ^ DAY 100 Figure 5. Mean condition index (±SE) in DW (Deep Water Shoal, James River), WP (Burtons Bay, Wachapreague) and MJ (Mobjack Bay) oysters at day (N = 30), day 35 (N = 20) and day 100 (N = 30). DWIO = DW oysters at 10 ppt water, DWYRW = DW oysters in York River Water, WPYRW = WP oysters in York River water, WP 32 = WP oysters at 32 ppt water, MJYRW = MJ oysters in York River Water. Disease Development in Three Populations of Oysters 25 lation were infected by the parasite when the experiment was terminated (Fig. 1). Oysters from Mobjacl^ Bay were found to be less susceptible to H. nelsoni than oysters from James River (An- drews 1984). but results of the present study indicates that they are equally susceptible to P. imirinus as are oysters from the James River. Results of the present study confirm that P. marinus can endure low salinity (Chu and Greene 1989. Ragone 1991 . Burreson 1990. 1991). At the beginning of the experiment, oysters from Deep Water Shoal of James River began with lower disease prevalence (bY/c at day and 507f at day 35 1 and intensity than both WP and MJ oysters. The increased disease prevalence in DW oysters at ambient salinity (10 ppt) at the end of the experiment is, appar- ently, a result of disease transmission between infected and unin- fected oysters maintained in the same tank. Thus, it is clear that low salinity did not restrict disease transmission among oysters. Under continuous disease pressure, salinity of 10 ppt did not in- hibit the progress of disease development. It has been shown that in vitro, only salinities lower than 6 ppt restrained P. marinus sporulation from prezoosporangia (Perkins 1966. Chu and Greene 1989). Oysters from Deep Water Shoal of the James River (DW oys- ters) and WP oysters may have responded differently to the low salinity treatment. It is surprising to note that DW oysters main- tained at ambient salinity ( 10 ppt) had higher P. marinus weighted incidence than those DW oysters at 20 ppt. whereas placing WP oysters at a salinity (20 ppt) lower than ambient salinity (32 ppt) reduced, relatively, the weighted incidence in these oysters. It is not known whether this difference is based on genetic dissimilar- ities between DW and WP oysters or whether it was an artifact. Further study is needed to verify this result. Restraint of disease progress has been noted in DW oysters infected by P. marinus maintained at low salinity (g(- nica). Elsewhere in Delaware Bay this set is welcomed, but ex- perimental groups, heavily set with spat, can quickly become overgrown by the overset. Besides impeding growth, fast growing natural set may be confused with yearling brood stock, potentially contaminating the specialized gene pool(s) of, for example, resis- tant strains. In the past overset spat were removed by scraping oysters individually, which is both tedious and inefficient. Various brine (salt) dips have been used to control oyster "en- emies" such as squirts iMogula sp.), boring sponges (Cliona sp.), and starfish (Asterias sp.) (Shearer and MacKenzie, I96I). Per- haps not surprisingly, there is little precedence for the use of brine dips to control oyster set, although Shearer and MacKenzie (1961), while testing the species above, observed mortality in 10-20 mm oysters at high brine concentrations. Preliminary ex- periments conducted in 1990 at HSRL indicated that overset might be controlled by treatments in concentrated brine solutions fol- lowed by some period of aerial exposure. Results of these preliminary experiments showed that adult oysters survived well (1-2% mortality) when treated for 10 min- utes in a 200 ppt brine solution, followed by 3, 6, or 12 hour aerial exposure. These treatments also resulted in 89-100% mortality of spat measuring 1 mm or less in shell length. Consequently, a field test was conducted in 1990 on oysters heavily set by Delaware Bay native spatfall. Three wire mesh trays, each containing overset adult oysters (age 3-5 yrs), were dipped in 200 ppt brine for 10 minutes; three similar trays were dipped in ambient seawater (22 ppt). Dips were followed by a three hour aerial exposure. Fifty days after treatments, numbers of spat on oysters were counted in both dipped and control trays: oysters in dipped trays averaged 3 spat per oyster; in controls — 22 spat/oyster. Yet other variables associated with this treatment were unex- plored. The objective of the research reported here was to refine the dip procedure so that we might incorporate it as a too! in the routine maintenance of our stocks. Variables explored were (1) tolerance of adults to brine treatments of various, durations, and aerial exposures; (2) optimum treatments for removing recently set spat at several durations and aerial exposures, and size related mortality; and (3) a field test of spat covered oysters. MATERIALS AND METHODS Two experiments and a field test were designed to better define the parameters of brine treatments. Treatments consisted of im- mersing oysters or spatted cultch contained in wire mesh trays in 200 ppt bnne solution for vanous lengths of time, followed by either a 3 or 6 hour aerial exposure. We chose 200 ppt brine for two principal reasons: First, preliminary work with brine dips indicated that 200 ppt brine seemed as effective as saturated brine ( — 300 ppt). Secondly, and more important to our experimental design, saturation of brine solutions is relative, depending on fac- tors such as temperature and humidity. It is almost impossaible to achieve uniformity in saturation from experiment to experiment. Aerial exposures were conducted in a shaded area to normalize for differences in the intensity of sunlight over the course of the ex- periments. Aerial exposures were conducted at ambient outside temperatures, ranging from 2I-28°C. Experiment I Since it is imperative that dip treatments not harm adult oys- ters, the first experiment was conducted to test the effect of im- 29 30 DeBrosse and Allen tnersion for 2, 5, or 10 minutes in a 200 ppt brine solution on survivorship for yearlings and for older adults (age: 2-3 yrs). Dips were followed by an aerial exposure of either 3 or 6 hours in a shaded location. For each of the two year class groups (age 1 and age >1), 16 groups of 50 oysters were tested: two replicates for each immersion duration (2, 5. and 10 minutes) x aerial exposure period (3 or 6 hours) combination (12 groups). Controls consisted of two replicates each of a 10 minute dip in ambient sea water followed by a 3 or 6 hour aerial exposure (4 groups). A total of 1600 adult oysters were used. Each group was examined 2, 4. and 6 days post treatment to assess survival. Experiment 2 The second set of experiments was conducted to test the effect of concentrated brine solutions on survivorship of recently set spat. We wanted to determine the maximum size that spat can be efficiently removed with a brine dip. Eyed larvae cultured in the hatchery were set on prepared shells of the Atlantic sea scallop, Placopecten magellaniciis (Gmelin, 1791 ). Spat were reared in the hatchery for 1-3 weeks. Prior to treatments, individual spat were measured to the nearest 0. 1 mm and surrounded by a numbered, pencil drawn circle. This made it possible to determine post- exposure mortality for individual spat. Two size classes of spat were tested: <5.0 mm (mean 2.3 mm; range 0.9-5.0 mm) and >4.0 mm (mean 6.1 mm; range 4.0-11.6 mm). Brine concentra- tion for all dips was 200 ppt. For spat <5 mm, 18 groups of 50 spat were tested: two repli- cates for each immersion duration (2, 5, and 10 minutes) x ex- posure period (3 or 6 hours) combination (12 groups). Controls consisted of one immersion for each duration (2, 5, and 10 min- utes) X aerial exposure period (3 or 6 hours) combination (6 groups). A total of 900 spat were used. For spat >4 mm, 6 groups of 50 spat were tested: two repli- cates and one control were dipped for 10 minutes and exposed to air for either 3 or 6 hours. A total of 300 spat were used. In both size classes (<5 and >4 mm) survival of individual spat was assessed 2, 4, and 6 days after treatment. Field Test For the final experiment we had planned a large scale field trial on trays of adult oysters that were fouled by native set. Ironically, spatfall in the summer of 1991 was particularly light at our Cape TABLE 1. Cumulative survival (percent) 6 days following treatment of either yearling (<1 year old) or adult oysters (2-3 years old). Aerial Exposure Age (years) Brine Duration (hours) (ppt) (min) 3hr 6hr 4 mm. Here a one- way ANOVA was run on controls (3 and 6 hour exposures pooled) and the two treatment (10 min x 200 ppt dip for 3 or 6 hour exposure). 100 Mean of two replicates. Figure 1. Top — Mean (of all replicates) cumulative mortality of spat ( 1-5 mm) 6 days after exposure to various brine treatments. Bottom — Difference in mean size of spat ( 1-5 mm) between day and day 6 after treatment with various brine dips. Differences in means between day and day 6 were tested by Students t-test *— P < 0.05; **— P < 0.01. Four hi.stograms on left (3 hour aerial exposure): CTL 3 hr — 22 ppt for 10 min; DIP 2/3—200 ppt for 2 min; DIP 5/3—200 ppt for 5 min; DIP 10/3 — 200 ppt for 10 min. Four histograms on right (6 hour aerial exposure): CTL 6 hr- 22 ppt for 10 min; DIP 2/6—200 ppt for 2 min; DIP 5/6—200 ppt for 5 min; DIP 10/6—200 ppt for 10 min. Control of Overset on Oysters 200/10/6 200/5/6 200/2/6 SIZE INTERVALS Imml Figure 2. Frequency distribution liistograms of spat (1-5 mm) in 0.5 mm size intervals on days (from back to front) 0, 2, 4, and 6 for various brine treatments. Left (3 hour aerial exposure), from bottom to top: control 3 hr— 22 ppt for 10 min; 200/2/3 — 200 ppt for 2 min; 200/5/3—200 ppt for 5 min; 200/10/3—200 ppt for 10 min. Right (6 hour aerial exposure), from bottom to top: control — 22 ppt for 10 min; 200/2/6—200 ppt for 2 min; 200/5/6—200 ppt for 5 min; 200/10/6—200 ppt for 10 min. RESULTS Experiment I Brine dips had no appreciable effect on survival of adult oysters for dip duration (P = 0.93). aerial exposure (P = 0.506), or the interaction of the two (P = 0.857). Cumulative mean survival after 6 days ranged from 94-98% in treated groups compared to 97% for the controls (Table 1 ). We therefore felt that none of our treatments were harmful to adult brood stock. Experiment 2 Spa! <5 mm Overall, the principal effect of brine dips on spat <5 mm (mean 2.3 mm; range 0.9-5.0 mm) was high, selective mortality (Fig. 1 , top). For 3 hour aerial exposure, mean mortality of spat 6 days after treatment for 2, 5, and 10 minute dips in brine was 58%, 10%. and 83%. respectively, and 28%), 18%), and 2A% for respective controls. In spat exposed to air for 6 hours, mean mortality for 2, 5, and 10 minute dips in brine was 64%. 85%, and 86%, respectively, and 38%, 22% , and 8% for respective controls. ANOVA demonstrated a significant effect due to dip duration (F, ,0 = 32.9, P < 0.001) but not for aerial exposure (F, jf, = 1.03, P = 0.335) or interaction (F, „, = 0.43, P = 0.734). An a posteriori test (Tukey's HSD) demonstrated that all treatments were significantly different from the control (Tukey's HSD pair- wise comparison, maximum P = 0.002); 10 minute dips differed significantly from 2 (P = 0.041), but not 5, minute dips. Mortality in all groups, including controls progressed over the course of the 6 days of observation (Fig. 2). All size classes under 5 mm experienced some mortalities, but mortality was size depen- dent. For each treatment, we compared the mean size of spat before the brine dip with the mean size of spat 6 days afterward. Mean spat size was significantly larger 6 days after treatment (Student's t-test, day 6 vs day 0) in bnne dips for 5 and 10 min for both 3 and 6 hour aerial exposure (Figure 1 , bottom). The obvious interpretation is small spat are more susceptible to brine dips than larger ones. (There was no significant difference among treatment groups at day 0: 2-way ANOVA. minimum P = 0.114). Spat >4 mm For testing spat over 4 mm (mean 6.1 mm; range 4.0-11.6 mm), we used the best treatment from previous tests: 10 min dip in 200 ppt brine with either 3 or 6 hour exposure. In this experiment, the 6 hour exposure caused higher mortality than the 3 hour exposure, but not significantly (Tukey's HSD 100 > 60 1.6 1 1.4 u= 1.2 > D 1 E 0.8 E uj 0.6 u 2 4 a: ff 0.2 Q CTL 3 hr DIP 3 hr CTL 6 hr DIP 6 hr Figure 3. Mean (of all replicates) cumulative mortality of spat (4-12 mm) 6 days after exposure to various brine treatments. Bottom — Difference in mean size of spat (4-12 mm) between day and day 6 after treatment with various brine dips. Differences in means between day and day 6 were tested by Students t-test * — P < 0.05. Two histograms on left (3 hour aerial exposure): CTL 3 hr — 22 ppt for 10 min; Dip 3 hr — 200 ppt for 10 min. Two histograms on right (6 hour aerial exposure): CTL 6 hr — 22 ppt for 10 min; DIP 6 hr — 200 ppt for 10 min. 32 DeBrosse and Allen pairwise comparison, P = 0.053) probably due to high variability in the data from the shorter exposure (Fig. 3. top). For 3 hour aerial exposure, mortality of spat 6 days after treatment was 36% and 58% for brine dips, and 22% for its control. For 6 hour exposure, mortality was 84% and 92% for brine dips, and 32% for controls. Mortality in controls occurred mostly during the first two days, perhaps corresponding to handling, whereas mortality in brine dips occurred gradually over the 6 day observation period (Fig. 4). As found in the experiment using smaller spat, mortality was size dependent. We compared the mean size of spat before the brine dip with the mean size of spat 6 days afterward in each treatment. Mean spat size was significantly larger 6 days after treatment in brine for the 6 hour aerial exposure only (Fig. 3, bottom). (There was no significant difference among treatment groups at day 0:F3 3 = 0.083, P = 0.963). Judging from the data shown in Figure 4, it appears that the size cutoff for these dips is about 7.5 mm. That is, a 10 min dip in 200 ppt brine, followed by a 6 hour aerial exposure will kill mostly those spat less than 7.5 mm; also, the smaller the spat, the higher the mortality. Field test Brine dips conducted on large spat (mean 11.3 mm; range 4.7-22.0 mm) in a small scale field test in October 1991 had no effect on overset. For these larger spat, there was no effect due to brine dips on mean number of spat/oyster 6 days after treatment. Mean number of spat per oyster decreased 30% in controls, but only 40% in treatments (F, 1.07, P = 0.332). Adult survival 6 days post treatment was slightly higher in treated groups (94%) than in controls (90%), confirming that dip treatments are not injurious to adults. DISCUSSION The removal of overset from our brood stocks has been an ongoing maintenance problem at HSRL. Heretofore, overset was removed by scraping oysters individually. Very small spat were eliminated with a wire brush; older spat, by scraping with an oyster knife. Both of these methods are tedious and inefficient. Also, scraping often damages the growing edge of the oyster shell. 200/10/6 > u z LU ID a LU > u o LU CC in >J=: ^ L r A r^ 00 in 05 K^ CO in o SIZE INTERVAL (mm) Figure 4. Frequency distribution histograms of spat (4-12 mm) in 0.5 mm size intervals on days (from back to front) 0, 2, 4, and 6 for various brine treatments. Left (3 hour aerial exposurel, from bottom to top: control — 22 ppt for 10 min: 200/10/3 — 200 ppt for 10 min. Right (6 hour aerial exposure), from bottom to top: control — 22 ppt for 10 min; 200/10/6 — 200 ppt for 10 min. Control of Overset on Oysters 33 The same problem would pertain to commercial grow-out any- where natural set occurs. As a means of removing oyster overset on containerized oysters, for example in wire trays or in plastic mesh bags, brine dips are attractive for their efficiency and low cost, and would be amenable to large scale application. Preliminary experiments conducted m 1490 indicated that over- set might be controlled in this way, followed by some period of aerial exposure. Initially, aerial exposure was conducted in un- shaded areas for 12 to 24 hours so that oysters could be returned at the next low tide or the next day's low tide. Early in the sum- mer, when temperatures were moderate and when adult oysters were healthy (i.e.. prior to spawning, disease pressures, food lim- itations, etc.). unshaded protracted aerial exposures were innocu- ous. Later in the summer when temperatures were higher and adult oysters less hardy, these same exposures began to cause higher mortalities. Fortunately, our preliminary trials and the data shown here demonstrate that only short aerial exposures, on the order of 3-6 hours, are required to kill overset. With such a short aerial exposure, brine treatments could be conducted within a tidal cycle, in most cases. Immersion in brine solutions has been used in the past to re- move fouling organisms from oysters. Loosanoff (1958) reported killing various soft-bodied invertebrates as well as egg cases of the oyster drill Eiipleura caudata using 300 ppt brine solutions. MacKenzie & Shearer ( 1961) reported that from 87 to 98% of the mud blister worm Polydora websteri were killed using a 10-15 minute dip in a saturated salt solution, followed by 15 or more minutes of air drying. In another study. Shearer & Mackenzie ( 1961 ) reported lOO'/c mortality of boring sponges (CUona celaia). starfish [Asterias forbesi). and tunicates [Molgula manhattensis) after immersion in 180 ppt brine solution for 10 minutes, followed by a 1 hour aerial exposure. When Crassostrea virginica spat measuring 10-20 mm in shell length were subjected to this treat- ment, 2.3% had died 14 days post treatment. More recently, Ar- akawa (1980) reported elimination of up to 59% of the fanworm Hydroides elegans using a 60 minute dip in saturated brine. Fi- nally, Dealteris et al (1988), while investigating alternative treat- ments to prevent the bio-deterioration of wood lobster traps by the wood-boring bivalve Xxlophaga atlantica. found that a 30 second dip in saturated brine resulted in 99% mortality of the bivalve. The studies above shared the goal of removing fouling organisms, but these organisms, unlike the oyster, were incapable of isolating their soft-body parts from the treatment. Treatments to kill oyster spat must necessarily be more rigorous. This is apparent from the data of Shearer and Mackenzie (1961); treatments that lead to 100% mortality in boring sponges, starfish, and tunicates caused only 2.3% mortality in C. virginica spat. Our tests in 1991 revealed that up to 86% of oyster spat mea- suring 5.0 mm or less can be removed using immersion for 10 minutes in a 200 ppt brine solution, followed by a 3 or 6 hour aerial exposure. These treatments did not hurt adults. In spat be- tween 4 and 12 mm. up to 88%' were removed with the same treatment, followed by a 6 hour exposure. However, for large spat (5-22 mm), removal by brine dip becomes ineffective. We con- clude, based on the results of our preliminary field trials and the data shown here, that best results can be obtained from treatment of 5-10 min in 200 ppt brine, followed by a 6 hour aerial expo- sure. But of key importance is treating spat at a very small size, below 5 mm. Better yet, treatments should be most effective if they begin within days of the overset. ACKNOWLEDGMENTS We thank Bob Wargo for preliminary data for this study and for assisting with experimental design. This work was partly spon- sored by the New Jersey Marine Science Consortium Mini-grant Program and partly by the National Coastal Resources Research and Development Institute (NCRI) AQ 106. 90-56 18-43. This is Publication No. D-32100-I-93 of the NJAES and Contribution #93-05 of the Institute of Marine and Coastal Sciences, Rutgers University. LITERATURE CITED Arakawa, K. 1980. Prevention and removal of fouling on cultured oysters: a handbook for growers. (Translated by R. B. Gillmore.) Maine Sea Grant Program, University of Maine, Orono. ME. Technical Report No. 56. 38 pp. Dealteris. J. T.. R. C. Bullock & W. L. Romey. 1988. Alternative treat- ments to prevent the biodeterioralion of offshore wood lobster traps by the wood-bonng bivalve, Xylophaga allannca. J . Shellfish Res. 7.445- 451. Loosanoff, V. L. 1958. New method for control of oyster enemies with common salt. U.S. Fish Wildl. Ser. Comm. Fish. Rev. 20(l):45-47. MacKenzie, C. L. & L. W. Shearer. 1961. Chemical control of Polydora websleri and other annelids inhabiting oyster shells. Proc. Nail. Shell- fish Assoc. 50:105-111 Shearer, L. W. & C. L. Mackenzie. 1961. The effects of salt solutions of different strengths on oyster enemies. Proc. Natl. Shellfish Assoc. 50:97-104. Sokal, R. R. & F. J. Rohlf. 1981. Biometry. W.H. Freeman and Com- pany. New York. 859 pp. Wilkinson, L, 1990. SYSTAT: The system for statistics. SYSTAT. Inc., Evanston. Illinois. 676 pp. Journal of Shellfish Rcst'unh. Vol 12, No. 1. 35-40. 1993. OBSERVATIONS ON THE PEARL OYSTER FISHERY OF KUWAIT S. M. ALMATAR, K. E. CARPENTER,* R. JACKSON, S. H. ALHAZEEM, A. H. AL-SAFFAR, A. R. ABDUL GHAFFAR AND C. CARPENTER* Kiiwaii Institute for Scientific Research Mariciilture and Fisheries Department P.O. Bo.x 1638. Sahniya 22017 Kuwait ABSTRACT The pearl oyster fishery of Kuwait was monitored daily from January 1989 to May 1990. Landings of pearl oysters in 1989 totalled 287 tons with a market value of U.S. $1 .0 million. Commercial pearls (>3 mm) were estimated to be present in one of every 4200 oysters. Most of the pearl oysters landed were new recruits with hinge lengths between 40-56 mm. There was a curvilinear relationship between total weight and size of oysters (length) and the sex ratio approached 1 ; 1 . Spawning occurs throughout the year, with a spat settlement peak in early fall. Over the size range examined there was no relationship between the size of oysters and the size of pearls and subsequent resource management strategies are discussed. A'£)' WORDS: pearl, oyster. Pinaada radiala, fishery INTRODUCTION Thriving from historic times until the 1930s, the traditional peari oyster fisherv- in the Arabian Gulf was large and revered, furnishing about SO'/t of the worid production of natural pearls, which were famous for their excellent shape and quality (Bowen 1951). Lorimer (1915) described the various peari oyster banks in the Arabian Gulf (Fig. 1) and estimated the average yeariy export values of pearls and mother-of-pearl (shells) to be Pounds Sterling 561.353 and 269.788. respectively, for the period 1873 to 1905. The annual catch for the entire Arabian Gulf was approximately 35,000 tons, a conservative estimate calculated from literature reports of catch rate, number of boats and number of fishermen (Lonmer 1915. Villiers 1969). Bowen (1951) described the early pearl diving techniques and discussed various aspects of the industry. Peari fishing in the Gulf was peiformed originally only during summer. May to September. Except for occasional inclement weather, diving was a continuous operation over this period. The traditional fishery declined steadily from 1930 onwards because of a world recession, the introduction of Japanese cultured pearls, and later with the discovery of oil in the area. In the late 1940s most people deserted the peari industry for more lucrative oil-related positions. In the late 1960s, pearl fishing was revived with the introduc- tion of modem diving equipment, such as air compressors and speedboats. Since 1980, peari oyster fishing is practised year round in Kuwait. A peari oyster market was re-established in Kuwait in 1982, and the first catch statistics were reported for a five-month period in 1983 (Almatar et al. 1984). The present pearl oyster market of Kuwait is based exclusively on natural pearls from Pinctada radiata (Leach), (Khamdan 1988). In the Arabian Gulf this species has variously been referred to as P. margaritifera (Steininger 1968, Anderiini et al. 1981, Almatar et al. 1984). P. fucata (Mohammad 1976) and P. radiala (Sadig and Alam 1989). The objective of this report is to review peari oyster landings. *Present Address: Food & Agriculture Organization of the United Nations, Via le delle Terme di Caaracalla. 00100 Rome. Italy. describe size composition and frequency of pearl occurrence and discuss resource management strategies in light of the present findings. MATERIALS AND METHODS Individual boat fishing activity and catches were monitored daily at the single pearl oyster market in Kuwait by interviewing fishermen in the market place. Fishing effort was calculated by multiplying the number of boats by average number of diving hours; the latter was estimated via interviews and direct observa- tion. Monthly size frequency distribution of the oyster hinge length (HL) were obtained from samples (200-300 oysters) purchased twice a month. AUometric measurements (maximum dorso-ventral height. (DVM). total oyster weight and wet meat weight) were obtained from subsamples. Shell measurements were measured to the nearest 0.1 mm using Vernier calipers. Oysters were cleaned of external fouling material and wiped dry before weighing to the nearest 0.1 g. Oyster meats were shucked from the shell and weighed individually. Sex was determined by gonad color: fe- males were yellow-orange throughout study and mature males were milky white when sexually active or brown-yellow in the resting stage. Oysters of undetermined sex were recorded as im- mature. Wet mounts of gonads were conducted frequently to con- firm sex. RESULTS The Current Fishery The diving fleet during this study consisted of 25 speedboats (3-8 m OAL). most with a single diver. Eleven major pearl oyster beds, varying in size from one to several square kilometers (10-20 m deep), were scattered within the fishing grounds (Fig. 2). An average of six 30-minute dives per day per diver were conducted using hookah air supply systems between 8 a.m. and 12 noon. Divers hand-picked oysters and placed approximately 6 kg in a 35 36 Almatar et al. SAUDI ARABIA 100 _I 200 -J st- UNITED ARAB EMIRATfS J L Figure 1. Location of traditional pearl oyster beds in the Arabian Gulf (from Lorimer 1915). mesh bag. Unsorted oysters were sold to buyers at the market who later opened the oysters to retrieve any pearls which were subse- quently resold. Catch Statistics and Fishing Effort The mean daily landing of pearl oysters in 1989 was 865 kg, and varied from 146 kg in January to 1716 kg in July (Fig. 3). Landings in 1989 totaled 287 tons or about 6.3 x 10'' oysters. worth approximately U.S. $1.0 million. Landings varied directly with effort; highest effort occurred between June and October. The poor weather/diving conditions between December and March accounted for the lowest effort (Fig. 3). The average catch perhour of diving (CPUE) in 1989 was 37 ± 17.4 kg (n = 12); this is a slight overestimate since some diving boats occasionally carried more than one diver. CPUE was lowest in January 1989 and highest in July 1989. Earlier data from 1983 also showed that landings and CPUE increased steadily from May to September 1983 (Fig. 3) (unpublished data). Size Composition Total shell and meat weight, wet flesh weight and hinge length (HL) are presented by size groups in Table 1. Quarterly size fre- quency histograms are shown in Figure 4. The HL of the majority of pearl oysters were unimodal and ranged between 40-56 mm. Oysters less than 40 mm HL were landed throughout the year, but were most abundant in fall and winter. A linear relationship exists between HL and maximum height (DVM) measurement; DVM = -16.863 -I- 1.619 (HL) (r^ = 0.79; n = 120) The size-weight data (Fig. 5) are best described by curvilinear relationships of the form Y = aL** (where Y is total weight in g and L is length in mm) as follows; Log (Wt) = -5.655 + 4.253 log (HL) (r^ = 0.78; n = 120) Log (Wt) = -4.246 -I- 3.228 log (DVM) (r- = 0.97; n = 120) Pearl Oyster Fishery of Kuwait 37 4erio' 20' Figure 2. Location of major pearl oyster beds in Kuwait waters; shaded areas offshore indicate oyster beds. Pearl Harvest Ninety-six of 4414 oysters sampled (2.2%) bore one or more pearls. Oysters with multiple pearls accounted for 17.7% of all pearl-bearing oysters. All pearls recovered from the study were too small (1.53 ± 0.88 mm: n = 132) to be of commercial value. Table 2 displays pearl harvest by size and location over a range of oyster sizes. Pearls found in the mantle were significantly larger (Student's t-test; t < 0.05) than those in the gonad. No pearls were found in oysters with HL less than 40 mm and there was no significant correlation between oyster size and pearl size (r = 0.003, df = 130) from the oysters examined. However, the prob- ability of pearl occurrence increases with size of oysters. Three percent of oysters less than 58 mm HL contained pearls whereas the frequency increased to 5 percent for those over 58 mm HL. From the 5.9 x lO*" oysters landed from June 1989 to January 1990, only 400 large pearls (>4 mm) and 984 small pearls (3^ mm) were sold in the market. Thus, the probability of landing a commercial-sized pearl is one in 4200. This estimate is slightly skewed because a few pearls were sold outside the oyster market. Maturity Figure 6 reveals that both sexes matured at the same size (50 mm HL) and there was no evidence of a sex change with size in P. radiata, as has been reported for other species (Tranter, 1958). Sex ratio over the period of sampling approached 1 ;1 . Because of JFMAMJJASONDJF Months Figure 3. Mean monthly landings and mean monthly catch per unit of effort (CPUE) of P. radiata for 1989 and from January to May 1990, Data from May to September 1983 are also shown for comparison. CPUE is defmed as amount (kg) of oyster harvested per hour. the high growth rate of oysters, maturity is probably reached in the first year, and for those spawned in early spring, possibly during the first six months (Tranter 1958; Rose et al. 1990). The presence of small oysters (<40 mm HL) throughout the year indicates that spawning is continuous with the greatest activity occurring in the summer and late fall (Fig. 4). DISCUSSION Compared with the harvest rates of pearl oysters in other trop- ical areas, Kuwaiti waters are highly productive (Pragasam and TABLE 1. Total weight and wet flesh weight (mean ± standard deviations) in relation to 2 mm size intervals of HL for subsample of P. radiata landings HL (mm) n Total Weight (g) Flesh Weight (g) 22-23 1 1.17 0.34 — 24-26 1 8.51 2.51 — 30-32 3 3.45 ± 1.64 0.96 ± 0.40 33-35 2 7.28 ± 3.11 2.10 ± 0.74 36-38 5 8.35 ± 2.54 2.38 ± 0.41 39-41 4 13.83 ± 2.16 3.90 ± 0.46 42-44 8 23.63 ± 14.18 7.23 ± 4.33 45^7 22 34.46 -*- 10.40 10.22 ± 2.91 48-50 14 44.23 ^ 18.35 13.62 ± 5.24 51-53 16 49.20 ± 18.58 15.92 ± 6.66 54-56 18 64.08 ± 19.05 21.89 ± 5.60 57-59 16 60.85 -h 14.72 21.25 ± 3.35 60-62 9 64.14 ± 25.56 24.08 ± 5.32 63-65 1 81.83 — 29.75 — (n number of oysters). 38 Almatar et al. > o c 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 JAN-MAR N-670 17 16 - 15 - 14 - 13 - 12 - 11 - to - g - 8 - 7 - 6 - 5 - 4 - 3 1 2 iTtMltffff ffM If? 1 Tt 1 JULY-SEPT N-865 ff f f f fMr r f ffTtll 17 OCT-DEC N-1042 16 - 15 - 14 - 13 - 12 - 11 - 10 - 9 - 8 - 7 - 6 - 5 - 4 - 3 1 2 1 1 ;.^.fiiuuuiul — Im+ 1 1 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 HL (mm) Figure 4. Quarterly hinge length frequency distribution at 2 mm intervals of P. radiata collected from landings in the pearl oyster market during 1989 and early 1990. Dev 1987; Easwaran et al. 1969; Dybdahl and Rose 1986). In- deed, catch data from this study are relatively high — 780 oysters or 37 kg per hour of diving. Direct observation and interviews indicate that our CPUE calculations may have been overestimated by 25 percent at most. Due to the high annual harvest rates, the fishery in Kuwait relies heavily on recruitment of young oysters. Narayanan and Michael (1968) reported a growth rate off. vulgaris in the Gulf of Kutch of 38.4 mm HL in the first year while Jeyabaskaran et al. (1983) reported a growth rate of P. fucala in the Gulf of Mannar Pearl Oyster Fishery of Kuwait 39 20 25 30 35 55 60 65 40 45 50 HL (mm) Figure 5. Total weight versus HL for P. radiata collected throughout the course of the study. The length-weight relationship is W = 0.00000221 L^ ' (r^ = 0.78, n = 120). of 41.2 mm in its first year. Nayaret al. (1992) concluded that the growth of P. radiata in Bahrain waters was higher than the growth of pearl oysters in the Gulf of Mannar (corroborates unpublished data of the present study). It appears that the majority of the commercial Kuwaiti catch is composed of 0+ or 1 + year-classes. Although increased effort yields increased landings, other fac- tors that affect landings are poorly understood. Annual fluctua- TABLE 2. Number, size (mean diameter ± s.d.) and location of pearls found in relation to 2 mm HL intervals of pearl oysters. Number Oyster with Pearl Locatior Mantle Gonad HL Oysters size size (mm) Searched Pearls n (mm) n (mm) <40 127 — — — — — — 41 376 1 3 1.0 ± 0.3 — — 45 355 4 3 1.4 ± 0.7 1 0.5 — 47 501 13 10 1.8 ± 1.0 8 1.3 ± 0.3 49 572 11 12 2.3 ± 1.7 2 1.0 ± 0.1 51 626 15 10 1.1 ± 0.8 7 1.2 ± 0.4 53 564 17 14 1.9 ± 0.8 8 1.3 ± 0.3 55 391 7 7 1.6 ± 0.3 2 1.1 ± 0.2 57 294 7 6 2.0 ± 0.8 6 1.4 ± 0.8 59 188 10 — — 13 1.5 ± 0.9 61 106 6 9 1.1 ± 0.1 5 1.1 ± 0.2 63 52 3 1 2.6 — 2 1.8 ± 0.1 65 28 1 — — 1 2.0 — >67 8 1 — — 1 2.5 — X Female N-840 -I- Male N-995 — Both sexes 22 26 30 34 38 42 46 50 54 58 62 66 HL (mm) Figure 6. Percent HL frequency distribution of male and female and cumulative HL frequency distribution of combined sexes for P. radiata collected during 1989 and early 1990. tions and sources of spat settlement are virtually unknown. Exten- sive beds in Saudi Arabian waters could be significant sources of spat, since they lie within protected zones near oil wells. There are no data to suggest that the present fishing pressure (about 1000 hours per month) is sustainable. Ongoing data collection subse- quent to the Gulf War (May-November. 1992) indicates no ap- parent change in fishing pressure, but there is a slight drop in catch volume. This decreased volume may be due to overharvest or environmental damage to oyster beds caused by oil spills and combustion products of oil fires during the war. The objective of managing the pearl oyster fishery is not to maximize the landings of oysters, but rather to maximize the value of pearls — via increased numbers or sizes of pearls. The present study found no relationship between pearl size and oyster size over the range 40-68 mm HL and agrees with earlier findings for the same species (Almatar et al. 1984). Large valuable pearls (>3 mm) observed in the market (not taken from our subsamples) were not found in unusually large oysters. This study did find an in- crease in occurrence of pearls in larger (>58 mm HL) versus smaller (<58 mm HL) oysters. Studies of other species have also related pearl yield to age of oyster (Pearson 1933; Easwaran et al. 1969). Results of this study indicate that the total value of pearls could be increased if the fishery were managed to promote harvest of oysters greater than 58 mm HL. ACKNOWLEDGMENTS This study was partially supported by the Public Authority of Agriculture and Fishery of Kuwait. Thanks to Dr. M. Saif, head of the Mariculture and Fisheries Department, for his support through- out the study and to Dr. J. Bishop for his critical review on the early drafts of the manuscript. 40 Almatar et al. LITERATURE CITED Almatar, S. M., G. R. Morgan & S. Hakim. 1984. The pearl oyster fish- ery of Kuwait. In: C. P. Mathews (ed.). Proc. of Shrimp and Fin Fisheries Management Workshop, 9-11 Oct., 1983. Kuwait Institute for Scientific Research No. 1366, 2:575-600. Anderlini. V. C, L. Al-harmi, B. W. De lappe, R. W Risebrough, W. Walker, B. R. T. Simoneit & A. Newton. 1981. Distnbution of hy- drocarbons in the oyster, Pinclada margaralifera . along the coast of Kuwait. Mar. Pollul. Bull. 12:57-62. Bowen, R. LeB. 1951 . The pearl fisheries of the Persian Gulf. Middle East yowrna/ 5(2): 161-180. Dybdahl, R. & R. A. Rose, 1986 The pearl oyster fishery in Western Australia. In: A. K. Haines, G. C Williams & C. Coates (eds.). Torres Strait Fisheries Seminar. Port Moresby. 11-14 Feb. 1985. Aust. Gov. Publ. Service. Canberra: 122-132. Easwaran,C. R.,K. R. Narayanan & M. S.Michael. 1969. Pearl fisheries of the Gulf of Kutch. J. Bombay Nut. His!. Soc. 66:338-344. Jeyabaskaran, Y., D. S. Dev. I. Nalluchinnappan & N. Radhaknshnan. 1983. On the growth of the pearl oyster Pinclada fucata (Gould) under farm conditions at Tuticorin, Gulf of Mannar. Proc. Symp. Coastal Aquacullure 2:587-589. Khamdan, S. A. 1988. The Bahrain pearl oyster: Their genetics and sys- tematics. Univ. College of North Wales. U.K. MS Thesis. 132 pp. Lonmer.J. G. 1915. Gazetteer of the Persian Gulf. Appendix C. The pearl and mother-of-pearl fishenes of the Persian Gulf, 2:2220-2293, Cal- cutta, India. Mohammad, M-B. M. 1976. Relationship between biofouling and growth of the pearl oyster Pinclada fucata (Gould) in Kuwait, Arabian Gulf. Hydrobiologia 51:129-138. Narayanan, K. R. & M. S. Michael. 1968. On the relation between age and linear measurements of pearl oyster, Pinclada vulgaris (Schuma- cher), of the Gulf of Kutch. J. Bombay Nat. Hist. Soc. 65:441-452. Nayar, K. N., M, Al-Rumaidh & H. Al-Sayed. 1992. Expenmental study of the settlement and collection of spat of the pearl oyster Pinctada radiata (Leach) from Bahrain waters. In: Proc. of Symp. on Maricul- lure Technology and Investment Opportunities. 9-10 May 1992. Bah- rain Centre for Studies and Research, 16 p. Pearson, J. 1933. The maximum pearl-yield of a pearl oyster bed. Ceylon J. Sci. (Vl:l-20. Pragasam, B. & D. S. Dev. 1987. Studies on the pearl oyster population in pearl oyster grounds off Tuticorin in the Gulf of Mannar. In: The Seas around India. Spl. Pub. Bull. Cent. Mar. Fish. Res. Inst. 42:79- 83. Rose, R. A., R. E. Dybdahl & S. Harders. 1990. Reproductive cycle of the Western Australian Silverlip Pearl Oyster, Pinclada maxima (Jame- son) (Mollusca: Ptenidae) J. Shellfish Res. 9:261-272. Sadig. M. & I. Alam. 1989. Metal concentrafion in pearl oyster, Pinclada radiata. collected from Saudi Arabian coast of the Arabian Gulf. Bull. Environ. Contam. Toxicol. 42:111-118. Steininger, F. 1968. Recent marine molluscs. In: W Fuchs, T. E. Gat- tinger & H. F. Holzer (eds. ). Explanatory texi to the Synoptic Geologic Map of Kuwait. Geologic Survey of Austria. Vienna. 87 pp. Tranter, D. J. 1958. Reproduction in Australian Pearl Oysters (Lamelli- branchia) I. Pinctada albtna (Lamarackl: Pnmary gonad development. Ausi. J. Mar. Freshwat. Res. 9:135-143. Villiers, A. J. 1969. Sons of Sinbad. Scnbner. N.Y. 414 pp. Journal oj Shellfish Research. Vol. 12. No. 1. 41-47. 1943. ULTRASTRUCTURAL STUDY OF GAMETOGENESIS IN THE FRENCH POLYNESIAN BLACK PEARL OYSTER PINCTADA MARGARITIFERA (MOLLUSCA, BIVALVIA). I— SPERMATOGENESIS. MARYSE THIELLEY,' MAURICE WEPPE/ AND CHRISTIAN HERBAUT' ^Universite Fram^aise du Pacifique BP 6570 Facia Airport Tahiti, French Polynesia ^Aqnacop. IFREMER Centre Oceanologique du Pacifique BP 7004 Taravao Tahiti. French Polynesia ABSTRACT Ultrastructure of the germinal cells is descnbed throughout the spermatogenesis, in the French Polynesian black pearl oyster. Pmctada margcirilifera (L.. 1758) var. cumingii (Jameson 1901). Special emphasis is given to the spermatozoon structure descnption. Abnormal spermatogenesis and processes of degeneration and resorption of residual germinal cells are also reported. Male germinal cells present a centnpetal evolution in the acini. Germinal cells denving from a same germinal lineage, are connected among themselves and among one auxiliary cell by cytoplasmic bndges. The mature sperm of this species is of the primitive type, with a short acrosome and without axial rod. The spermatozoa are 45-50 (xm long. Midpiece contains two centrioles along with satellites bodies and four or five mitochondria. KEY WORDS: spermatogenesis, ultrastructure. mollusc, bivalve. Pinclada margaritifera INTRODUCTION / . Normal Spermatogenesis Initial study of spermatogenesis of the commercially important black pearl oyster was carried out with light microscopy on Aus- tralian specimens (Tranter 1958) and on French Polynesian spec- imens (Thielley 1989). In French Polynesia, the black pearls provide the major source of exportation revenues for the Territoi7. As a result of this eco- nomic importance, a wide program of research focused on the biology of this species has recently been carried out. The present study is part of this program and describes the ultrastructure of the germinal cells including normal and abnormal spermatogenesis with emphasis on the spermatozoon. MATERIALS AND METHODS Pinclada margaritifera adult specimens were collected monthly between April 1990 and December 1990, from the natural stock and from a farm at Takapoto atoll's lagoon (Tuamotu, Ar- chipelago French Polynesia). Samples of male gonad were fixed for 3 hours in 3% glutaral- dehyde in 0.4 M cacodylate buffer (pH 7.4; 570 mosM). The tissues were then washed in the buffer solution and postfixed for one hour in l'7c osmium tetroxide in the buffer. After dehydration by ethanol. the pieces were embedded in Spurr resin. Semi thin sections (1 mm) were stained with toluidine blue. Ultrathin sec- tions (600 A) were contrasted with uranyl acetate and lead citrate and examined under a JEOL TEM 200 transmission electron mi- croscope. A few sections collected on gold grids, were treated for the detection of glycogen (Thiery and Rambourg 1974). RESULTS The male germinal cells are gathered in acini and present a centripetal evolution. Spermatogonia Stem Cells The spermatogonia stem cells stick largely to the acinus wall (Fig. 1). They are oval in section, with an average size of 14 (xm X 8 ji,m. Their nucleus can reach 9 jxm in length and 5 n.m in width. The chromatine is uniformly dispersed in small aggregates, and gives a fine granular aspect. One or two nucleoli are present. In their abundant cytoplasm, a large number of mitochondria, often oval in section, are gathered in two heaps at two poles of the cell. Rough endoplasmic reticulum, a few dictyosomes, ribosomes and few dense inclusions are present. Spermatogonia Primary and secondary spermatogonia are very similar in as- pect. They often adhere to the acinus wall. The spermatogonia are spherical or oval and measure about 7 fjim in length. The nucleus is 5 |jLm in diameter, containing a single nucleolus and small clumps of electron dense chromatine. Their cytoplasm is reduced and contains some dictyosomes, osmiophilic inclusions, a large number of mitochondria, ribosomes and endoplasmic reticulum (Figs. 1. 2). Auxiliary cells, adhering to the acinus wall, can be observed between the spermatogonia (Fig. 2). Their nucleus is elongated, about 5 Jim in length and 1.5 iJim in width. They present a scat- tered chromatin essentially peripherical. The cell is polymorphic and notably emits cytoplasmic digitations that infiltrate between the spermatogonia. Their cytoplasm contains mitochondria, a lot of glycogen particles, endoplasmic reticulum and sometimes my- elinic formations and dense lysosomial formations. Desmosome- like junctions can be observed between auxiliary cells and sper- matogonia. One auxiliary cell can bridge several spermatogonia deriving from the same germinal lineage. These spermatogonia are connected between themselves by cytoplasmic bridges. 41 42 Thielley et al. P^l^ '' ^;^f: . b .'*-i|;.-V r ■♦ Spermatogenesis in Black Pearl Oyster 43 Spermatocytes Primary and secondary spermatocytes lie at the periphery of the acini, within groups of 10 to 20 synchronous cells. Primary spermatocytes: The cells have a similar size to spermatogonia. Their cytoplasm also contains a complement of organelles very similar to them. In addition, centrioles can be sometimes observed. The nucleus un- dergoes complex morphologic changes during the early meiotic prophase. Distinguishable stages of prophase are; Leptoteii stage: the chromatine is set out in very dense clus- ters. Some vacuoles appear in the 5 |jim wide nucleus. Zygoten and pachyten stages: the zygoten stage is character- ized with the rise of synaptonemal complexes formation. As soon as these formations are fully completed all along the chromosome, the spermatocyte enters the pachyten stage. Electron-dense chro- matin forms a network, where synaptonemal complexes can be observed. The nucleus size increases to 6 |jim in diameter (Fig. 3). Diploten-diacinese stage: the nucleus has a similar size as in the previous stage, but chromatin appears more condensed. Intra- nuclear vacuoles can be distinguished. Secondar>' spermatocytes: The average size of the secondary spermatocyte is about 5 |jim. Its nucleus has a variable shape and the nuclear envelope is not easily observed. The highly dense chromatin fills in most of the space in the nucleus. Its cytoplasm contains some mitochondria, ribosomes, vacuoles and golgi bodies (Fig. 4). Spermatids The size of the spermatid is about 4 jj-m. Its nucleus is spherical and about 2 (im in diameter. Condensation of the chromatin takes place throughout the spermatogenesis. During the process, the few clear areas between the chromatin masses reduce in size and then disappear. The spermatid cytoplasm contains ribosomes, a single dictyosome, mitochondria and two centrioles. In the early stages of spermatid development, the mitochondria forms a collarette all around the nucleus (Fig. 5). They gather towards the basal pole of the future spermatozoon and fuse into only four or five voluminous mitochondrial spheres. The centrioles appear at the basal pole and move to an orthog- onal position (Fig. 6). Very soon, the distal centriole produces a caudal flagellum. The acrosomial vesicle appears at the basal pole. Originally spherical, about 0.6-0.7 (jim in diameter, it becomes flat and then slightly incurved against the nucleus when it migrates to the apical pole, where it takes a half sphere shape. Two major different electron-dense regions are discemable within the acroso- mial vesicle (Fig. 7). During spermiogenesis, most of the cytoplasm evaginates from the maturing spermatids and is eliminated as free masses into the acinus lumen. Spermatozoa The spermatozoon of Pinctada margantifera is of the primitive type according to Franzen ( 1983). It is 45-50 jxm in length. It can be divided into three parts: sperm head consisting of the nucleus and acrosome, middle piece consisting of two centrioles and mi- tochondria, and tail. The acrosome is invaginated at its adnuclear surface and forms a conical structure which is 0.9 (xm in diameter and 0.5 |xm in height (Fig. 8). The acrosome consists of three major electron- dense materials. One has an electron-low density and makes up the enlarged basal part of the cone. The apical part of the acrosome is formed of an highly electron dense material that includes a lamel- lar structure (Fig. 9). A third substance of intermediate density covers the both other materials, constituting so the walls of the acrosomial cone. Between the plasma membrane and the cone, as well as in the central lumen of the acrosome, a fine granular material accumulates. The spherical electron-dense nucleus is 1.7 (xm in diameter, and presents a large anterior invagination, 0.3 |xm in depth, where fine granular material accumulates (Fig. 8), and a smaller posterior invagination, 0.2 jxm in depth (Fig. 4). The midpiece contents a ring of four or five mitochondria of about 0.8 |xm in diameter (Figs. II, 12) around two centrioles (Fig. 10). The two centrioles are connected to each other at right angles, and show the classical nine triplets of microtubules. The proximal centriole is joined to the nuclear envelope by a satellite body found in the post-nuclear fossa (Fig. 14). The distal centriole forms the basal body of the flagellum. It is connected to the plasma mem- brane by radiating satellite bodies (Figs. 10, 14). Granules of glycogen are detected by the reaction of Thiery, essentially be- tween mitochondria but also around the nucleus (Fig. 13). The flagellum is about 45 |jim long and shows the classical structure of nine external and one internal microtubule doublets (Fig. 15). 2. Abnormal Spermatogenesis Abnormal cells are more or less numerous according to the specimens. Plurinuclear cells can be observed at most of stages of the spermatogenesis. Up to six nucleus appear more particularly at the spermatogonia stages (Fig. 16). Figure 1. Section through an acinus. Acinus wall (W); stem cell attached to the acinus wail (S); spermatogonia (Spg); primary spermatocytes in metaphase (Spc). Bar = 3 (i.m. Figure 2. Auxiliary cell (AC), attached to the acinus wall (W). Myelinic formation (my); spermatogonia (Spg); desmosome-like junction (arrow). Bar = 2 )im. Figure 3. Primary spermatocytes (Spcl) in zygoten-pachyten stage characterized by the presence of synaptonemal complexes (SC). Bar = 2 p.m. Figure 4. Secondary spermatocyte (Spc2). Bar = 2 (im. Figure 5. One of the early stages of spermatid development. Numerous mitochondria (m) form a collarette all around the nucleus (N). Acrosomial vesicle (a) is spherical. Bar = 500 nm. Figure 6. Spermatid. Mitochondria (m) are in the basal pole around proximal centriole (pc) and distal centriole (dc) which begin to elaborate a caudal flagellum (f); acrosomial vesicle (a); nucleus (N). Bar = 500 nm. Figure 7. One of the last stages of the spermatid. Mitochondria (m); proximal centriole (pc); distal centriole (dc); acrosomial vesicle (a) in the apical pole of the future spermatozoon; cytoplasm (c) is still abundant; nucleus (N). Bar = 1 (i,m. 44 Spermatogenesis in Black Pearl Oyster 45 Binucleatcd spermatocytes and spermatids arc frequent (Figs. 17. 18). An intracytoplasmic flagellum was observed in some sperma- tids (Fig. 19) and more scarcely in the spermatozoa (Fig. 20). Occasionally, two tlagcllum complexes were observed inside the same plasma membrane (Fig. 21). 3. Gametic Degeneration and Resorption Male germinal cell degeneration can occur at any developmen- tal stages. Main degeneration aspects are caryolyses and cyto- plasm alterations. The nucleus can present an hypercondensation of its chromatin (Fig. 22) or sometimes diffused chromatin with lysis of the nuclear envelope (Fig. 23). Main alterations of the cytoplasm arc numerous vacuoles, huge lysosomial inclusions and altered mitochondria. Such aspects of male germinal cells degen- eration make it difficult for their classification into a particular cellular type. The degenerative germinal cells, more or less deg- radated. can be driven out by the genital duct. Some residual cells can also be resorbed in situ: macrophage cells (12-15 |j.m in length) are often observed inside the acini (Fig. 24). Degenerative cells and residual bodies of the gametes, can be phagocyted by these cells. DISCUSSION The processes of spermatogenesis described in Pinctada mar- garitifera are similar to other studies reported on other bivalves molluscs (Hodgson and Bernard 1986, Dorange and Le Pennec 1989). Four or five large mitochondria may be the result of a fusion of smaller mitochondria (Dorange and Le Pennec 1989, Hodgson and Bernard 1986). The spermatozoon of Pinctada margaritifera is typically of the primitive type (Franzen 1983). The spermatozoon type is in direct relation with oocyte's reproduction and morphology (Franzen 1983). According to this author, spermatozoa of the primitive type are usually associated with species having external fertilization and small oocytes. Such is the case in the Pinctada margaritifera species. The spermatozoon head differs in size, form and structure from the one described in many other bivalves in T.E.M. In the Mytil- idae species, the acrosome structure seems to be more complex, particularly with the presence of an axial rod (Bourcart et al. 1965, Hodgson and Bernard 1986). This axial rod is also present in Crassostrea virginica (Daniels et al. 1971) and Crassostrea an- gulata (Gutierrez et al. 1978). In Chama macerophylla and Spisula solidissima spermatozoa, Hylander and Summers (1977) report the presence of two major constituants of the acrosomial vesicle: an electron-dense acroso- mial material as the "basal ring", and a less dense homogeneous material in the central and anterior portion of the acrosome. The acrosome of Pinctada margaritifera contains three major materials of different electron density, but, unlike the two previous species, the electron dense material occupies the apex of the acrosome and the less dense zone forms the basal ring of the conical acrosome. In many mollusc species, the acrosome shows a lamellar struc- ture (Popham et al. 1974, Dorange and Le Pennec 1989, Franzen 1983). This type of structure has been observed in P. margaritif- era. According to Hylander and Summers (1977). the acrosome structure can be correlated with the oocyte vitelline envelope. Accumulation of granular material around the acrosome and especially in the central lumen, has often been described in many bivalve species (Hodgson and Bernard 1986, Dorange and Le Pennec 1989, Popham et al. 1974, Hylander and Summers 1977). We also observed this granular material in P . margaritifera. which possibly binds the acrosomal vesicle to the nuclear envelope, ac- cording to Popham et al. (1974). Franzen (1983) describes the midpiece as a stable structure in bivalve molluscs. The number of mitochondria is variable between and inside species. The spermatozoa of Mytilus galloprovinciatis and Aulacomya described by Hodgson and Bernard (1986), present five or six mitochondria. Crassostrea virginica spermato- zoon has four (Daniels et al. 1971 ). Mytilus perna has five mito- chondria, very rarely four (Bourcart et al. 1965). Hodgson and Bernard ( 1986), in Choromytilus meridionalis and Dorange and Le Pennec (1989) in Pecten maximus observed four mitochondria, rarely five. According to the latter authors, the presence of five mitochondria is abnormal. In Pinctada margaritifera. four or five mitochondria were observed, with a 1/1 ratio. Therefore, it is difficult to conclude that four or five mitochondria give abnormal or normal spermatozoon. A satellite body found in the postnuclear fossa, as a connection between the proximal centriole and the nuclear envelope, has been described by Popham et al. (1974) in Bankia australis and Bankia carinata and by Daniels et al. (1971) in Crassostrea virginica. A similar structure is observed in Pinctada margaritifera . The pres- ence of this satellite body has not been reported in Mytilidae by Hodgson and Bernard (1986) and Bourcart et al. (1965). Franzen (1983) in his study of three Bivalve species and Dorange and Le Pennec (1989) in Pecten maximus did not describe this structure. Otherwise, satellite bodies form connections between the distal centriole and the plasma membrane, at the basal part of the distal centriole. This structure is widely described by many authors about many species (Popham et al. 1974, Dorange and Le Pennec 1989, Franzen 1983). Our study and other observations on bivalves spermatozoon Figure 8. Longitudinal section through the acrosome showing the three major electron-dense materials. Anterior invagination (Al) where fine granular material (gm) accumulates. Bar = 300 nm. Figure 9. Transverse section through the acrosome showing the three major electron-dense materials and lamellar structure (LS). Bar = 200 nm. Figure 10. Longitudinal section of spermatozoon. Acrosome (a): nucleus (N); mitochondria (m); proximal centriole (pc); distal centriole (dc); radiating satellite bodies (rs); flagellum (f). Bar = 500 nm. Figure 11. Transverse section through the midpiece showing four mitochondria )m). Bar = 400 nm. Figure 12. Transverse section through the midpiece showing five mitochondria (m). Bar = 400 nm. Figure 13. Longitudinal section through a spermatozoon. Granules of glycogen (G) are detected by the reaction of Thiery. Nucleus (N); mitochondria (mi; acrosome (a). Bar = 500 nm. Figure 14. Longitudinal section through the midpiece showing a satellite body (s) in the posterior invagination (PI) and radiating satellite bodies (rs) connected to the plasma membrane (pm); flagellum (f); noyau (N). Bar = 500 nm. Figure 15. Transverse section through a flagellum showing the classical structure 9 external and 1 internal microtubule doublets. Bar = 50 nm. 46 Vf ^'' ^^f '¥ Spermatogenesis in Black Pearl Oyster 47 Figure 16. Multinuclear spermatogonium. Nucleus (N). Bar = 2 (im. Figure 17. Multinuclear spermatocytes. Bar = 4 (im. Figure 18. Binuclear spermatid. Bar = I |im. Figure 19. Spermatid with intracytoplasmic flagellum (f). Bar = I pm. Figure 20. Spermatozoon with intracytoplasmic flagellum. Bar = 500 nm. Figure 21. Transverse section through a flagellum with two flagellum complexes. Bar = 200 nm. Figure 22. Degenerating male germinal cells (dgc) showing an hypercondensation of their chromatin. Bar = 2 (im. Figure 23. Atretic spermatozoon. Flagellum Ifl. Bar = 500 nm. Figure 24. Macrophage. Nucleus (N). Bar = 2 ^m. structures, show that many differences in general morphology and structure are evident between the different species of the same family. Our results are in accordance with earher studies that suggest the ultrastructure of the speim can be used for identifica- tion purposes, and represents a significant taxonomic and phylo- genic criterion (Franzen 1983; Hodgson and Bernard 1986; Daniels et al. 1971; Popham et al. 1974). Multinuclear cells are reported during gamete evolution in Pecten maximus by Dorange and Le Pennec (1989) and in Mya arenaria by Allen et al. (1986). Dorange and Le Pennec (1989) have also observed atypical spermatozoa with intracytoplasmic flagellum. According to Fain-Maurel (1966) and Dohmen (1983), these abnormalities are probably the result of accidental deviations in spermatogenesis, for example with abnormal multiplication of centrioles that can give numerous flagella, rather than the result of a pathological condition. In Piiuliuhi inarganlifera. degenerating multinuclear cells are frequently observed, but bicephal sperma- tozoa have never been noted. In the same way, spermatozoa with intracytoplasmic flagellum are often observed in advanced degenerative stages. These obser- vations lead us to think that abnormal cells are rapidly eliminated and such degenerating cells might be driven out by the genital orifice. In the acini, phagocytes have frequently been recorded in bi- valves studies (Dorange and Le Pennec 1989; Mathieu 1987). These macrophages are thought to be the result of a differenciation from hemocytes. Such transformation of hemocytes into macro- phages has been followed by Houtteville ( 1974) in Mytilus edidis. With presence of lysosomial inclusions in certain auxiliary cells, also reported in Pecten maximus (Dorange and Le Pennec 1989), we can suppose that these cells can also be involved in the resorption of degenerative germinal cells. In the both types of resorption in situ, the products of cellular lysis can be recovered by the organism. Recuperation of this material is possible by absorb- ing cells in the gonoducts or digestive tract (Dorange and Le Pennec 1989). ACKNOWLEDGMENT The authors would like to acknowledge Mrs. D. Chagot and Mrs. A. Fougerouse for their helpful technical assistance in elec- tron microscopy preparations and observation. We also thank the E.V.A.A.M. (Tahiti), for providing us the living material. LITERATURE CITED Allen, S. K., Jr., H. Hidu & J. G. Stanley. 1986. Abnormal gametoge- nesis and sex ratio in triploid softshell clams [Mya arenana). Biol. Bull. 170;198-210. Bourcart, C R. Lavallard & P. Lubet. 1965. Ultrastructure du sperma- tozoide de la moule [Mytilus perna von Ihering). C R. Acad. Sc. Paris 260:5096-5099. Daniels, E. W., A. C. Longwell, J. M. Niff & R. W. Wolfgang. 1971. Ultrastructure of spermatozoa from the American oyster Crassostrea virginica. Trans. Amer. Micros. Soc. 90(3):275-282. Dohmen, M. R. 1983. Gamelogenesis. In: Verdonk. N. H., J. A. M. Van Den Biggelaar & A. S. Tompa, (ed). The Mollusca. Vol. 3. Devel- opment. Academic Press, New York, London, Paris: 1-48. Dorange, G. & M. Le Pennec. 1989. Ultrastructural characteristics of spermatogonesis in Pecten maximus {Mollusca, Bivalvia). Invert. Re- prod. Dev. 15(2):109-117. Fain-Maurel, M. A. 1966. Acquisitions recentes sur les spemiatogeneses atypiques. An. Biol. 11-12:514-564. Franzen, A. 1983. Ultrastructural studies of spermatozoa in three bivalve species with notes on evolution of elongated sperm nucleus in primitive spermatozoa. Gamete Res. 7:199-214. Gutierrez. M., J. Perez Crespo & E. Pascuala. 1978. Ultrastructura de ovocitos y espermatozoides del oslion, Crassostrea angulala Lmk. de la costa sudatlantica de Espana. Inv. Pesq. 42(11:167-178. Hodgson, A. N. & R. T. F. Bernard. 1986. Ultrastructure of the sperm and spermatogenesis of three species of Mytilidae (Mollusca, Bi- valvia). Gamete Res. 15:123-135. Houtteville, P. 1974. Contribution a Tetude cytologique et experimentale du cycle annuel du tissu de reserve du manteau de Mxtilus edulis. These doct. spec Universile de Caen, [France). 98 pp. Cited by Dorange and Le Pennec, op. cit. Hylander, B. L. & R. G. Summers. 1977. An ultrastructural analysis of gametes and early fertilization in two Bivalve Molluscs, Chama ma- cerophylla and Spisula solidissima, with special reference to gamete binding. Cell. Tiss. Res. 182:469^89. Mathieu. M. 1987. Etude experimentale des controles exerces par les ganglions nerveux sur la gametogenese et les processus metaboliques associes chez la moule Mytilus edulis L. (Mollusque Lamellibranchel. These Universile de Caen. [France). 218 pp. Popham, J. D., M. R. Dickson & C. K. Goddard. 1974. Ultrastructural study of the mature gametes of two species of Bankia [Mollusca: Teredinidae). Aust. J. Zool. 22:1-12. Thielley, M, 1989. Etude histologique et cytochimique de la gametogenese chez la nacre Pinctada margaritifera (L.) var. cumingii (Jameson). D.E.A. Universite Frangaise du Pacifique. 22 pp. Thiery, J. P. & A. Rambourg. 1974. Cytochimie des polysaccharides. J. Microscopic 21:225-232. Tranter, D. J, 1958. Reproduction in Australian peari oysters (Lamelli- branchia). IV. Pinctada margaritifera (Linnaeus). Aust. J, Mar, Freshw. Res. 9:509-525. Journal of Slu-lllhli Research. Vol. 12. No. 1. 49-58. 1993. INVESTIGATIONS INTO THE TRANSMISSION OF PARASITES OF THE BAY SCALLOP, ARGOPECTEN IRRADIANS (LAMARCK, 1819), DURING QUARANTINE INTRODUCTION TO CANADIAN WATERS SHARON E. MCGLADDERY,' BRENDA C. BRADFORD,^ AND DAVID J. SCARRATT^ ^Department of Fisheries and Oceans P.O. Box 5030 Moncton. N.B.. EIC 986. Canada 'Department of Fisheries and Oceans P.O. Box 550 Halifax. N.S.. B3J 2S7. Canada ABSTRACT The potential impact of bay scallop Argopecten irradians (Lamarck) parasites on commercially important bivalve species in Canadian Atlantic waters was assessed using two transmission experiments. The first was a parallel flow-through system passing water from the bay .scallops over five species of native bivalves. The second was a synchronous spawning of infected bay scallops and uninfected blue mussels Mytihis ediiHs. to determine if larval bivalves are more susceptible to parasite transmission than adults. Zoospores of Perkinstis karlssoni were observed adhering to D-stage larvae of bay scallops approximately 48 hours post- spawning, suggesting this to be the method of transmission. Surface sterilization of fertilized bay scallop ova with \'7c iodophor for 15 minutes failed to destroy the zoospores. No evidence of transmission of bay scallop parasites to adults of other species was found during the ten month expenmental period. Results of the second expenment are inconclusive. No P karlssoni zoospores were seen among the larvae, and no tissue-stages have been detected subsequently in the exposed mussels. KEY WORDS: scallop. Argopecten. parasites, transmission, quarantine INTRODUCTION The bay scallop Argopecten irradians (Lmk). occurs in the shallow tidal lagoons of the northeastern United States but does not occur naturally north of Maine. Due to interest in this species as a candidate for aquaculture, it was introduced to Canada in 1979 when broodstock were held in quarantine on Prince Edward Island (PEI) (Townshend and Worms 1983). Histological examination revealed rickettsial and chlamydial infections (Morrison and Shum 1982, 1983) which were monitored closely over the next 4-6 generations to determine their significance to both the bay scallops and native species. Since the rickettsial and chlamydial infections declined over this period, and native species were found to harbour similar prokaryotes. the F4 generation of bay scallops was released in 1983 for grow-out at specific sites around PEI (Townshend and Worms 1983). The transplanted seed grew well during the summer and autumn but did not survive the winter. Further generations of bay scallops were maintained in low numbers by overwintering broodstock in hatcheries while potential aquaculture sites were evaluated (Mallet and Carver 1987. 1988). In 1987. a commercial enterprise began growing and marketing adult bay scallops, thus stimulating interest in their culture as a cash crop. By 1989. com- mercial quantities of seed were produced at private hatcheries in Nova Scotia for grow-out in PEI. In accordance with regional guidelines for introduction and transfer of live aquatic organisms, samples of bay scallop brood- stock and seed (2 mm long) were checked in May 1989. prior to transfer to PEI. Nothing of concern was found in the spat, however a previously undescribed apicomplexan parasite. Perkinsus karlssoni (McGladdery et al. 1991) was found in the broodstock. Re-examination of histological sections from the original bay scal- lops introduced in 1979 revealed the same parasite which had been marked by a strong hemocyte encapsulation response. No similar parasite has been observed in native molluscs from Atlantic Can- ada. This information, together with histological evidence of the same parasite in bay scallops from Rhode Island (Karlsson 1991), indicated that the parasite had persisted in hatchery bred popula- tions for at least 10 generations. Since P. karlssoni is related to the known oyster pathogen Perkinsus inariniis. concern was raised about its potential for transfer to native bivalves. An additional observation that seed retained in the hatchery developed P. karlssoni infections similar to those in stocks which had been in open water suggested that the infection had been transmitted either in the egg or during the few minutes that newly-spawned gametes were exposed to infected broodstock. No other perkinsiid species has been reported to trans- mit directly from infected broodstock to their offspring. Transmis- sion in other perkinsiids. where known, is reported as being from moribund hosts to neighbouring hosts, i.e., lateral proximal trans- mission (Ray and Chandler 1955, Andrews 1965, Goggin et al. 1989). Although individual parasites have been infrequently ob- served within bay scallop ova (Karlsson 1991) it is unlikely these ova maintain their viability. Infected ova are associated with an extensive hemocyte infiltration, and the parasite occupies a sig- nificant proportion of the cell volume. In April 1989. shortly before the discovery of P. karlssoni. staff from the Department of Fisheries & Oceans, Canada, intro- duced a second bay scallop broodstock from Cape Cod, U.S.A. This introduction was in response to concern that repeated breed- ing from small numbers of broodstock had resulted in genetic impoverishment (Dr. M. Helm, pers. comm.). Figure 1 outlines the chronology of events in the present study in relation to those reported by McGladdery et al. (1991). MATERIALS AND METHODS Wild bay scallops (n = 123) were harvested in April 1989, at Osterville, Cape Cod, Massachusetts, and transported to the quar- antine laboratory at DFO, Halifax, for disease screening in accor- dance with ICES Guidelines (Turner 1987). Thirty specimens were submitted for bacteriological and virological examination as 49 50 McGladdery et al. SYNCHRONOUS STUDIES Stock derived from original introduction from Cape Cod in 1979 Discovery of the "new" parasite in broodstock- May 1989 July 1989 -Oct. 1990 Investigation of identity & development of the parasite in juvenile bay scallops derived from infected broodstock Broodstock introduction from Cape Cod - May 1989 PRESENT STUDY June 1989 -Start of cross-species transmission expenment using quarantined broodstock and warm, flow- through, filtered seawater October 1989 - Detection of Perkinsus karlssoni in quarantined broodstock April 1990 -End of cross-species transmission experiment Identification of parasite as a new species - Perkinsus karlssoni (McGladdery et al. 1991) July 1990 - transplantation of Fl bay scallop to grow- out sites in Nova Scotia Oct, 1990- April 1991 Overwintering of Fl bay scallops at Halifax July 1991 :i) Spawning of Fl bay scallops and observation of P. karlssoni zoospores amongst D-stage F2 larvae ii) synchronous spawning of mussels and scallops April 1992 - Examination of mussels from synchronous spawning experiment and F2 bay scallops. Figure 1. Chronology of events associated with the present bay scallop study and synchronous studies described in McGladdery et al. (1991). soon as they arrived in Halifax. Tissue samples were also extracted for microscopic examination and thioglycollate culture. The re- mainder were cleaned of macroscopic fouling organisms and placed in quarantine for conditioning and breeding. Sea water for the quarantine laboratory is drawn from Halifax Harbour at a depth of 20 m. filtered through gravel, sand, and activated charcoal filters, and passed through heat exchangers. Temperatures in the experimental tanks were maintained by blending water from the different temperature lines. Waste water was injected with gaseous chlorine to give a minimum concentration of 3 ppm for 30 min- utes. Transmission Experiment #1 In June 1989, the newly introduced bay scallops were placed in wooden, mcsh-bottom trays floating in a 1000 litre tank of flowing sea water at 17°C (±1°C). The 1000 litre tank was fitted with air lifts at either end which discharged water into the floating trays to maintain circulation and ensure adequate aeration. A third air lift discharged water from the tank into one of two shallow. 200 litre tanks which held samples of the test species. A second 200 litre tank (the control) was fed seawater at the same temperature directly from the laboratory supply, thereby isolating it from water which had passed through the bay scallop holding tank (Fig. 2). After a one-week acclimation period and initial histological examination of the stocks being used, 150 eastern (American) oyster {Crasso- streii virginicci). 150 edible (European) oyster {Oslrea edulis), 50 soft-shell clams {Mya arenaria), 150 mussels {Mytilus edulis). and 50 giant sea scallops (Placopecten magellanicus) were divided between the two 200 litre tanks. Bivalves m all three tanks were fed cultured strains of Chaetocenis gracilis and hochrysis gal- bana, augmented as required with commercially prepared spray- heated, filtered seawater Control bivalves I^ Exposed bivalves Quarantined Bay Scallops chlorinalion 1 water which has not been in contact with bay scallops ^^^^H water exposed to broodstock bay scallops Figure 2. Diagram of the quarantine holding faciUties for Transmis- sion Experiment #1. Bay Scallop Parasite Transmission 51 TABLE I. Collection schedule for Transmission Experiment #1. July 10 Jan 16 Species: 1989 Aug 10 Sep 14 Oct 26 Nov 30 1990 Feb 23 Apr 04 Aug 15 Total Ostrea edulis 10 10(c) 10(c) 10(c) 5(c) 10(c) 5(c) 10(c) 14(e) 144 10(e) 10(e) 10(e) 5(e) 10(e) 5(e) 10(e) Crassoslrea 10 10(c) 10(c) 10(c) 5(c) 10(c) 5(c) 2(c) 3(e) 125 virginicti 10(e) 10(e) 10(e) 5(c) 10(e) 5(e) 10(e) Mytilus edulis 10 10(c) 10(c) 10(c) 5(e) 10(c) 5(c) 9(c) 1(e) 125 10(e) 10(e) 10(e) 10(e) 5(e) 10(e) Mya arenaria 2 5(c) 5(e) 5(c) 5(e) — — 5(c) 4(e) 5(c) 2(c) 38 Placopecten 5 5(c) 5(c) 5(c) 3(e) 1(c) 2(e) — — 42 magellanicus 5(e) 5(e) 5(e) 1(e) Total 37 80 80 70 28 71 37 53 18 474 Key: (c) = control animals held in water bypassing the bay scallop holding tank; (e) bay scallop holding tank. exposed animals held in a tank fed by effluent water from the dined Tetraselmis sp. (Cell Systems Ltd'). Bay scallops were fed to excess to maintain breeding condition, while the other species were fed a maintenance ration only. At approximately monthly intervals (Table 1 ) exposed and con- trol specimens were selected at random, a 3 mm transverse section of the tissues was excised, preserved in modified Davidson's so- lution (Howard and Smith 1983), paraffin-infiltrated, sectioned and stained with Harris's Hematoxylin and Eosin for light microscopy. Bay scallops were removed from the tanks, as required for an experimental breeding program, individually heat stimulated, and spawned in a separate spawning facility within the quarantine lab- oratory. These scallops were subsequently returned to the trans- mission experiment. On one occasion, water in the bay scallop holding tank rose to 20°C and stimulated a mass spawning. The scallops then required about 4 weeks reconditioning at 17°C before experimental spawning could recommence. Breeding continued throughout the summer. The last spawning was in October 1989, after which tissues from the remaining 10 scallops were excised and processed for histology (Table 2) and thioglycollate culture. Water flow to the holding tank with the "exposed" experi- mental animals was switched to a blended laboratory supply at 17°C and sampling continued until the last exposed and control animals were removed for light-microscopical examination in Au- gust, 1990. Spawning Observations Bay scallop broodstock were spawned in quarantine as de- scribed above. The resultant F, spat were retained m the quaran- tine system until July 1990 at which time they were examined and transferred to 6 sites around Nova Scotia for growth to market size. In November 1990 they were returned to the Halifax Labo- ratory. Fifty bay scallops of the F, generation were conditioned for spawning in a 1000 litre tank (T = 18°C; S = 30 ppt) for ap- proximately 8 weeks, beginning in February 1991 . They were fed cultured algae consisting of Chaetoceros muelleri, Isochrysis gal- bana (Tahitian strain), Thalassiossira sp., and Tetraselmis sp. At week nine. 20 of the broodstock were selected for batch-spawning, producing approximately 10 million eggs. Eggs and sperm were 'No longer available examined microscopically in order to count the eggs, assess fer- tilization rate and determine presence or absence of P. karlssoni zoospores. Water samples from holding tanks which did not con- tain bay scallops (control tanks) were also checked for P. karlssoni. A second spawning (March 1991) produced fertilized eggs which were surface-disinfected with a 1% iodophor solution for 15 minutes. They were rinsed and allowed to develop normally. Un- fortunately the experiment terminated 10 days post-spawning due to a technical malfunction. A further spawning was induced at the end of July, and the larvae raised and planted out as before. Eggs from this spawning were not surface-sterilized. These F, bay scal- lops were returned to the Halifax Laboratory for overwintering. In April 1992, tissue samples were collected from these bay scallops for histological examination and thioglycollate culture (Table 2). Transmission Experiment #2 In July 1991, mature adult blue mussels, Mytilus edulis. were spawned at the same time as bay scallops, and the larvae from both species were reared together. No bay scallop spat survived past metamorphosis, but the mussels thrived, and a sample was exam- ined histologically and using thioglycollate culture in April 1992 (Fig. 1). RESULTS Parasites of Bay Scallop Broodstock. 1989 Microscopic examination and tissue culture in thioglycollate medium (Mr J. W. Comic, pers. comm.) of the bay scallop brood- stock on arrival in Halifax from Cape Cod in May 1989 revealed no evidence of perkinsiid infections. Infections by rickettsia-like organisms were observed (Fig. 3, Table 2), but these were not identical to those reported from the specimens originally intro- duced in 1979 (Morrison and Shum 1982, 1983). Five bay scallops examined in October 1989 showed extensive tissue lesions identical to those observed in bay scallops examined by McGladdery et al. ( 1991 ) and attributed to Perkinsus karlssoni (Apicomplexa: Perkinsea). All five scallops were also positive for perkinsiid parasites, using thioglycollate culture of soft-tissue samples. The last 10 bay scallops remaining from the quarantined 52 McGladdery et al. TABLE 2. Parasite prevalence (%) in bay scallops, Argopecten irradians, examined during the present study. Date May 89 Oct 89 Nov 89 April 90 Nov 90 July 91 Apr 92 Generation Quarantine Broodstock Quarantine Broodstock Quarantine Broodstock Fl Juveniles Fl Adults Fl Broodstock F2 Adults Sample Size 30 5 10 12 7 9 30 Perkinsus karlssoni Pseudoklossia-hke coccidian Gill ciliates Gill rickettsias Digestive tubule rickettsias 0% 20% 100% 100% 0% 0% 100% 14% 89% 0% 57%. 0% 0% 0% 0% 89% 0% 0% 0% 0% 57% 0% 0% 0% 0% 0% 0% 0% 0% 0% broodstock were collected at the end of November and yielded the same results (Table 2). Since bay scallops normally die shortly after spawning, post spawning mortality could not be attributed conclusively to infection by P. karls.wni. Spawning broodstock. despite parasite loads, were apparently healthy when sacrificed for histological examination. Transmission Experiment #1 There was no evidence from either histology or thioglycollate culture of transmission of P. karlssoni from the bay scallop to the "exposed" specimens, despite the likelihood that the broodstock were already infected when introduced in May (even though the parasite was not detected until later). Similarly there was no evi- dence of transmission to native species of any of the rickettsial or chlamydial organisms observed in the bay scallop broodstock ex- amined in May, 1989 (Tables 3-7). Rickettsia-like inclusions were observed in exposed samples of edible oyster, eastern oyster and giant scallop, however, histologically identical inclusions were also found in samples collected prior to exposure and in controls. All other parasites and prokaryote inclusions observed are com- monly found throughout Atlantic Canada in the species examined (McGladdery 1990, McGladdery and Stephenson 1991, Morrison J, 'Ji>*' U _^ ^ ^^' ^ " s? \ '^: ■* ^^ V^.vi.. fe Figure 3. Intracellular Rickettsia-iike inclusion bodies in quarantine bay scallop, Argopecten irradians, from Cape Cod. (Scale bar = 50 p,m). Bay Scallop Parasite Transmission 53 TABLE 3. Histological observations from control blue mussels, Mytilus edulis, and blue mussels exposed to efHuent water from bay scallops, Argopecten irradians. Observation July 1989 August September October November January 1990 February April August Mytilus edulis: Control Time Zero Sample size 10 Internal Turbellana 9c? 0.0 I A Aniistrtim mMili %P 0.0 Sphenophryid-like %P gill-ciliate I A Mytilus edulis: Exposed Sample size Internal Turbellana 9cP I A Ancistrum mylili %P I A Sphenophryid-like %P gill-ciliate I A 0.0 10 10 10 10.0 0.0 0.0 1.0 0.1 0.0 0.0 10.0 1.0 0.1 0.0 0.0 0.0 10 10 10 0.0 0.0 0.0 0.0 10.0 I.O 0.1 0.0 0.0 0.0 20.0 1.0 0.2 5 0.0 0.0 0.0 10 0.0 0.0 0.0 10 0.0 0.0 0.0 5 9 0.0 0.0 0.0 0.0 — 20.0 0.0 1.0 0.2 5 10 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Key: %P = prevalence; I = mtensity (mean number ot parasites per tissue section ot infected mdividuals) and A = abundance (mean number of parasites per tissue section for all individuals in a sample). and Shum 1983) and show no correlation with areas used to culture bay scallops since 1982. A decrease in prevalence (percentage of histological sections containing evidence of infection) of rickettsial-like inclusions in bay scallops, edible oysters and giant sea scallops was observed within two months of being placed into the quarantine facility. This may be due to the relatively small numbers examined or, possibly, reflect a trend similar to that observed during the 1979 introduction of bay scallops, where the number of rickettsia-like lesions declined over time (Townshend and Worms 1983). Eastern oysters showed no distinct decline in similar lesions, with a 10% prevalence being observed in the last sample examined in April 1990. Although mass mortality of bay scallops has been attributed to infection by these prokaryotes (Leibovitz 1989). no pathology was associated with any of the infections listed in Tables 3-7. Moreover, nearly all specimens showed evidence of feeding prior to being collected. Bay Scallop Spawning Observations Prior to spawning the F, generation bay scallops, a sample of 9 was examined in July, 1991. All showed extensive P. karlssoni TABLE 4. Histological observations from control edible oysters, Ostrea edulis, and edible oysters exposed to effluent water from bay scallops, Argopecten irradians. Observation July 1989 August September October November January 1990 February April August Ostrea edulis: Control Time Zero Sample size 10 10 10 10 5 10 5 10 Rickettsia-like inclusions %P 30.0 20.0 20.0 0.0 0.0 0.0 0.0 0.0 — Gymnophallid-like %P 0.0 0.0 0.0 0.0 0.0 0.0 20.0 0.0 — metacercaria I A 1.0 0.2 Ostrea edulis: Exposed — Sample size 10 10 10 5 10 5 10 14 Rickettsia-like inclusions %P — lO.O 0.0 10.0 0.0 0.0 0.0 0.0 0.0 Gymnophallid-like %P — 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 metacercaria I A Key; %P = prevalence; I = intensity (mean number of parasites per tissue section of infected individuals) and A = abundance ( mean number of parasites per tissue section for all individuals in a sample). 54 McGladdery et al. TABLE 5. Histological observations from control eastern oysters, Crassostrea virginica. and eastern oysters exposed to effluent water from bay scallops, Argopecten irradians. Observation July 1989 August September October November January 1990 February April August Control (Sample size) 10 10 10 10 5 in 5 2 Rickettsia-like inclusions %P lO.O lO.O 0.0 0.0 20.0 0.0 0.0 0,0 — Internal Turbellaria %P 10.0 1.0 0.1 0.0 0.0 0,0 0.0 0.0 0.0 0,0 — A Ancistrocoma-Wke %P 10.0 10.0 20.0 10. 40.0 0.0 10.0 0,0 — digestive gland ciliate I 1.0 1.0 1.0 6.0 2.5 61.0 A 0.1 0.1 0.2 0.6 1.0 12.2 Sphenophryid-like %P lO.O 10.0 20.0 10.0 20.0 50.0 80.0 0,0 — gill-ciliate I 3.0 2.0 1.0 2.0 9,0 3.8 7.0 A 0.3 0.2 0.1 0.2 0.45 1.9 5.6 Exposed (Sample size) — 10 10 10 5 10 5 10 3 Rickettsia-like inclusions %P — 0.0 20.0 0.0 0.0 0.0 0.0 10,0 0,0 Internal Turbellaria %P I A 20.0 1.0 0.2 0.0 0.0 0.0 0.0 0.0 0,0 0,0 Ancistrocoma-Wke %P — 10.0 40.0 10.0 20.0 0.0 10. 20,0 0,0 digestive gland ciliate I 14.0 1.75 1.0 10 5.0 5.5 A 1.4 0.7 0.1 0.2 1.0 1,1 Sphenophryid-like %P — 0.0 10.0 10.0 20.0 60.0 80.0 60,0 66,7 gill-ciliate I I.O 1.0 1.0 6.0 18.7 17,3 41,0 A 0.1 0.1 2 3.6 15.0 10.4 27.3 Key: %P = prevalence; I = intensity (mean number of parasites per tissue section of infected individuals! and A per tissue section for all individuals in a sample). abundance ( mean number of parasites lesions (Fig. 4. Table 2) as well as heavy kidney infection by a Pseudoklossia-like coccidian (Fig. 5, Table 2). Pseudoklossia-Vike coccidians have been found in bay scallops descended from the original introduction (McGladdery 1990), but this was the first observation of this parasite in the progeny from the 1989 intro- duction. Following spawning of the F, broodstock, zoospores of P. karlssoni were detected among the F, larvae (Figs, 6. 7), Sutface sterilization of fertilized bay scallop eggs with 1% iodophor appeared to have no effect on the parasite: free swim- ming zoospores were observed among, and possibly attached to the surface of healthy D-stage lai^'ae 48 hours post treatment, and subsequently during larval development. Unfortunately problems with the heating system caused the loss of these larvae after 10 days, but development to that time had been normal. No zoospores were seen among the larvae from the final breed- ing trial in late July, 1991 , suggesting that they might be parasite- free. These F, spat were outplanted in late August, In November 1991 stock from this outplanting were returned to the Halifax Laboratory where they were maintained overwinter. Tissue sam- ples collected in April 1992 from these bay scallops were infected with P. karssom (Table 2), Transmission Experiment 2 Results of the mussel-scallop larval rearing trial are equivocal. The bay scallop larvae did not survive past metamorphosis in the same tanks with the mussels, and no P. karlssoni zoospores were detected among the growing mussel larvae. The mussels were examined histologically in April 1992 following grow-out in open water. There was no evidence of transmission of P . karlssoni or Pseudoklossia to the mussels, Thioglycollate culture of tissues from the same mussels was negative for perkinsiid protozoans. TABLE 6. Histological observations from control giant sea scallops, Placopeclen magellanicus, and giant sea scallops exposed to effluent water from bay scallops, Argopecten irradians. Observation July 1989 August September October November January 1990 February April P. magellanicus: Control Sample size Rickettsia-like inclusions %P Time Zero 5 80,0 5 20,0 5 40.0 5 0.0 — 1 0.0 — — P. magellanicus: Exposed Sample size Rickettsia-like inclusions %P — 5 60,0 5 20.0 5 0.0 3 0.0 1 0.0 2 0.0 — Key: %P = prevalence; I = intensity (mean number of parasites per tissue section of infected individuals) and A = abundance (mean number of parasites per tissue section for all individuals in a sample). Bay Scallop Parasite Transmission 55 TABLE 7. Histological observations from control soft-shell clams, Mya arenaria, and soft-shell clams exposed to effluent water from bay scallops, Argopecten irradians. Observation July August September October November January February April Mya arenaria: Control Sample size Gymnophallid-like nietacercaria VrP I A Mya arenaria: Exposed Gymnophallid-like nietacercaria %P A Ti me Zero on 5 00 5 0.0 — 5 0.0 5 0.0 5 5 0.0 20.0 2.0 0.4 4 — 0.0 — 0.0 Key: %P = prevalence; I = intensity (mean number of parasites per tissue section of infected individuals) and A per tissue section for all individuals in a sample). abundance ( mean number of parasites DISCUSSION Histological examination and thioglycoliate culture of bay scal- lop broodstock introduced in May 1989 revealed no sign of P. karlssoni or Pseudoklossia-Vike protozoans (Table 2), however, rickettsia-like organisms were found which initiated cross-species transmission studies. Five months later (October 1989). a second sample of the quarantined bay scallop broodstock revealed 100% prevalence off. karlssoni. This reinforces previous reports (Mc- Gladdery et al. 1991) stating that certain stages of P. karlssoni may not be detected by routine histology or thioglycoliate culture. This cryptic period can be shortened at elevated water tempera- tures ( I7-20°C), but no attempt was made to increase the temper- ature above 17°C during the transmission experiment. Conditions may, therefore, have been inadequate for transmitted P. karlssoni to develop in the native species to a detectable stage by the end of the ten month expenment. It is known that P. marinus can escape routine detection in eastern oyster, Crassostrea virginica, for a full year (Ray 1954, Andrews 1965). Experiment 1 was not repeated since a longer (two year) experiment commenced in November 1990 at "^^ Figure 4. Tissue lesions containing Perkinsus karlssoni in quarantine bay scallop, Argopecten irradians, from Cape Cod. (Scale bar = 50 |xm). 56 McGladdery et al. «». ■*%*' >«" ASl Figure 5. Kidney coccidia (Pseudoklossia sp.) in F, generation bay scallops, Argopecten irradians. (.Scale bar = 50 (im). the Atlantic Veterinary College (AVC). in which bay scallops will epidemiology and histological appearance of P. karlssoni. com- be held in a closed-circulation system and in the same tanks as pared to other perkinsiid species, may have been due to ten years several native species (Dr. R. J. Cawthorn. pers. com.). of transmission via hatchery-manipulated spawning. The lesions in McGladdery et al. (1991) suggested that slight variations in quarantined broodstock imported directly from Cape Cod (Fig. 3), Figure 6. Perkinsus karlssoni zoospore attached to surface of D-stage bay scallop larva. Bay Scallop Parasite Transmission 57 Figure 7. Diagrammatic representation of Perkinsus karlssoni zoo- spore attached to surface of D-stage bay scallop larva. however, were identical to those observed in hatchery-bred stock. Moreover. Perkinsus infected material generously provided by Karlsson (pars, comm.) from Rhode Island bay scallops is indis- tinguishable from specimens collected from Canadian bay scallops descended from either the 1979 or 1989 introductions. Precautions taken during spawning in 1989 to minimize the likelihood of transmitting parasites from broodstock bay scallops to their offspring involved removal of the adults from the dishes containing their spawn and the subsequent raising of the fertilized eggs in a separate section of the quarantme facility. Clearly, how- ever, transmission of P. karlssoni to the F, generation still oc- curred. The observation of zoospores among healthy D-stage lar- vae, reinforces the suggestion of McGladdery et al. (1991) that persistance of the parasite in the previously introduced stock was due to exposure of offsprmg to mfected adults during breeding. The possibility of transmission via infected ova is discussed be- low. The question of the timing of transmission and its linkage to spawning is especially important for bay scallop culture in Cana- dian waters, where spawning is confined to hatcheries. The fact that bay scallops are spawned separately from other bivalve spe- cies within these hatcheries may reduce the impact of this parasite on native species if transmission is found to be possible. The lower proportion of parents having the parasite during the early part of the spawning season suggests that there may be some lateral transmission between adult scallops. Alternatively, the par- asite may persist in low numbers until the bay scallop grows and only proliferates to a detectable level with the maturation of the gonad. Under artificial conditions the parasite was observed in the tissues of immature scallops (<20 mm shell height) (McGladdery et al. 1991), however, no such development has been observed in bay scallops growing in ambient Canadian waters. The question of transmission of bay scallop parasites to native bivalve species and the timing of such transmission is at least partially answered. Experiment 1 showed no evidence of trans- mission to other species held directly downstream of infective bay scallops. Scallops which spawned, accidentally, upstream from "exposed" native species, as well as moribund and dead scallops, were left in situ between June and October in order to enhance any potential for cross-species transmission (Ray 1954, Andrews 1965). Progeny from induced spawning were subsequently shown to be in- fected indicating that within-species transmission had taken place. Initial observations of P. karlssoni zoospores adhering to hold- mg-dish surfaces and demonstrating negative buoyancy suggested that the flow-through system might not have been optimal for testing the transmission potential of this parasite. Subsequent de- tection of zoospores throughout the water column in tanks used for conditioning F, broodstock, however, demonstrated that zoo- spores could be carried from tank to tank, with the upwelling system helping maintain the zoospores in suspension. The "ex- posed" bivalve species showed no evidence of transmission of any of the bay scallop parasites during the nine months of the exper- iment (July 1989 to April 1990). The additional 14 edible oysters, 3 eastern oysters and one blue mussel, maintained for a further 4 months, also showed no sign of infection attributable to exposure to the bay scallops. Breeding in quarantine failed to prevent transmission of the Pseiuloklossia-hke coccidian from one generation to the next, al- though this has evoked less concern than the persistence of P. karlssoni. due to the widely-held belief that most Pseiidoklossia species are non-pathogenic (see review by Lauckner 1983). Re- cently, however, Cawthom et al. ( 1991 ) reported a mass mortality of experimentally held bay scallops caused by an unusually heavy infection of the same Pseudoklossia-Wke parasite. The appearance of this coccidian in F, bay scallops reinforces the question of the efficacy of sub-sampling quarantined broodstock and breeding in quarantine as methods for preventing parasite introduction. These Pseudoklossia-Wke parasites are commonly found in bay scallops from the eastern US (Getchell 1991 . Karlsson 1991 ) but have also been reported from bay scallop descendants of the original ( 1979) Canadian introduction (McGladdery 1990). No similar coccidians have been reported from native bivalve species. Regardless of the pathogenicity of these parasites, there is a need to reassess disease screening protocols and techniques, es- pecially for introduction of species for which little base-line in- formation is available. For example, the 1979 introduction of bay scallops into Canadian waters preceded publication of reviews of parasites and diseases of scallops (including bay scallop) (Leibo- vitz et al. 1984, Getchell 1991, Karlsson 1991). Moreover, the assumption of disease-free status of F, generations produced from broodstock found to be "parasite-free" by current diagnostic tech- niques may be erroneous, at least for certain bivalve parasites. Since the F, broodstock were found to be infected by two different species of protozoan, the biflagellate zoospores found among the F, generation spat were examined carefully and com- pared to samples from bay scallops which had shown no evidence of the Pseudoklossia-Wke infection (McGladdery, et al. 1991 ). The zoospores were identical, and no other zoospores or oocyst-like stages were detected among the 8-month old juveniles. The bi- flagellate zoospores observed were, therefore, assumed to be ex- clusively oi P. karlssoni. Observation of the zoospores but not the zoosporangia of P. karlssoni among the F2 spat leaves the precise mechanism of infection open to speculation. The zoosporangia of P. marinus and P. atlanticus develop within the host tissue, and it is from these that the motile zoospores are released (Perkins 1976, Azevedo 1989). Bay scallops spawned in Canadian hatcheries are only in contact with their spawn for up to 4 hours, indicating that zoospore release may occur during that period. Alternatively, the zoosporangia may be released from infected broodstock tissues and the zoospores emerge later. Histological sections of the in- fected F| broodstock showed a marked localization of P. karlssoni around the mantle margin and other surface epithelia. Some of the mantle lesions appear to open to the outside of the scallop, but there was no evidence of zoospore release (Figure 4). Karlsson (1991) observed P. karlssoni inside individual ova. A large pro- portion of the egg volume was displaced by the protozoan which 58 McGladdery et al. casts doubt on the viability of infected ova and trans-ovarian trans- mission. No infected ova were observed in the present study. Additional evidence for extra-cellular transmission is that the in- fective stages of all Perkinsus species described to data are motile, biflagellate zoospores (Perkins 1976, Azevedo 1989, Gogginet al. 1989). The possibility that bivalve larvae may be the most susceptible age group for P. karlssoni transmission was tested, based on the observation from this experiment that motile, adhesive, zoospores are present among D-stage larvae. Synchronous spawning of bay scallops and blue mussels, however, revealed no evidence of cross-species infection. Scallops from the same broodstock grown under similar conditions were infected. Cross-species transmission has been demonstrated experimentally for other species of Per- kinsus (Goggin et al. 1989), although host-specificity appears to be the rule in the wild (Ray 1954). The ability off. karlssoni zoospores to survive the 15 minute 1% iodophor treatment may have been achieved by avoiding ex- posure inside a zoosporangial stage, or it may indicate that the zoospores themselves are highly resistant. Tissue stages of other species of Perkinsus have been reported to withstand 6 ppm chlo- rine treatment for up to two hours, although free prezoosporangia (stage prior to expansion into the zoospore-containing zoospo- rangium) lasted less than 30 minutes in the same treatment (Gog- gin et al. 1990). Surface sterilization using this concentration of chlorine would kill the bay scallop ova. Investigation of alternative treatments is required. ACKNOWLEDGMENTS We thank Dr. K. Freeman, Ms. R. Outerbridge, Dept. Fish- eries and Oceans, Halifax, and Ms. M. F. Stephenson, Dept. Fisheries and Oceans, Moncton, for valuable assistance with col- lection of samples and maintenance of our animals. We are also endebted to Mr. J. Cornick and the Fish Health Unit for thio- glycollate culture analyses and advice on surface sterilization pro- cedures. Drs. S. M. Bower, G. Olivier and T. W. Sephton pro- vided valuable critiques of the draft manuscript. We also gratefully acknowledge the valuable discussion of this project provided by Dr. R. J. Cawthom, Atlantic Veterinary College. LITERATURE CITED Andrews, J. D. 1965. Infection experiments in nature with Dermocystid- ium marinum in Chesapeake Bay. Chesapeake Sci. 6:60-67 . Azevedo, C. 1989. Fine structure of Perkinsus atlanlicus n.sp. (Apicom- plexa, Perkinsea) parasite of the clam Rudimpes decussatus from Por- tugal, y. Parasilol. 75:627-635. Cawthom, R. J., R. J. MacMillan & S. E. McGladdery. 1991. Epidemic of Pseudoklossia sp. (Apicomplexa) in bay scallops Argopecten irra- dians. 14th Regional Fish Health Workshop, Nov, 6-8, 1991, Halifax (Absl. only). Getchell, R. G. 1991. Diseases and Parasites of Scallops. In: Shumway. S. E (ed). Scallops: Biology, Ecology and Aquaculture. Develop- ments in Aquaculture and Fisheries Science #21, Elsevier, pp. 471- 494. Goggin, C. L., K. B Sewell & R. J. G Lester 1989. Cross infection experiments with Australian Perkinsus species. Dis. Aquat. Org. 7: 55-59. Goggin, C. L., K. B. Sewell & R, J. G. Lester. 1990. Tolerances of Perkinsus spp. (Protozoa, Apicomplexa) to temperature, chlorine and salinity. J. Shellfish Res. 9:145-148. Howard, D. W. & C. S. Smith. 1983. Histological techniques for marine bivalve mollusks. NOAA Tech. Memo. NMFS-F/NEC-25. 95 pp. Karlsson, J. D. 1991 . Parasites of the Bay Scallop, Argopeclen irradians (Lamarck. 1819). In: Shumway, S. E. and P. A. Sandifer (eds). In- ternational Compendium of Scallop Biology and Culture. 1991. pp. 180-190. World Aquaculture Society and National Shellfisheries As- sociation. Lauckner, G. 1983. Diseases of mollusca: Bivalvia. In: Kinne O. (ed). Diseases of Marine Animals. Biologische Anslalt Helgoland. Ham- burg. 2:477-961. Leibovitz, L. 1989. Chlamydiosis; a newly reported serious disease of larval and postmetamorphic bay scallop, Argopecten irradians (Lama- rck). J. Fish. Dis. 12:125-136. Leibovitz, L., E. F Schott & R. C. Kamey. 1984. Diseases of wild, captive and cultured scallops. J. World Marie. Soc. 15:269-283. Mallet. A. L. & C. Carver. 1987. Feasibility of bay scallop Argopecten irradians culture in Nova Scotia: a preliminary study. ERDA Rept #5, Nova Scotia Dept. Fish. 37 pp. Mallet, A. L. & C. Carver. 1988. Within and among site variability in bay scallop Argopecten irradians production. ERDA Rept #13, Nova Scotia Dept. Fish. 26 pp. McGladdery. S. E. 1990. Shellfish Parasites and Diseases on the East Coast of Canada. Bull. Aquacult. Assoc. Can. #90-3:14—18. McGladdery. S. E , R. J. Cawthom & B. C. Bradford 1991 Perkinsus karlssoni n.sp. (Apicomplexa) in bay scallops Argopecten irradians. Dis. Aquat. Org. 10:127-137. McGladdery, S. E. & M. F. Stephenson. 1991. Parasites and Diseases of Suspension- and Bottom-Grown Shellfish from Eastern Canada. Bull. Aquacult. Assoc. Can. #91-3:64—66. Morrison, C. & G. Shum. 1982. Chlamydia-like organisms in the diges- tive diverticula of the bay scallop, Argopecten irradians (Lmk). J. Fish. Dis. 5:17-V184. Morrison, C. & G. Shum. 1983. Rickettsias in the kidney of the bay scallop. Argopecten irradians (Lamarck). J. Fish. Dis. 6:537-541. Perkins, F. O. 1976. Dermocystidium marinum infection in oysters. Mar. Fish. Rev. 38:19-21. Ray, S. M. 1954. Biological studies of Dermocystidium marinum. Rice Inst. Pamph. #41(Spec. lssue):l-l 14. Ray, S. M. & A. C. Chandler. 1955. Dermocystidium marinum. a parasite of oysters. Expl. Parasit. 4:172-200. Townshend, E. R. & J. M. Worms. 1983. Introduction of a new Pectinid species Argopecten irradians irradians to the Gult of St. Lawrence, Canada. ICES CM 1983/K:44. Turner, G. E. 1987 Codes of Practice and Manual of Procedures for Consideration of Introductions and Transfers of Manne and Freshwater Organisms. ICES Doc. No. F:35A. Journal of Shellfish Research. Vol. 12. No. 1, 5Q-64, 1993. GAMETOGENIC CYCLE OF THE CHILOE SCALLOP (CHLAMYS AMANDl) R. JARAMILLO,' J. WINTER,' 'institute de Embriologia Universidad Austral de Chile Casilla 567 Valdivia Chile ^Institute de Biologia Marina Universidad Austral de Chile Casilla 567 Valdivia ^Instituto de F omenta Pesquero Ancud J. VALENCIA,^ AND A. RIVERA' ABSTRACT The gametogenic cycle of the Chiloe scallop Chlamys amandi from Hueihue Bay. Chiloe Chile, was examined for one year. Chiloe scallops were collected at 4-5 week intervals between October. 1989 and December, 1990, Histological sections of the gonad were prepared and the gonadal index was determined. A semiannual spawning cycle was observed; i.e.. scallops were in an active or ripe stage throughout most of the year. Spawning and spat were observed in conjunction with high food availability. KEY WORDS: Molhisca. Bivahia. Pectinidae, reproduction histology INTRODUCTION Reproductive cycles of marine bivalves are comprised of a gametogenic phase, spawning, larval development and growth. The cycle may be annual, semiannual or continuous, depending upon the species and location (Sastry 1979). Well documented patterns of energy storage and utilization are often associated with these cycles, although the role of endogenous and exogenous fac- tors and their interactions in the synchronization of gamete devel- opment and release within a population are still not fully under- stood (Bayne 1976, Sastry 1979, Mac Donald and Thompson, 1986, Barber and Blake 1991). Several environmental factors may influence the timing of re- production in bivalve molluscs. The most commonly cited are water temperature, food availability and tidal influence (Sastry 1966, Machell and De Martini 1971). Many authors have at- tempted to explain reproductive timing in bivalves primarily in terms of water temperature and its variation with latitude (Loosanoff 1937, Ropes and Stickney 1965, Newell et al. 1982. Sastry 1966, 1970, Malachowsky 1988). The gametic production in several species of marine bivalves requires a great deal of energy suggesting a close relationship between the reproductive cycle and energy available for growth (Bayne 1985, Mac Donald and Thompson 1986). Time of spawn- ing may also be related to food availability. Most bivalves tend to spawn during periods when food is available for developing prog- eny and for replenishing the energy adults spend in spawning (Bayne 1976). It is possible therefore that temporal and quantita- tive differences in the food supply have a greater influence on the reproductive cycle than water temperature or latitude (Emmett et al. 1987). For restocking or mariculture purposes it is important to know the life cycle of the target species, and documentation of the re- productive cycle is one logical step in determining when recruit- ment might occur. The Chiloe's scallop {Chlamys amandi) is an unexploited, but potentially valuable resource. However, to date no research has been conducted on the biology or life history of this species. The present paper is a brief study describing the gametogenic cycle and its relationship to temperature, salinity and food availability. MATERIALS AND METHODS Environmental Parameters and Phytoplankton Analysis Temperature and salinity were measured monthly at the culture site (20 m in depth) Hueihue Bay (4r,54'S: 73°,3rW) in Chiloe Island, Chile (Fig. 1) using a YSI Model 33 SCT meter. For phytoplankton, samples were taken with a net having a Kitahara mesh size of 100 jxm. Phytoplankton were fixed in 5% formalin and subsequently decanted. Decanted phytoplankton volume was measured in a graduated test-tube (0.1 ml accuracy) and consid- ered as an indirect index of primary productivity. Gonad Index Analysis Monthly samples of 28 to 30 gonads were used for determina- tion of Gonad Index (GI), prior to fixation. At this sample size, the standard error of the mean Gonad Index remained below 4%, which was regarded as highly precise (see Table 3). The following relationship was used to determine the Gonadal Index Gl (fresh wet weight of gonad) (fresh wet weight of soft parts) X 100 The high average values of GI are coincident with gonadal matunty. Minimal average values following high average values are considered as indicating of spawning (Akaboshi and Illanes 1983). Histological Procedures The gonads were removed from 440 scallops at monthly inter- vals from October 1989 to November 1990. Samples ranged from 28 to 30 individuals. Shell height ranged from 3.8 to 5.2 cm and shell length ranged from 4.5 to 6.0 cm. Scallops were cultured in both lantern nets and pearl-nets suspended at 15 to 20 m depth in Hueihue Bay. Gonadal tissue was fixed in Hollande Bouin (picric-formol- 59 60 Jaramillo et al. '•('•'r'S'r'S'r'Jf 73'30'w !"-!'\:!v!:!v!-H Manao ••'. '• 'r'. '■ :;. '• :;.'•': '.'■:,'.'• J H u e i h u e J G* O e -'^^ :i Linao Figure 1. Map showing the culture site at Hueihue Bay in Chiloe Island Chile. acetic plus cupric II acetate mixture) (Ganter and JoUes 1970) for 24 hours. The samples were then dehydrated using a series of decreasing ethanol solutions. The embedded tissue was sectioned at 5 (xm to 7 (j.m and placed on slides. Tissue was processed using a series of increasing ethanol solutions and sections were stained with hematoxylin-eosin (Humanson 1962). Gonadal tissue was qualitatively examined following the schemes of Ropes and Stickney (1965), Avilez y Lozada (1975), Ramorino (1975) and Malachowsky (1988) to assess developmen- tal stage. The definition of each stage (active, ripe and spent) is stated in Table 1. Examples of each stage arc pictured in Figs la, lb, Ic, 2a, 2b, 2c. Photomicrographs were taken with Standard Leitz and Nikon microscopes. RESULTS The Chiloe scallop Chlamys amandi is a gonochoristic species, commonly found inhabiting gravel or sand bottoms and scattered in small beds along the coast in depths of 15 to 30 m. This species is found exclusively in the South of Chile (Chiloe Island). Environmental Parameters and Phytoplankton On Hueihue Bay water temperature at 20 m depth ranged from 1 1 .6°C in Oc- tober 1989 to a maximum of 13.9°C in February 1990. A high temperature period of over 1 3°C was recorded during the summer months (January to March (1990)). Temperature during all other months was about 1 TC (see Table 2). Salinity measured (20 m depth) in Hueihue Bay fluctuated between 32 and 33.5 ppm throughout the study period (see Ta- ble 2). The phytoplankton data obtained during this study are pre- sented in Fig. 3. Densities (measured in volume) recorded during this study are presented in Table 2 and considered as an Indirect primary productivity index. A seasonal fluctuation in phytoplankton was recorded during summer months (December 1989 to March 1990), for Autumn and Winter months (April to September 1990) relatively low values were observed but for October, newly values were increasing. Gonad Index The mean gonad index (GI) values for the study period are presented in Table 3 and Fig. 4. Major peaks were observed in October 1989. and January, and August 1990. All these peaks were followed by decreases in GI, representative of spawnings events. Gametogenic Cycle The Chiloe scallop showed a semiannual gametogenic cycle summarized in figs 5a, 5b, 5c. Scallops were either in an active (Fig. 5a). ripe (Fig. 5b) or spent phase (Fig. 5c) throughout the year. Microscopic observations of male and female gametic conditions revealed a tendency to maintain an active re- generation of the gametes during the year. Active males were found from October (1989) to December (1989), February to June (1990) and November to December (1990), in female scallops active stage was observed in all months sampled with exception of September 1990. Ripe individuals of both sexes were found throughout the year. However ripe females were most numerous in February. June and September (1990). Meanwhile the lowest values of ripe female during the sampled period were seen during November-December 1989 and February 1990. Spent stage was recorded in female during November- December 1989, May, September and November-December 1990. Spent male were recorded in both November-December 1989 and November-December 1990. That is four resting periods for females, including two short resting phases during May and TABLE 1. Principal histological characteristics of different gonadal maturity stages. Female Male Active (Fig. lA) The active phase is characterized by the presence of ova in all stages of development, from oogonia on the follicle wall to stalked oocytes characterized by a large nucleus. Some fully developed oocytes are also free in the lumen. Ripe (Fig. IB) The ripe ovary exhibits distended follicles with detached mature oocytes, their cytoplasm contains large amounts of yolk platelets of different sizes. Only a few stalked oocytes remain. Spent (Fig. IC) The follicles are empty except for devclopmg oogonia lining the walls Some follicles show free oocytes in the lumen. (Fig. 2A) Stem cells, spermatogonia, spermatocytes, spermatids and a few spermatozoa are present extending from the follicle wall to the center of the lumen. (Fig. 2B) The follicles are distended, the lumen is filled with mature spermatozoa. Spermatogonia, spermatocytes and spermatids are found on the follicle wall. (Fig. 2C) The follicles are collapsed or decreased in size. A few follicles contain a small amount of unspent spermatozoa in the lumen. Spermatogonia and spermatocytes are found on the follicle wall. Gametogenic Cycle of the Chii.oe Scallop 61 •y^ '^»'- Photomicrographs of gonadal sections of female and male C/i/am.vs amanrfi Fig. I A. active female (80 x ), IB. ripe female (80x), IC. spent female (200x ), 2A. active male (200x i, 2B. ripe female (320x ), 2C. spent (320x ). female. September 1990 and two notorious during November-December nng after spawning (Fig. Ic). Males in the spent phase usually 1989 and 1990. Male appears with seasonal resting period during retained a few spermatozoa in a small number of seminal tubules November-December each year (1989 and 1990). (Fig. 2c). Developing oogonia were present in the follicular walls of fe- Lanae and spat It is possible that the Chiloe scallop may males in the spent phase indicating that redevelopment was occur- spawn many times throughout the year in Hueihue Bay. At least 62 Jaramillo et TABLE 2. AL. Temperature S.D. Phytoplankton S.D. Salinity S.D. Month °C + cmVm' ± ppm ± October 1989 11.16 0.61 4.86 1.85 32.78 0.55 November 1989 11.33 1.02 6.67 2.35 33.40 0.26 December 1989 11.17 0.32 2.10 0.90 33.83 0.42 January 1990 13.28 1.64 5.13 4.57 33.18 0.66 February 1990 13.90 1.47 1.96 0.81 32.90 0.44 March 1990 13.50 0.54 2.43 0.98 32.73 0.31 April 1990 11.90 0.12 2.00 0.10 33.05 0.29 May 1990 11.70 0.22 0.34 0.26 33.22 0.37 June 1990 11.15 0.31 0.30 0.14 33.03 0.13 July 1990 10.66 0.29 0.56 0.26 32.92 0.22 August 1990 10.65 0.10 1.00 0.86 33.03 0.33 September 1990 10.50 0.16 4.80 2.55 32.83 0.30 October 1990 10.47 0.21 0.20 0.08 32.73 0.32 November 1990 11.15 0.34 0.55 0.45 32.87 0.34 December 1990 12.22 0.32 4.40 1.10 33.12 0.32 Temperature and Phytoplankton values (included Standard Deviation) recorded for the study period at Hueihue Bay, Chiloe, Chile four tiines are suggested by the recorded resting periods. Veliger larvae of the Chiloe scallop were found in plankton samples in November 1989, January, February, September and December 1990. Settlement may be dependent on certain environmental conditions (food, temperature). Spat settled on onion-bags collec- tors were found in late November 1989, and late February and September (1990), and early December. DISCUSSION The histological changes in the gonadal cycle of female Chiloe scallops {Chlamys ainandi) collected from October 1989 to De- cember 1990 showed that maturity and spawning were occurring in a semiannual pattern. Males were more or less continually spawning throughout the year. Similar observations have been described by Borden (1928), Naidu (1970) and Du Paul et al. (1989) for Placopecten magellanicus. In bivalves the reproductive cycle is generally closely linked to several environmental factors. The most commonly cited are water temperature, food availability, tidal influence and depth (Sastry 1966, Machell and DeMartini 1971, Mac Donald and Thompson 1985a). In Hueihue Bay the water temperature was highly constant during several months of the study period, but there was a marked Oct N(x Dec Jan Fetj Mar Apf May Jun Jul Aug Sep Oct Now Deo I 1969 I 1900 I Figure 3. Phytoplankton represents cm' of phytoplankton decanted for each m' of sample. seasonal peak during 1990 summer months just when gonads were in ripe stage. However, in spring 1990 when the gonads were ripe again, the water temperature was relatively low. Because high temperature appeared related to the ripe phase rather than the active phase, it is possible that another environmental factor was influencing gonad growth and gametogenesis. Gonadal develop- ment of Chiloe scallop is probably initiated when temperature is low, but food availability was high. Low temperature does not inhibit gametogenesis, at least for Chlomys amandi, (this study) Placopecten magellanicus (Thompson, 1977) and Pecten novae- zelandiae (Bull, 1976). Abundance of food has been generally associated with breeding period of marine invertebrates and thought to ensure adequate nutritional availability for planktotrophic larvae (for review see Thorson 1950, Giese 1959 and Sastry 1966). More recent studies have emphasized the importance of food availability (for review see Bayne and Newell 1983, Broom and Mason 1978, Emmett et al. 1987). In the Chiloe scallop, gonad growth appears to coincide with periods in which there are high levels of food available in Hueihue TABLE 3. Month G.I. 100% S.D. ± October 1989 12.5 2.90 November 1989 6.08 1.97 December 1989 8.12 1.01 January 1990 14.6 2.63 February 1990 10.24 3.82 March 1990 10.84 4.06 April 1990 8.62 3.98 May 1990 7.35 3.31 June 1990 9.03 2.72 July 1990 7.32 1.99 August 1990 10.56 2.87 September 1990 8.27 2.06 October 1990 5.05 1,63 November 1990 5.78 1,14 December 1990 2.53 1.54 G.l. values recorded (included S.D.) for the studied period at Yaldad Bay. Gametogenic Cycle of the Chiloe Scallop 63 io« 0% Oot Nov Dec Jen Fab Mar A()r M«y Jw Ju( Aug Sap Oct Ncm Dae I 1989 I 1900 I Figure 4. Variation of Gonadic Index for the sampled period. Oct Nov D«o Jan Fob Mar Apr May Jun Jul Aug Sep Oct Nov Dae I 1988 I 1990 I Active 5a Bay. Thus it is probable that food supply is more related to go- nadal development in Chlamys amandi than temperature. Observations reported by Sastry (1966) for bay scallops Ae- quipecten irradians exposed to vanous temperatures during the period of gonad growth, without food supply, showed a decrease in both gonad and digestive gland index. This shows that addi- tional food is essential for gonad growth. Enimett et al. (1987) conclude that temporal and quantitative differences in food supply has a greater influence on reproductive cycles than water temper- ature or latitude. Time of spawning may also be related to food availability. Most bivalves tend to spawn during periods when food is available for developing progeny (Bayne 1976). Disalvo et al. (1984) in- duced spawning m A. purpuratus in winter at 13°C by simply rinsing a concentrate of phytoplankton into the culture tanks. Thus Wolff (1988) suggests that high temperature, although favouring maturation and spawning, might be less critical for a successful spawning than food availability. In Chlamys amandi the spawning time also appeared related to high food levels rather than water temperature in Hueihue Bay. The presence of larvae of Chlamys amandi registered in No- vember ( 1989), January, February, April, September and Decem- ber (1990) may be explained by the timing of female spawning (mainly spring and fall) and a food availability. The spat which settled during late November 1989, late Feb- ruary, April, September and December (1990) coincided with high food values and female spawning peaks. Success and rate of larval development of many marine species that have a planktonic larval stage are affected by physical and endogenous parameters. Among physical parameters, temperature is probably the most frequently investigated because it can be easily manipulated and has a significant effect on growth and survival (Davis and Calabrese 1964, Lought and Ganon 1973, Tettelbach 1979. Falmagne 1984, Wolf 1988). These and another studies have shown that growth to settling size, and therefore completion of the larval period, generally is more rapid as tem- perature increases to some optimun level, and then declines with further temperature increases (Bayne 1983). The presence of lar- Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sop Oct No^ Doc I 1989 I 1990 I Ripe 5b Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Doc I 1989 I 1990 I Spent -^ Ivlale —*~ Female OC Figure 5A,5B,5C. Gametogenic phases of Chiloe scallop, the values represents the percentage frequency of scallop in each phase. vae of Chiloe scallops under favorable temperatures (over 1I°C) after the spawning time agree with these suggestions. ACKNOWLEDGMENTS This work was supported by a FONDECYT N* 158-1987. Research Project and also by contributions of DID-UACH. LITERATURE CITED Akaboshi, S. & J. E. Illanes. 1983. Estudio experimental sobre la capta- cion, pre-cultivo y cultivo, en ambiente natural de Chlamys iar- gopecten) piirpurata. Lamarck 1819, en Batiia Tongoy, IV Region Coquimbo. Symp. Intemac. de Acuac. Coquimbo, Chile. Sept. pp: 233-254. Avjies, S. & E. Lozada. 1975. Estudio histologico del cicio reproductive de Concholepas concholepas Brugiere (1789) en Punta Saliente. Co- quimbo Boletin de la Sociedad de Biologi'a de Concepcion, 49:207- 218. Barber, B. J. & N. J. Blake, 1991. Reproductive Physiology. 377-428 In 64 Jaramillo et al. Scallops; Biology, Ecology and Aquaculture Ed. By S. E. Shumway. Elsevier Science Pub. The Netherlands. Bayne. B. L. 1976. Marine Mussels: Their ecology and Physiology. Cam- bridge, England. 506 pp. Bayne. B. L. 1983. Physiological ecology of marine molluscan larvae. Verdonk, N. H., J. A. M. van der Biggelaar, and A. S. Tompa, eds. The Mollusca. New York. NY: Academic Press, vol. 3 Development pp. 299-343. Bayne, B. L. & R. C. Newell: 1983. Physiological energetics of marine molluscs. In: The Mollusca. Vol. 4(11. pp 407-515. Ed. by A. S. M. Saleuddin and K. M. Wilbur. New York: Academic Press. Bayne, B. L. 1985. Responses to environmental stress: tolerance, resis- tance and adaptation. Proc. 18th Eur. mar. Biol. Symp. 331-349. (Ed. by J. S. Gray and M. E. Chnstiansen. Chichester: Wiley Interscience). Borden, M. A. 1928. A contribution to the study of the giant scallop Placopecten grandis(S), Fisher. Res. Board Canada, Manuscript Re- ports of the Biological Stations, N°350, 24 pp. Broom, M. J. & J. Mason. 1978. Growth and spawning in the pectinid Chlamys opercularis in relation to temperature and phyloplankton con- centration. Mar. Biol. 47:277-285. Bull, M. P. 1976. Aspect of the Biology of the New Zealand scallop, Pecten novaezelandiae Reeve 1853. in the Malborough Sounds. Ph. D. thesis. Victoria University. Wellington. New Zealand, 175 pp. David, H. C. & A. Calabrese. 1964. Combined effect of temperature and salinity on development of eggs and growth of larvae of A^. mercenana and C. virginica U.S. Fish, and Wild!. Serv., Fish, Bull, 63:64.V655, DiSalvo, L, H,, E. Alarcon, E. Martinez & E. Uribe. 1984. Progress in mass culture of Argopecten purpuraius with notes on its natural his- tory. Rev. Chilena de Hist. nat. 57:33-45. Du Paul, W. D., J. E. Kirkley, & A. C. Schmitzer. 1989. Evidence of semiannual reproductive cycle for the sea scallop Placopecten magel- lanicus (Gmelin) in the Mid-Atlantic region. J. Shellfish Res 8(11: 173-178. Emmett, B., K. Thompson & J. D. Popham. 1987. The reproductive and energy storage cycles of two populations oi mylitus ediilis (Linne) from British Columbia. J. Shellfish Res. 6(l):29-36. Falmagne, C, M. 1984. The combined effect of temperature/salinity on survival and growth of Mytilus californianus larvae ( A response sur- face analysis). Seattle. WA: Univ. of Washington. 85 p. Thesis. Ganter. P. & G. Jolles. 1970. Histochimie Nomiale et Patologique vol 2. Ed. Gauthiers-Villars Paris. 1390 pp. Giese, A. C. 1959. Comparative physiology: annual reproductive cycles of marine invertebrates. Annual Review of Physiology, vol 21:547- 576. Humanson. G. L. 1962. Animal Tissue Techniques. W. H. Freeman and Company. San Francisco. Ca 468 pp. Loosanof, V, L, 1937, Spawning of Venus mercenaria Ecology 18:506- 515. Lough, R. G. & J. J. Ganon. 1973. A response-surface approach to the combined effects of temperature and salinity on the larval development of Adula californiensis (Pelecypoda: Mytilidael 1 survival and growth of three and fifteen-day old larvae. Mar. Biol. {Berl.) 22:241-250. Machell, J. R. & D. DeMartini. 1971. Annual reproductive cycle of gaper clam, Tresiis capax in South Humboldt Bay California. Calif. Fishy. Gam. 57(4):274-282. MacDonald, B, A. & R, J. Thompson. 1985a. Influence of temperature and food availability on the ecological energetics of the giant scallop Placopecten magetlanicus . II Reproductive output and total produc- tion. Marine Ecology-Progress Series Vol. 25:295-303. MacDonald, B. A. and R. J. Thompson. 1986. Influence of temperature and food availability on the ecological energetics of the giant scallop Placopecten magellaniciis. Mar. Biol. 93:37-48. Malachowski, M. 1988. The Reproductive cycle of the rock scallop Hin- nites giganteiis (Grey) in Humboldt bay California. Journal of Shellfish Research. Vol. 7 N° 3, 341-348. Naidu. K. S. 1970. Reproduction and breeding cycle of the giant scallop Placopecten magellanicus (Gmelin) in Port au Port Bay. Newfound- land. Can. J. Zool. 48:1003-1012, Newell, R, I,. T, J. Hilbish, R. K. Koehn & C. J. Newell. 1982. Temporal variation in the reproductive cycle of Mylilus edulis (Bivalve:Mytilidae) from localities on the East coast of the United States. Biol. Bull. 162:229-310. Ramorino. L. 1975. Cicio reproductivo de Concholepas concholepas en la zona ed Valparaiso. Rev. Biol. Mar. Valparaiso. 15(2):I49-177. Ropes, J. W. & A. P. Stickney. 1965. Reproductive cycle o{ Mya are- naria in New England. Biol. Bull. 128:315-327, Sastry. A, N, 1966 temperature effects in reproduction of the bay scallop, Aequipecten irradians Lamarck, Biol, Bull, 130:118-134, Sastry, A. N. 1970, Reproductive physiological vanation in the latitudi- nally separated populations of the bay scallop, Aequipecten irradians Biol. Bull. 138:56-65, Sastry, A. N. 1979. Pelecypoda (excluding Ostreidae). In: Reproduction of marine invertebrates. Vol. V. pp 113-292. Ed. by A. C. Giese and J. S. Pearse. New York: Academic Press. Tettelbach. S. T. 1979. The combined effects of temperature and salinity on embryos and larvae of the northern bay scallop. Argopecten irra- dians irradians (Lamarck). Seattle, WA: Univ. Washington. 74 pp. Thesis. 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(1I):2104-21I6. Thorson, G. 1950. Reproduction and larval ecology of marine inverte- brates. Biol. Rev. 25:1-45. Wolff, M. 1988, Spawning and recruitment in the Peruvian scallop Ar- gopecten purpuratus Marine Ecology Progress Series Vol. 42:213- 217. Journal of Shellfish Research. Vol. 12. No. 1, 65-64. 149.1. DISSEMINATED SARCOMAS OF SOFT-SHELL CLAMS, MYA ARENARIA LINNAEUS 1758, FROM SITES IN NOVA SCOTIA AND NEW BRUNSWICK CAROL M. MORRISON, ANNE R. MOORE, VIVIAN M. MARRYATT AND DAVID J. SCARRATT Deparimcni of Fisheries and Oceans Benthic Fisheries and Aquacidture Division Halifax Laboratory Halifax. N.S. B3J 2S7 ABSTRACT Over a one year penod, 896 soft-shell clams from 22 locations along the Bay of Fundy coast of New Brunswick and Nova Scotia and the Eastern shore of Nova Scotia were exammed for the possible occurrence of sarcomas. Biopsies revealed disseminated sarcomas in clams from five sites around the Bay of Fundy; the first records for Canadian waters. The data were insufficient to show whether there was any correlation between incidence of sarcomas and either pollution or recent declines in clam abundance. KEY WORDS: sarcoma, neoplasia, soft-shell clam. Mya INTRODUCTION Disseminated sarcomas have been reported in 15 species of marine and estuarine bivalve molluscs world-wide (Peters 1988) including the soft-shell clam, Mya arenaria Linnaeus, a species of importance to east coast fisheries in Canada and the U.S. These sarcomas consist of abnormal anaplastic cells that have a distinc- tive apf)earanee; being enlarged and rounded, with large, hyper- chromatic, often lobed nuclei containing one or more prominent nucleoli, surrounded by little cytoplasm. Usually, mitoses are common in these cells, which are found throughout the tissues, including the blood vessels. This distribution, together with im- munocytological evidence (Smolowitz and White 1992) indicate that the neoplastic cells are of haemocytic origin. However, neo- plastic cells have different antigens from normal haemocyles, sug- gesting that they may not have the same origin (Reinisch et al. 1983). Disseminated sarcomas have been reported in soft-shell clams along the Atlantic coast of the United States from the Hudson River to the northern part of Maine, near the U.S. /Canada border (Brousseau 1987. Brown et al. 1976. 1977 and 1979. Gardner et al. 1991, Peters 1988, Reinisch et al. 1984, Sherburne and Bean 1983. Yevich and Barszcz 1977), but not hitherto from Canadian waters. No sarcomas were reported in clams from Chesapeake Bay until 1979. These are believed to be derived from clams trans- ferred from New England to Chesapeake Bay after hurricane Agnes decimated local stocks in 1972. Only isolated cases were reported until 1983, when the numbers reached epizootic propor- tions of up to 90% (Farley 1969, 1989. Farley etal. 1986a. 1986b. 1991). This relatively recent appearance and sudden increase of the numbers of affected clams indicates that an infectious etiology is involved, and a virus similar to a B-type retrovirus has been reported in Rhode Island clams with disseminated sarcomas (Coo- per, Brown and Chang 1982a, Oprandy et al. 1981. Oprandy and Chang 1983). Sarcomas have been found in relatively unpolluted as well as polluted areas; and there does not seem to be a clear-cut correlation between prevalence of the disease and presence of pollutants (Far- ley 1989. Mix 1986. Reinisch et al. 1984. Yevich and Barszcz 1977). There are. however, some indications that stress can en- hance the spread of the infectious agent (Brown 1980. Peters 1988), which may explain the occurrence of more neoplasias at some sites where clams have been exposed to oil (Yevich and Barszcz 1977); herbicides (Gardner et al. 1991) or mixtures of pollutants (Reinisch et al. 1984); or where clams contain high levels of the pesticide chlordane (Farley et al . 1 99 1 ) or polynuclear aromatic hydrocarbons (Brown et al. 1979). There have been few studies of neoplasias in bivalves in Can- ada. Haematopoietic neoplasms were found in mussels Mylilus edulis from several sites along the shore of Vancouver island (Mix 1986). and in the truncate soft-shell clam Mya iruncata and the chalky macoma. Macoma calcarea from Baffin Island (Neff et al. 1987). On the east coast "Malpeque disease"', originally reported in 1915 from Malpeque Bay. Prince Edward Island, killed over 90% of the eastern oyster Crassostrea virginica in 3 years. The disease spread to other parts of P.E.I, between 1915 and 1937, causing similar epidemics (Drinnan and Medcof 1961). In 1955 the disease had spread to the mainland, and by the end of 1960 all major areas where oyster beds in New Brunswick and west of Cape George in Nova Scotia had experienced epizootics. Mori- bund oysters contain secondary invaders, and many agents, in- cluding bacteria, Hexamila sp. and fungi have been suggested as the etiologic agent of Malpeque disease (Logic 1958). However, Drinnan et al. (1960-61) described cells similar to the neoplastic cells of disseminated sarcomas in diseased oysters. It was not known at the time whether these cells were normal oyster cells or pathogens, but it seems likely that they were the cause of the mortalities in view of more recent work on neoplasias in bivalves. It is interesting that it was possible to rebuild oyster populations in the stricken areas of Nova Scotia and New Brunswick by trans- planting P.E.I, oysters that had survived the epizootic and were resistant to the etiologic agent (Drinnan and Medcof 1961 ). Clams resistant to sarcomas were also found in Rhode Island, where sarcomas were chronic in some parts of the population, and showed remission in others (Cooper et al. 1982a). Oysters in the Bras d'Or lake. Cape Breton Island were never infected by Mal- peque disease, but when moved to areas where it had spread de- veloped disease symptoms and suffered high mortalities (Drinnan and England 1965). Apparently soft-shell clams in the areas where there were diseased oysters did not show typical symptoms and were considered to be unaffected. There have been no published studies of sarcomas in soft-shell clams from Canada, so in 1985 it 65 66 Morrison et al. TABLE. Geographical locations, site number, sampling dates, and number of specimens found to have disseminated sarcomas. Site Number Number Positive by Stage Location Date Collected Early Intermediate Advanced % Positive Lepreau harbour 1 16.12.1985 Bocabec Bay 2 16.12 1985 St. Croix River 3 16.12 1985 Lepreau harbour 1 18.12 1985 Thome's Cove 4 3.1 1986 , Yarmoulh area 5 3.1 1986 Advocate 6 6.8 1986 2 3 2 21.9 Lower Five Islands 7 6.8 1986 3 2 5 31.3 Smith's Cove 8 6.8 1986 Upper Clements 9 6.8 1986 Cook's Beach 10 6.8 1986 1 1 6.3 Pottery Creek 11 20.8 1986 1 1 6.3 Magaguadavic 12 20.8 1986 L'Etete 13 20.8 1986 Lepreau harbour 1 20.8 1986 1 3.1 Stuarttown 14 20.8 1986 McCanns Cove 15 20.8 1986 Goat Island 16 10.9 1986 Cole Harbour 17 26.11 1986 Oak Point, Annapolis 18 9.12 1986 Clementsport 19 9.12 1986 Thome's cove 4 9.12 1986 The Joggins. Annapolis 20 9.12 1986 Economy Point 21 20.1 1987 Economy Point 21 20.1 1987 Advocate 6 20.1 1987 Five Islands 22 22.1 1987 Five Islands 22 23.1 1987 was decided to monitor sites close to the U.S. border, to determine if sarcomas were present in Canadian waters. METHODS Sampling was restricted to 22 sites in New Brunswick and Nova Scotia; most being in the Bay of Fundy (Table, Map). Most samples of clams were obtained from commercially harvested clam flats. Some of these flats are subject to closure to harvesting as a result of high faecal colifonm counts in the overlay waters. The coliforms may originate from populated areas, farmland or wild animals such as gulls, and do not usually appear to stress the clams. The second sample from Lepreau harbour (Site 1) was taken one day after a diesel oil spill which resulted in some clam flats having to be closed due to detectable hydrocarbon odours. Thome's Cove (Site 4) is in the Annapolis Basin, where the clam population had been depleted by overfishing. Clams were sampled from Oak Point (Site 18), near Goat Island, because there was an unexplained disappearance of clams between July 1986 and April 1987 (Prouse et al. 1988, Rowell and Woo 1990, Rowell in press). Clams were sampled from two sites at Five Islands (Site 22) be- cause one was open for harvest, and one was closed because of contamination by faecal coliforms. The sample from each site consisted of thirty-two clams, rang- ing in length from 2.9-8.5 cm, except for one of the Economy Point sites, where no small clams could be found, so only sixteen large clams were taken. Clams were biopsied using an in vivo bleeding technique slightly modified from that described by Farley et al. (1986a). This technique detects neoplastic cells circulating in the haemolymph (Cooper et al. 1982b), and is quicker than pre- paring histological sections. Cooper et al. ( 1982b) found that 94% of diseased clams were detected using the bleeding technique. 46t 67 66 65 64 63 Map. Locations and site numbers of sampled sites. Solid circles indi- cate the sites of positive cases. Sarcomas of Soft-Shell Clams 67 compared to the number found to be positive from histological sections. The accuracy was greater as the disease progressed, be- ing 66-71% in clams with a light infection, rising to 100% in more advanced infections. Farley et al. (1986b. 1989) found that the biopsy technique was more sensitive than examination of histo- logic sections, the latter only being reliable for the more advanced stages. Since this survey was conducted with limited resources only the biopsy technique was used, without histological compar- isons. Blood samples were put into eight-chambered tissue-culture slides (Lab-Tek. Miles Scientific. Illinois. U.S.A.) rather than a single chamber attached to a standard slide as used by Farley et al. (1986a). The former produces a smaller area of cells from each sample for observation, but more samples can be processed at one time, and a poly-L-lysine coating is not necessary to ensure ad- herence of neoplastic cells (Dr. R. A. Sonstegard. pers. comm). We also used ambient sea-water filter-sterilized through a 0.45 \x membrane filter instead of artificial sea-water. The fixing and staining procedures were as described by Farley, and positive sam- ples were staged according to the percentage of neoplastic cells into: early (0.01-0.9%), intermediate (1^9%) and advanced (50- 100%) stages (Farley et al. 1986a). RESULTS Neoplastic clams were found in the early, intermediate and advanced stages (Table). The neoplastic cells in the intermediate and advanced stages were very distinct from normal haemocytes, being large (3-7 ^.m in diameter) with a round or oval nucleus contaimng clumps of chromatin, and a distinct nucleolus (Fig.). Nuclei from the early stages had a similar appearance, but were less distinct because they were only about 3^ jji.m in diameter. Neoplastic cells were found in 7 clams from Advocate. 10 clams from Lower Five Islands. 2 clams from Cooks Beach, 2 clams from Pottery Creek and 1 clam from Lepreau Harbour (Map). All positive cases were found in August. Two of the cases at Advocate were advanced, three intermediate and two early. Five of the cases at Lower Five Islands were advanced, two intermediate and three early. One case at Cook's Beach and Pottery Creek was early. Figure. Neoplastic cells from a clam with a heavy infection. Abbre- viations: Nu — nucleolus of neoplastic cell, .N — nucleus of normal haemocyte. Bar = 20 nm. one intermediate, and the case at Lepreau Harbour was interme- diate. DISCUSSION Having discovered that neoplasias were present in the Bay of Fundy, it had originally been intended to extend this study to soft-shell clams from other sites around the Maritime Provinces of Canada, to sample at different times of year, and to use histolog- ical sections to verify the occurrence of neoplasias found using the bleeding technique, and to study the course of development of neoplastic cells. Unfortunately, lack of time and loss of personnel made this impossible. However, although limited in scope, this study clearly demonstrates the presence of sarcomas in soft-shell clams at several sites around the Bay of Fundy. In other studies of disseminated sarcomas it has been found that prevalence differs among sites, and may vary seasonally. Usually the prevalence is low. as found here, although epizootics may occur (Peters 1988). In Chesapeake Bay. the epizootics seemed to develop in the fall and reach more advanced stages by spring. Laboratory-held animals having these sarcomas have experienced 100% mortality, and in the field few sarcomas were found from June to August, indicating that infected clams had died (Farley et al. 1986b, Farley 1989). The reason for this cycle is unknown (Peters 1988). The cycle in the Bay of Fundy seems to be differ- ent, because no neoplastic cells were found in December or Jan- uary, although 52% of the clams were sampled during these months; and all infected animals were found in August. A sample size of 30 animals gives a 95% chance of detecting one or more infected specimens when the detectable infection rate is 10% or more in a population over 100,000 (Ossiander and Wedemeyer, 1973). Neoplastic cells in the intermediate and advanced cases are easily seen using the biopsy method, so the detection efficiency of these cases should be high, although early infections, which are more difficult to detect, could be present during the winter. Most mortalities from Malpeque disease occurred in summer and early fall (Needier and Logic 1947), so possibly colder winter temper- atures in Canada delay the progress of disseminated sarcomas in bivalves. Further studies of sarcomas in soft-shell clams at differ- ent times of the year are needed, to see if there is a seasonal variation in the progression of the disease. The sample sizes from Lepreau harbour (Site I ) were not big enough to establish any correlation with the presence of oil from the oil spill and neoplasias, although one neoplastic clam was found 8 months after the clam flats were contaminated by oil. In the present study it was not possible to link mortalities to the occurrence of neoplasias, but this kind of correlation is difficult since so many factors can cause large mortalities among soft-shell clams. For example, reductions in numbers of clams have been attributed to silting of clam-beds, harvesting methods, overfish- ing, predation by the green crab, Carcinus maenas. the clam drills Limalia triseriata and L. heros: and flounders and gulls (Hart 1954 and 1955, Needier 1947 and 1953, Emerson et al. 1990, Robinson and Rowell 1990), At the Oak Point site in the Annapolis Basin (Site 18), where there had been an unexplained reduction in the clam population, we found no neoplasias. The decline has since been shown to be due to the synergistic effects of predation by the nemertean worm Cerebraiulus lacieus Verrill, and silting which apparently prevented larval settlement. The silting probably re- sulted from a change in hydrographic conditions as a consequence of the installation of a tidal power facility (Prouse et al. 1988, Rowell and Woo 1990, Rowell in press). 68 Morrison et al. ACKNOWLEDGEMENTS Mr. A. Farley's advice was invaluable in carrying out this work, and he also sent a slide of a biopsy from a clam with a sarcoma and verified some of our slides. Mr. S. Sherburne also provided valuable comments, and sent us slides, and micrographs of sectioned material of clams with sarcomas. Members of the Inspection Division, Fisheries Operations Branch and Mr. P. Woo of the Habitat Ecology Division, Biological Sciences Branch ob- tained and delivered samples, and co-workers in the Fish Health Unit looked after samples, processed biopsies, and provided ad- vice and supplies. Mr. J. Black set up the computer program for the map. Mr. T. W. Rowell provided a thorough review of the manuscript, and provided helpful advice. LITERATURE CITED Brousseau, D. J. 1987. Seasonal aspects of sarcomatous neoplasia in Mya arenaria {soft-shell clam) from Long Island Sound. /. Invert. Path. 50:269-276. Brown, R. S. 1980. The value of the multidisciplinary approach to re- search on marine pollution effects as evidenced in a three-year study to determine the etiology and pathogenesis of neoplasia in the soft-shell c\am, Mya arenariu. Rapp. P. -v. Reun. Cons. int. Explor. Mer. 179: 125-128. Brown, R. S., R. E. Wolke & S, B. Saila. 1976. A preliminary report on neoplasia in feral populations of the soft-shell clam, Mya arenaria, prevalence, histopathology and diagnosis. Proc. Int. Coll. Invert. Path. 1:151-158. Brown, R. S., R. E. Wolke, C. W. Brown & S. B Saila 1979. Hydro- carbon pollution and the prevalence of neoplasia in feral New England soft-shell clams {Mxa arenaria). pp. 41-51 . In Animals as monitors of environmental pollutants. Nat. Acad. Sci. Washington, D.C. Brown, R. S., R. E. Wolke. S. B. Saila & C. W. Brown. 1977. Preva- lence of neoplasia in 10 New England populations of the soft-shell clam (Mya arenaria). Ann. New York Acad. Sci. 298:522-534. Cooper, K. R. R. S. Brown & Pei W. Chang. 1982a. The course and mortality of a hematopoietic neoplasm in the soft-shell clam, Mya arenaria. J. Invert. Path. 39:149-157. Cooper, K. R, R. S. Brown & Pei W, Chang, 1982b, Accuracy of blood cytological screening techniques for the diagnosis of a possible hema- topoietic neoplasm in the bivalve mollusc. Mya arenaria. J. Invert. Path. 39:281-289. Drinnan, R. E. & L. A. England. 1965. Further progress in rehabilitating oyster stocks. Fisheries Research Board of Canada Biological Station, St. Andrews, N.B. General Series Circular No. 48. Drinnan, R. E., E. B. Henderson & B. T. Khouw 1960-61 Malpeque Disease, p. 31-36. In Annual report and investigators' summaries. Fisheries Research Board of Canada, Biological Station, St, Andrews, N B Drinnan, R, E, and J, C. Medcof. 1961. Progress in rehabilitating disease affected oyster stocks. Fisheries Research Board of Canada. Biological Station, St. Andrews, N.B, General Series Circular No. 34. Emerson, C. W, J. Grant & T. W. Rowell. 1990. Indirect effects of clam digging on the viability of .soft-shell clams, Mya arenaria L. Nether- lands J. Sea Res. 27:109-118. Farley, C. A. 1969. Probable neoplastic disease of the hematopoietic sys- tem in oysters, Crassostrea virf^imca and Crassostrea gi^as. National Cancer Institute Monograph 31:541-555, Farley, C. A. 1989. Selected aspects of neoplastic progression in niol- lusks, pp. 23-31 In H. E. Kaiser [ed.] Comparative aspects of tumor development. Kluwer Academic Publishers, Dordrecht. Farley, C. A., S. V. Otto & C. L Reinisch. 1986a. New Occurrence of epizootic sarcoma in Chesapeake Bay soft shell clams, Mya arenaria Fishery Bullelm. 84:851-857. Farley, C, A., S, V. Otto & C L, Reinisch, 1986b Epizootic sarcoma in Chesapeake Bay soft-shell clams — A virus?, pp, 436-440 In R, A, Sampson, J. M. Viak and D Peters (ed.) Fundamental and applied aspects of invertebrate pathology. Foundation of the Fourth Interna- tional Colloquium of Invertebrate Pathology, Wageningen, the Neth- erlands, Farley, C, A,, D, L, Plutschak & R, F, Scou. 1991, Epizooliology and distribution of transmissible sarcoma in Maryland softshell clams, Mya arenaria. 1984-1988. Environmental Health Perspectives. 90:35^1. Gardner, G. R., P. P. Yevich, J. Hurst, P. Thayer, S Benyi, J. C, Harsh- barger & R. J. Pruell, 1991. Germinomas and teratoid siphon anom- alies in softshell clams. Mya arenaria. environmentally exposed to herbicides. Environmental Health Perspectives, 90:43-51, Hart, J. L. 1954. Report of the Atlantic Biological Station for 1954, Fish- eries Research Board of Canada, Hart, J. L. 1955, Repon of the Atlantic Biological Station for 1955. Fish- eries Research Board of Canada. Logie, R, R, 1958. Epidemic disease in Canadian Atlantic Oysters (Cras- sostrea virginica). Thesis submitted in partial fulfillment of the re- quirements for PhD., Graduate School of Rutgers University, The State University of New Jersey. Fish. Res, Bd. Can. Ms. Rept. Senes (Biological) No. 661. Mix, C, M, 1986, Cancerous diseases in aquatic animals and their asso- ciation with environmental pollutants: a critical literature review, Ma- nne Environmental Research 20:1:141. Needier, A. W. H. 1947. Report of the Atlantic Biological Station for 1947. Appendices 19 and 20. Fisheries Research Board of Canada. Needier, A. W. H. 1953. Report of the Atlantic Biological Station for 1953. Fisheries Research Board of Canada. Needier, A, W, H, & R. R, Logie. 1947. Senous mortalities in Prince Edward Island oysters caused by a contagious disease. Trans. Roy. Soc. Canada. XL 1:73-89. Neff, J. M., R E Hillman, R. Scott Carr, R. L. Buhl & J. I Lahey. 1987. Histopathologic and biochemical responses in arctic manne bi- valve molluscs exposed to experimentally spilled oil. Arctic 40:220- 229. Oprandy, J. J. & P. W. Chang. 1983. 5-bromodeoxyuridine induction of hemopoietic neoplasia and retrovirus activation in the soft-shell clam, Mya arenaria. J Invert. Path. 42:196-206, Oprandy, J. J., P W Chang, A D. Pronovost, K. R, Cooper. R, S. Brown & V, J. Yates. 1981. Isolation of a viral agent causing hema- topoietic neoplasia in the soft-shell clam, Mya arenaria. J . Invert. Pathol. 38:45-51, Ossiander, F. J. & C Wedemeyer. 1973. Computer program for sample sizes required to determine disease incidence in fish populations. Jour- nal of ihe Fishenes Research Board Canada. 30:138.V1384, Peters, E, C, 1988, Recent investigations on the disseminated sarcomas of marine bivalve molluscs, American Fishenes Society Special Publica- tion 18:74-92. Prouse. N, J, T. W. Rowell, P. Woo, J. F. Uthe, R. F. Addison, D. H. Loring, R. T. T. Rantala, M. E. Zinck & D. Peer. 1988. Annapolis Basin soft-shell clam (Mya arenaria) mortality study: a summary of field and laboratory investigations. Can. Man. Rep. Fish. Aquat. Sci. No. 1987:vii -H 19pp. Reinisch, C. L., A. M. Charles & A. M. Stone. 1984 Epizootic neopla- sia in soft shell clams collected from New Bedford Harbor Hazardous Waste. 1:73-81. Reinisch, C. L., A. M. Charles & J. Troutner. 1983 Unique antigens on neoplastic cells of the soft shell clam Mya arenaria. Dev. and Comp. Immun. 7:33-39. Robinson, S, MK. C. & T. W. Rowell. 1990. A re-examination of the incidental fishing mortality of the traditional clam hack on the Sarcomas of Soft-Shell Clams 69 soft-shell clam, Mya arenaha Linnaeus, 1758. J Shellfish Res. 9:283- 389. Rowell, T. W Destruction of a clam population [Mxa arenaria Linne) through the synergistic effects of habitat change and predation by a nemerlean (Cerebrantlus lacteus Verrill) Proc. 25th Europ. Mar. Biol Symp. (In press). Rowell, T. W. & P. Woo. 1990. Predation by the nemertcan worm, Cer- ehrutulus liicleus Verrill. on the soft-shell clam, Mya arenaria Lin- naeus, 1758. and its apparent role in the destruction of a clam flat. J. Shellfish Res. 9:291-297. Sherburne, S. W. & L. L. Bean. 1983. Hematopoietic neoplasms, acces- sory siphons and gross tumours in soft-shell cjams, Mya arenaria. from Dennysville, Maine. Research Reference document 83/18. Maine Department of Marine Resources, Fisheries Research Laboratory. West Boothbay Harbor. Maine 04575. Smolowitz, R. M. & M. K. White. 1992. Monoclonal antibodies recog- nizing normal hemocytes cross react with a sub-population of leukemia cells, suggesting a common origin in Mya arenaria. Abstract. 12th Annual Shellfish Biology Seminar. Nat. Oceanic and Atmospheric Administration, Nat. Mar. Fish. Service, Northeast Fisheries Science Center. Milford Laboratory Milford, Connecticut 06460. Yevich, P. P. & C. A. Barszcz. 1977. Neoplasia in soft-shell clams (Mva arenaria) collected from oil-impacted sites. Ann. N.Y. Acad. Sci. 298: 409-426. Journal of SheUfish Research. Vol. 12. No. 1. 71-75. 1993. GONADAL COMPARISON OF MASCULINIZED FEMALES AND ANDROGYNOUS MALES TO NORMAL MALES AND FEMALES IN STROMBUS (MESOGASTROPODA: STROMBIDAE) SHAWNA E. REED* Department of Marine Sciences University of Puerto Rico P.O. Box 908 Lajas. Puerto Rico 00667 ABSTRACT Gonad and other reproductive tissues were taken from masculinized females and sexually undeveloped individuals of the West Indian fighting conch. Sirombus piigili.s. and from two masculinized female queen conch. 5. f;it;a.s, for comparison to normal males and females of their respective species. Masculinized females were indistinguishable from normal females, except for the presence of a small, deformed verge which resembled, microscopically, that of a normal male. Sexually undeveloped specimens did not possess a verge or an egg groove. Microscopic examination of the gonad tissue revealed some undeveloped, inactive spermatogenic tissue, showing that these individuals were androgynous males. KEY WORDS: Stromhus. masculmization INTRODUCTION The Stronibidae are tropical marine gastropods with the ma- jority of species found in the Indo-Pacific region (Abbott 1960). The family is represented in the Caribbean by six species found: Strombus gigas, the queen conch. S. co.siaiiis. the milk conch, 5. raninus. the hawkwing conch. 5. gaUus. the rooster-tail conch. S. ahitus. the Florida fighting conch, and 5. pugilis. the West Indian fighting conch. All conch are gonochoristic and exhibit sexual dimorphism; males have a penis (termed "verge") and females have an egg groove. Intersexes can occur in any gonochoristic species due to genetic or environmental causes or a combination of genetic-environmen- tal interactions. For example, meiotic nondysjunction of sex- determining chromosomes can lead to a variety of intersexes, and mitotic nondysjunction can cause mosaics, as found in Droso- philo melanogaster (see Ayala and Kiger 1984). Chemical toxins are known to cause imposex in dog whelks (Bryan et al. 1988. Gibbs et al. 1988) and masculinization in mosquitofish (Howell et al. 1980). Intersexes are found during transition from one sex to the other in those species that change sex. There are also cases of hermaphrodites occurring, at low percentages, in dioecious spe- *Present address: 6H4. % 24265— 60th Ave., Langley. B. C. Canada V.IA cies (e.g. brachiopods: Culter and Simon 1987, clams: Ropes 1982). Individuals have been found in several species of Slrombus that exhibit hermaphroditic characteristics in that they possess both an egg groove and a verge (R. S. Appeldoom, pers. comm.; pers. obs.). The verge was, however, dysfunctional due to underdevel- opment and deformity, often being multilobed. Kuwamura et al. (1983) described a similar condition in female 5. luhuanus but assumed that this secondary sexual characteristic was a "clitoris" rather than any abnormality. Botero (1984) depicted a male S. gigas with three verges that appear to be of functional size but did not mention whether an egg groove was also present. No studies were ever done to determine the extent of the intersexual condition found in strombids. This study was undertaken to determine the incidence and clar- ify the nature of intersexual strombids. The study concentrated on masculinized female Strombus pugilis taken from two populations in the La Parguera area of Puerto Rico, and includes two androg- ynous males. Two masculinized female S. gigas specimens were also found in this area and included for study. METHODS Specimens for study came primarily from two populations of Strombus pugilis. designated CdP and MIG, that were separated by 1 .5 km. Both inhabited muddy-sand bottoms, at a depth of 10 A Figure 1. Histological section through A) normal and B) masculinized female Strombus pugilis ovaries (xlOO); F— follicles, Sg— signet cells. 72 Reed Figure 2. Histological section tlirougli A) normal and B) masculinized Strombus gigas ovaries ( x 100); F — follicles, Sg — signet cells (signet cells in B) appear empty and partially collapsed as spawning )iad just been completed at time of capture. meters, with mixed patches of algae such as Hatimeda. Ulva. and Penicillis. Strombus pugilis were also examined from a third pop- ulation from which conch were taken as part of another study (Sanders 1988); this population inhabited a mangrove bay with similar bottom characteristics (maximum depth 4 meters). Samples of conch were collected haphazardly from both pop- ulations and brought back to the laboratory for processing. They were kept in a free-flowing holding tank for 1-7 days, as needed. All animals were sexed by observing the foot when extruded from the shell. Intersexes, which possessed both an egg groove and small verge, were separated from the rest of the sample; excess males and females not required for study were returned to their respective field sites. Two specimens of Strombus gigas were found that possessed both an egg groove and verge. Also, included for study were two adult 5. pugilis from the CdP population that lacked secondary sexual development altogether. All specimens were dissected to examine internal anatomy be- fore processmg. Gonads were excised from both Strombus gigas specimens and ten S. pugilis intersexes taken from the MIG pop- ulation as well as one asexual specimen, and fixed in Davidson's solution prior to histological processing. Normal males and fe- males of both species (six of each sex) were included for compar- ison. Tissues were dehydrated in 95% ethanol and embedded in paraffin. Serial sections were cut 6-10 p-m in thickness and mounted on albuminized slides. Staining was with hematoxylin and eosin according to Harris' regressive method in Howard and Smith (1983). Tissues from the reproductive tracts of some S. pugilis intersexes were also processed in the same way except that they were fixed in Bouin's solution prior to dehydration and em- bedding. The slides were then examined using light microscopy to determine sexual condition and state. Several Strombus pugilis intersexes were kept in a holding tank with normal individuals prior to processing. They were observed to copulate with normal males and to spawn. Egg masses were weighed and examined for abnormalities. Several masses were allowed to hatch in aerated aquaria in order to get an idea of the fertility and viability of the spawn. Intersexual individuals were also observed in their natural habitat. RESULTS All intersex conch were normal females in terms of reproduc- tive ability. Reproductive tracts were completely feminine in na- ture except for the development of a dysfunctional verge on the foot where it would normally be found in a male (see Reed, in press a). The verges of such Strombus pugilis females never ap- proached the size of a normal male's (25 mm) and were often deformed by splitting into two or more. These verges ranged in size from nibs of no more than 2 mm to a well-developed 12 mm. There was no interference with egg laying, as the verge developed from the lip of the egg groove, rather than from the interior. All egg masses spawned by such females were normal in appearance and weight (10.05 ± 5.72 g, n = 10, compared to 9.78 ± 5.84 g, n = 10 for normal females), and hatched within five days, with no apparent differences in hatchability between the two groups. Figure 1 shows histological sections of the ovaries of a normal R^'-^ ■ / -I. ' ■ J • SC, :. .| 1 Us ^f ,,»v. ' . ' • ■*■>' ;' * JBk^ IK^'^^^^M *n;v. ■ v,. s . ; " i ■ ' . •*■■ ■4 ^JeS i ■^, ^^;- ■•: w i'. a^^ -■ A ■ i-r '■^^■: ^\ , ^^ . -V. ,^ ^^ Figure 3. Histological section through A) sexually undeveloped and B) sexually developed male Strombus pugilis testes; Us — undeveloped sexual tissue, Fs — mature sperm free in lumen, Sg — signet cells. Gonadal Comparisons in Strombus 73 Figure 4. Histological section through the verges of A) masculinized female Strombus pugilis and B) male S. pugilis for comparison (xlOO); CE — ciliated epithelium, GC— goblet cell, M— muscle. 74 Reed female (A) and a masculinized female (B) Strombiis pugilis . There are no differences in the sexual or connective tissues. Both females were reproductively active at the time of capture. Comparison of the gonads (Fig. 2) of a normal female Strom- bus gigas (A) to that of a masculinized specimen (B) revealed no differences in the sexual tissue. Note, however, that the signet cells of the masculinized female appear empty and partially col- lapsed, indicating that she had just completed spawning at the time of capture (she was found within 30 cm of a freshly-laid egg mass). Her verge was split into two at the base, and both measured 35 mm (normal male, 60 mm). Both verge tips resembled those of a normal mature male in appearance and structure, except for size (see Reed in press b), as did that of the other specimen. Figure 3 shows histological sections through the gonads of the sexually undeveloped specimen (A) and a normal male (B) Strom- bus pugilis for comparison. The undeveloped specimen is a male, with typical testicular tissue (Egan 1985), but there is no sper- matogenesis taking place. No feminine sexual tissue was found in either the gonads or other reproductive organs. Internal examina- tion revealed the presence of an undeveloped prostate gland. One other asexual specimen found was kept in an aquarium for a year. Eventually, minor verge growth did begin, indicating that both these specimens were sexually retarded males. Histologically, the verge of a masculinized female is no dif- ferent from that of a normal, mature male (Fig. 4), except for underdevelopment. Goblet cells are not visible in the section shown for A), but all cell types can be found in both masculinized female and male verges. Other reproductive glands were indistin- guishable from those of a normal female (see Reed in press c) and are not reproduced here. Thus, these females are only masculin- ized, rather than true intersexes. Masculinized females were found in the mangrove bay population but were not further studied. The incidence of masculinization was 2.3% in the CdP colony and 13% m the MIG colony (see Table 1 ). The sex ratio was found to differ significantly (p < 0.05) from 1:1 in both populations when masculinized females were excluded from the calculations, showing a deficit of males in both populations. No masculinized female Strombus costatus have been found as yet, although over 5(X) individuals have been examined. Neither have other aberrant types such as the three-pronged male of Botero (1984) been found in any of the species sampled. DISCUSSION Anatomically and histologically, all masculinized females ex- amined were reproductively normal females. None of their verges ever approached functional size, and were often deformed. This small verge development is similar to that of immature Strombus gigas, S. costatus. and 5. pugilis males (see Appeldoom 1988). None of these individuals were ever observed acting in any way as a male, such as attempting to copulate with another female. These TABLE 1. Sex ratios and incidence of masculinized females in the two populations of Strombus pugilis. Population Males Normal Females Masculinized Females MIG CdP 254(31%) 542 (44%) 416(56%) 660(54%) 104(13%) 29 (2%) females spawned normally and their condition did not in any way interfere with reproductive activities. The correct term for this phenomenon in female conch is masculinization. The incidence of masculinization was not confined to just one population, but was found in three separate colonies that have spatial and geographic barriers between them. Chemical mutagens have been found that cause gastropods to develop both male and female reproductive tissues (Bryan et al. 1988, Gibbs et al. 1988) when present in the water; however, in the case of these Strombus pugilis colonies, there is no likely source of chemical contami- nants in the area, and no way to determine if exposure possibly occurred during the planktonic or juvenile stage. Masculinization has been found in female mosquitofish (How- ell et al. 1980) subject to chemical effluents; however, males were affected as well in that they exhibited precocious sexual develop- ment. It is not known whether males of the Strombus pugilis populations studied developed precociously or not, as these pop- ulations consist entirely of mature individuals. Masculinization has been recorded in other normally gonochoristic species as well (e.g. brachiopods; Culter and Simon 1987) with no apparent cause. In 5. pugilis. at least, this condition is limited to morpho- logical development that does not interfere with normal behavior or reproduction, but does appear to affect growth as maculinized females were larger than normal females which, in turn, were larger than males (see Reed in press a). The latter indicates that some genetic factor is involved, such as non-dysjunction of the sex chromosomes which leads to a variety of aberrant sexual condi- tions and morphological differences (Ayala and Kiger 1984). Pre- liminary electrophoretic observations (Reed and Juste, unpubl. data, cited in Reed 1992) revealed differences in enzymatic ex- pression between males, and normal and masculinized females, further supporting the hypothesis of a genetic basis for sex deter- mination in conch, abberations of which can lead to masculiniza- tion and sexual retardation. ACKNOWLEDGMENTS I would like to thank A. Roman, CODREMAR, for the prep- aration of the gonad sections; V. Juste for her help in preparation of the other tissues; and R. S. Appeldoom for review of this manu- script. LITERATURE CITED Abbott, R, T. 1960. The genus Strombus in the Indo-Pacific. Indo-Pac. Mollusca 1(2): 33- 144, Appeldoom. R. S. 1988. Age determination, growth, mortalily and age of first reproduction in adult queen conch. Strombus gigas L.. off Puerto Rico. Fisheries Res. 6:363-378. Ayala, F. J. & J. A. Kiger. 1984. Modem Genetics. 2nd ed The Ben- jamin/Cummings Publishing Co.. Inc.. Menlo Park. California. 923 pp. Botero. L 1984. Observaciones sobre una poblacidn de Strombus nigas L. en la Ensenada de Nenguange. Caribe Colombiano. An. Inst. Inv. Mar. Puma de Betin 14:47-66. Bryan. G. W,. P. E. Gibbs & G, R. Burt. 1988. A conipanson of the effectiveness of tn-N-butyltin chlonde and five other organotin com- pounds in promoting the development of imposex in the dog-whelk. Nucella lapillus. J Mar. Biol. Ass. U.K. 68:733-744. Culter. J. K. & J. L. Simon. 1987. Sex ratios and the occurrence of Gonadal Comparisons in Strombus 75 hemiaphrodites in the inarticulate brachiopod, Glolliilia pyramidata (Slinipson) in Tampa Bay, Flonda. Bull. Mar. Sci. 40:193-197. Egan. B. D. 1985. Aspects of the reproductive biology of Strimihu.t ^iga.s. M. Sc. thesis, Univ. of British Columbia, Vancouver. Canada, 147 pp. Gibbs, P. E., P. L. Pascoe & G. R. Burt. 1988. Sex change in the female dog-whelk, Niuelta lapillus. induced by tnbulyltin from antifouling paints. J. Mar. Biol. Ass. U.K. 68:715-731. Howard, D. W. & C. S. Smith. 1983. Histological techniques for manne bivalve mollusks. NOAA Tech. Memo. NMFS-F/NEC-25, 64 pp. Howell, W. M.. D. A. Black & S. A. Bortone. 1980. Abnormal expres- sion of secondary sex characters in a population of mosquitofish, Gam- husia uffinis holhrooki: evidence for environmentally-induced mascu- linization. Copeia 4:676-681. Kuwamura,T., R. Fukao, M. Nishida, K. Wada & Y. Yanagisawa. 1983, Reproductive biology of the gastropod Strombus luhuanus (Strom- bidae). Publ. Selo Mar. Biol. Lab. 28:433^43. Reed, S. E. 1992. Reproductive anatomy, biology and behavior of the genus Strombus in the Caribbean with emphasis on Strombus pugilis. Ph.D. Thesis, University of Puerto Rico, Mayagiiez, PR., 149 pp. Reed, S. E. In press a. Masculinized females in the genus Strombus: aspects of their biology and possible advantages for mariculture. In "The Biology, Fishenes, Mariculture and Management of the Queen Conch" (R. S. Appeldoom, B Rodriguez Q,, eds.), Fundacion Cien- tifica Los Roques. Caracas, Venezuela. Reed, S. E. In press b. Reproduction anatomy and biology of the genus Strombus in the Canbbean: I. Males. Proceedings of the 44th Annual Meeting of the Gulf and Caribbean Fisheries Institute, Nassau, Baha- mas, Nov. 3-8, 1991. Reed. S. E. In press c. Reproductive anatomy and biology of the genus Strombus in the Canbbean: 11. Females. Proceedings of the 44th An- nual Meeting of the Gulf and Canbbean Fishenes Institute, Nassau, Bahamas, Nov. 3-8, 1991. Ropes, J. W. 1982. Hermaphroditism, sexuality and sex ratio in the surf clam, Spisula solidissima. and the soft-shell clam, Mya arenaria. Nau- tilus 96:\4\^\46. Sanders, I. M. 1988. Energy relations in a population of Strombus pugilis. Ph.D. Thesis, Univ of Puerto Rico. Mayaguez, PR., 130 pp. Journal of Shellfish Research. Vol. 12, No. 1. 77-79. 1993. SIZE DIFFERENCES BETWEEN SEXES (INCLUDING MASCULINIZED FEMALES) IN STROMBUS PUGILIS (MESOGASTROPODA: STROMBIDAE) SHAWNA E. REED* Department of Marine Sciences University of Puerto Rico P.O. Box 908 Lajas. Puerto Rico 00667 ABSTRACT The discovery of masculinized female Slrombus pugilis provided a third morph which allowed sexual dimorphism in size to be studied in more detail. Masculinized females were found to be larger in overall size than normal females which were in turn larger than males. The difference in size among morphs was not due to differences in shape but only to growth to a larger average size. Lip thickness did not differ among the morphs indicating that masculinized females grow to a larger size than normal females which grow larger than males in the same amount of time. The apparent cause is a genetically-based sex determination system, operating in this species. KEY WORDS: Slromhiis. dimorphism, masculinization INTRODUCTION There appears to be sexual dimorphism in size for most strom- bid species (Abbott 1960). Two Indo-Pacific species studied showed females to be longer on average than males (Abbott 1949). For Strombus gigas. females are known to be larger than males based on shell length frequencies, and may also be broader (Ran- dall 1964. Alcolado 1976. Blakesley 1977). Robertson (1959) mentions that females are larger than males in S. costatus and S. raninus but presents no data. Strombus pugilis and S. alatus also apparently show size dimorphism (Colton 1905. Goodrich 1944). No studies have been done that examine whether size differences between males and females are based on differences in shape, age at maturation, or rates of shell and tissue production, and whether sexual dimorphism in general is controlled by environmental and/ or genetic factors. Conch are characterized by the formation of a flaring lip on the outer edge of the last body whorl of the shell at maturity. Growth •Present Address: % 24265 — 60th Ave. 6H4. Langley. B.C.. Canada V3A in length ceases at this point, and may only decrease, due to abrasion, but never increase. Full sexual matunty is not attained until after lip formation. The conch continues to add shell material to the lip such that it becomes quite thick as the conch ages. Masculinized female strombids. those that possess both an egg groove and a small verge, have been found (Kuwamura et al, 1983. Reed, in press, R. S. Appeldoom. f)ers. comm.). Reed (1992) found several colonies of Slrombus pugilis which had large numbers of such females. Masculinized females offer the oppor- tunity to look at sexual dimorphism in stombids in more detail. No companson of masculinized females has ever been done on the basis of size due to a paucity of specimens. The presence of a third sexual morph may allow hypotheses to be developed that explain not only the proximal causes of sexual dimorphism, but possibly their underlying causes as well. This study was undertaken to characterize sex-based size dif- ferences in Strombus pugilis. METHODS Two populations of Strombus pugilis. designated CdP and MIG for identification, were studied off the southwest coast of Puerto Measurement TABLE 1. Means and standard deviations for the MIG population of Strombus pugilis (prob. level: 0.05). Males (n) Normal Females (n) Masculinized Females (n) LENGTH (mm) LIP (mm) WIDTH (mm) SPIRE (mm) SHELL wt. (g) ANIMAL wt. (g) TOTAL wt. (g) WIDTH/LENGTH ANIMAL/SHELL 79.5-' ± 4.1 (45) 5.1 ± 0.7" (46) 36.4' ±3,1 (51) 26,2' ± 1.5' (47) 76.6' ±11.5 (50) 12.7' ± 2,2 (50) 89.5' ± 13.0(51) 0.46 ± 0.03 (45) 0.16 ± 0.02 (50) 82.8'" ± 4.2 (59) 4.7 ± 0,5'" (53) 38,0'^ ± 2.5 (61) 28.0' ± 2.1' (59) 86.2'" ±11.9 (62) 14.3'" ± 2.3 (62) 00,4'" ± 13,4 (62) 0,46 ± 0.03 (59) 0.16 ± 0.02 (62) 87.6" ± 2.6 (79) 4.8 ± 0.7" (67) 39.6" ± 2.4 (79) 28.5 ± 2.2 (80) 93.9" ± 10.7 (7) 17,4" ± 2.2 (7) 111.3" ± 11.5(7) 0,46 ± 0.03 (79) 0.16 ± 0.02 (7) ' Significant difference between males and normal females. " Significant difference between normal and masculinized females. 77 78 Reed Measurement TABLE 2. Means and standard deviations for the CdP population of Strombus pugilis Iprob. level: 0.05). Males (n) Normal Females (n) Masculinized Females (n) LENGTH (mm) LIP (mm) TOTAL wt. (g) 79.0 ± 5.0(10) 5.6 ± 0.7 (9) 76. 4'' ± 7.9 (9) 78.8" H- 3.2 (12) 5.3 -t- 0.6 (10) 86. 0-"" ± 10.2 (12) 86.3" ± 4.5 (28) 4.9 ± 0.6 (23) 100.8" ±11.1 (28) * Significant difference between males and normal females. " Significant difference between normal and masculinized females. Rico in the vicinity of La Parguera. Samples of conch were col- lected haphazardly from both populations and brought back to the laboratory for processing. They were kept in a free-flowing hold- ing tank for 1-7 days, as needed. All animals were sexed by observing the foot when extruded from the shell. Masculinized females were separated from the rest of the sample; excess males and females not required for study were returned to their respective field sites. The MIG population was chosen for size comparison due to the disappearance of the CdP colony, most likely due to fishing, be- fore sufficient samples were collected for analyses (only a limited amount of data is presented). Measurements taken were LENGTH, length of the shell from siphonal canal to tip of spire (mm), LIP, thickness of lip midway along shell (mm), WIDTH, width of the body whorl (mm), SPIRE, height of the spire (mm), TOTWT, total weight of live animal and shell (g). and ANIMWT, weight of animal after removal from shell (g). SHELLWT, weight of shell only (g), was determined by subtraction of ANIMWT from TOTWT. Some measurements were not possible due to shell damage, especially in cases of broken lips and spires. Data were processed using the SYSTAT package on an IBM PC. Analyses of variance were used to test for differences among the three morphs. Tukey's HSD multiple comparison test was used to distinguish which morphs differed. In cases where variances were not homogeneous among morphs, separate variance t-tests were used to test for differences between means in a pair-wise fashion. For all statistical analyses, the significance level chosen was 0.05. RESULTS All means and standard deviations are presented in Table I for the MIG population. Data collected on CdP individuals prior to their disappearance are presented in Table 2. Comparisons be- tween populations and sexes within population should be inter- preted with reservation for those cases where sample size was small. In the MIG population, males were smaller than females, which were smaller than masculinized females. Males were sig- nificantly shorter in LENGTH and SPIRE, and lighter in all weights (TOTWT, ANIMWT. SHELLWT) than females, which were in turn shorter in LENGTH and SPIRE, and lighter in all weights than masculinized females. In the CdP population, mas- culinized females exceeded normal females in LENGTH and TO- TAL weight. LIP was not significantly different among the sexes within the MIG colony or within the CdP colony, but was between the two populations. CdP conch had thicker lips than MIG conch indicat- ing that the colonies were different in overall collective age. The CdP colony was comprised of smaller animals than the MIG col- ony. Ratios of shell width to length and animal weight to shell weight were not significantly different among sexes in the MIG colony. DISCUSSION Size dimorphism is apparent between Strombus pugilis sexes, as has been noted in several other strombid species (Abbott 1949. I960). The longer overall shell length, also reflected in spire height, implies that females grow faster than males (Abbott I960, Alcolado 1976, Webber 1977). Females are also heavier in tissue weight indicating they grow to a larger body size than males. Masculinized females were found to be larger than normal females in all respects, even though they are functional females and are characterized only by small verge development, much like that of immature males (see Appeldoom 1988), but with severe deforma- tion (verge development does not account for differences in weight as its weight is negligible). Lip thickness does not vary significantly among males, fe- males, and masculinized females, indicating they are of the same age class and probably derived from the same stock. Extra shell weight, thus, cannot be accounted for by excess shell deposition with age, but rather by a larger shell in general. Females were found to be broader than males, as Colton (1905) found in his study. However, the ratios of shell width to length and animal weight to shell weight were not significantly different, indicating consistency in shape among the three groups. Consistent differ- ences in length, width, and weight, among the three groups, cou- pled with similarities in shape and age. indicate that sexual dimor- phism arises from differential rates of productivity. The most probable explanation for this gradation in size be- tween the sexes is genetic. A heterotic effect can account for masculinized females growing faster than males. This phenome- non has long been exploited in agricultural breeding programs to produce animals and plants that grow faster and larger in a shorter time period than others (Mitton and Grant 1984). In such a case, masculinized females would have to be more heterozygous than normal females, which would be more heterozygous than males, and could best be explained if sex-determining chromosomes are present. Heterosis in the female would result in faster growth to a larger size, and the presence of an extra dose of genetic material in masculinized females could cause further heterotic effects. Pre- liminary electrophoretic observations (Reed and Juste, unpubl. data, cited in Reed 1992) indicate that masculinized females do have extra genetic material, not possessed by normal females. ACKNOWLEDGMENTS I would like to thank Dr. R. S. Appeldoom for his review of this manuscript. Sexual Dimorphism in Strombus 79 LITERATURE CITED Abbott. R. T. 1949. Sexual dimoq^hism in Indo-Pacific Strombus. Nau- tilus bH2):5»-6\ . Abbott. R. T. 1960. The genus Siromhus in the Indo-Pacitle Indit-Pac. Mollusca 1(2):33-144. Alcolado. P. M. 1976. Creclmiento. vanaciones moil'ologicas de la con- cha y algunos datos biologicos del cobo Strombus gigas L. (MoUusca. Mesogastropoda). Acad. Cienc. Cuba Ser. Oceanol., No. 34, 36 pp. Appeldoom. R. S. 1988. Age deteimination. growth, mortality and age of first reproduction in adult queen conch. Strombus gigas L., off Puerto Rico. Fish. Res. 6:363-378. Blakesley. H. L. 1977. A contnbution to the tlshene.s and biology of the queenconch. Strombus gigasL. inBehze. (Abstr.l 107"'Annu. Meet. Am. Fish. Soc, Sept. 15-17. 1977, Vancouver. Canada, p, 12. Colton, H. S. 1905. Sexual dimorphism in Strombus pugilus. Nautilus I8(12):138-140. Goodrich, C. 1944. Variations in Strombus pugilis aUitus. Occas. Pap. Mus. Zool. Univ. Mich., No. 490, 10 pp. Mitton. J. B. & M. C. Grant. 1984. Associations among protein heterozy- gosity, growth rate, and developmental homeostasis. Ann. Rev. Ecol. Syst. 15:479-499. Randall, J. E. 1964. Contributions to the biology of the queen conch, Stombus gigas. Bull. Mar. Sci. GulfCaribb. 14{2):246-295. Reed, S. E. 1992. Reproductive anatomy, biology and behavior of the genus Strombus in the Caribbean with emphasis on Strombus pugilis. Ph.D. Thesis. University of Puerto Rico. Mayaguez. P.R.. 149 pp. Reed. S. E. In press. Masculinized females in the genus Strombus: aspects of their biology and possible advantages for manculture. In "'Queen Conch Biology. Fisheries and Mariculture" (R. S. Appeldoom and B, Rodriguez Q.. eds.). Fundacion Cientitlca Los Roques. Caracas. Ven- ezuela. Robertson. R. 1959. Observations on the spawn and veligers of conchs {Strombus) in the Bahamas. Proc. Malacol. Soc. Lond. 33(4): 164- 171. Webber. H. H, 1977. Gastropoda: Prosobranchia. Chap. 1 In "MoUusca: Gastropods and Cephalopods" (Reproduction of Marine Invertebrates. vol. IV; A. C. Giese and J, S. Pearse. eds.). Academic Press. New York, New York, pp. 1-97. Journal of Shellfish Research. Vol. 12. No 1 . 81-87. 1993. SPATIAL STRUCTURE OF THE PINK SHRIMP PANDALUS BOREALIS KR0YER, 1838 FROM THE FAR-EASTERN SEAS AS PROVED BY METHODS OF POPULATION GENETICS AND MORPHOMETRICS Y. P. KARTAVTSEV,t K. A. ZGUROVSKY,* AND Z. M. FEDINA* "t Institute of Marine Biology Far East Division of Russian Academy of Sciences Vladivostok 690041, Russia *Pacific Research Institute of Fisheries and Oceanography Vladivostok 690600. Russia ABSTRACT Polymorphic allozyme loci GPI, PGM. MDH. FDH and 1 1 morphological traits were examined in 12 samples of the pink shnmp from the Sea of Japan, the Okhotsk Sea. and the Bering Sea to determine the specific population structure. Data obtained suggest within-sea-basin genetic homogeneity and, vice versa, statistically significant heterogeneity among shrimp samples from different seas. Three major clusters based on allele frequency data each representing a different sea were seen on a dendrogram. The discriminant and factor analyses used for morphological classification of shrimp individuals and populations support the results of genetic investigation. It is assumed that every sea in general is inhabited by genetically homogeneous local shrimp population, which in this case is an equivalent of Mendelian population. Differences in shnmp morphology within the seas give us an opportunity to suppose an existence of subpopulation structure on this level as well. KEY WORDS: pink shnmp. Parulalus. genetics, population structure INTRODUCTION The pink shrimp Pandalus borealis Kroyer is an important commercial fisheries in many countries of the Pacific and Atlantic basins. This has stimulated detailed studies of its distribution, population recruitment, growth rates and other features of the life cycle (Butler 1964, Ivanov 1972, Balsiger 1979, Shumway et al. 1985). Population structure of this species still remains insuffi- ciently studied. Today the knowledge of population structure is considered to be a key for organizing a rational fishery and un- derstanding the plasticity of stock reaction against fishing. Data on genetic composition of the pink shrimp cohorts from the Bering Sea, the Sea of Japan and the Barents Sea were presented but largely in Russian literature (Kartavtsev et al. 1991a). Morpho- metric investigations of the pink shrimp population structure were performed mostly in the Barents Sea (Kuznetsov 1964, Briazgin 1970, Briazgin, Rusanova 1974, Berenboim 1978, Teigsmark 1983) and without application of modem multivariate statistical approaches. Here we summarize earlier published genetic data on the Far Eastern seas (Kartavtsev et al. 1991a) in combination with mor- phometric data using multivariate statistical analysis. Such analy- sis was presented in oral form but only an abstract has been pub- lished (Kartavtsev et al., 1990). MATERIALS AND METHODS Shrimp samples were taken in 1987 and 1988 from the catches of off-bottom trawling in the Sea of Japan (JS), the Okhotsk Sea (OS) and the Bering Sea (BS). The exact geographic coordinates of samples were given earlier (Kartavtsev et al. 1991a). The dis- tribution of the 12 samples is shown in Fig. 1. We managed to perform individual genotyping in only one (OSl) of the three OS samples. Electrophoretic studies of enzymes were conducted in starch gel (14-15%). Tris-EDTA-boric buffer, pH 8.1 (Korochkin 1977) and tris-EDTA-maleate buffer, pH 7.4 (Shaw, Prasad 1970) were used. Four enzymes from over 50 screened appear to be polymor- phic and were included in the analysis; 1) Glucose phosphate isomerase (GPI, EC 5.3.1.9, abbreviation of locus is the same as enzyme), 2) phosphoglucomutase (PGM, EC 2.7.5.1), 3) malate dehydrogenase (MDH, EC 1.1.1.37), 4) formaldehyde dehydro- genase ( FDH , EC 1.2.1.1). More detailed information on the elec- trophoresis and staining are given earlier (Kartavtsev et al. 1991a,b). For each individual 1 1 traits of external body morphology were measured ( ± 1 mm): 1 ) carapace length (CL), 2) body length (BL), 3) carapace width (CW), 4) width of pleura of the second abdom- inant segment (S2W), 5) length of the left scaphocerite (LSL), 6) length of the right scaphocerite (RSL), 7) telson length (TL), 8) length of the left exopodite of the uropode (LUEL), 9) length of the right exopodite of the uropode (RUEL), 10) length of the left endopodite of the uropode (LUENL), 11) length of the right en- dopodite of the uropode (RUENL). Ten indices-ratios were in- cluded in the analyses as well. To continue the numbering we indicated them in the following way: 12) CL/BL, 13) LSL/BL, 14) RSL/BL, 15) S,W/BL, 16) CW/BL. 17) TL/BL, 18) LUEL/BL, 19) RUEL/BL, 20) LUENL/BL, 21) RUENL/BL. When performing morphological analysis the following prin- ciples were taken into account: 1. Traits complex should represent different morpho- functional structures. 2. All individuals should be characterized with identical set of traits. 3. Normalized traits, i.e. divided by BL, should minimize size variability of individuals (allometry) and has their own meaning. 4. Estimates of differences should be based on biologically homogeneous material, i.e. considered sex dimorphism, and age variability. Statistical analysis was performed mainly using BMDP soft- ware (Dixon 1982), which permit all necessary transformations. 81 TABLE 1. Allele frequencies at the GPI. PGM. MDH, FDH loci in samples of the pink shrimp Pandalus borealis and x'-values of goodness of fit of the observed and expected genotype frequencies. Figure 1. Map showing location of the pink shrimp Pandalus borealis samples in the Far Eastern seas. JSl-JSS = samples from the Sea of Japan, OSI-OS3 = samples from the Okhotsk Sea, BS1-BS4 = sam- ples from the Bering Sea. standardizations and normalizations. Some other details of mor- phological analysis are given elsewhere (Kartavtsev et al. 1993). RESULTS AND DISCUSSION Genetic data presented below in a schematic form to outline main concept concerning the population genetic variability in the pink shrimp cohorts. More thorough genetic examination are given elsewhere (Kartavtsev et al. 1991a. b). For each of the four studied loci, all samples showed a close agreement between observed and expected Hardy-Weinberg equi- librium frequencies of genotypes the x" did not exceed the critical levels (Table 1). At the studied loci allele frequencies within any sea were rather similar but they greatly differ between the sea basins. Examination of allele frequencies confirm this suggestion (Fig. 2). Taking into account the observed variability of allele frequen- cies in the area, we can assume that shrimp samples collected from the same sea basin are genetically homogeneous (Fig. 2). These results are in accordance with the above mentioned Hardy- Weinberg equilibrium in the individual shrimp samples (Table I). Moreover, we can speak about the equilibrium between gametic (allele) and genotype frequencies in total shrimp samples from N Allele Frequency Sample Pl p2 p3 X^ Locus GPI JSI 82 0.079 0.921 — 0.57 JS2 113 0.049 0.951 — 0.32 JS2' 70 0.050 0.936 0.014 0.69 JS3 117 0.064 0.927 0.009 + JS4 26 0.077 0.923 — * JS5 49 0.071 0.929 — 0.24 SJS 457 0.063 0.934 0.003 4.22 OSl 42 0.012 0.988 — 0.00 BSl 51 — 1.000 — — BS2 88 — 1.000 — — BS3 55 — 1.000 — — BS4 96 — 0.990 0.010 0.00 SBS 290 7 0.996 0.004 1.47 Locus PGM JSI 84 0.047 0.292 0.661 1.61 JS2 113 0.014 0.265 0.721 2.25 JS2' 70 0.014 0.257 0.727 2.27 JS3 116 0.018 0.284 0.698 1.84 JS4 26 — 0.346 0.654 2.71 JS5 43 — 0.244 0.756 1.36 SJS 452 0,018 0-276 0.706 7.44 OSl 15 — — 1.000 — BSl 51 0.019 0-069 0.912 0.38 BS2 87 0.046 0.040 0.914 * BS3 55 — 0.055 0.945 3.72 BS4 95 0.032 0.052 0.916 0.82 SBS 288 0.028 Locus 0.052 MDH 0.920 * JSI 65 0.815 0-185 — 2.20 JS2 113 0.889 0.1 11 — 0.34 JS2' 70 0.871 0.129 — 1.56 JS3 117 0.897 0.103 — 3.34 JS4 26 0.885 0,115 — 0.40 JS5 49 0.847 0.153 — 0.02 SJS 440 0.873 0.127 — 1.53 OSl 27 0.426 0.574 — 0.75 BSl 15 0.598 0.402 — 1.68 BS2 88 0.534 0,466 — 0.22 BS3 55 0.655 0,345 — 0.06 BS4 96 0.552 0,448 — 3.11 IBS 290 0.574 0,426 — 1.68 Locus FDH JSI 26 0.923 0.019 0.058 0.23 JS2 78 0.820 0.052 0.128 3.68 JS2' 70 0.836 0.079 0.085 6.69 JS3 108 0.875 0-051 0,074 2.62 JS4 25 0.920 0.040 0.040 0,15 iJS 307 0.860 0.057 0.083 5.58 BSl 47 1.000 — — — BS2 85 0.988 — 0.012 0.01 BS3 54 1,000 — — — BS4 92 0.984 0.005 0.011 0.01 SBS 278 0.991 0.002 0.007 0.01 Note: Asterisk means that x"-values were not calculated because of insuf- ficient digital filling of some cells in the frequency table; N = number of studied animals; JSl-JSS = shrimp samples from the Sea of Japan; OSl = the sample from the Okhotsk Sea; BS1-BS4 = samples from the Bering Sea; iJS and iBS = total samples for the corresponding seas. Population Structure of Pink Shrimp 83 GPI Pgm Mdh Fdh JS BS OS JS BS OS JS BS OS JS BS '®0(D QOO «€€ • o 2(r)2Q GO «€ i^m 2^30 O© • « • o ■ ''l aO^O ©O • C • o DP^ 4© © • • nP3 5® © « P= 933 P= 997 P=,988 P = .705 P^.920 P=1. D P = ,87I P^.574 P = . 126 p. 860 ( >- 991 vi , " I 8? <;^ = 3 89 >■:,,- ■: ,, - 115 82- ^ 5 xU = 589 X?-w = ■* 72 X-; i - 6.35 x^-. - 3 03 n^ ,, = 199 7?: * x; a = ajoi ■ ' Figure 2. Variability of predominant allele frequencies at the MDH, GPI, PCM and FDH loci in the pink shrimp Pandalus borealis among three Far-Eastern seas. Size of sectors in the circles shows the vari- ability of frequencies of three alleles. The variability of fastest allele (P,) is indicated with black color, the intermediate allele (Pj) with white color, and slowest allele (P,) with shading. Arabic figures show the number of samples. JS, OS, BS = shrimps from the Sea of Japan, the Okhotsk Sea, the Bering Sea. Locations of samples over the area shown in Fig. 1. For the JS samples 2 and 2' — is a large sample split into two sub samples when analyzed for allele and genotype frequen- cies. Below circles are the mean values of predominant allele frequen- cies and x^-test of their heterogeneity (Workman, Niswander 1970) for the different population complexes: JS = x5-6> BS = xJ_4- JS + OS -^ BS = x'-,i. JS + BS = xU. + + = P < 0.001. both the Sea of Japan and the Bering Sea (Table 1). The differ- ences in allele frequencies of samples from the same sea collected in different years are not statistically significant. In the Sea of Japan samples were taken in 1987 and 1988 and studied over all 4 loci (Table 1 . JS5 and JS1-JS4). These results are in good agree- ment with different data set (Kartavtsev et al. 1991b|. Four sam- ples from the Bering Sea taken in its western part in 1988 (p = 0.920 ± 0.01 1 , Kartavtsev et al. 1991a) and the sample taken in 1972 in the offshore waters of Alaska (p = 0.924 ± 0.009; Johnson et al. 1974) did not differ at PGM — the only comparable locus. Data presented above agree well with a postulate that in time stability of allele frequencies is a valuable feature for large and self reproducing populations (Mattler, Gregg 1972). which is one of main statements of Hardy-Weinberg law. Analysis of the whole set of samples reveals the opposite ten- dency in spatial variability of allele frequencies. For each of the four studied loci, a statistically significant heterogeneity of allele frequencies was observed (Fig. 2). The amount of genetic differentiation within and between pop- ulation units show an order or more increase from first to second level in three different scales: D,„ F',„ and D^ (Nei 1987). Av- erages at GPI, PGM. and MDH. (compared for the whole set of samples) were obtained for the three statistics above: JS - D„ = 0.0009, F\, = 0.0031, D„, = 0.0029 ± 0.0037: BS - D„ = O.OOII, F'„ = 0.0046, D„ = 0.0006 ± 0.0032; JS -H OS -t- BS - D„ = 0.0160, F'„ = 0.0506. D„ = 0.0298 ± 0.0165. Relationship between the samples in a scale of genetic dis- tances may be presented in a graphic form (Fig. 3). On the basis of both the intuitive cluster division and an exact approach, which uses the formation of step value clusters (Rao 1980), the main conclusions drawn from the analysis of these dendrogram are iden- tical: 1. There are three main clusters in the dendrograms which include the JS, BS and OS shrimps. 2. Samples from the same sea basin form sufficiently homogeneous clusters (Fig. 3). By analyzing the variation distributions of morphological traits of the pink shrimp samples we found that mean values of over- whelming majority of both the traits and the indices were different for females (F), males (M) and hermaphroditic (FM) individuals. As an example, the differences in body length and the S^W/BL index values are illustrated (Table 2). The integral differences in complexes of traits between females, males and hermaphrodites were also high-valued. This suggests the necessity of separate morphological analysis of three groups F, M, FM but inference that our pooled data on genotypic variation are representative tak- ing into account absence of difference among these groups in allele frequencies (Kartavtsev et al. 1991a, b). For shortness let us con- sider the data on morphological variability and among sample differences in females — the most representative group in our ma- terial. The complexes of traits and indices chosen for comparison of the shrimp populations form 16 correlation clusters under step value r = 0.9; 7 of them belong to the traits complexes and 9 to indices (Fig. 4). As a whole, the differences between the shrimps of three studied seas are statistically significant for both traits and indices complexes (Table 3). The results of discriminant analysis give a clear view on the ratio of intra- and inter-basin morphological differences. The dis- tribution of individuals, taken from three Far Eastern seas, as the values of canonical variables (CV, and CV,, which estimate in the discriminant analysis the main integral information about differ- entiation of the elements classified) is shown in Fig. 5. In that case the analysis included 10 indices and individuals were combined according to their belonging to three seas: JS, OS and BS. Coor- dinates of mean values of CV, and CV, are indicated by circled numbers (Fig. 5). Total discriminative accuracy or the average accuracy of the classification of an individual to its group (sea) was 99.1%. Clusters of JS and BS do not overiap at all at CV, values. The JS and OS clusters are overiapping in the projections of individual values of the CV, axis by 5.7%, and to BS-OS clusters overiapping by 65.6% (overiapping here is a percent of D m .05 .03 .01 .00 ^ 1 8 91011 2 7 6 3 5 4 OS BS JS Figure 3. Dendrogram showing the integral differences in allele fre- quencies at the PGM, GPI, MDH loci between the samples of the pink shrimp Pandalus borealis. Along the axis — minimal unbiased genetic distance (Nei 1978). I = sample from the Okhotsk Sea (OSl), 2-7 = samples from the Sea of Japan (JS5, JSI-JS4), 8-9 = samples from the Bering Sea (BSI-BS4). Dotted line shows the step value D„ = 0.006 defined at the level of mean standard errors values in the pop- ulations of JS and BS (see text). 84 Kartavtsev et al. TABLE 2. An example of comparison of two morphological traits (BL, SjW/BL) in males (Ml, females (F) and hermaphrodites (FM) of the pink shrimp Pandalus borealis. BL SjW/BL Compared Traits M F FM M F FM Mean 93.66 104.75 10111 13.03 15.36 13.79 SD 13.12 11.28 7.21 1.15 1.27 1.16 SE 1.05 0.67 0.99 0.13 0.08 0.16 Sample size 77 286 53 77 286 53 Max 116.0 146.0 122.0 20.0 18.7 18.5 Min 67.0 82.0 79.0 10.4 12.2 11-4 'M-F 7.39 (p < 0.01) 14.61 (p < 0.01) 'M-FM 4.16 (p < 0.01) 8.68 (p < 0.01) •F-FM 3.04 (p < 0.01) 8.40 (p < 0.01) Note: SD = standard deviation; SE = standard error; BL = body length; SiW/BL statistics; Max, Min = maximal and minimal value of traits; * = P < 0.01. ratio of abdominal segment width to body length; / = Student's individuals having the same CV values or laid in the scope of intermediate values of this variable in two compared clusters). Thus, as it was previously shown by allozyme data, shrimps from the Oidiotsk Sea are closer to those from the Bering Sea than to those from the Sea of Japan. Using for sample grouping of individuals by means of the method of canonical variables, belonging as stated above to dis- criminant analysis, we can obtain a spatial distribution of the cor- responding vector values without an a priori assumption of the existence of high hierarchical level, i.e. the sea basin. As a result of the analysis of indices complex we manage to get more detailed 1.0 r 11 7 6 5 12 1613 1418 19 20 21 15 17 0.6 10.2 -0.2 L Figure 4. The correlation clusters of the studied traits and indices of the pink shrimp Pandalus borealis. On the axis the values of the cor- relation coefficient are shown. Designations of traits from 1 to 11 were as follows: 1) CL, 2) BL, 3) CW, 4» S^W, 5) LSL, 6) RSL, 7) TL, 8) LUEL, 9) RUEL, 10) LUENL, 11) RIIENL. Indices were designated in the following way: 12) CL/BL, 13) LSL/BL, 14) RSL/BL, 15) SjW/BL, 16) CW/BL, 17) TL/BL, 18) LUEL/BL, 19) RUEL/BL, 20) LUENL/ BL, 21) RUENL/BL. Explanation of the abbreviations are given in the section Materials and Methods. understanding of features and degree of phenotypic differentiation of shrimps. Under this approach there were formed by a natural way the following clusters: BS88, JS88, OS87 and JS87 (Fig. 6). Indices and traits itself are not mutually complementary because of weak correlation of most of them (see Fig. 4) and make somewhat different contribution to the diagnostics of samples. In particular the samples taken in the offshore waters of the Eastern Sakhalin (ES) and the Western Kamchatka (WK) are clearly discriminated by the traits complex. In view of this we repeated the discriminant analysis according to the above mentioned scheme but combine the traits and indices complexes (Fig. 7). As we see the discrim- inative power of the method increased and OS87 cluster separated into two — ES87 and WK87 (Fig. 7). It is important that when using this approach the accuracy of inter-basin discrimination re- mains the same, and equal on the average 97.2%, while it is impossible to distinguish individuals from different samples of the same sea basin in most cases. However, in contrast to allozymic data showing the gene pool homogeneity of populations from each sea, the morphological traits were not indicative of a full uniformity of shrimps within the same sea. For example, in the above mentioned discriminant anal- ysis the accuracy of the classification of the individuals to the JS4 TABLE 3. Evaluation of the differences between the vectors of mean values as shown by the traits and indices complexes of the pink shrimp Pandalus borealis from three Far Eastern seas. 1. Traits (1-11) Sea Basin JS OS BS JS OS BS 89.59* 290.81* 29.15 62.89* 51.32 21.15 Sea Basin 2. Indices (12-21 JS OS BS JS OS BS 93.77* 238.61* 29.56 67.94* 41.09 22.28 Note: Above the diagonals are D'-Makhalanobis distance, below it Fish- er's F values, * = P < 0.001. Abbreviations of the sea basins as before. Population Structure of Pink Shrimp 85 3 - - B ± B •— B B B B B o B BBBe B BB „ - B B Bb B ©^„%B B B Np, B„B R b1 B BB „ B „ ' ^BsiB^BBB^BB B B^ B ^B B B \ OS JS \, J I I L J L I I I I I I L _L J L I -6.0 -4.8 -3.6 -2.4 -1.2 0.0 1.2 2.4 3.6 6.8 6.0 Figure 5. Female pink shrimp Pandalus borealis distribution at tlie first two canonical variables (CV) for the Sea of Japan (JS), the Okhotsk Sea (OSl, and the Bering Sea (BS). Numbers in circles are mean values of CV, and CV2 coordinates for the three seas. Each letter (B, O, J) corresponds to a classined individual from the given sea. The broken lines show the borders of clusters from these seas. 7.50 - 6.25 - 5.00 3.75 2.50 1.25 0.00 -1.25 -2.50 -3.75 5.00 D B '^ CD° d' >°tD©0|BBA C D BS88 ^B ° °B B@' B C ^ "N' FE H H H G--^ G / H — H "h H -H JS87 JS88 I I I I I I I I I I I L J I I I I I I I L -6.45 -5.25 -3.75 -2.25 -1.50 .750 2.25 3.75 5.25 6.75 8.25 Figure 6. Female pink shrimp Pandalus borealis distribution on two canonical variables (CV) as exemplified by the for-sample, for-individual discriminative procedure based on 10 indices complex. The spread in values of CV for individuals and samples is shown by the corresponding letters: BS88 — A, B. CD,; JS88— E, F, G, H: JS87— I; ES87— K. L. The asterisks indicate an overlap of two or more pairs of values. Other signs and abbreviations are identical to those in Fig. 5. 86 Kartavtsev et al. 3.00 3.75 h 2.50 1.25 0.00 -1.25 -2.50 -3.75 -5.00 -6.25 -7.50 JS88 HH-H-H--" HGhH H H ^H Fh E HH H \i \h-, Y H^V-E^fe EG I H p^HHH H\yFF° / y^ \j JS87 / \,^- I I I I I I I I I I L J I I L J I L_L -7.50 -6.00 -4.50 -3.00 -1.50 0.00 1.50 3.00 4.50 6.00 7.50 9.00 Figure 7. Female pink shrimp Pandalus borealis distribution on two canonical variables (CV) as exemplined by the for-sample, for individual discriminative procedure based on the combine morphological traits and indices complex. The signs and abbreviations are the same as in Fig. 5-6. sample as to "its own" ones reached 87.5%, to the JS2 sample — 73.5%. Taking into account the statistically significant differ- ences of traits complex between some samples from the same sea, it is possible to speak about the intrapopulation differentiation of the pink shrimp. Evidently morphological differentiation of native shrimp cohorts is based on the effect of disruptive local selection at the larval and early juvenile developmental stages, the differences in the growth conditions at biotopes occupied by separate groups and due to the known certain territorial residency of adult individuals. Such morphological and ecological differentiation was also observed for shrimps from the Barents Sea (Bcrenboim 1978. 1982. Teigsniark 1983). The differences in morphological traits of shnmps may also depend on age structure of cohorts (Skuladottir et al. 1978). From the information presented above on variability and sim- ilarity/distance data both for allele frequencies and morphological traits between the samples of the pink shrimp Pandalus borealis from three Far-Eastern seas two main conclusions can be drawn: 1 . Local groups of individuals or cohorts of shrimps from the same sea are genetically homogeneous and, evidently, they are the members of the same Mendelian population. 2. Neighboring seas are inhabited by local populations which are basically different genetically and phenetically. How do these conclusions correspond to the previously known data on population structure of the pink shrimp? As it was men- tioned above, we and other authors investigated population struc- ture of this species in the Barents Sea. It was proposed that in the Barents Sea there is only one superpopulation of the pink shrimp (Berenboim 1982). Our data on genetic composition of shrimp cohorts in this sea as well as in the Bering Sea and in the Sea of Japan (Kartavtsev et al. 1991a. b) correspond very well to this notion. Combined analysis presented here support all major state- ments made above. In general it is possible to claim that large, weakly differentiated populations are a common phenomenon among marine invertebrates species with a long-term planktonic larva, including crustaceans (Hedgecock et al. 1982). Data presented in the paper do not infer that in other popula- tions of the species their structure will be identical. For example in fjords or other regions with restricted gene flow and (or) differ- entiating natural selection more complex division is possible. In any case observed morphological differentiation within the sea basin should be taken into consideration in fishing of the pink shrimp populations. This investigation was partially subsidized by Sozos Founda- tion grant. LITERATURE CITED Balsiger. J. W. 1979. A review of pandalld shrimp fisheries in Northern Hemisphere. In: Proc. Intemat. Shrimp. Symp.. No. 81-3. pp. 7-35. Ed. T. Frady. Kodiak. Univ. Alaska: Sea Grant Rept. Baiter. T. H. 1964 Growth, reproduction and distribution of Pandalid shrimps in the British Columbia J Fish- Res. BiHirdCan. 21(6):1403- 1452. Berenboim, B. 1. 1978. On the population differences in the shrimp Pan- dalus borealis from the Barents Sea. GidrohiologicheskiyZhurnat 14,1: 44-47 (In Russian). Berenboim, B. I. 1982. Reproduction of the populations of the shrimp Pandalus borealis in the Barents Sea. Okeanotogiya 22,1:1 18-124 (In Russian). Bryazgin, V. F. 1970. On the distribution and biology of the shrimp Pan- dalus borealis (Kr. ) in open regions of the Barents Sea. In: the Results of Economical Investigations of Fisheries in the Northern Basin. Mur- mansk, No. 16, part 2. pp. 93-108. Bryazgin, V. F. & M. N. Rusanova. 1974. Distribution patterns and pop- ulation variability of Pandalus borealis Kr. in open regions of North- Population Structure of Pink Shrimp 87 Eastern Atlantic. In; Hydrobiology and Biogeography of the shelf of the cold and temperate waters of the world Ocean. Leningrad, Nauka Pub!., pp. 88-89 (In Russian). Dixon, W. E. (Ed.). 1982 BMDP: Biomedical computer programs. Univ Calif. Press. Los Angeles. 2:283-403. Hedgecock, D., M. Tracey & K. Nelson. 1982. Genetics In; The Biology of Crustacea. Acad. Press, N.Y.. 2;283-403 Ivanov, B. G. 1972. Geographical distnbulion of the northern shrimp Pandatus borealis Kr. (Crustacea. Decapoda). Estimated productivity of the World Ocean. Trudy VNIRO 77(2);93-I09 (In Russian). Johnson, M. S.. F. M. Utter & O. Hodgins. 1974. Electrophoretic com- parison of five species of Pandalid shrimps from the Northern Pacific Ocean. Fish. BulL 72;799-803. Kartavtsev, Y. P.. K. A. Zgurovsky & Z. M. Fedina. 1990. Analysis of the spatial structure of the pink shrimp (Pamktlus borealis) from the Far Eastern seas using methods of population genetics and phenetics. In; Abstr. Sci. Papers a. Posters. Shellfish Life Hist. & Shellfishery Models Symp.. Moncton. Canada. ICES. Palaegade 2-4. DK-1261. Copehagen K, p. 61. Kartavtsev, Y. P., K. A. Zgurovsky & Z. M. Fedina. 1991a. Allozyme variability and differentiation of the pink shrimp Pandalus borealis from the Far-Eastern seas. Genelica 28(2);1 10-122 (In Russian). Kartavtsev. Y. P.. B. I. Berenboim & K. A. Zgurovsky. 1991b. Popula- tion genetic differentiation of the pink shrimp Pandalus borealis from the Barents and Benng seas. J. Shellfish Res. 10(2);333-339. Kartavtsev, Y. P., K. A. Zgurovsky & Z. M. Fedina. 1992. Morpholog- ical variability of the pink shrimp Pandalus borealis from the Far Eastern seas and its relationships with population structure of the spe- cies and allozyme heterozygostiy. Biol. Moria. (In Press, In Russian) Korochkin, L. 1. (Ed.) 1977. Genetics of isozymes. Moscow; Nauka Publ., 257 pp. (In Russian). Kuznetsov. V. V. 1964. Biology of mass and common crustacean species from the Barents and White seas. Nauka Publ.. Moscow-Leningrad, 241 pp. (In Russian). Mettler. L. & T. Gregg. 1972. Population genetics and evolution. Mir Publ,. Moscow, 323 pp. (In Russian). Nei. M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89;583-590. Nei. M 1987. Molecular evolutionary genetics. Columbia Univ. Press. NY.. 512 pp. Rao, C. R. 1980. Cluster analysis as applied to the study of racial inter- mixing in human populations. In: Classification and Clustering, pp. 148-167. Ed. J. Van Ryzin. Mir. Publ., Moscow (Russian Ed.). Shaw. C R. & R. Prasad. 1970. Starch gel electrophoresis of enzymes. A compilation of recipes. Biochem. Genet. 4:292-520. Shumway. S. E., H. C. Perkins. D. F. Schick & A. P. Stikney. 1985. Synopsis of biological data on the pink shrimp. Pandalus borealis Kroyer 1838. FAO Fish. Synopsis 144:1-157. Skuladottir. U.. E. Johnsson & I. Hayrimsson. 1978. Testing for hetero- geneity of Pandalus borealis at Iceland. ICES CM. I978/K78, 8 pp. Teigsmark, G 1983. Populations of the deep-sea shrimp Pandalus bor- ealis Kroyer in the Barents Sea. Fisheridir. skr. Her. Havunders 17; 377^30. Workman, P. Z. & J. D. Niswander. 1970. Population studies of south- western Indian tribes. II. Local genetic differentiation in Papago. Amer. J. Hum. Genet. 1:24-29. Journal of Shellfish Research. Vol. 12. No 1. 89-92, 1993. PREDATION BY THE CRAB, CANCER OREGONENSIS DANA, INSIDE OYSTER TRAYS SYLVIA BEHRENS YAMADA, HEIDI METCALF, AND BART C. BALDWIN Zoology Department. Oregon State University. Cordley Hall. 3014 Corvallis. Oregon 97331-2914. USA ABSTRACT Cancer oregonensis is a predator of sub-market size oysters iCrassoslrea gigas). Crabs enter oyster trays as megalops larvae between May and October, and attain a carapace width (CW| of 30 mm within a year. Despite its small size. Cancer oregonensis has powerful chelae; molar teeth and sharp tips are well adapted for crushing and puncturing oysters. In laboratory experiments the largest C. oregonensis (43 mm CW) was able to open market size oysters larger than 60 mm in length, while even a 20 mm wide crab consumed oysters 30 mm in length. Medium size crabs (20-35 mm CW) consumed an average of one young oyster (20-40 mm in length) per day. A field experiment was set up in which 15 trays, each containing 315 ± 23 seed oysters, received 5. 2 or newly settled C. oregonensis. Ten months later the average survival of oysters in the two crab treatments was 63% and 69% versus 90% for the control treatment. We recommend that crabs be manually removed during sorting operations. KEY WORDS: crab, oyster culture, predation, Cancer oregonensis. Crassosirea gigas INTRODUCTION Crabs have been identified as major predators in shellfish cul- ture on the shores bordering the Atlantic and the Gulf of Mexico (Menzel and Hopkins 1955. Parsons 1974, Walne and Davies 1977. Dare et al. 1983). Quayie ( 1988) recognized five species of Northeastern Pacific crabs as potential predators on the Pacific oyster. Crassosirea gigas (Thunberg): Hemigrapsus niidis (Dana). Hemigrapsus oregonensis (Dana), Cancer magister (Dana). Can- cer productus (Randall) and Cancer gracilis (Dana). We now add Cancer oregonensis (Dana) to this list. While the crabs listed by Quayie (1988) primarily attack newly-planted oysters on the sea- bed, C oregonensis feeds on a wide size range of oysters inside suspended trays. Cancer oregonensis is found in the subtidal and low intertidal zones from the Bering Sea to Santa Barbara. California (Hart 1982). Cancer oregonensis is a small crab, attaining a maximum carapace width (CW) of 45 mm (Morris et al. 1980). Megalops larvae settle in interstitial habitats such as rock crevices, mussel beds, barnacle patches, kelp hold-fasts, bumper tires on floating docks, and oyster trays (Hart 1982. Orensanz and Gallucci 1988. personal observation). While larger Cancer species leave their nursery habitats as adults, C. oregonensis remain in these refuge- rich habitats their entire life (Orensanz and Gallucci 1988). Cancer oregonenesis is an opportunistic forager, feeding on barnacles, snails, bivalves, worms and algae (Knudscn 1964. Behrens Ya- mada. personal observation). Peak settlement of megalops larvae occurs during late spring and early summer (Jamieson and Phillips 1988. Lough 1975). Growth is rapid with some females attaining sexual maturity by the fall, just a few months after settlement (Orensanz and Gallucci 1988). The small size of C. oregonensis megalops (2 mm CW; De- Brosse et al. 1989) allows them to enter oyster trays through the 6 mm holes provided for water circulation. Tray-raised oysters are thinner-shelled, and thus more susceptible to crab predators than intertidally raised oysters (C. Sanford. Innovative Aquaculture Products Ltd., Lasqueti Island, British Columbia, personal com- munication). Of all the crab species that settle inside oyster crabs. C. oregonensis has the most powerful chelae for its size (Lawton and EIner 1985, Behrens el al.. in preparation). With stout molar teeth on the occlusal surfaces and pointed tips, the chelae appear well adapted for crushing and puncturing growing oyster (Figure 1 ). C. oregonensis is common at oyster farms off the west coast of Vancouver Island, in the northern Hood Canal, northern Puget Sound and in the Strait of Georgia where salinity remains high throughout the year. Oyster growers from these areas report pre- dation rates on young oysters exceeding 40% and as high and 90%. The objectives of this study were; 1) To determine the maximum size at which Pacific oysters are vulnerable to C. oregonensis of a given size. 2) To determine the feeding rates of these crabs on oysters in the laboratory. 3) To quantify predation damage of known densities of crabs inside oyster trays. 4) To make recommendations for crab control. MATERIALS AND METHODS /) Critical Size of Oysters Laboratory trials were set up to determine the largest oyster a given size C. oregonensis could crush. Crabs and oysters were obtained from Westcott Bay Sea Farms. San Juan Island. Wash- ington and transported to Oregon State University where they were kept in recirculating sea water at 14°C with 12 hour light;dark cycle. Sixteen crabs of either sex. ranging from 11^3 mm CW, were placed inside individual plastic sandwich boxes (5 x 15 x 15 cm) with mesh sides to allow for water circulation. Four single oysters ranging from 12 to 40 mm length were offered to each crab. Consumed oysters were replaced by slightly larger ones, while non-feeding crabs received oysters of a smaller size range. Containers were monitored three times a week from February 1 1 to March 6. 1991. Feeding trials were repeated with 17 fresh crabs from April 10 to May 28, 1991. All feeding crabs (N = 28) were sexed and the average of the two largest oyster eaten per crab was plotted against crab CW. 89 90 Yamada et al. Figure 1. Right cheliped of Cancer oregonensis (32 mm carapace width). Scale bar = 10 mm. 2) Laboratory Feeding Rates Feeding rates of crabs were determined in water tables with an open sea water system at the University of Washington Friday Harbor Laboratories, and in a re-circulating sea water system at Oregon State University (Table 1). In a preliminary trial, 16 large crabs (30-40 cm CW) of either sex and 60 oysters ranging from 27 to 40 mm length were intro- duced into a sea water table ( 1 50 x 1 50 x 20 cm; water temper- ature = 14°C) at Friday Harbor Laboratories and covered with a sheet of black plastic on August 28, 1990. The number of oysters eaten in the first 9 hours was noted. Fifty more live oysters were then added to the water table and the number of oysters eaten in the subsequent 20 hours was determined. In the next trial, 5 small crabs (19 to 28 mm CW) and 10 small oysters (21-36 mm in length) were introduced into each of 5 large plastic boxes (21x21x9 cm) in a water table (water temperature of 15° C) at Friday Harbor Laboratories. Boxes were checked daily for 5 days and consumed oysters replaced. The subsequent trial was carried out with the same crabs at two locations. At Friday Harbor 32 crabs ranging from 17 to 44 mm CW were placed inside mdividual plastic sandwich boxes contain- ing 4 oysters each. Crabs smaller than 30 mm received oysters ranging from 15 to 40 mm in length, while larger crabs received 30 to 50 mm oysters. Boxes were kept in a water table at 15°C. Feeding boxes were checked every day from June 24 to June 28 1991 and consumed oysters replaced. On June 29 all feeding crabs were transported inside a cooler to Oregon State University. For the next 3 days crabs were fed cracked oysters and allowed to acclimate to the new conditions (water temperature = 16°C). Feeding trials resumed July 3 and continued until July 18. This time the boxes were monitored every second day. Daily feeding rates were determined for each feeding crab. 3) Crab Predation Inside Oyster Trays To assess the predation pressure of C. oregonensis on oysters under natural conditions, we set up an experiment inside oyster trays at Westcott Bay Sea Farms on August 29, 1991. Fifteen Mexican oyster trays (56 x 57 x 7.5 cm) each received 3 liters of seed oysters (mean number per tray = 315; standard deviation = ±23), ranging in length from 28 to 35 mm. Either 5, 2 or juvenile C. oregonensis (10 to 20 mm CW) were added to each tray. Survival of oysters, growth of oysters and crabs, and settlement of juvenile crabs were monitored on October 12, 1991, February 7, and June 22, 1992. An average daily consumption rate per crab was estimated for all 10 crab trays by taking the number of dead oysters (difference between the number of live oysters at the be- ginning and the end of the experiment), subtracting 31 (the aver- age number of dead oysters in a control tray) and dividing by the mean number of crabs in a tray (total of initial number and final number divided by 2) and by 297 d. The arcsine transformation was used on percent oyster survival before performing ANOVA on treatment effect (Sokal and Rohlf 1981). RESULTS /) Critical Size of Oysters The average length of the largest two oysters consumed by crabs of various CW is given in Fig. 2. No sex difference in crushing ability was detected. Since oysters vary in shape, length should not be interpreted as an absolute measure of critical size. Nevertheless, Cancer oregonensis of all sizes are able to crush and feed on oysters longer than their own carapace width. Thus, a crab of 20 mm carapace width can successfully attack oysters 30 mm in length, while the largest crab can open market size oysters (>60 mm). 2) Laboratory Feeding Rates Over short time periods crabs are capable of consuming over 3 oysters (within their critical size range) per day. An average long- TABLE 1. Feeding rates of Cancer oregonensis in laboratory trials at Friday Harbor Labs (FHL) and Oregon State University (OSU). Experim. Crab CW Oyster Length Feeding Rate (#/day/crab) Trial Units # crabs/unit (mm) (mm) Duration Mean (SD) Maximum FHL 14°C 1 water table 16 30^0 27^0 9h 20 h 5.5 2.0 FHL I5°C 5 large boxes 5 19-28 21-36 5d 0.62 (0.25) 1 FHL 15°C 19 boxes 1 17-29 15^0 4d 1.06 (0.76) 4 13 boxes 1 31-44 30-50 4d 0.77 (0.68) 3 OSU 16°C 16 boxes 1 17-29 15-40 15 d 1.16 (0.50) 3 9 boxes 1 31^14 30-50 15 J 0.99 (0.18) 3 For details on experimental design see text. Crabs in Oyster Trays 91 E 70- 60- 50 X I- C3 40 1- ?o- cn >- O 10- » MALES ° FEMALES 1 20 30 40 50 CRAB CARAPACE WIDTH (mm) Figure 2. Average of two largest oysters crushed by crabs of various carapace width. No significant difference was found between male and female crabs. term feeding rate of 1 oyster per day. however, is more realistic in that this rate includes data from molting crabs that ceased feeding for up to 6 d. 3) Crab Predation Inside Oyster Trays Survival of experimental crabs inside oyster trays was 77% over the 297 days of the experiment. Since crabs died at various times throughout the year, no significant regression between num- ber of surviving crabs and number of dead oysters was found. Their average carapace width was 23 mm in October. 26 mm in February and 30 mm in June. At termination of the experiment the oysters had attained an average length of 60 mm. Cancer oregonensis open oysters by progressively chipping away at the shell margin or, more commonly, by puncturing the shell. Small, rapidly-growing oysters with fragile shells and thin lips are particularly vulnerable. The average feeding rate of crabs inside the 10 oysters trays was estimated to be 0.09 (standard deviation = ±0.05) oysters per crab per day. This value is one order of magnitude lower than those observed in laboratory trials. One reason for this discrepancy is that throughout the field experiment, the average oyster was over twice the carapace width of the average crab. Survival of oysters inside control trays was significantly greater than in trays containing an initial 2 or 5 crabs (ANOVA for arcsine transformed percentages: F = 12.939, d.f. = 2, p < 0.001) (Figure 3). Survival in control trays ranged from 88 to 93% while trays with crabs ranged from 37 to 81%. Only 3 newly-settled juveniles entered the \5 trays between August 29 and October 12. while none were recovered in Febru- ary. On June 22, eighteen newly-settled Cancer oregonensis (3-5 mm CW) were recovered. DISCUSSION Cancer oregonensis has the potential to be an important pred- ator inside suspended oyster trays. Since tray-raised oysters are thin-shelled, they do not attain an absolute size refuge from these powerful crab predators as is the case for sea bottom-reared oysters on Atlantic shores. Eggleson (1990) and EIner and Lavoie (1983) found that American oysters (Crassostrea virginica (Gmelin)) < > > 3 t/3 d UJ W >- o 100 90 80 70 60 50 40 30 100 200 300 TIME (days) Figure 3. Mean survival rate (%) of seed oysters inside trays with and without crabs. The range of values are indicated by error bars. larger than 30 mm in length were rarely opened by lobster (Homarus americanus (Milne-Edwards)), rock crab (Cancer ir- roratus (Say)) or blue crab iCallinecles sapidiis (Rathbun)). The largest C. oregonensis. however, can open market size oysters over 60 mm in length, while even 20 mm CW crabs consume 30 mm length oysters. At an average consumption rate of 0. 1 oyster per day. five crab inside a tray could eat 120 oysters in the 8 mo during which the oysters are normally kept in trays. That represents a 40% reduction in oyster survival and profit. The winter and spring of 1992 were unusually mild, with oysters growing and surviving well. In years when oysters grow more slowly, crabs would gain a size advan- tage over the oysters and could cause more devastatmg effects than we measured. Cancer oregonensis megalops larvae with a carapace width of 2 mm (Lough 1975) can easily enter the 6 mm diameter holes of Mexican oyster trays. Once inside, the larvae metamorphose into first stage crabs of 3 mm CW (Orensanz and Gallucci 1988, and personal observation). Oyster trays are ideal crab habitats, with abundant food and protection from predators such as the octopus. In addition to oysters, growing crabs can feed on fouling organ- isms such as algae, sponges, tunieates, sea cucumbers, gunnels, barnacles and mussels. Settlement of C. oregonensis megalops larvae off the west coast of Vancouver Island occurs from April to August, with a peak abundance in late June (Jamieson and Phillips 1988). Our observations suggest that the settlement peak in 1991 occurred during late June, but that in June 1992 it occurred two weeks earlier. In our experimental oyster trays, we observed some newly settled C. oregonensis between late August and early October, none between October and February, and a moderate settlement during June 1992. While Lough (1975) reports some megalops larvae in Oregon plankton samples during the winter, the chance of a commercially important settlement to occur from October to April appears low. Recommendations for Crab Control Since C. oregonensis attack oysters larger than their own car- apace width, and since newly-settled crabs become oyster preda- tors within 3 months, crabs of all sizes should be removed from oyster trays. Growers at Skerry Bay on Lasqueti Island, use a freshwater bath to rid their oyster trays of sea stars and crab pred- 92 Yamada et al. ators (C. Sanford, personal communication). Cancer crabs, are osmoconformers (Dehnel and Carefoot 1965) and are thus intolerant of low salinities. Smaller C. oregonensis, with higher surface area to volume ratios, would be especially susceptible to fresh water baths. Growers at Westcott Bay Sea Farms manually remove larger crabs (>20 mm in CW) from their trays when oysters are sorted. They report an increase in the survivorship of young oysters since this predator control measure was started four years ago (B. Peo- ples, personal communication). Crab predation, however, remains a problem during the winter when oysters grow more slowly and trays are not checked as frequently. Since crabs continue to feed during the winter, a special effort should be made to rid trays of all crabs during the last oyster sorting operation in the fall. ACKNOWLEDGMENTS We thank Buz Peoples, Westcott Bay Sea Farms, for providing the inspu-ation for this study and for enthusiastically helping us set up and monitor our experiments. We also thank O.S.U. Zoology Club members, Jennifer Yamada, and Kyle Yamada for helping with field sampling, and June Mohler for drawing Figure 1. Jim Eagleton, Ted Kuiper, Ron Logan, Cathy Sanford, and Roger Sardina informed us on the incidence of Cancer oregonensis set- tlement in their growing areas. Mary Ann Asson-Batres, Robert W. Elner, Jeff Gonor, Bob Malouf, Dan B. Quayle, Deirdra Rob- erts and three anonymous reviewers made suggestions for improv- ing the manuscript. Heidi Metcalf and Bart C. Baldwin were supported by the Native Americans in Marine Science Program (National Science Foundation Grant OCE-9016300 to Oregon State University). Other expenses were covered by R. S. Yamada and an O.S.U. Research Council Award. This research could not have been car- ried out without the cooperation of Bill and Do Webb of Westcott Bay Sea Farms and the facilities provided by the Director and staff of the University of Washington Friday Harbor Laboratories. LITERATURE CITED Dare, P. J., G. Davies & D. B. Edwards. 1983 Predation on juvenile Pacific oysters iCrassoslrea gigas Thunber) and mussels (Mylilus edii- Us L. ) by shore crabs (Carcinum maenas L. ). Fisheries Research Tech- nical Report, Ministry of Agriculture, Fisheries and Food Directorate of Fisheries Research, Lowestoft, No 73, 15 pp. DeBrosse, G. A., A. J. Baldinger & P. A. McLaughlin. 1990. A com- parative study of the megalopal stages of Cancer oregonensis Dana and C. producliis Randall (Decapoda: Brachyura: Cancridae) from the northeastern Pacific, Fishenes Bulletin. U.S. 88:39^9. Eggleston, D. B. 1990. Foraging behavior of the blue crab, Callinectes sapidus. on juvenile oysters, Crassoslrea virginica: Effects of prey density and size. Biiltelin of Marine Science 46:62-82. Elner, R. W. & R, E. Lavoie. 1983. Predation on Amencan oysters (Crassoslrea virginica Gmelin) by American lobsters (Homarus amer- icanus Milne-Edwards), rock crabs (Cancer Irroratus Say), and mud crabs (Neopanope sayi Smith). J. Shellfish Res. 3:129-134. Gross, W. J. 1957. An analysis of repsonse to osmotic stress in selected decapod Crustacea. Biol. Bull. 112:43-62. Hart, J. F. L. 1982, Crabs and their relatives of British Columbia. British Columbia Provincial Museum No 40, 267 pp. Jamieson, G. S. & A. C, Phillips. 1988. Occurrence of Cancvr crab (C. magisier and C. oregonensis) megalopae off the west coast of Van- couver Island, British Columbia. Fishery Bulletin 86:525-542. Knudsen, J. W. 1964. Observations of the reproductive cycles and ecol- ogy of the common brachyura and crab-like anomura of Puget Sound, Washington, Pacific Science 18:3-33. Lawton. P & R. W. Elner. 1985. Feeding in relation to morphometries within the genus Cancer: Evolutionary and ecological considerations. pp. 357-379. In Proceedings of the symposium on Dungeness crab biology and management. B. R, Melteff (ed.). University of Alaska, Alaska Sea Grant Report No 85-3. Lough, R. G. 1975. Dynamics of crab larvae (Anomura, Brachyura) of the central Oregon coast, 1969-1971. Ph.D. thesis, Oregon State Univer- sity, Corvallis, 299 pp. Menzel, R. W. & S. H. Hopkins. 1955. Crabs as predators of oysters in Louisiana. Proceedings of the National Shellfisheries Association 46: 177-184, Moms, R. H,, D, Abbott & E, C, Haderiie, 1980, Intertidal invertebrates of California, Stanford University Press, Stanford. California, 690 pp. Orensanz, J. M. & V. Gallucci, 1988. Comparative study of postlarval life-history schedules in four sympatric species of Cancer (Decapo- da:Brachyura:Cancridae). Journal of Crustacean Biology 8:187-220. Parsons, J. 1974. Advantages of tray culture of Pacific oysters (Crassos- trea gigas) in Strangford Lough, N. Ireland. Aquaculture 3:221-229. Quayle, D. B. 1988. Pacific oyster culture in Bntish Columbia. Fisheries Research Board of Canada Bulletin 218, 241 pp. Sokal, R, R, & F J, Rohlf, 1981, Biometry, the pnnciples and practice of statistics in biological research. Second Edition, W, E, Freeman and Company, New York, 859 pp, Walne, P. R. & G. Davies. 1977. The effect of mesh covers on the survival and growth of Crassoslrea gigas Thunberg grown on the sea bed, Aquaculture 11:313-321. Journal of Shellfish Research. Vol 12. No 1.93-94. 1993 INTERSEX AUSTRALIAN RED CLAW CRAYFISH (CHERAX QUADRICARINATUS) PAUL B. MEDLEY* AND DAYID B. ROUSE Department of Fisheries and Allied Ac/iiacultiires Auburn University Auburn Alabama 36H49, USA The Australian red claw crayfish has recently been considered a candidate for commercial aquaculture in the United States be- cause of characteristics, such as large size, ease of reproduction. multiple spawnings, high fecundity, gregarious behavior, and higher percentage of tail meat than red swamp crayfish (Procam- barus clorkit) (Jones 1990. Medley et al. 1991 . Rouse et al. 1991 ). In 1989 and 1990. red claw juveniles were purchased from com- mercial hatcheries in Queensland, Australia, and Missouri, USA, respectively. These animals and their progeny were used in aqua- culture production experiments conducted from May to October, 1990. at the Alabama Agricultural Experiment Station, Fisheries Research Unit, Auburn University, Alabama. During experimen- tal culture, some red claw were noted to possess both male and female secondary sexual characteristics. Normally, crayfish are distinguished by distinct dimorphic secondary sexual characteris- tics. Males have two genital openings at the base of the fifth pereiopods and females have two genital openings at the base of the third pereiopods. Intersex crayfish or "pseudohermaphro- dites", are characterized by aberrant secondary sexual character- istics of both male and female (Sokol 1988, Huner and Barr 1991). In production trials, red claw crayfish averaging 3.7 g were stocked into three. 0.02-ha, fertilized ponds at a rate of 1/m' and fed hay at a rate of 500 kg/ha/mo (Medley 1991). The crayfish were cultured for 165 days. Average survival (mean ± SE) of red claw at harvest for the three 0.02-ha earthen ponds was 85.5 ± 4.6%. At harvest, crayfish were sampled (n = 326), sexed, and weighed. Total average weight (n = 513) for harvested crayfish was 70.0 ± 1.9 g (mean ± SE). Mean weights among the three production ponds were not significantly different (P > 0.05). Red swamp crayfish averaging 3.3 g were also cultured under condi- tions identical to those used to raise red claw crayfish. At harvest, no red swamp crayfish possessed aberrant second- ary sexual characteristics (n = 1 16). Females with male secondary sex characteristics have been reported for red swamp crayfish, but this type of pseudohermaphroditism is not common (Huner and Black 1977). To date, only one case of true hermaphroditism has been documented in red swamp crayfish (Huner and Black 1977). Pseudohermaphroditism is more common among other native American crayfish of the genera Cambarus and Orconectes (Turner 1935, Huner and Barr 1991). Among red claw the following combinations of gonopore placement were observed: (1) one male opening on right or left side and two normally-positioned female openings, (2) one male opening on nght or left side and one female opening on opposite side, (3) one male opening on right side and one female opening on right side, (4) two normally-positioned male openings and one female opening on right or left side, (5) two normally-positioned female and male openings. Individuals with only one male or one female genital opening were also found, but were not classified as intersex. Of the above intersex types, combination ( 1 ) was the most common, comprising 31% of sampled intersex crayfish, whereas combination (3) was the least common, comprising only 2% of intersex crayfish. Normal males occurred most frequently in the culture ponds, comprising 60% of sampled crayfish, while inter- sex crayfish were the least common comprising 17%. Average weights among the red claw crayfish sexes were sig- nificantly different (P < 0.01). Mean weights for male, female, and intersex crayfish from the three ponds were 75.7 g, 58.2 g, and 77.8 g, respectively, with a mean square error (MSE) of 3.6. Intersex crayfish were larger than females {P < 0.01), whereas no significant difference could be detected between males and inter- sex crayfish (P > 0.05) (Fisher's protected LSD). Although intersex individuals do occur in several species of Australian parastacids (Sokol 1988. Lake and Sokol 1988), noth- ing has been mentioned about this phenomenon in red claw cray- fish. Several studies have been conducted with red claw crayfish that present information on sexual differentiation, reproduction, and biology (Sammy 1988, Jones 1990. Merrick and Lambert 1991), but none of these specifically mentioned the presence of intersex red claw crayfish. Memck and Lambert (1991) gave a brief mention of intersex Australian crayfish possessing both male and female openings, and functioning as males. Whether intersex red claw are functional hermaphrodites or in a transitional phase between sexes is not certain. Turner (1935) mentioned that a crayfish cannot be considered a true hermaphro- dite unless both testicular and ovarian tissue are present. Of the intersex crayfish sacrificed and examined internally, one had a complete testis on one side and what appeared to be undeveloped ovarian tissue on the opposite (Fig. 1). However, in the absence of histological, anatomical, and endocrinological data, we cannot address the reasons for the observed intersex situation reported ♦Present address: School of Forestry-, Wildlife and Fisheries, Louisiana Agricultural Experiment Station. Louisiana State University Agricultural Center. Baton Rouge. LA 70803, USA. Figure 1. Dorsal view of dissected red claw crayrish showing ovarian tissue (o) on right and testicular tissue (t) on left. 93 94 Medley and Rouse here. Determination of whether the occurrence of intersex crayfish is a genetic or hormonal phenomenon, or a condition triggered by environmental factors will require further study. Since red claw exhibit intersexuality. there may be merit to sex-reversal using hormone treated feeds similar to those used for cultured fishes such as tilapia (Oreochromis spp.) (Clemens and Inslee 1968, Pandian and Varadaraj 1988). Production of all-male populations would eliminate spawning in grow-out ponds thus allowing red claw to divert more energy to growth. Male red claw are also larger, on average, than females. LITERATURE CITED Clemens, H. P. & T. Inslee. 1968. The production of unisex broods of Tilapia mossambica sex reversed with methyltestosterone. Trans. Am. Fish. Soc. 97:18-21. Huner, J. V. & J. B. Black. 1977. Aberrant secondary sexual characters in the crawfish Procambanis ctarkii (Girard) (DecapodaiCambaridae). Southweslern Naluralisi 22(2):271-275. Huner, J. V. & J. E. Barr. 1991. Red swamp crawfish: biology and ex- ploitation. Center for Wetland Resources, Louisiana Sea Grant Pro- gram, Baton Rouge, Louisiana, Sea Grant Publication No. LSU-T-80- 001. Jones, C. M. 1990. The biology and aquaculture potential of the tropical freshwater crayfish Chenix quadncarinanis. Queensland Department of Pnmary Industnes. Information Senes No. Q190028. 130 pp. Lake, P. S. & A. Sokol. 1986. Ecology of the yabby Cherax desiruclor Clark (Crustacea: Decapoda: Parastacldael and its potential as a senti- nel animal for mercury and lead pollution. Australian Water Resources Council Technical Paper No. 87. Austrialian Government Publishing Service, Canberra. 186 pp. Medley, P. B. 1991 . Suitability of the Australian red claw, Cherax quad- ricarinatus (von Martens) for aquaculture in the southeastern United States. M.S. Thesis, Auburn University, Alabama. 113 pp. Medley, P. B., R. G. Nelson, D. B. Rouse & L. U. Hatch. 1991. Eco- nomic feasibility and nsk analysis of pond produced Australian red claw crayfish (Cherax quadricannatus) in the southeastern United States. Auburn University Department of Agricultural Economics and Rural Sociology Working Paper Series No. 91-6. 31 pp. Merrick, J. R. & C. N. Lambert. 1991 . The yabby, marron and red claw: production and marketing. Macarthur Press Pty. Ltd. NS.W., Aus- tralia. 180 pp. Pandian, T, J. & K. Varadaraj. 1988. Techniques for producing all-male and all-triploid Oreochromis mossambicus. In: The Second Interna- tional Symposium on Tilapia Aquaculture. R. S. V. Pullin, T. Bhukaswan, K. Tonguthai and J. L. Maclean (eds.), ICLARM Con- ference Proceedings 15:243-249. Rouse, D. B., C. M. Austin & P. B. Medley. 1991. Progress toward profits'? Information on the Australian crayfish. Aquaculture Magazine 17(3):46-56. Sammy, N. 1988. Breeding biology of Cherax quadricarinaius in the Northern Territory. In: Proceedings of the First Australian Aquacul- ture Conference. L. H. Evans and D. O'Sullivan (eds), Curtin Uni- versity of Technology, Perth, Western Australia, pp. 79-88. Sokol, A. 1988. The Australian yabby. In: Freshwater Crayfish: Biology. Management, and Exploitation. D. M. Holdich and R. S. Lowery (eds.), Croom Helm, London/Sydney. Timber Press, Portland, Ore- gon, pp. 401^25. Turner, C. L. 1935. The aberrant secondary sex characters of crayfishes of the genus Cambarus. American Midland Naturalist 16:863-882. Joiirmil of Shellfish Research. Vol, 12, No. 1. 95-100. 1993. EVALUATION OF MICROBIAL INDICATORS FOR THE DETERMINATION OF THE SANITARY QUALITY AND SAFETY OF SHELLFISH PATRICK M. REGAN,' ^ * AARON B. MARGOLIN,' AND WILLIAM D. WATKINS^ ' Winchester Engineering and Analytical Center U.S. Food and Drug Administration Winchester. Massachusetts 01890 'Departmemt of Microbiology University of New Hampshire Durham. New Hampshire 03824 ^Northeast Technical Senices Unit U.S. Food and Drug Administration North Kingstown. Rhode Island 02852 .ABSTRACT Shellfish consumed either raw or partially cooked have been implicated in the transmission of viral gastroenteritis and hepatitis A. The effectiveness of bacterial indicators to signal the presence of human pathogenic viruses has been questioned. Earlier viral assays made it impractical to monitor shellfish for viral contaminants. There exists a need for rapid and sensitive assays for human enteric viruses to ensure the sanitary quality of shellfish. Sample collections of hard-shell clams (Mercenaria mercenaria) were taken from approved, conditionally approved and prohibited shellfishing areas in Narragansett Bay. Rhode Island between July 1989 and May 1990. Clams were assayed for poliovirus and other microbial indicators (total coliforms. fecal coliforms, Clostridium perfringens. enterococci and male-specific bactenophage ) to evaluate their usefulness as viral indicators. Of these indicators, bactenophage were most consistently recovered from each of the collection areas, and enterococci were recovered with the least frequency. Polovirus was detected in clams from the conditionally approved and prohibited area primanly dunng the fall and winter months. On one occasion in the prohibited area, the coliform standards for water and shellfish were not exceeded, although poliovirus was detected by a hybndization probe assay. A viral indicator system based on bactenophage levels would require further development and evaluation to determine the correlation of specific human enteric viruses and phage. New advances in nucleic acid technology may soon enable routine monitoring of shellfish for enteric viruses. KEY WORDS: male-specific bactenophage. poliovirus and hybridization probe INTRODUCTION Shellfish have been widely recognized as a means of transmis- sion of foodbome enteric disease since early this century, when a number of serious shellfish associated typhoid fever outbreaks were reported (Guzewich and Morse 1985). Edible bivalve mol- luscs of the class Pelecypoda (oysters, clams, and mussels) are the only molluscan shellfish of commercial importance for which san- itary controls are currently required (Metcalf 1975). Although the National Shellfish Sanitation Program (NSSP) bacterial indicator system has decreased the incidence of shellfish-associated enteric disease, its efficacy as a reliable indicator for protecting against the presence of human enteric viruses is questionable (Wait et al. 1983). One of the principal concerns with the present indicators and standards are that coliform bacteria are much more sensitive to chlorine than are a number of human enteric viruses, such as hepatitis A virus (Engelbrecht and Greening 1978). Also, the sur- vival of certain human enteric viruses in environmental water, during the winter months, is substantially greater than that of coliforms. During the last several decades, viral infections appear to ac- count for the majority of foodbome illnesses in the U.S. During 1982 there were 103 well-documented cases of gastroenteritis as- sociated with the consumption of raw shellfish involving 1.017 individuals in New York. The predominant etiological agent was determined to be Norwalk virus (Guzewich and Morse 1985, *Corresponding Author Morse et al. 1986). Other outbreaks of viral gastroenteritis and hepatitis A related to the consumption of raw or partially cooked shellfish have been reported (Gill et al. 1983, Portnoy et al. 1975. Richard 1985) as well. There is good likelihood that the incidences of individual cases and isolated outbreaks of shellfish-associated viral illnesses are significantly underreported. All viruses known to be normally transmissible through foods are derived from the human intestine (Blackwell et al. 1985). The discharge of both treated and untreated sewage into waterways, being utilized as sources of seafood, has gained much attention in regard to viral contaminated shellfish (Gerba and Goyal 1978, Landry et al. 1983). All species of commercially important shell- fish have been shown to enteric viruses from environmental sea- water during routine feeding activities (Metcalf et al. 1980). Since ordinary wastewater treatment does not always completely remove or disinfect such viruses, there is a need to be able to assess the efficacy of current indicators and standards. Currently, there is no one organism that can be considered to be the ideal indicator. Since it is impractical, indeed impossible to test for each individual bacterial or viral pathogen, the use of an alternative indicator, one that best correlates with the survivability and occurrence of the most resistant human enterovirus is probably the most practical means to ensure the sanitary quality of shellfish. The feasibility of using other indicator organisms such as fecal streptococci (Berg and Metcalf 1978) Clostridium perfringens (Emerson and Cabelli 1982) and bacteriophage (Havelaar et al. 1986) have been discussed. Assays involving the detection of en- teroviruses in shellfish by cell culture (Bemiss et al. 1989, Idema 95 96 Regan et al. et al. 1991) and by the use of hybridization probes (Bruce et al. 1989, Jiang et al. 1986, Margolin et al. 1986) have been evalu- ated. The objective of this study was to determine the levels of and compare the relationships between bacterial indicators, male- specific bacteriophage, and poliovirus found in shellfish collected from approved, conditionally approved, and prohibited waters. MATERIALS AND METHODS Shellfish and Water, Collection and Handling Hardshell clams {M. mercenaria) for this study were harvested from Narragansett Bay, Rhode Island. Samples were collected at approximately one month intervals. Clams were obtained with a long handled shellfish-rake from approved, conditionally ap- proved, and prohibited waters and held in the polypropylene bags on ice. Samples from each of the three collection areas were taken from approximately the same sites over the course of the study. Clams were not segregated by size prior to analyses; therefore, large sized and also those typically eaten raw Ciittle necks") were analyzed together. The clams were divided into two equal por- tions, one assayed for poliovirus, and the other assayed for male- specific bacteriophage (MSB) and other bacterial indicators. Sur- face water samples were obtained at each site when shellfish were harvested. Water samples were collected in sterile, 500 ml, poly- propylene screw cap bottles (Nalgene Laboratories Inc., Roches- ter, NY), and were held on ice until examined in the laboratory. Water samples were analyzed for total coliforms and fecal coliforms. Microbiological Analyses (i) Shelirish Approximately 10 clams were used in each analysis. Clams were scrubbed with a sterile brush, opened, and the entire contents (meat and liquor) were placed in sterile blender jars (Waring Corp., Coming, NY). Samples were blended at high speed for two minutes and held on ice (up to 60 minutes) until assayed. Total and fecal coliform densities in shellfish were determined by a most-probable-number (MPN) procedure, using lauryl tryp- tose broth (Difco) as the selective enrichment medium prescribed in Recommended Procedures (American Public Health Associa- tion 1970). Fecal coliforms were confirmed in EC-MUG medium (Difco) (Rippey et al. 1987). Enterococci densities were deter- mined by a 5-tube MPN procedure, utilizing azide dextrose broth (Difco) as the selective enrichment medium. Confirmation of tubes exhibiting growth was carried out at 24 and 48 hours; all positive tubes were streaked onto membrane filters (HC filters; Millipore Corp., Bedford, MA) placed onto Me (Levin et al. 1975) agar as previously described (Dufour 1980) with indoxyl- P-D-glucoside (Sigma. St. Louis. MO). The modified Me plates were incubated for 24 hours at 4rc. and tubes positive for en- terococci were confirmed by the presence of blue growth along the streaks. The levels of C. perfrin^ens in shellfish were determined by an iron milk MPN procedure (Abeyta 1983). MSB levels were determined using a modified double-agar-overlay procedure (Ca- belli 1988). utilizing £. coli strain (HS|pFamp]R). Plaques were counted after 18 to 24 hours of incubation at 35°C. MSB densities were calculated per 100 g of shellfish; determined by the number of plaques per volume of supemate assayed times the total volume of supemate obtained times 100 g divided by the number of g of homogenate examined. (ii) Water Samples were analyzed utilizing a multiple tube fermentation technique with lauryl tryptose broth as the selective enrichment medium (Difco). according to the Recommended Procedures (American Public Health Association 1970). All tubes exhibiting gas production were confirmed for coliforms in brilliant green lactose bile broth (Difco) and for fecal coliforms in EC broth (Difco). Poliovirus Elution from Shellfish Meats Clams were scrubbed with a sterile brush, opened, and 200 g of meat was transferred to a stainless steel canister (Omni Corpora- tion, Waterbury, CT). Two hundred ml of elution medium, con- sisting of 3% beef extract. 3.2% NaCl and 90 mM glycine, at pH 9.5, was added to the sample (Deleon et al. 1986). The sample was homogenized usmg a Omni-Gen homogenizer (Omni-Gen), the pH checked and adjusted to 9.5 with 1 N NaOH, then centri- fuged (10,000 X g for 10 minutes) (Beckman model J2-21M, Fullerton, CA). The supematant was decanted. pH adjusted to 7.0 with 1 N HCl. and then divided into two aliquots. One aliquot (non-flocculated) was used for direct analysis by a hybridization probe and cell culture. The second aliquot was concentrated by acid precipitation (flocculated) (Katzenelson et al. 1976) prior to analysis by probe and cell culture techniques. The pellet generated by flocculation was resuspended in 0. 1 M Na^POj buffer at pH 9.5. The sample pH was checked and adjusted to 9.5 with 1 N NaOH, mixed for 5 minutes, and then centrifuged ( 10,000 x g for 10 minutes). The supemate was adjusted to pH 7.0 and the final volume adjusted to 30 ml. Phenol Chloroform Extraction of Viral Nucleic Acid Viral nucleic acid was liberated from both the flocculated and non-flocculated samples as follows; approximately 50 ml of the non-flocculated and 10 ml of the flocculated sample were individ- ually mixed with phenol:chloroform;isoamyl alcohol (25;24;1). Samples were vortexed for two minutes, centrifuged ( 10,000 x g for 10 minutes), and the aqueous phase was removed and trans- ferred to new phenol;chlorofonn:lsoamyl. The original tube was extracted two more times by the addition of diethyl-pyrocarbonate (DEPC) treated water, and each time the aqueous phase was re- moved and transferred to new phenol:chloroform;isoamyl. The aqueous phases from the original tubes were extracted until there was a minimal amount of protein present. Residual phenol was removed by chloroform extractions followed by an ether extraction to remove the residual chloroform. Filtered air was passed through the sample using DEPC treated pipet tips to evaporate off the remaining ether. The samples were applied to a Genescreen Plus hybridization membrane (New England Nuclear) using a vacuum manifold dot blot apparatus (Bio-Rad, Richmond, CA). Two ml of extracted sample was applied to each well, the membranes were baked in an 80°C incubator for two hours. Hybridization Probe Preparation and Hybridization Fragments of poliovirus cDNA, from poliovirus cDNA (bp 115-7440) (kindly supplied by David Baltimore) cloned into the Pst 1 site of pBR322 and transformed in £. co/ZHB-lOl were used as the probe. Following amplification, the recombinant plasmid Evaluation of Microbial Indicators for Shellfish 97 was isolated (Maniatis et al. 1989). The insert was excised from the vector by performing a Pst 1 digest (Boehringer Mannheim Corp.. Indianapolis. IN). Two bands corresponding to poliovirus cDNA. the 1174 base pair and the 16S9 base pair bands were excised from the gel and purified by either a commercially avail- able kit (Gene Clean II. La Jolla. CA) or by electroelution (Schlei- cher and Shuell. Keene, NH). These fragments were used as the probe in this study. The cDNA fragments were labeled with ^"P dCTP using a random primer extension labeling kit (New England Nuclear, Bos- ton, MA). Probe activity was determined by a scintillation counter, activities of 5 x 10* to I x 10"^ were obtained. Membranes containing the extracted nucleic acid samples were first prehybridized and then hybridized in heat sealed poly bags as described in Gene Screen Plus protocols (New England Nuclear, Boston. MA). The membranes were prehybridized at 42°C for 2 hours in a reciprocating water bath and hybridized for 36 hours at 42°C in the same solution, except for the addition of 10^ to 10^ cpm of the heat denaturated radiolabeled probe. The membranes were washed twice in a solution of 2x sodium chloride/sodium citrate (SSC) \% sodium dodecyl sulfate (SDS) for 15 minutes with constant agitation at room temperature, and once in 2x SCC/ 0.\% SDS with constant agitation at 52°C. The membranes were then air dried, and placed in a cassette with Dupont Cronex inten- sifying screens and Kodak XAR-5 film for 36 hours at -70°C. Cell Culture Analysis The viral assay was performed using a continuous cell line of Buffalo Monkey Green Kidney (BGM) cells. Cells were grown in minimal essential media (MEM) (Sigma) supplemented with %7c fetal calf serum. 292 mg/L glutamine. 0.075% sodium bicarbon- ate. 100 U/ml penicillin, 100 ug/ml streptomycin, 50 ug/ml kana- mycin and 25 U/ml mycostatin. Three 75 cm'^ tissue culture flasks with confluent monolayers of BGM cells were each inoculated with 3 ml of sample. Adsorption of virus was allowed to proceed for 2 hours at 37°C, flasks were rocked every 15 minutes. Fol- lowing adsorption, the cells were washed with phosphate buffered saline (PBS) and overlaid with maintenance medium. Flasks were incubated at 37°C and examined periodically for the presence of cytopathic effects (CPE) up to 14 days following inoculation. Flasks that exhibited CPE were confirmed by passage to new monolayers of BGM cells. RESULTS Samples of surface waters and clams from Narragansett Bay were obtained during the period of July 1989 to May 1990. A total of 9 collection trips were made for each of the three different shellfish classification areas. The bacteriological quality of clams and their overlying waters from the approved area are presented in Table 1 . from the conditionally approved area in Table 2, and from the prohibited area in Table 3. Water from the approved area exceeded the total coliform stan- dard (70/100 ml) once (April), and the fecal coliform standard (14/100 ml) was exceeded in another collection (May). Water quality for the majority of the conditionally approved area samples exceeded the coliform and fecal coliform levels found in the ap- proved area (note: this area was conditionally closed during all but the 8/30/89 sample collection). From the conditionally approved area. 6 of 9 water samples exceeded the total coliform standard for approved areas, whereas 5 of 9 samples exceeded the fecal coliform standard. Coliform and fecal coliform NPNs from the prohibited area were greater than levels found for both the ap- proved and conditionally approved areas. In the prohibited area 8 of 9 of water samples exceeded the coliform standard, and 7 of 9 samples exceeded the fecal coliform standard. One of 8 clam samples obtained from the approved area and the conditionally approved area exceeded the fecal coliform market guideline (230/100 g). whereas 2 of 8 prohibited area samples exceeded the guideline. Clostridium perfringens was detected in the prohibited area with the greatest frequency, and levels there remained detectable throughout all the collection times. Overall, levels of C. perfrin- TABLE L Microbial indicator levels in waters and clams from the approved area. Clams Water" Poliovirus ___,„„.. ._ Probe Cell Culture Sample Date Conforms Coliforms Coiiforms'' Coliforms'' perfringens*' Enterococci'' Bacteriophage' Eluent Floe Eluent Floe Male- specific 7/31/89 <1.8 <1.8 nd" 8/30/89 <1.8 <1.8 20 10/3/89 <1.8 <1.8 110 1 1/7/89 49 7.8 68 1/10/90 2 2 20 2/12/90 9.3 2 <20 3/13/90 <1.8 <1 8 <20 4/16/90 130 2 20 5/31/90 39 17 140 nd 20 110 68 20 <20 20 78 1 .300 nd 140 200 <20 <20 <20 <20 790 2.200 nd <20 <20 <20 <20 <20 <20 <20 <20 nd 15 3 127 7 14 15 <16 83 na= na na na na na na na na ' MPN per 100 ml " MPN per 100 g, ' Densities per 100 g calculated from plaque counts. '' Not determined. ' Not analyzed due to toxicity. 98 Regan et al. TABLE 2. Microbial indicator levels in waters and clams from the conditionally approved area. Clams Water" Poliovirus Total Fecal Total Fecal Clostridium Male- specific Probe Cell Culture Sample Date Coliforms Coliforms Coliforms'" Coliforms'' perfringens^ Enterococci'' Bacteriophage' Fluent Floe Fluent Floe 7/31/89 4^5 <1.8 nd" 8/30/89 240 49 78 10/3/89 33,000 2,300 5.400 11/7/89 79 11 490 1/10/90 2 2 <20 2/12/90 920 220 <20 3/13/90 <1 8 <1.8 <20 4/16/90 2.400 790 68 5/31/90 79 33 790 nd <20 790 140 <20 <20 <20 <20 110 nd 1 ,300 2.400 20 20 78 <20 110 490 nd 20 110 68 <20 <20 <20 <20 <20 nd 25 501 180 38 322 147 198 3 -I- + na na na na na na na na na " MPN per 100 ml. " MPN per 100 g. ' Densities per 100 g calculated from plaque counts. "* Not determined. ' Not analyzed due to toxicity. gens were seen to decrease from the prohibited to conditionally approved to the approved area, and the latter exhibited the most samples with levels in clams below the detectable limits. Entero- cocci densities were generally below detectable levels throughout all the collection times in all the areas. The highest levels of enterococci were detected in clams from the prohibited area, and densities appeared to decrease in clams from the conditionally approved area. No enterococci were detected in any clam samples from the approved area. The occurrence of MSB and results obtained for poliovirus in the approved area are shown in Table 1 ; Table 2 shows the results for the conditionally approved area; and those obtained for the prohibited area given in Table 3. Phage levels detected were great- est in the prohibited area, and these were notably higher than those detected in the conditionally approved area; the lowest levels de- tected were found in the approved area. MSB in clams (per 100 g) were detected in 7 of 8 samples examined from the approved area. In the conditionally approved and prohibited area MSB (per 100 g) were detected in all samples analyzed. In the approved area, hybridization probe results for poliovirus for both the non-flocculated and flocculated sample portions, were negative at all times. In the conditionally approved area, 3 of 9 non-flocculated clam samples were positive for viral nucleic acid, by probe analysis, whereas none of the flocculated samples were found to be positive. In the prohibited area. 4 of 9 non-flocculated samples were found positive by the probe assay, but only 2 of these 4 were positive using the flocculated samples. Cell culture analyses of clam samples for poliovirus were per- TABLE 3. Microbial indicator levels in waters and clams from the prohibited area. Water" Clams Total Fecal Clostridium Male- specific Poliovirus Probe Sample Total Fecal Cell Culture Date Coliforms Coliforms Coliforms'' Coliforms" perfringens'' Enterococci'' Bacteriophage' Fluent Floe Eluent Floe 7/31/89 > 1.600 1,600 5,400 nd-* 5.400 460 3,042 - - na'' - 8/30/89 > 1 ,600 > 1 ,600 9,200 1.100 2,400 230 465 - - na 10/3/89 49,000 2,300 > 16,000 3,500 3.500 3,500 5,078 -1- -1- na - 1 1/7/89 1,100 170 790 220 490 230 1,027 -1- - na - 1/10/90 49 6.8 20 . <20 140 <20 1,036 + - na -1- 2/12/90 130 2 <20 <20 110 <20 994 + -1- na - 3/13/90 490 130 <20 <20 78 <20 2,700 - - na - 4/16/90 22,000 2,300 1,700 45 330 140 1,180 — - na - 5/31/90 350 49 2,200 93 3,500 <20 124 — — na - "MPN per 100 ml. " MPN per 100 g. ' Densities per 100 g calculated from plaque counts. '' Not determined. ' Not analyzed due to toxicity. Evaluation of Microbial Indicators for Shellfish 99 formed using the flocculated portions only, since the non- flocculated portions were toxic to the BGM cells. Results of clams from the approved and conditionally approved area failed to detect cytopathic effects (CPE) with any of the samples. One of the 9 samples from the prohibited area was found to cause CPE on the BGM cells. DISCUSSION Periodic outbreaks of non-bacterial gastroenteritis and hepatitis A have indicated that the current means of evaluating the sanitary quality of shellfish and their harvesting waters requires re- evaluation. This study compared several bacterial indicators and MSB. to the occurrence of poiiovirus in shellfish collected from approved, conditionally approved, and prohibited areas over about a one year period. Poiiovirus was used because of a recently de- veloped nucleic acid probe technique, its ease of detection by cell culture techniques, and the generally higher degree of poiiovirus prevalence in sewage, relative to that expected for other enteric viruses. Earlier studies, (Margolin, unpublished results), demon- strated that this hybridization probe assay was able to detect virus with a sensitivity comparable to cell culture. The nucleic acid hybridization assay for poiiovirus permitted analysis of shellfish non-flocculated portions directly. This could not be done with cell culture due to the toxic effects of the non-flocculated portion on BGM cells. Consequently, all samples evaluated by cell culture were further processed by flocculation of the sample. Although flocculation adequately reduces sample toxicity, it also provided results with a reduced level of detectable poiiovirus, indicating the procedure is not 100% efficient for poiiovirus recovery. Water samples were analyzed for coliforms only, as this currently is a monitoring tool in determining the sanitary quality of shellfish harvesting areas. As expected, the approved area exceeded the coliform standard the least number of times. Results for the conditionally approved area show a wide variation in coliform levels, likely due to the affects from rainfall events which occurred prior to many of the collection periods. During most of the sample collections (all col- lection times except 8/89), the area was temporarily (condition- ally) closed due to excess rainfall. Samples taken from the pro- hibited area usually exceeded the coliform standard. However, in one instance (January), coliform levels detected in waters from the prohibited area were found to be acceptable, while the shellfish non-flocculated portion was positive for poiiovirus by the hybrid- ization probe. The advent of molecular detection techniques, has greatly enhanced our ability to detect specific types of bacteria and viruses. Future modifications for probe procedures should focus on minimizing virus loss during sample processing. Since viruses apf)ear to have a longer survivability time at lower temperatures than vegetative bacteria, and since standards based on levels of bacteriophage do not exist, it seems advisable to verify the sanitary conditions of shellfish, particularly those in conditionally managed areas, using a multifaceted approach which includes assays for viruses. Clostridium perfringens is a spore forming, obligate anaerobe. This organism is widespread in the environment and is not solely of fecal origin. Compared to the coliform bacterial indicators, C. perfringens spores have a significantly longer survival time in estuarine and are less susceptible to environmental stresses. The levels of C. perfringens spores in the prohibited area were mark- edly higher than those found in the conditionally approved and approved areas. Compared to the coliforms and enterococci. C. perfringens levels are less severely affected during the colder months in the conditionally approved and prohibited areas. In the approved area C. perfringens levels were similar to the coliform indicators, in that levels were undetectable during the colder months (November-March). While levels of C. perfringens spores were detectable more often than the coliforms or enterococci, there is no reliable relationship between C. perfringens and the occur- rence of human enteric viruses. Enterococci levels were significantly lower than those found for coliforms, C. perfringens and MSB. In the approved area enterococci densities were below the assay detection limit at all times. The lower levels of enterococci determined in this study are likely reflective of the lower numbers found in wastewater efflu- ent. These data demonstrate that enterococci levels would no fur- ther ensure the sanitary quality of shellfish than the present coliform system. The hybridization probe assay to detect coliform was per- formed with both the non-flocculated and flocculated sample por- tions. Non-flocculated portions were found to be positive more often than the flocculated samples, and in no instances were floc- culated samples found positive where non-flocculated portions were negative. This indicates that the flocculation procedure is not 100% efficient in recovery of virus particles, loss of virus could occur resulting in loss of sensitivity. Considering the low level of viral contamination expected in the clams, it appears preferable to rely on results from non-flocculated portions to yield the greatest degree of sensitivity. All clam samples were assayed for poiiovirus by cell culture. Comparison of the hybridization probe and cell culture results for the non-flocculated portion was not possible, due to sample tox- icity on BGM cell monolayers. Therefore, only the flocculated portions of the clam samples were examined by cell culture. Floc- culation followed by resuspension in phosphate buffer was suffi- cient in detoxifying the samples for cell culture analysis. In the approved area none of the flocculated samples were positive by cell culture, this correlates with the hybridization probe results for this area. Similar results were obtained for flocculated samples from the conditionally approved area, in that all cell culture and hybndization probe assay results for poiiovirus were negative. In the prohibited area, there was only one sample (January) that showed CPE in cell culture however, this sample was probe pos- itive with the non-flocculated sample portion, but not with the flocculated portion. One explanation for the discrepancies between tissue culture and probe analysis would be that the lower virus concentrations in the flocculated sample were below the limits of detection for the nucleic acid hybridization assay. MSB have received consideration as indicators of enteric viral pathogens. Though not regularly detected in fresh fecal material, this group of bacterial viruses are consistently present in sewage and sewage polluted waters (Debartolomeis and Cabelli I99I). Also, some members of the MSB group have been shown to be as resistant to disinfection by chlorination as Norwalk virus (Keswick et al. 1985). Relating to the occurrence of other indicators in this study, MSB were the only indicators consistently detected at all of the collection sites. However, MSB were present in all waters, including approved waters precluding their use in a simple pres- ence/absence test, since this would probably result in the unnec- essary closure of safe shellfish beds. Further data is needed to determine if there is a correlation between certain levels of MSB and the presence of human enteric viruses, also to establish a predictive index and protective MSB standards. Such studies are essential before MSB is considered as an indicator organism with- out excessively restricting shellfish waters. 100 Regan et al. In summary, the indicators in this study exhibit widely ranging results. Enterococci exhibited the lowest numbers throughout all of the collection sites and periods, followed by the fecal coliforms and total coliforms. These vegetative bacterial indicators are greatly affected by environmental stresses, such as water temper- ature, low nutrients, and salinity. C. perfringens and the MSB were present at greater frequencies than the coliform and entero- cocci groups, with MSB being detected more often and at greater levels than any of the other indicators. The use of hybridization probes in this study demonstrates an alternative technique to detect the presence of enteric viruses in clams. Even so, the direct detection of viral nucleic acid by probe analysis appears to require improvements in its detection limits. Detection of low levels of viruses in the shellfish is compounded by the low efficiency of viral recovery from clam meats. Addi- tionally, the procedures used in this study are time-consuming and labor-intensive. The results of this study support those reported by others, which suggest that bacterial indicators and standards while serving to adequately protect against bacterial pathogens in shellfish, they do not reliably predict the presence of human enteric viruses. The development of better extraction procedures and rapid, inexpen- sive, automated molecular techniques, may soon allow for the direct detection of most if not all pathogens potentially present in shellfish. Consequently, a more complete approach in evaluating shellfish sanitary quality and the safety of shellfish is currently needed. This includes indicator organisms along with analyses to detect certain viral pathogens directly using a molecular based assay. ACKNOWLEDGMENTS We thank Jack L. Gaines of the U.S. Public Health Service for his assistance in the collection of samples for this study. LITERATURE CITED Abeyta, C. 1983. Companson of iron milk and official AOAC methods for enumeration of Clostridium perfringens from fresh seafoods. J. Assoc. Ojf. Anal. Chem. 66:1175-1177. American Public Health Association. 1970. Recommended procedures for the examination of sea water and shellfish, 4th ed. Amencan Public Health Association, Washington DC. Bemiss, J. A., M. M. Logan, J. D. Sample & G. P. Richards. 1989. A method for the enumeration of poliovirus in selected moUuscan shell- fish. J. Virol. Methods 26:209-218. Berg, G. & T. G. Metcalf. 1978. Indicators of viruses in waters, pp. 267-296. In G Berg (ed.). Indicators of viruses in water and food. Ann Arbor Science Publishers, Ann Arbor, Mich. Blackwell, J. H., D. O. Oliver, J. J. Callis, N. D. Heidelbaugh, E. P. Larkin, P. D. McKcrcher & D. W. Thayer. 1985. Foodbome viruses: their importance and need for research. J . Food Proi. 48:717-723.2. Bruce, C, W. Al-Nakib, M. Forsyth, G. Stanway & J. W. Almond. 1989. Detection of enteroviruses using cDNA and synthetic oligonu- cleotide probes. J. Virol. Methods 25:233-240. Cabelli, V. J. 1988, Microbial indicator levels in shellfish, water, and sediments from the upper Narragansetl Bay conditional shellfish grow- ing area. Report to the Narragansett Bay Project, Providence, Rl. Debartolomeis, J. & V. J. Cabelli. 1991 . Evaluation of an Escherichia coli host strain for enumeration of F male specific bactenophages. Appl. Environ. Microbiol. 57:1301-1305. Deleon, R H.,H. A. Payne &C. P. Gerba. 1986. Development of a method for poliovirus detection in freshwater clams. Food Microbiol. 3:345-349. Dufour, A. P. 1980. A 24-hour membrane filter procedure for enumerat- ing enterococci, abstr. Q 69, p. 205. Abstr. 80th Annu. Meet. Am. Soc. Microbiol. 1980. Emerson, D. J. & V. J. Cabelli. 1982. Exlr^ction of Clostridnim perfrin- gens spores from bottom sediment samples. Appl. Environ. Microbiol. 44:1144-1149. Engelbrecht, R. S. & E. O. Greening. 1978, Chlorine-resistant indicators, pp. 24.3-265. In G, Berg (ed.). Indicators of viruses in water and food. Ann Arbor Science Publishers, Ann Arbor, Mich. Gerba, C. P. & S. M. Goyal. 1978. Detection and occurrence of entenc viruses in shellfish: a review. J. Food Prot. 41:743-754. GUI, O. N., W. D. Cubitt, D. A. McSwiggan, B. M. Watney & C. L. R Bartletl. 1983, Epidemic of gastroenteritis caused by oysters contaminated with small round structured viruses. Br Med. J. 287:1532-1534. Guzewich, J. J. & D. L. Morse. 1985, Sources of shellfish in outbreaks of probable viral gastroenteritis: implications for control, J. Food Prot. 49:389-394. Havelaar, A. H., K. Furuse & W. M. Hogcbooni. 1986. Bactenophages and indicator bacteria in human and animal faeces. J. Appl. Microbiol. 60:255-262. Idema, G. K,, B. W. Bateman, R Kfir & W. O. K. Grabow. 1991. A comparison of methods for the isolation of a wide range of viruses from shellfish. Wat. Sci. Tech. 24:427-430. Jiang, X., M. K. Estes, T. G. Metcalf & J. L. Melnick. 1986. Detection of hepatitis A virus in seeded estuanne samples by hybridization with cDNA probes. Appl. Environ. Microbiol. 52:711-717, Katzenelson, E.. B. Fattal & T. Hostovesky. 1976. Organic fiocculation: an efficient second step concentration method for the detection of vi- ruses in tap water. Appl. Environ. Microbiol. 32:638-639, Keswick, B. H., T. K. Satterwhite, P. C. Johnson, H. L. DuPont, S, L. Secor, J, A, Bitsura, G. W, Gary & J. C. Hoff. 1985. Inactivation of Norwalk virus in drinking water by chlorine. Appl. Environ. Micro- biol. 50:261-264. Landry. E. F.. J. M. Vaughn, T. J. Vicale & R, Mann, 1983, Accumu- lation of sediment-associated viruses in shellfish, Appl. Environ. Mi- crobiol. 45:238-247. Levin, M. A., J. R. Fischer & V. J. Cabelli. 1975. Membrane filter technique for enumerating enterococci in marine waters. Appl. Microbiol. 30:66-71. Maniatis, T., E. F. Fritsch & J. Sambrook. 1989. Molecular Cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Har- bor, NY, Margolin, A, B.M, JHewlen&C, P. Gerba, 1986, Use of a cDNA dot blot hybndization technique for detection of enteroviruses in water, p, 87-95, Water Quality Technol, Conf, Proceedings, Amer, Water Works Assoc. Denver, CO, Metcalf, T, G, 1975, Evaluation of shellfish sanitary quality by indicators of sewage pollution. In Pergamon Press. Great Britain. Metcalf, T. G., D. Eckerson, E. Moulton & E. P. Larkin. 1980. Uptake and depletion of paniculate-associated poliovirus by the soft shell clam. J. Food Prot. 43:87-88, Morse, D, L., J. J. Guzewich, J. P, Hanrahan, R, Stricof, M, Sheyegani, R, Deibel, J, C, Grabau, N. A, Nowak, J. E. Herrmann, G, Cukor & N, R, Blacklow, 1986, Widespread outbreaks of clam and oyster associ- ated gastroenteritis: role of norwalk virus, N. Engl. J. Med. 3 14:678-681 . Portnoy, B. L., P A Mackowiak, C. T. Caraway, J. A. Walker, T. W. McKinley & C. A. Klein. 1975. Oyster associated hepatitis — Failure of shellfish certification programs to prevent outbreaks. JAMA 233: 1065-1068. Richard, G. P. 1985. Outbreaks of shellfish associated enteric virus in the United States: requisite for development of viral guidelines. J. Food Prot. 48:815-823, Rippcy, S, R,, A, Chandler & W, D, Watkins, 1987, Flourometric method for enumeration o( Escherichia coli in molluscan shellfish, J. Food Prot. 50:685-690. Wait, D. A., C. R. Hackney, R. J. Camck, G. Lovelace & M D. Sob- sey. 1983. Enteric bacterial and viral pathogens and indicator bacteria in hard shell clams, J. Food Prot. 46:493-496. Journal of Shellfish Research. Vol, 12. No. 1, 101-115, 1493. ABSTRACTS OF TECHNICAL PAPERS Presented at the 13th Annual Meeting MILFORD AQUACULTURE SEMINAR Milford, Connecticut February 22—24, 1993 101 Milford Aquaculture Seminar, Milford, Connecticut Absiracis. 13th Annual Meeting, February 22-24, 1993 103 CONTENTS Walter Blogoslawski Historical perspectives — Shellfish Biology Seminar/Milford Aquaculture Seminar 105 Standish K. Allen, Jr. Development of high survival resistant lines in oysters using MSX-resistant strains 105 Lee Anderson, Dave Jones and Standish K. Allen, Jr. Interactive spreadsheet on the economics of oyster fanning 105 Ann Arseniu, Lyu Suifen and Standish K. Allen, Jr. Optimizing metamorphosis and survival for lab studies of Mulinia lateralis 106 Sebastian Belle The role of research and development in a competitive domestic aquaculture industry 106 Joseph Buttner, Pei Chang, Paul Bowser, Frank Hetrick, Philip McAllister, Bruce Nicholson and Paul Reno Detection of fish pathogens 106 Gerald M. Capriulo, Robert Troy, Marcelo Morales, Kathleen Beddows, Helen Budrock, Gary Wikfors and Charles Yarish Possible eutrophication-related enhancement of the microbial loop in Long Island Sound and consequences for shellfish 10'7 Gregory A. Debrosse and Standish K. Allen, Jr. Control of overset on cultured oysters using brine solutions 107 C. Austin Farley and Earl J. Lewis, Jr. Juvenile oyster mortality studies — 1992: Histopathology , pathology, epizootiology 107 Susan E. Ford Recent outbreaks of dermo disease in the northeast; new introductions or climate change? 108 Susan E. Ford, Francisco J. Borrero and Walter J. Blogoslawski Studies of juvenile oyster mortality on Long Island Sound, NY in 1992 108 Ximing Guo and Standish K. Allen, Jr. Reproductive genetics of triploid Crassostrea gigas 108 Robert E. Hillman Effect of trematodes on east coast populations of the blue mussel , Mytilus edulis 109 Robert E. Hillman Relationship of environmental contaminants to occurrences of neoplasia in Mytilus edulis populations from east and west coast mussel-watch sites 109 Ya Ping Hu and Standish K. Allen, Jr. Cytological and cytogenetic examination of gametogenesis in triploid Crassostrea virginica and Crassostrea gigas .... 109 John Karlsson Parasites of the bay scallops, Argopecten irradians 109 John Karlsson and Arthur R. Ganz Occurrence of Oslrea edulis in Rhode Island 1 10 Stephen J. Kleinschuster and Sharon L. Swink A simple method for the in vitro culture of Perkinsus marinus 110 Kenneth P. Kurkowski Overview of the operations of Atlantic Littleneck Clamfarms 110 Earl J. Lewis, Jr. and C. Austin Farley Preliminary results of laboratory attempts to transmit a disease affecting juvenile oysters in the northeastern United States 1 10 Wenyu Lin, Michael A. Rice and Paul K. Chien The differential effects of three heavy metals on particle filtration and amino acid uptake by the Pacific oyster, Crassostrea gigas HI Mark Luckenbach, Sandra E. Shumway and Kevin Sellner "Non-toxic" dinoflagellate bloom effects on oyster culture in Chesapeake Bay: Preliminary results Ill Victor J. Mancebo Northeastern Regional Aquaculture Center: An update Ill Victor J. Mancebo Shrimp culture in the Philippines: Birth of the industry 112 Harold C. Mears Aquaculture in the northeast region of NMFS 112 104 Abstracts, 13th Annual Meeting, February 22-24, 1993 Milford Aquaculture Seminar, Milford, Connecticut Daniel J. Medina, Gregory E. Paquette, Eiken C. Sadisar and Pei W. Chang Isolation of infectious particles having reverse transcriptase activity and producing hematopoietic neoplasia in Mya arenaria 112 Sidney K. Pierce Differences in the salinity tolerance mechanisms between Chesapeake Bay and Atlantic Coast oyster: Genetics or disease-induced effects on mitochondrial metabolism? 113 Robert B. Rheault Food-limited growth and condition index in Crassostrea virginica and Argopecten irradians 113 Eileen C. Sadasiv, Pei W. Chang and Wenyu Lin IPNV antibody as a means of detection of possible virus carriage in Atlantic salmon surviving virus challenge 113 Sandra E. Shumway and Allan D. Cembella Impact of harmful algal blooms on scallop culture and fisheries 114 Sheila Stiles and Walter Blogoslawski Viability and genetic effects on oyster embryos exposed to bacterial and effluent from diseased juvenile oysters from a Long Island hatchery 114 Gary H. Wikfors, Roxanna M. Smolowitz and Barry C. Smith Effects of Prorocentrum isolate upon the oyster, Crassostrea virginica: A study of three life-history stages 1 14 Milford Aquaculture Seminar, Milford. Connecticut Abstracts. 13th Annual Meeting, February 22-24, 1993 105 HISTORICAL PERSPECTIVES— SHELLFISH BIOLOGY SEMINAR/MILFORD AQUACULTURE SEMINAR. Walter J. Blogoslawski, National Oceanic and Atmospheric Administra- tion, National Marine Fisheries Service, Northeast Fisheries Sci- ence Center, Milford Laboratory, 212 Rogers Avenue, Milford, CT 06460. In May 1992, the federally-sponsored Joint Subcommittee on Aquaculture published a report, "Aquaculture in the United States: Slocks, Opportunities, and Recommendations", which stated that United States aquaculture production for 1990 exceeded 860.8 million pounds with a value of $761 million dollars. This is a four- fold mcrease in production from 1980. The aquaculture industry accounts for more than 290,000 jobs with a total eco- nomic impact of $8 billion dollars. In recognition of the impor- tance of aquaculture to the northeast, the Milford Laboratory has sponsored shellfish biology seminars since 1975. The first meeting of the Milford Shellfish Biology Seminar occurred as a technical exchange between staff at the Milford Laboratory and shellfish managers of the F. M. Flower Company, Bluepoints Company, and Long Island Oyster Farms. There were nine industry attendees and six Milford staff. In subsequent years the forum broadened in technical scope, covering topics of algal rearing, genetics, and water quality control, and organized as a technical exchange from Government scientists to shellfish indus- try representatives. Since 1980, the scope of the seminar has fur- ther expanded to include presentations by aquaculture scientists from Sea Grant, academia, and other state and federal agencies, as well as the commercial aquaculture companies. The purpose of these annual gatherings of shellfish biologists was to share current ideas and innovative methodologies in shell- fish research. This year the name was changed from "shellfish" to "aquaculture" seminar to focus attention on other marine species of commercial interest. For example, cultured marine Atlantic salmon, a new multi-million dollar industry in the Northeast, will soon exceed in value that of wild-harvested Maine lobsters. Under a multi-species approach, innovation and fiexibility can be applied to the northeast aquaculture industry. The industry can adopt new methods, use disease-resistant strains, or new species of animals, thereby offering greater value in the marketplace. Some of these adaptations, including the use of genetically manipulated or new species of oyster broodstock, are the focus of this year's meeting. We will also discuss possible causes of juvenile oyster mortalities, effects of shellfish disease, pollutants and noxious bloom organisms on shellfish, and effects of diseases of cultured fish. DEVELOPMENT OF HIGH SURVIVAL RESISTANT LINES IN OYSTERS USING MSX-DISEASE RESISTANT STRAINS. Standish K. Allen, Jr., Haskin Shellfish Research Laboratory, Department of Marine and Coastal Sciences, Rutgers University, Box B-8, Port Noms. NJ 08349. Since 1958 Rutgers has been breeding American oysters for resistance to MSX-disease. Rutgers maintains a commitment to continuing these strains and to strategies for further improvements in the American oyster. The advent of Dermo in the northeast has caused us to reevaluate the future and role of the MSX-disease resistant (RR) strains. RR oysters are not resistant to Dermo. Ad- ditionally, because genetic variability in RR stocks has been con- strained through intense selection pressure and population bottle- necks, they may be more susceptible. Almost certainly, there is enough loss of genetic variability in RR oysters to question the wisdom of selecting for Dermo resistance from any one of the RR strains per se. Our progress toward creating two new strains of resistant oysters is reported here. These strains are collectively to be called High Survival Resistant Lines (HSRL). From June 10 to July 21, 1992, we produced two geographic races of HSRL; a Delaware Bay (DB HSRL) and a northeast race (NE HSRL). Broodstock for DB HSRL comprised four strains of resistant oys- ters, three years old, and wild stock from Delaware Bay. The broodstock populations have been exposed to Dermo pressure for only one generation. Constituent populations for NE HSRL in- cluded two succeeding generations of Long Island RR strains (BLA and CLA), and F. M. Flowers, Inc. and Ocean Pond, Inc. varieties of the BLA line. We did not introduce Long Island wild stock genes into NE HSRL. Founder populations were produced by controlled matings among (but not within) each constituent population. Matings were made for both DB HSRL and NE HSRL using as many pairs as possible from each constituent population. Five founder sub-populations for each race (DB or NE) were pro- duced, but we lost one of the NE sub-populations in the hatchery stage. Through a series of matings among (but not within) strains, we produced a total of 2388 families comprising the five DB HSRL sub-populations and 3287 families comprising the five NE HSRL sub-populations The reason for subdividing the lines is so that each sub-population can be cross bred to other sub- populations (but not to itselO in future generations. Such a cross- ing scheme prevents matings between closely related individuals. Also, because the high survival line comprises five sub- populations, high effective population sizes are maintained, min- imizing genetic drift. Breeding adjacent sub-populations recipro- cally (but not to themselves) will produce a new generation (F,) also consisting of five sub-populations in each geographical race. INTERACTIVE SPREADSHEET ON THE ECONOMICS OF OYSTER FARMING. Lee Anderson.' Dave Jones,^ and Standish K. Allen, Jr.,^ 'College of Manne Studies. University of Delaware, Newark, DE 19716 and "Haskin Shellfish Research Laboratory. Department of Marine and Coastal Sciences. Rutgers University. Box B-8. Port Norris, NJ 08349. From 1990-1992, NCRI sponsored a research program at the Haskin Shellfish Research Laboratory, "Oyster grow-out tech- niques for the mid- Atlantic: a Delaware Bay model". In our proj- ect, we explored intensive (rack and bag) aquaculture using "cultchless", MSX-disease resistant spat. The overall goal of our 106 Abstracts. 13th Annual Meeting, Febnjary 22-24, 1993 Milford Aquaculture Seminar, Milford, Connecticut demonstration project was to determine the economic feasibility of oyster grow-out using rack and bag culture of MSX-resistant, cultchless oysters. Specifically, we wanted to estimate the biolog- ical parameters and the economic feasibility of oyster farming for this area. The principal question we address here: How do the biological parameters affect the economic model? What are the sensitive features of the model? We developed a computer based spreadsheet program in the user friendly spreadsheet environment of Microsoft Excel (computer adaptation to Excel was done by Ken Cooper, Kingston. WA). The program is demonstrated in this poster session. Profitability of a commercial venture depends on a myriad of factors. For aquaculture, these factors include biological characters as well as more classic ones such as labor, capital, etc. In this program we have tried to incorporate all the economic elements that we have identified over the course of our pilot scale oyster farm. We have also tried to keep the spreadsheet as flexible as possible, enabling the operator to experiment with the econom- ics of a "Ma-and-Pa" operation or with the economy of scale of a corporate giant. We have tried to incorporate several features to the economic analysis that are not immediately obvious to a nov- ice. For example, on the east coast of the US, overwintering oysters, even in the mid-Atlantic, is a critical consideration. In this program we provide a worksheet for figuring these costs as part of the overall costs. The purpose of this program is to provide the prospective oyster farmer with a preliminary estimate of the prof- itability of a particular operation. Basically, the program is de- signed to demonstrate if a particular aquaculture project is a good investment. From a financial point of view, does it make sense to tie up one's money and time in the enterprise? OPTIMIZING METAMORPHOSIS AND SURVIVAL FOR LAB STUDIES OF MULINIA LATERALIS. Ann Arseniu, Lyu Suifen, and Standish K. Allen Jr., Haskin Shellfish Re- search Laboratory, Department of Marine and Coastal Sciences, Rutgers University, Box B-8. Port Norris. NJ 08349. In 1969, Calabrese described the dwarf surfclam, Mulinia lat- eralis, as the "molluscan fruit fly", an observation still meaning- ful today. Its value to our lab is the short generation time (60-90 days) depending on conditions that allows genetic studies. For example, we are using Mulinia as a model system for the inves- tigation of gynogenesis in bivalves. In general, Mulinia is easy to raise in the lab. But optimizing the conditions for rearing Mulinia to adulthood are important to eliminate excessive selection of lab stocks and for rearing rare genetic variants to adulthood. We in- vestigated the effect of sediment on the rate of metamorphosis by rearing seven replicate crosses simultaneously. The seven replicate cultures were reared for eight days and each was split between 15<' culture containers with or without sediment. Sediment did not increase the rate of metamorphosis significantly; replications with sediment had slightly fewer spat than those without. However, clams measured two weeks after settlement were significantly larger and there were fewer dead. We then compared the survival and growth of newly set juveniles in either static seawater. static seawater with air. or flowing seawater. Flowing seawater (—1(1 hour) produced significantly more and larger juveniles after 2 weeks. Our conclusion from these studies is that conditions that mimic natural conditions for Mulinia are preferable to standard hatchery techniques for bivalves. Studies to maximize survival from juveniles to adulthood are ongoing. THE ROLE OF RESEARCH AND DEVELOPMENT IN A COMPETITIVE DOMESTIC AQUACULTURE INDUS- TRY. Sebastian Belle, New England Aquarium. Central Wharf. Boston. MA 02110. Aquaculture is an increasingly technical field. Artisanal culture methods are rapidly evolving into a science. Research and devel- opment plays a critical role in this transition. The impact of re- search and development on a viable domestic aquaculture industry is discussed. The U.S. experience is compared with that of other countries. Pure and applied research are characterized and their relative contributions to commercial development examined. Re- search strategies in the private and public sectors are compared. The relationship between initial innovation, commercialization and competitive viability is discussed. DETECTION OF FISH PATHOGENS. Joseph Buttner,' Pel Chang, ^ Paul Bowser,' Frank Hetrick,'' Philip McAllister,^ Bruce Nicholson,* and Paul Reno,' 'Department of Biological Sciences. SUNY College at Brockport. Brockport. NY 14420; "Department of Fisheries. Animal and Veterinary Science. Uni- versity of Rhode Island. Kingston. RI 02881; ''Department of Avian and Aquatic Animal Medicine, College of Veterinary Med- icine, Cornell University, Ithaca, NY 14853; ''Fish Disease Lab- oratory, Department of Microbiology, University of Maryland, College Park, MD 20742; 'NFRHL, U.S. Fish and Wildlife Ser- vice, Kearneys ville, WV 25430; ''Aquatic Animal Health Labo- ratory, Maine Animal Health Laboratory, University of Maine, Orono, ME 04469; 'Hatfield Marine Science Center, Oregon State University, Newport, OR 97365. Diseases, particularly bacterial and viral, cost the United States aquaculture industry millions of dollars in losses annually. Good management practices can avoid many disease problems, but not all. When a disease problem appears, an accurate and prompt diagnosis is essential to initiate an effective and appropriate cor- rective action. Current disease diagnostic procedures almost al- ways require sacrificing the fish. Non-destructive techniques are being developed and evaluated for by the Northeastern Regional Aquaculture Center. Investigators have used blood, ovarian fluid, or mucous and survival-surgery procedures to obtain kidney and liver specimens. Quick and accurate immunological assay meth- ods, recombinant DNA technologies, and viral plaque assays are being developed. Milford Aquaculture Seminar. Milford. Connecticut Ahsirucis. 13th Annual Meeting, February 22-24, 1993 107 POSSIBLE EUTROPHICATION-RELATED ENHANCE- MENT OF THE MICROBIAL LOOP IN LONG ISLAND SOUND AND CONSEQUENCES FOR SHELLFISH. Gerald M. Capriulo,' Robert Troy,' Marcelo Morales,' Kathleen Beddows,' Helen Budrock,' Gary Wikfors.^ and Charles Yarish,"' 'Environmental Science Department. State University of New York Purchase. NY 10577; "National Oceanic and Atmo- spheric Administration. National Marine Fisheries Service. North- east Fisheries Science Center. Milford Laboratory. 212 Rogers Avenue. Milford. CT 06460; 'Department of Ecology and Evo- lutionary Biology, University of Connecticut. 641 Scofieldtown Road. Stamford. CT 06903. Laboratory and field studies have shown that changes in water column chemistry related to absolute levels and relative rations of important nutrients, alter the species composition of water column plankton communities towards smaller, microbial forms. In ma- rine/estuarine waters, nitrogen, and to a lesser extent silicate, lev- els control phytoplankton growth, and N/P and N/Si ratios affect species composition. Therefore, anthropogenic inputs of N stim- ulate excessive algal growth, and select for smaller-sized phyto- planktonic species. A fundamental question arising from this is why the excess algal and other biomass is not enhancing secondary production of a quality leading to finfish and shellfish production in coastal systems such as Long Island Sound. We believe the answer to this question lies in the fact that food web dynamics in the western Long Island Sound have been anthropogenically shifted toward a microbial loop dominated system as compared to the more traditional food web dynamics of the central to eastern Long Island Sound. If such a fundamental shift has occurred in the western Sound, then both fmfish and shellfish production are be- ing negatively affected. We herein report preliminary results of an ongoing, comprehensive, baseline data study, in which we are comparing microbial loop and related planktonic food web dynam- ics in the western versus central Long Island Sound. CONTROL OF OVERSET ON CULTURED OYSTERS US- ING BRINE SOLUTIONS. Gregory A. Debrosse and Standish K. Allen, Jr., Haskin Shellfish Research Laboratory. Department of Marine and Coastal Sciences. Rutgers University. Box B-8. Port Norris. NJ 08349. HSRL has a long standing program in oyster genetics and breeding. One of the worst scenarios for maintenance of brood- stocks is overset by native oysters. Preliminary experiments in 1990 indicated that overset might be controlled simply by immers- ing animals in a concentrated brine solution; such treatments in 1990 resulted in 89-100% mortality of <1 mm spat. In 1990 field tests, overset on broodstocks was reduced to 3 spat/oyster using 200 ppt immersions compared to 22 spat/oyster in controls. In 1991 we refined the parameters for effective brine dips. First, we tested survival of oysters (potential substrate for overset) im- mersed for 2, 5, or 10 minutes in 200 ppt brine followed by either 3 or 6 hours aerial exposure. For juveniles, cumulative mortalities ranged from 3-6% compared to 5% in controls; for adults, 2^% died after brine immersion and 2-3% died in controls. Second, we tested survival of hatchery set oyster spat immersed in 200 ppt brine. For spat with shell lengths <5.0 mm and immersed in 200 ppt brine for 2, 5, or 10 min, 57%, 70% and 83% died after 3 hr aerial exposure and 64%-, 85%!, and 86% died after 6 hr aerial exposure. Control mortality averaged about 23% in both 3 and 6 hr aerial exposures. For larger spat immersed in 200 ppt brine for 10 minutes, cumulative mortality was 47% and 88%^ for 3 and 6 hr aerial exposure, respectively, and 22%: and 32% for controls. Re- sults of 1990 field tests and 1991 experiments demonstrate that brine solutions will be effective and save considerable labor. JUVENILE OYSTER MORTALITY STUDIES— 1992: HIS- TOPATHOLOGY, PATHOLOGY, EPIZOOTIOLOGY. C. Austin Farley and E. J. Lewis, National Marine Fishenes Ser- vice. NCAA. Northeast Fisheries Science Center, Cooperative Oxford Laboratory, Oxford, MD 21654. Studies of cytology, pathology, and population characteristics were conducted in relation to mortalities of Long Island Sound hatchery-reared juvenile oysters. Studies included major mortality periods of July-September in both 1991 and 1992. Data have been analyzed and support information reported previously by others suggesting size and temperature in relation to onset of disease and mortality. Dead oysters typically were less than 30 mm in length (mean 16-20 mm). Depending upon water temperature, mortali- ties in oysters occurred 3 to 8 weeks after being transplanted from the hatchery and maintained in trays in the nursery. Oysters from the nursery experienced 4—66%) mortality with conchiolin deposi- tion. Representative oysters from each spawning batch kept in the hatchery, in 25-p.m filtered ambient water diluted with high sa- linity well water, suffered 0-8%; mortalities with conchiolin de- position. Epizootiology studies of variously treated juvenile oyster populations further suggest that an infectious entity is responsible for mortalities. As in our earlier studies, histological tissues re- vealed the presence of small, round intracellular bodies in lesions of the mantle epithelium in 60-90% of populations experiencing >50% mortality. We believe these bodies to be a parasite, not autophagic vacuoles or necrotic host cells as others have sug- gested. Tissues stained with Feulgen picromethyl blue revealed that many of these bodies possess multiple dense staining Feulgen- positive structures resembling developmental life cycle stages of protists. particularly ciliates. Intracellular parasites with protistan characteristics were found by electron microscope studies. Mitochondria with tubular cristae. small nuclei, indications of a pellicle in some, and suggestions of endogenous budding similar to that seen in suctorian ciliates were seen. Similar intracellular organisms were seen in large commen- sal ciliates in spaces between the mantle and shell, suggesting a possible carrier host role. These large ciliates would not pass a 25-(jLm filter, explaining the protection of comparable populations held in the hatchery. 108 Abstracts. 13th Annual Meeting, February 22-24, 1993 Milford Aquaculture Seminar, Milford, Connecticut RECENT OUTBREAKS OF DERMO DISEASE IN THE NORTHEAST NEW INTRODUCTIONS OR CLIMATE CHANGE? Susan E. Ford, Haskin Shellfish Research Labora- tory, Department of Marine and Coastal Sciences, Rutgers Uni- versity, Box B-8, Port Norris, NJ 08349. Since 1990, the protozoan Perkinsus marinus. cause of Dermo disease in the eastern oyster, Crassostrea virginica, has appeared in numerous northeastern areas where it had not previously been detected or caused mortality. Hypotheses offered to explain the phenomenon include recent introductions by movement of infected oysters from southern waters; the appearance of a new low tem- perature tolerant strain of the parasite; and a change to a more favorable environment. Importation of a large number of P. mciriniis-mfecled oysters into Delaware Bay during the 1950s failed to establish a self- sustaining parasite population, which declined after importations ceased. Occasional findings of infected oysters over the following 35 years, however, suggest that the parasite remained in the Bay, but at undetectable levels. The P. marinus introductions of the mid-1950s occurred during a period of average or below average temperatures. In contrast, the 1990-1992 epizootic coincided with a period, beginning in Janu- ary 1990 and lasting until March 1992, in which monthly mean air temperature in southern New Jersey, adjacent to Delaware Bay, was consistently higher than average, often by several degrees Celsius. Similar temperature deviations were recorded in more northern sites where epizootics of P. marinus have occurred for the first time. Above average winter temperatures appear to cor- relate better with P. marinus outbreaks than do high summer tem- peratures. It is difficult to believe that infected oysters have been intro- duced suddenly into multiple locations from New Jersey to Mas- sachusetts, including sites condemned for the harvest of shellfish, over the last two or three years. Historical records do, however, show that for many years during the century oysters were moved from south (where P. marinus is enzootic) to north along the Atlantic coast. The simplest explanation consistent with historical knowledge and current observations is that extraordinarily high temperatures beginning in 1990 stimulated the proliferation of ex- isting small foci of infection, which may have been present, but undetected, for years. STUDIES OF JUVENILE OYSTER MORTALITY ON LONG ISLAND SOUND, NY IN 1992. Susan E. Ford,' Fran- cisco J. Borrero,'^ and Walter J. Blogoslawski,^ 'Haskin Shell- fish Research Laboratory, Department of Marine and Coastal Sci- ences, Rutgers University, Box B-8, Port Norris, NJ 08349; "State University of New York, Marine Science Research Center, Stony Brook, NY 1 1794; 'National Oceanic and Atmospheric Adminis- tration, National Marine Fisheries Service, Northeast Fisheries Science Center, Milford Laboratory, 212 Rogers Avenue, Mil- ford, CT 06460. A study of juvenile oyster mortality was conducted over the summer of 1992 at two sites on Long Island, New York: F. M. Flower and Sons Oyster Co. (north shore of Long Island) and the Bluepoints Co. (south shore of Long Island). Major objectives of the study mcluded determining whether mortalities were associ- ated with a particular broodstock (either a genetic problem or a source of pathogen transmitted to offspring) or growout site, whether mortalities could be stimulated by experimental temper- ature elevation, and to document the association of tissue and shell abnormalities (or their absence) with mortalities (or their absence) in the various experimental treatments. High mortalities occurred at the F. M. Flower and Sons Oyster Co. even though a new broodstock (wild oysters from the Thames River, CT) was submitted for Oyster Bay, NY stock, offspring of which had suffered high mortalities in 1990 and 1991. Mantle lesions were found in oysters just before mortalities began but no evidence of a pathogenic protozoan was found. Oysters main- tained inside the F. M. Flower and Son Co. hatchery at elevated temperature (25°C) in a mixture of well water and 25-tJLm filtered bay water, supplemented with cultured algae, did not experience unusual mortalities. Wild spat from Connecticut (1991 year class) placed in Oyster Bay did not experience typical juvenile oyster mortalities, but some individuals were found to have abnormal conchiolin ring deposits. No losses occurred at the Bluepoints Co. site in offspring of both Oyster Bay, NY and Thames River, CT broodstock. Results of this and previous studies lead us to conclude that broodstock is not the problem and to suspect that affected juvenile oysters are reacting to a toxin, probably of bacterial or microalgal origin, which irritates the mantle edge causing abnormal shell matrix secretion, tissue damage, and eventual death. REPRODUCTIVE GENETICS OF TRIPLOID CRASSOS- TREA GIGAS. Ximing Guo and Standish K. Allen, Jr., Haskin Shellfish Research Laboratory, Department of Marine and Coastal Sciences, Rutgers University, Box B-8, Port Nortis, NJ 08349. Crassostrea gigas has been variously proposed as a replace- ment or supplement species for C. virginica in several east coast situations. Triploids potentially offer a "'safe" way to test C. gigas in the field. Are triploid C. gigas sterile? The genetics of repro- duction in triploid Pacific oyster, Crassostrea gigas. was exam- ined in matings between diploids (D), triploids (T), and their re- ciprocal crosses (D x T and T x D). Meiotic metaphases were examined in eggs of diploid and triploid eggs. Ploidy of embryos of all matings were determined by karyology and flow cytometry. Sperm from triploids showed a single distribution of DNA content at 1 .49c, as determined by flow cytometry; no haploid peaks were observed. Before fertilization, eggs from diploids had ten syn- apsed chromosomes. In eggs from triploids, chromosome numbers varied considerably within and among females: some were com- pletely synapsed to form ten trivalents, but most had between 11-13 trivalent and bivalent chromosomes. Gametes from trip- Milford Aquaculture Seminar. Milt'ord. Connecticut Ahsinuis. 13th Annual Meeting, February 22-24, 1993 109 loids were capable of fertilization and fertilization was about the same in all groups, probably limited only by the maturity of ga- metes. After fertilization, eggs of triploids went through two mei- otic divisions, releasing two polar bodies. Ploidy of embryos from the four types of niatings was determined by both flow cytometry and karyology to be 2n for D x D, 2.5n for D x T and T x D. and 3n for T x T. Survival to D-stage was about the same in all crosses, ranging from 32-66%. Survival to seven days post- fertilization was 40% for D x D, 0.5% for D x T. 8% fori x D. and 0.4% for T x T. Percent metamorphosis to spat was 23% for D x D, 0.001% for D x T, 0.058% for T x D. and 0.0% for T x T. EFFECT OF TREMATODES ON EAST COAST POPULA- TIONS OF THE BLUE MUSSEL, MYTILUS EDULIS. Rob- ert E. Hillman, Battelle Ocean Sciences, 397 Washington Street, Duxbury, MA 02332. During the course of histological examinations of gonadal de- velopment of Mytilus edulis populations for the National Oceanic and Atmospheric Administration's (NOAA) Mussel Watch Proj- ect, apparent gonadal abnormalities were observed in connection with the presence of trematodes in the tissues. In most cases, either the gonadal follicle tissues were disorganized and did not progress past an early stage of development, or no gonads were present at all. These observations prompted a more detailed look at the ef- fects of the trematode infestations on the mussel populations being monitored for the NOAA study. This paper is a preliminary report on the distribution and effects of the trematodes in mussel popu- lations on the U.S. east coast monitored over a seven-year period. A total of 49 sites have been sampled once annually from Maine to Delaware since 1986, although not each site has been sampled every year. At this point, no attempt has been made to thoroughly identify the species involved in the infestations, but there are at least two types of trematodes involved. The most common type, by far. appears to infest all tissues except the foot, and is responsible for much of the observed abnormal gonad de- velopment. The second type is found almost exclusively in the foot and does not appear to affect gonadal development. The intensity of infestation is highest in the New York Bight and Long Island Sound region. Gonadal abnormalities are from 10 to 26% higher at sites where trematode infestations occur than at sites where there were no observed trematodes. inorganic contaminants, and examined histologically for evidence of neoplasia. Neoplasias were found in mussels from 6 east and 12 west coast sites. With the exception of 2 germinomas, all neopla- sias were disseminated neoplasias. Significantly higher concentra- tions of combustion-related and total PAHs, f/i-chlordane, pesti- cides, and cadmium were found at east coast sites where neopla- sias occurred than at sites where no neoplasias were found. Arsenic was found at higher concentrations at non-neoplasia sites than at neoplasia sites. On the west coast, significant differences were observed for combustion-related and total PAHs, total PCBs, and lead. Cadmium, chromium, and mercury were higher at non- neoplasia sites than at neoplasia sites. A step-wise regression anal- ysis was performed to determine those contaminants whose con- centrations significantly affected the presence of neoplasia. On the east coast, the negative effect of arsenic was overwhelmingly sig- nificant compared to the effects of all other contaminants. On the west coast, combustion-related PAHs contributed to high proba- bilities of neoplasia, while chromium and indene-based pesticides were significantly negatively correlated with the occurrence of neoplasia. CYTOLOGICAL AND CYTOGENETIC EXAMINATION OF GAMETOGENESIS IN TRIPLOID CRASSOSTREA VIR- GINICA AND CRASSOSTREA GIGAS. Ya-Ping Hu and Standish K. Allen, Jr., Haskin Shellfish Research Laboratory, Institute of Marine and Coastal Sciences, Rutgers University. Box B-8, Port Norris, NJ 08349. Crassostrea gigas has been variously proposed as a replace- ment or supplement species for C. virginica in several east coast situations. Triploids potentially offer a "safe" way to test C . gigas in the field. Are triploid C. gigas sterile? Gametogenesis of trip- loid C. gigas were compared to a triploid C. virginica group with both flow cytometric and histological techniques. A total of 1 18 C. virginica and 144 C. gigas were sampled from early May to Au- gust, 1992, at bi-weekly intervals. No haploid sperm were re- corded in any male triploid, but all male triploids were capable of producing 1.5 N aneuploid sperm. Overall the gonadal develop- ment and gamete production in both male and female triploids were significantly decreased in terms of quality and quantity rel- ative to diploids. The present study also documented a correspon- dence between the male gamete type (histology) and the relative DNA content (fiow cytometry). RELATIONSHIP OF ENVIRONMENTAL CONTAMI- NANTS TO OCCURRENCE OF NEOPLASIA IN MYTILUS EDULIS POPULATIONS FROM EAST AND WEST COAST MUSSEL-WATCH SITES. Robert E. Hillman, Battelle Ocean Science, 397 Washington Street. Duxbury, MA 02332. Over 8,000 mussels, Mytilus edulis, collected for the National Oceanic and Atmospheric Administration's Mussel Watch Project from approximately 80 sites along the U.S. east and west coasts from 1986 through 1991 were analyzed for levels of organic and PARASITES OF THE BAY SCALLOP, ARGOPECTEN IR- RADIANS. John D. Karlsson, Rhode Island Coastal Fisheries Laboratory. 1231 Succotash Road. Wakefield, RI 02879. Parasites observed during histological examination of approx- imately 2500 bay scallops, Argopecten irradians Lamarck, col- lected between 1983 and 1985 from wild populations in two coastal ponds in Rhode Island, are reported. In addition to a re- view of previously reported bay scallop diseases, occurrence of these diseases in Rhode Island waters is documented, and several 110 Abstracts. 13th Annual Meeting, February 22-24, 1993 Milford Aquaculture Seminar, Milford, Connecticut newly discovered disease conditions are described, including two ovarian parasites which appear to cause parasitic castration. OCCURRENCE OF OSTREA EDULIS IN RHODE ISLAND. John D. Karlsson and Arthur R. Ganz, Rhode Island Coastal Fisheries Laboratory, 1231 Succotash Road, Wakefield, HI 02879. In recent years a number of European oysters, Ostrea edulis. have been found in Rhode Island waters. There is evidence that observed occurrences result from larval settlement rather than the release of post-metamorphic animals. A SIMPLE METHOD FOR THE IN VITRO CULTURE OF PERKINSUS MARINUS. S. J. Kleinschuster and S. L. Swink, Haskins Shellfish Research Laboratory, Rutgers University, Box B-8, Port Norris, NJ 08349. Cells tentatively identified as Perkinsus mariims were origi- nally identified as a contaminant of primary tissue culture explants of visceral ganglia from Crassostrea virginica. Following sterile isolation, various mixtures of Leibowitz's medium (L-15), oyster hemolymph and fetal calf serum (FBS) were tested for cell growth potential. The osmolarity of each constituent was adjusted to 750 mOs/kg by the addition of sea salts. The pH of each medium and/or constituent was then adjusted to 7.6 and sterilized when necessary. As might be expected, cells cultured with a high proportion of hemolymph in the medium (50%), displayed vigorous propaga- tion. Alternately, those cells cultured with a high proportion of L-15 and/or FBS tended to attain sporangial morphology with many daughter cells. Anomalous morphology of cells was com- mon in cultures with high L-I5/FBS concentrations, especially in older cultures. Original cultures were subcultured several times over several months and the medium exchanged (50%) weekly. Cultures so obtained were challenged with sterile oyster tissue which became infected within 2-3 weeks. Additionally, cells from challenged and infected tissue formed hypnospores and tested pos- itive upon exposure to Lugol's solution following culture in fluid thioglycolate. The optimum medium for growth and differentia- tion in this study consisted of 5.0 mg taurine, 50.0 mg glucose, 30.0 mg galactose, 50.0 mg fructose, 50.0 mg trehalose, 300.0 mg lactalbumin, 100.0 mg yeast extract, 1.0 ml vitamin mixture (Sigma Chemical Co,), 0.1 ml lipid mixture (Sigma Chemical Co.), 20.0 ml FBS (Sterile Systems Inc.) and 80.0 ml L-15 Var- ious aliquots of hemolymph may be substituted for L-15. The authors wish to thank Dr. F. O. Perkins for his invaluable assis- tance with this study. OVERVIEW OF THE OPERATIONS OF ATLANTIC LIT- TLENECK CLAMFARMS. Kenneth P. Kurkowski, Atlantic LittleNeck ClamFarms, P.O. Box 12139, Charleston, SC 29422. Atlantic LittleNeck ClamFarms is a fully integrated commer- cial aquaculture company located south of Charleston, S.C. It began production in mid-1991, with facilities consisting of a 15,000-ft- hatchery building, an 11,000-ft- nursery, a 1 ,500-ft- field house and an 8,000-ft' fabrication shop. Seawater from Folly Creek is settled for 36 hours and filtered to 1 micron for the hatchery. It is further pasteurized prior to use in algal culture. Three-thousand broodstock clams are kept at 19°C in the condi- tioning system to provide for year-round spawning. Clams se- lected for spawning are suspended in trays in a larval tank at 28°C and allowed to spawn for two hours before being removed. After seven days in larval culture the pediveligers are transferred to downwellers in the post-set system for three to five weeks. The algal system produces 13,000 liters daily for feeding larvae, post- set and broodstock. Thirty-million 1-mm seed are transferred to either a land-based or pond nursery monthly. Two to four months in passive upwellers yield 200 million 4 to 6-mm clams ready for field growout in clam pens annually. Initially the clams are stocked at 1.000 per ft". They are harvested 10 months later at 20-25 mm and replanted at 80 per ft". Sixteen months later 140 million 50-mm clams will be harvested annually. Upon harvest, the clams are depurated for 48 hours and tested for heterotrophic bacteria by our own certified shellfish lab before being sent to market. PRELIMINARY RESULTS OF LABORATORY AT- TEMPTS TO TRANSMIT A DISEASE AFFECTING JUVE- NILE OYSTERS IN THE NORTHEASTERN UNITED STATES. Earl J. Lewis, Jr. and C. Austin Farley, NCAA National Marine Fisheries Service, Northeast Fisheries Science Center, Oxford Laboratory, Oxford, MD 21654. Since the late 1980s, juvenile oysters from areas in northeast- em United States have experienced heavy mortalities. As yet, the cause of these mortalities has not been resolved, although many possible causes have been hypothesized. Our hypothesis is that this is an infectious disease process, with mortalities possibly re- sulting from pathology associated with a protistan parasite. Based on this, transmission experiments were designed to determine if the disease could be transmitted under controlled laboratory con- ditions. Preliminary results of ongoing experiments support this hypothesis. Depending upon water temperature, Maryland hatchery-reared oysters held in recirculating aquaria showed mortalities with char- acteristic heavy conchiolin deposition within 3 to 7 weeks of ex- posure to infected oysters from Long Island Sound. Cumulative mortality in experimentally challenged oysters ranged from 40% {18°C) to 74% (24°C). Associated conchiolin deposition was present in 26% of dead oysters held at 18°C, compared to a high of 40% at 24°C. No indications of dinotlagellates were evident in water samples examined upon completion of the study. "Little round bodies'" resembling what we have considered previously to be a parasite were observed in "gapers" processed for histology. No conchiolin or comparable mortalities were observed in con- trol animals. Gross symptoms of the disease were observed to recur in sur- Milford Aquaculturc Seminar. Milford. Connecticut Ahslnuts, 13th Annua] Meeting. February 22-24, 1993 111 vivors of the 1990 and 1991 mortalities when held in aquaria for 10 months. THE DIFFERENTIAL EFFECTS OF THREE HEAVY MET- ALS ON PARTICLE FILTRATION AND AMINO ACID UP- TAKE BY THE PACIFIC OYSTER, CRASSOSTREA GIGAS. Wenyu Lin and Michael A. Rice, Department of Fisheries. An- imal and Veterinary- Science. University of Rhode Island. Kings- ton. Rl 0288 1 ; Paul K. Chien, Department of Biology. University of San Francisco, San Francisco. CA 941 17. The effects of copper, cadmium and zinc on rates of particle filtration and glycine uptake by Crassostrea gigas were studied. Constant filtration rates were induced in oysters by irrigating the mantle cavity with flowing seawater from a peristaltic pump at a rate of 2.5 t/h. The filtration rate (volume of water completely cleared of colloidal carbon per unit time) by control oysters was 26.60 m<^/gh ± 7.68 (SD). Filtration rates decreased with increas- ing concentrations of cadmium and zinc. In lower concentrations of copper (8-16 mg/f ) filtration rates were significantly higher than the control, but higher copper concentrations reduced filtra- tion. Influx of glycine is characterized by Michaelis-Menten Ki- netics with }^^^ and K, values of 1.85 )j.mol/gh and 33.7 ^.M respectively. The degree of inhibition of glycine uptake in oysters exposed to metals was in the order of copper > cadmium > zinc. At 128 mg/f copper, glycine uptake was reduced to 10.5% of the control. The rate of glycine uptake by filter feedmg bivalves is highly dependent on water pumping rate. The volume-specific glycine transport (amount of glycine transported/unit volume of seawater completely cleared of colloidal carbon) by control oysters in 1 jjlM glycine concentrations was 1 .03 (xmolf . The volume- specific glycine transport remained constant at increasing zinc concentrations, and declined at increasing copper concentrations, suggesting differential effects of the metals on particle filtration and the amino acid carriers. The apparent volume-specific glycine transport increased to 2.14 jjLmol/€ in 128 mg/( cadmium. This volume-specific transport which was greater than the glycine con- centration in the medium suggests that there may have been uptake of cadmium-complexed glycine by the oysters. "NON-TOXIC" DINOFLAGELLATE BLOOM EFFECTS ON OYSTER CULTURE IN CHESAPEAKE BAY: PRELIM- INARY RESULTS. Mark Luckenbach,' Sandra Shumway,^ and Kevin Sellner,' 'Virginia Institute of .Marine Science. Col- lege of William & Mary, Wachapreague, VA 23480; "Bigelow Laboratory for Ocean Sciences, West Boothbay Harbor, ME 04575; 'Benedict Estuarine Research Laboratory, The Academy of Natural Sciences, Benedict, MD 20612. Dinoflagellate blooms appear to be increasing in frequency, magnitude and duration in Chesaf)eake Bay. During 1992 we doc- umented dinoflagellate blooms of unprecedented intensity and dis- tribution in the southern portion of Chesapeake Bay. Though the species involved in these blooms are generally termed nontoxic (from a public health perspective), their effects on suspension- feeding bivalves, including oysters, may be anything but benign. As the oyster aquaculturc industry continues to grow in Chesa- peake Bay. the effects of these blooms may become increasingly important. Field and laboratory experiments were conducted to evaluate impacts of several dinoflagellate bloom species on the feeding, growth and survival of Crassostrea virginica. Hatchery-spawned oysters from a single cohort were deployed in off-bottom culture at twelve locations exhibiting varying degrees of bloom develop- ment, and growth and survival monitored. Laboratory experiments of 4 to 6 weeks duration were used to evaluate growth and survival of juvenile oysters fed monocultures of dinoflagellates and the diatom Thalassiosira weissflogii. Flow cytometry was used to de- termine grazing rates in short-term feeding trials, and fecal and pseudofecal composition to assess utilization of dinoflagellates by oysters. Results from both the field and laboratory suggest that growth and survival of juvenile oysters are affected by these di- noflagellate blooms. Our findings indicate potentially significant impacts on oyster culture in this region as a consequence of di- noflagellate blooms. NORTHEASTERN REGIONAL AQUACULTURE CEN- TER: AN UPDATE. Victor J, Mancebo, University of Massa- chusetts Dartmouth, Research 201. North Dartmouth. MA 02747. The Northeastern Regional Aquaculturc Center (NRAC), head- quartered at the University of Massachusetts Dartmouth, is one of five Regional Aquaculturc Centers (RACs) established by the U.S. Congress. Funded by the U.S. Department of Agriculture at an annual level of approximately $750, (X)0, and representing 12 states and the District of Columbia. NRAC develops and sponsors co- operative regional research, development and extension projects in support of the aquaculturc industry in the northeastern United States. A Board of Directors representing the region's aquaculturc in- dustries, academic institutions, and government agencies provides overall direction and management of NRAC. NRAC programs, like those of all the RAC"s are industry-driven, i.e.. industry com- municates research and technology transfer priorities to NRAC through bi-annual industry summits and through NRAC's 12- member Industrial Committee. A 12-member Technical Commit- tee provides technical oversight for NRAC's projects. Projects supported by NRAC are developed and carried out by Cooperative Regional Work Groups with researchers, extension specialists and industry representatives working together with multi-state and multi-institutional participation on each project. Projects are eval- uated annually for achievement of technical and industry objec- tives. NRAC has recently completed four major projects on genetic improvement and manipulation of oysters, finfish economics, and the development of a regional aquaculture extension program. Ter- mination reports are being prepared and relevant findings will be 112 Abstracts, 13th Annual Meeting, February 22-24, 1993 Milford Aquaculture Seminar, Milford, Connecticut disseminated. Ten regional projects are ongoing with areas includ- ing genetic manipulation of striped bass, domestication of striped bass, government regulations affecting aquaculture, finfish nutri- tion, commercial field trials of selected oyster strains, fish health, water quality and waste management, marketing options, oyster larval development and a regional industry situation and outlook report. Tiiree smaller projects designed to avail of rapid response funds are also ongoing. NRAC is currently reviewing three pro- jects for 1993 funding including an economic impact study of government regulations, a followup regional extension project, and a computer communication network. A project on quality assurance in aquaculture will soon be developed for funding in 1993. Total NRAC funding commitment to projects in progress or pending is approximately $1.8 million. NRAC also publishes "Northeastern Aquaculture", a quarterly newsletter highlighting NRAC projects and other topics of interest to the northeastern aquaculture community. SHRIMP CULTURE IN THE PHILIPPINES: BIRTH OF THE INDUSTRY. Victor J. Mancebo, University of Massachu- setts Dartmouth, Research 201. North Dartmouth. MA 02747. Production of farmed tropical shrimp underwent explosive growth in the decade of the 80s. A major contributing factor to the growth of the industry was the successful larval rearing and de- velopment of feeds for Penaeus monodon which occurred in Tai- wan and was led by Dr. 1-Chiu Liao. In 1978 a successful technology transfer for P. monodon hatch- ery, grow-out and feed-production methodology took place from Taiwan to the Philippines. The technology transfer was negotiated by Dr. Liao, President, Enterprise Corporation in Taiwan and San Miguel Corporation in the Philippines. Followmg a time lag for site selection and construction, a show-case facility was built in the Philippines and the first successful harvest demonstrating Tai- wan technology took place in 1982. At the time of the first harvest the total production of P. monodon in the Philippines was esti- mated at 1,000-1,500 metric tons. Within six years (1988) pro- duction of P. monodon had increased to approximately 40,000 met- ric tons, an increase largely attributed to San Miguel's technology demonstration and dissemination. AQUACULTURE IN THE NORTHEAST REGION OF NMFS. Harold C. Mears, National Oceanic and Atmospheric Administration, National Marine Fisheries Service, Northeast Re- gional Office, One Blackburn Drive, Gloucester, MA 01930. The genesis of aquaculture research in the Northeast Region of the National Marine Fisheries Service (NMFS) was prior to Re- organization Plan No. 4 of 1970 at which time the Bureau of Commercial Fisheries was renamed NMFS, and transferred to the Department of Commerce. The first permanent assignment of a full-time biologist and plans to establish a laboratory in Milford, Connecticut, occurred in 1931. The development of methods for commercial shellfish cultivation began in 1944. During that de- cade, federal biologists established procedures for conditioning eastern oysters to ripeness, inducing spawning and fertilization, rearing larvae, determining food requirements, and growing newly-set spat. Activities at this facility have continued on a va- riety of subjects including natural diets, genetics, disinfection techniques for hatchery water, and culture methods for eastern oysters, bay scallops, and Atlantic surtclams. During the past 25 years, approximately S18 million have been devoted by NMFS-administered programs to aquaculture-related research. In cooperation with state fishery resource agencies, aca- demia, the fishing industry, and other private interest, NMFS grant programs have supported projects covering a range of activ- ities from bay scallop culture techniques to eastern oyster shell planting to assessment of hatchery wastewater systems. During 1964—1986, under the Commercial Fisheries Research and Devel- opment Act (P.L. 88-309). aquaculture was divided among six primary categories: marine fish and shellfish culture (71.8%); en- vironmental monitoring (10.7%); aquaculture systems (7.8%); freshwater fish and invertebrate culture (3.8%); restoration/ fisheries enhancement (3.2%); and processing technology (2.7%), Aquaculture-related activities have also included NMFS par- ticipation and responsibilities in the areas of habitat conservation, trade and industry services, and fishery restoration programs. The NMFS Strategic Plan calls for the reduction of impediments to U.S. aquaculture. and a re-evaluation of the NMFS' role in achieving that goal. An internal Task Force is currently conducting that assessment. ISOLATION OF INFECTIOUS PARTICLES HAVING RE- VERSE TRANSCRIPTASE ACTIVITY AND PRODUCING HEMATOPOIETIC NEOPLASIA IN MYA ARENARIA. Daniel J. Medina, Department of Medical Oncology. Yale Uni- versity. New Haven. CT 06510; Gregory E. Paquette, Eileen C. Sadasiv, and Pei W. Chang, Department of Fisheries. Animal and Veterinary Science. University of Rhode Island, Kingston, RI 02881. The causative agent of hematopoietic neoplasia (HN) of soft shelled clams. Mya arenaria. has not been defined, though most investigators agree that the disease is caused by a small transmis- sible agent Oprandy and Chang (1983) isolated a transmissible agent from HN clams. The agent measured 100 nm and had a buoyant density of 1.18 g/cm". When it was passed through a 450-nm filter, it induced HN in clams after a latent period of 2 months. The virus has been reisolated from moribund clams experimentally infected with the virus and reinjected into normal clams which subse- quently developed neoplasia, with the same virus again isolated. Oprandy described the agent as a retrovirus, based upon the mor- phology of the isolated particles. However, electron micrographs of clam tissues and neoplastic hemocytes have not been able to demonstrate budding virus particles and the PTA-stained particles found in purified preparations have not shown the surface pep- Milford Aquaculture Seminar, Milford, Connecticut Ahsimcts. 13th Annual Meeting, February 22-24, 1993 113 lomeres which can be characteristic of retrovirus. Attempts to unequivocally demonstrate retroviral presence in cells has not been successful so far. Since retrovirus can demonstrate reverse transcriptase (RT) ac- tivity, we have tested purified virus for the presence of RT. RT has been found. It was active at 6°C and was inactive above 25°C. Neoplasia was accompanied by metabolic alterations: increases in uric acid, aspartate transaminase and triglycerides; decreases in hemolymph urea. The neoplastic hemocyte cell membranes showed differences in lectin binding proteins, indicating a change in cell surface glycoproteins. DIFFERENCES IN THE SALINITY TOLERANCE MECH- ANISMS BETWEEN CHESAPEAKE BAY AND ATLANTIC COAST OYSTER: GENETICS OR DISEASE-INDUCED EF- FECTS ON MITOCHONDRIAL METABOLISM. Sidney K. Pierce, University of Maryland at College Park, Department of Zoology, College Park, MD 20742. Crassostrea virginica from Florida to Cape Cod respond to increased external salinity by increasing intracellular concentra- tions of several amino acids, primarily taurine, and the quaternary amine, glycine betaine. Chesapeake Bay oysters from several pop- ulations use different amino acids, primarily glycine and alanine, and in addition, do not synthesize glycine betaine in response to high salinity stress. Since the synthesis of both the amino acids and glycine betaine occurs in the mitochondria, we have been com- paring isolated mitochondrial metabolism of Bay and Atlantic oys- ters. The respiratory coupling ratios (RCR) of Bay oysters is al- ways higher than in Atlantic oysters, regardless of biochemical substrate. Bay oyster RCRs are highest with a-ketoglutarate, while malate is preferred by Atlantic mitochondria. In addition, mitochondria from low salinity adapted oysters take up choline (glycine betaine precursor) faster than high salinity adapted oysters and Atlantic mitochondria take it up faster than Bay mitochondria. The synthesis of glycine betaine is faster in high salinity adapted Atlantic oysters. We are currently measuring synthesis in Bay oyster mitochondria. These differences in amino acid production, RCRs and glycine betaine metabolism indicate major biochemical differences between the mitochondria of the two oyster groups. Since all of our Bay oysters are likely parasitized with Dermo, it is not clear if the differences are due to genetics, the presence of the parasite or some other environmental factor. FOOD LIMITED GROWTH AND CONDITION INDEX IN CRASSOSTREA VIRGINICA AND ARGOPECTEN IRRADI- ANS. Robert B. Rheault, Spatco, Ltd., 264 Foddering Farm Rd., Narragansett, RI 02882. Oysters {Crassostrea virginica) and bay scallops (Argopecten irradians) were held in flumes with flowing seawater pumped from Point Judith Pond for six weeks in August and September of 1992. Flow to each flume was held constant and ration available was continuously monitored with flow-through fluorometry. The experiment was designed to define the relationships of growth and condition index (CI) to both the available and consumed ratios. These relations can then be used to derive a horizontal seston flux model for oyster growth. Filter feeding rapidly depleted down- stream food concentration in the flumes reducing fluorescent par- ticulate material by 90% or more. Downstream oysters and scal- lops responded to the depletion in available ration with significant decreases in incremental growth and condition index. Simultaneous experiments were conducted in nearby bottom cages to evaluate the growth and CI response to seston depletion caused by varying the initial planting density in a commercial aquaculture setting. Oysters planted in mesh bags at high densities (6.8-10.5 kg/m*) demonstrated significantly lower CI and growth of the animals planted at 2.7-3.6 kg/m~. Data indicate the possibility of a synergistic feeding interaction between scallops and oysters when cultured in close proximity. Scallops grown with oysters appear to grow faster than when grown with scallops alone. Individual condition index response to shell irritations caused by Polydora websterii (mud blisters), 'unexplained juvenile oyster mortality syndrome" and "bag scars' will also be discussed. Oys- ters with bag scars or survivors of the 'unexplained mortality' had significantly elevated (33% higher) CI over normal or Polydora infested animals. IPNV ANTIBODY AS A MEANS OF DETECTION OF POS- SIBLE VIRUS CARRIAGE IN ATLANTIC SALMON SUR- VIVING VIRUS CHALLENGE. Eileen C. Sadasiv, Pei W. Chang, and Wenyu Lin, Department of Fisheries, Animal and Veterinary Science, University of Rhode Island, Kingston, RI 02881. Prudent aquacultural practice suggests the use of salmonid broodstock which have not been exposed to certain infectious agents which might be capable of causing high levels of mortality in progeny. Certification of broodstock to be free of the pathogens can require sacrifice of a portion of the stock. This study was undertaken to determine the feasibility of non-lethal pathogen de- tection methods. We propose that antibody can be an indicator of infection or exposure. Atlantic salmon Salmo salsar L (AS) from presumptively virus- free broodstock were raised in aquaria supplied with deep well water and acclimated to 6°, 10° and I6°C. A total of 96 fish, aged either 17 or 27 months were used. Fish initially had no antibody and no virus was isolated from them. They were immersed in 10^ TCID5f| of tissue culture produced (CHSE) infectious pancreatic necrosis virus, strain WB (IPNV) per ml of water and monitored for 193 days. Fish at all three temperatures produced a similar level of antibodies, with some retardation noted at lower temper- atures. IPNV was found to persist in the kidneys of fish having circulating antibody, along with measurable levels of what appears to be virus-specific antibody, as detected by both virus neutraliza- tion in tissue culture and by ELISA. 114 Abstracts. 13th Annual Meeting, February 22-24, 1993 Milford Aquaculture Seminar, Milford, Connecticut Testing of a limited number of sexually-mature AS has shown variation in antibody. In at least four of five ocean-returned post- spawn AS. antibody was detected at higher levels than that found in similarly aged hatchery-held AS. IMPACT OF HARMFUL ALGAL BLOOMS ON SCALLOP CULTURE AND FISHERIES. Sandra E. Shumway, Bigelow Laboratory for Ocean Sciences, Department of Marine Resources, West Boothbay Harbor, ME 04575; and Allen D. Cembella, Na- tional Research Council. 1411 Oxford Street. Halifax. Nova Scotia. Canada B3H 3Z1. Harmful algal blooms occur worldwide and their associated phycotoxins are accumulated by filter-feeding bivalve molluscs. Since only the adductor muscle of scallops has been traditionally marketed, scallops are not usually included in routine monitoring programs. A renewed interest in marketing both whole and "roe- on" scallops from various geographic regions along with intensi- fied aquaculture ventures in areas prone to toxic blooms have provoked public health concerns regarding the safety of this re- source. Our studies have focused on the sequestering and biotransfor- mation of phycotoxins in scallops. Our results, coupled with a review of historic data, indicate that: 1 ) toxins are not distributed evenly throughout the scallop tissues; most toxin is usually con- centrated in the mantle and digestive gland; 2) some scallop tis- sues, e.g. digestive glands and mantles remain highly toxic throughout the year; 3) toxicity varies considerably (±43.5%) be- tween individual animals collected in the same area; 4) no corre- lations could be made between toxicity levels in gonadal tissue and other tissues. Scallop culture and commercial fisheries can thrive in areas prone to toxic algal blooms if only the adductor muscle is utilized. Safe marketing of "roe-on" scallops is feasible only under strict regulatory regimes. Marketing of mantles or whole scallops poses a high risk to public health and should only be undertaken after extensive monitoring. Scallop mariculturists should be acutely aware of the potential risks associated with phycotoxins. Further, public health guidelines with particular emphasis on toxin levels in individual tissues is necessary if scallops are to be marketed whole or in conjunction with tissues other than adductor muscles. VIABILITY AND GENETIC EFFECTS ON OYSTER EM- BRYOS EXPOSED TO BACTERIA AND EFFLUENT FROM DISEASED JUVENILE OYSTERS FROM A LONG ISLAND HATCHERY. Sheila Stiles and Walter Blogoslawski, National Oceanic and Atmospheric Administration, National Ma- rine Fisheries Service, Northeast Fisheries Science Center, Mil- ford Laboratory, 212 Rogers Avenue, Milford, CT 06460. Viability and cytogenetic effects were invesfigated in oyster embryos exposed to disease-causing organisms isolated from a commercial hatchery having high mortalities of juvenile oysters iCrassoslrea virginica). Vibrio bacteria isolated from TCBS me- dia as yellow and green colonies at different concentrations, as well as a seawater wash from diseased juvenile oysters, were used to challenge embryos. No normal larvae developed after 48 hours in cultures at highest concentrations of bacteria; some normal lar- vae developed at lower concentrations, however, these were , smaller than control larvae. In these assays, vibrios from some colonies appeared more toxic than others. Although a greater percentage of embryos de- veloped into normal larvae in the seawater wash than in the cul- tures challenged with bacteria, mortality was higher than in the cultures challenged with bacteria. Cytogenetic and cytological ob- servations showed delayed and arrested development as evidenced by elevated frequencies of non-dividing cells in embryos. Results suggest that selective mortality occurs early in oysters after expo- sure to bacteria or wash water from diseased oysters. EFFECTS OF A PROROCENTRUM ISOLATE UPON THE OYSTER, CRASSOSTREA VIRGINICA: A STUDY OF THREE LIFE-HISTORY STAGES. Gary H. Wikfors,' Rox- anna M. Smolowitz,'^ and Barry C. Smith,' 'National Oceanic and Atmospheric Administration, National Marine Fisheries Ser- vice. Northeast Fisheries Science Center. Milford Laboratory, 212 Rogers Avenue, Milford, CT 06460; "LMAH, School of Veteri- nary Medicine, University of Pennsylvania, Marine Biological Laboratory, Woods Hole, MA 02543. Evidence that some strains of the dinoflagellate genus Proro- centnim are harmful to shellfish has been obtained from both field and laboratory studies. Our previous laboratory exposures of one Prorocentrum minimum isolate (strain EXUV) to hard clams and bay scallops demonstrated clear differences in responses of the two bivalves; hard clams survived but did not grow, whereas scallops experienced complete mortality in 1^ weeks. Histological evi- dence suggested effects of an enterotoxin upon scallops. The present study was undertaken to determine possible toxicity of cultured P minimum (EXUV) to several life-history stages of the eastern oyster: embryos, feeding larvae, and juveniles. Embryos exposed to whole EXUV cells, spent medium from EXUV cultures, and filtrates from heat-killed and sonicated cells showed no differences from controls in survival, development, or histology (light and electron microscopy). Forty-eight-hr larvae were fed EXUV alone and as a '/» or % portion of a mixed ration with Isochrysis sp. (strain T-ISO); controls of T-ISO alone and unfed larvae also were included. Differences in survival and growth were obtained, with larvae fed 100% EXUV performing only slightly better than unfed larvae; no EXUV-fed larvae sur- vived to set. P. minimum EXUV cells were filtered poorly, rela- tive to T-ISO; some ingestion, but limited digestion was noted by epifluorescence microscopy. Mixed diets produced intermediate results. Histologic examination revealed clear differences between unfed, T-lSO-fed, and EXUV-fed larvae. EXUV-fed larvae Milford Aquauulliirc Seminar, Milford, Connecticut Abslnicts. 13th Annual Meeting, February 22-24, 1993 115 showed more development than unfed animals, but not the vigor- ous development nor the cellular lipid reserves of T-lSO-fed lar- vae. Digestive glands of EXUV-fed larvae contained a very dis- tinct phagolysosomic/residual body. Post-set oysters (ca. 3 mm) were evaluated in the same treatments as larvae. Oysters fed 100% EXUV produced abundant pseudofeces for 3 wk, following which well-formed fecal strands were seen; oysters fed T-ISO filtered normally. After 6 wk, no mortalities were noted, and slight growth was obtained in most treatments. Differences in histologic appear- ance and condition of the digestive system were again observed. In summary, although acute toxicity off. minimum EXUV to oysters was not found, there was strong evidence for nutritional deficiency or interference with digestion. This study underscores the great variation in pathological effects that a single dinoflagel- late can produce in different life-history stages and different bi- valve species, i.e., oysters, clams, and scallops. Journal of Shellfish Research. Vol 12. Ncv 1. 117-157. 199.1. ABSTRACTS OF TECHNICAL PAPERS Presented at the 85th Annual Meeting NATIONAL SHELLFISHERIES ASSOCIATION Portland, Oregon May 31 -June 3, 1993 117 National Shellfisheries Association, Portland, Oregon Abstracts, 1993 Annual Meeting, May 31-June 3, 1993 119 CONTENTS PARASITES AND DISEASES I Standish K. Allen, Jr. Triploids for field study? The good, the bad. and the ugly 125 Bruce J. Barber and R. Mann Comparative physiology of Crassostrea virginica and C. gigas: growth, mortality, and infection by Perkinsus marinus 125 Eugene M. Burreson and Lisa M. Ragone Calvo The effect of wmter temperature and spring salinity on Perkinsus marinus prevalence and intensity: a laboratory study 125 Eugene M. Burreson and Lisa \t. Ragone Calvo Overwintering infections of Perkinsus marinus in Chesapeake Bay oysters 125 Eugene M. Burreson, Victor Vidal-Martinez and Raul Sima-Alvarez Perkinsus marinus as a source of oyster mortality in coastal lagoons in Tabasco, Mexico 126 David Bushek Evaluation of Perkinsus marinus quantification techniques using fluid thioglycollate media 126 Gustavo W. Calvo and Eugene M. Burreson Chemotherapy of Perkinsus marinus-mfecled oysters: a two week bath treatment experiment with amprolium, cyclohexamine, malachite green, and sulfadimethoxine 126 Fu-Lin E. Chu, Carrie S. Burreson, Aswani Voltey and Georgetta Constantin Perkinsus marinus susceptibility in eastern (Crassostrea virginica) and Pacific (Crassostrea gigas) oysters; temperature and salinity effects 127 Dawn E. Dittman The quantitative effects of Perkinsus marinus on reproduction and condition in the eastern oyster, Crassostrea virginica 127 William S. Fisher, James T. Winstead, Leah M. Oliver and Patrice Edwards Physiological and immunological measures of Appalachicola Bay oysters during a one year period 127 Susan E. Ford and Robert D. Barber Spores of Haplospundium nelsoni: findings and speculations 128 Julie D. Gauthier and Gerardo R. Vasta In vitro continuous culture of Perkinsus marinus trophozoites: optimization of the methodology 128 John E. Graves and Jan R. McDowell Genetic differentiation among strains of disease challenged oysters McDowell 128 George E. Krantz Chemical inhibition of Perkinsus marinus in an in vitro test 129 Jerome F. LaPeyre, Mohamed Faisal and Eugene M. Burreson Propagation of the oyster pathogen Perkinsus marinus in vitro 1 29 Roger Mann Population models to evaluate the impact of diseases and management options for the James River oyster fishery 129 Harold C. Mears The oyster disease research program of the National Marine Fisheries Service (NMFS): an overview 129 Gary F. Smith and Stephen J. Jordan Utilization of a Geographical Information System (GIS) for the timely monitoring of oyster population and disease parameters in Maryland's Chesapeake Bay 130 Aswani K. Volety and Fu-Lin E. Chu Infectivity and pathogenicity of two life stages, meront and presporangia, of Perkinsus marinus in eastern oysters, Crassostrea virginica 1 30 GENERAL BIOLOGY V. Monica Bricelj, Susan Bauer and Shino Tanikawa-Oglesby Contrasting foraging tactics of two predators of juvenile bay scallops, Argopecten irradians. in the eelgrass canopy... 130 Albert F. Eble, J. Ramsbottom and B. Burkhardt Role of fecal elimination during uptake and depuration of 65ZN and 109CD in the hard clam 131 Mohamed Faisal, Jerome F. LaPeyre and Morris H. Roberts Jr. Development of confluent monolayers from tissues of the eastern oyster, Crassostrea virginica 131 120 Abstracts, 1993 Annual Meeting, May 31-June 3, 1993 National Shellfisheries Association, Portland, Oregon Frank E. Friedl and Marvin R. Alvarez Oxygen uptake, oxidant production, and luminol-enhanced chemiluminescence by hemocytes of eastern oysters 131 Gunadi Kismohandaka, Carolyn S. Friedman, Wendy Roberts and Ronald P. Hedrick Investigations of physiological parameters of black abalone with withering syndrome 131 Tracy Potter, Bruce A. MacDonald and J. Evan Ward Studies of the sporatic release of epithelial cells by the sea scallop, Placopecten magellanicus 132, Joan L. Reudiger and Glenn R. VanBlaricon Abalone withering syndrome at San Nicolas Island, California 132 Bradley G. Stevens, J. Haaga, J. E. Munk and W. E. Donaldson Morphometric maturity and aggressive mating behavior of tanner crab, Chionoecetes bairdi (DecapodaiMajiadae), sampled by scuba and submersible 132 REPRODUCTION AND RECRUITMENT Kwang-Sik Choi, Eric N. Powell and Donald H. Lewis Instantaneous reproductive effort of the American oyster, Crassostrea virginica. in Galveston Bay, Texas 132 Margaret M. Dekshenieks, Eileen E. Hofmann, John M. Klink and Eric N. Powell A modelling study of the environmental and behavioral factors controlling the vertical distribution of oyster larvae — 133 S. R. Fegley, J. N. Kraeuter, S. E. Ford and H. H. Haskin Estimating the survival of Delaware Bay oyster larvae within and between years 133 Robert A. McConnaughey and David A. Armstrong A juvenile critical stage in the dungeness crab {Cancer magister) life history 133 Robert A. McConnaughey, D. A. Armstrong, B. M. Hickey and D. R. Gunderson Coastal advective processes and recruitment variability in dungeness crab (Cancer magister) populations 134 Kennedy T. Paynter, Scott Gallager and Dennis Walsh Protein, carbohydrate and lipid levels associated with metamorphic success in larvae of the eastern oyster, Crassostrea virginica 134 David Rouse Growth of microcultched and remote set oysters in coastal waters of Alabama (ROUSE) 134 Janzel R. Villalaz Laboratory study of reproduction in Argopecten ventricosus 134 PARASITES AND DISEASES II R. S. Anderson, L. L. Brubacher, L. M. Mora, K. T. Paynter and E. M. Burreson Hemocyte responses in Crassostrea virginica infected with Perkinsus marinus 135 Susan M. Bower, Gary R. Meyer and Jim A. Boutillier Diseases of spot prawns {Pandalus platyceros) caused by the intracellular bacterium and a Hematodinium-Vike protozoa 1 35 Drew C. Brown, Brian P. Drew and Kennedy T. Paynter The physiological effects of protozoan parasitism on the eastern oyster, Crassostrea virginica: induction of stress proteins 135 Dominique Hervio, Susan M. Bower and Gary R. Meyer Detection, isolation, and host specificity of Microcytos mackini. the cause of Denman Island disease in Pacific oysters Crassostrea gigas 1 36 James D. Moore and R. A. Elston Pathogenesis of disseminated neoplasia in eastern Pacific Mytilus trossulus 136 Roger I. E. Newell, Christine J. Newell, K. Paynter and Gene Burreson The physiological effects of protozoan parasitism on the eastern oyster, Crassostrea virginica: feeding and metabolism 1 36 Kennedy T. Paynter, Christopher Caudill, and Eugene M. Burreson The physiological effects of protozoan parasitism on the eastern oyster, Crassostrea virginica: introductory overview 137 National Shellfisheries Association, Portland, Oregon Abstracts. 1993 Annual Meeting, May 31-June 3, 1993 121 Kennedy T. Paynter, Sidney K. Pierce and Eugene M. Burreson The physiological effects of protozoan parasitism on the eastern oyster, Crassostrea virginica: effects on cellular free amino acid levels '37 S. K. Pierce, L. A. Perrino and L. M. Rowland-Faux Several mitochondrial functions in Chesapeake Bay oysters are different in Atlantic oysters; disease or genetics? 137 Bob S. Roberson, Tong Li and Christopher F. Dungan Flow cytometric enumeration and isolation of immunofluorescent Perkinsus marinus cells from estuaiine waters 138 AQUACULTURE, ECOLOGY AND MANAGEMENT William D. Anderson and Arnold G. Eversole Over exploitation and signs of recovery; analysis of an offshore whelk fishery 138 Brian F. Beat Effects of initial clam size and type of protective mesh netting on the survival and growth of hatchery-reared individuals of Mya arenaria in eastern Maine 138 Bonnie L. Brown, Arthur J. Butt and Kennedy T. Paynter Growth of the eastern oyster, Crassostrea virginica. in floating rafts in North Carolina 139 T. Jeffrey Davidson, Rod McFarlane and Judy Clinton On farm computer program for mussel farms 139 Dorset H. Hurley and Randal L. Walker Factors of mesh size, stocking size, stocking density and environment which affect growth and survival of Mercenaria mercenaria (Linnaeus, 1758) in a maricultural growout application in coastal Georgia 139 Philip S. Kemp and Alfred J. J. Evans Development of the chub ladder oyster culture method 140 W. S. Perret, R. Dugas, J. Roussel and C. Boudreaux Effects of Hurricane Andrew on Louisiana's oyster resources 140 Junggeun Song and Eric N. Powell Health assessment of oyster reefs in Galveston Bay, Texas 140 HARMFUL PHYTOPLANKTON AND SHELLFISH INTERACTIONS Allan D. Cembella, Nancy I. Lewis and Sandra E. Shumway An interspecific comparison of paralytic shellfish poisons in marine bivalves; antatomical and spatio-temporal variation in toxin composition 141 Ann S. Drum, Terry L. Siegbens, Eric A. Crecelius and Ralph A. Elston Domoic acid in the Pacific razor clam Siliqua patula 141 Rita A. Horner and James R. Postel Domoic acid in western Washington waters 141 J. M. Kelly Ballast water and sediments as mechanisms for unwanted species introductions into Washington State 142 Mark Luckenbach, Sandra Shumway and Kevin Sellner ■'Non-toxic'" dinoflagellate bloom effects on oyster culture in Chesapeake Bay 142 Paul A. Montagna, Dean Stockwell and Greg Street Effect of the Texas brown tide on Mulinia lateralis populations and feeding 142 John E. Rensel Factors controlling paralytic shellfish poisoning (PSP) in Puget Sound, Washington 142 D. L. Roelke, G. A. Fryxell and L. A. Cifuentes Effects on the oyster Crassostrea virginica caused by exposure to the toxic diatom Nitzschia pungens f. multiseries. . . 143 Donald D. Simons and Dan L. Ay res Fisheries management and toxic phytoplankton; the razor clam example 143 M. C. Villac, G. A. Fryxell, F. P. Chavez and K. R. Buck Pseudonitzschia australis Frenguelli and other toxic diatoms from the west coast of the U.S.A.; distribution and domoic acid production 143 122 Abstracts, 1993 Annual Meeting, May 31-June 3, 1993 National Shellfisheries Association, Portland, Oregon NON-TRADITIONAL SHELLFISHERIES Bruce E. Adkins Management of the commercial fishery for spot prawns {Pandalus platyceros) in British Columbia 144 Frances V. Dickson The intertidal clam fishery in British Columbia; a fishery under review 144 Sue Farlinger and Greg Thomas Management of the British Columbia abalone fishery: a square peg in a round hole 144 Rick Harbo Dig a duck — the commercial geoduck clam fishery in British Columbia 144 Steve Heizer "Knob cod" — management of the commercial sea cucumber fishery in British Columbia 144 J. Eric Munk and R. A. Macintosh Continuing studies of green urchin growth and recruitment near Kodiak, Alaska 145 Catherine Pfister and Alex Bradbury Exploitation in natural populations: a case study of a "new" fishery 145 Shawn Robinson The soft-shell clam fishery in the Canadian maritimes; an industry in change 145 Robert E. Sizemore and Lynn Y. Palensky Fisheries management implications of new growth and longevity data for pink (Chlamys rubida) and spiny scallops (C. hastata) from Puget Sound. Washington 145 Greg Thomas Management of an expanding red sea urchin fishery in British Columbia 146 INTEGRATED PEST MANAGEMENT Kenneth M. Brooks Impacts on benthic invertebrate communities caused by aerial application of carbaryl to control burrowing shrimp in Willapa Bay , WA 146 Brett R. Dumbauld, David A. Armstrong and Kristine L. Feldman A proposal to take a closer look at burrowing shrimp recruitment to oyster culture areas in Washington coastal estuaries 146 Kristine L. Feldman, David A. Armstrong, David B. Eggleston and Brett R. Dumbauld Burrowing shrimp recruitment to intertidal shell habitat; substrate selection, post-settlement survival, and the impact on shell longevity 146 Daniel P. Molloy Approaches to the biological control of zebra mussels 147 John L. Pitts An integrated pest management plan for the control of burrowing shrimp populations on oyster beds in southwestern Washington State 147 ALASKAN SHELLFISH INDUSTRY PANEL Raymond RaLonde, J. Cochran, J. Hetrick, M. Soares, M. Ostasz and J. Burleson Promise and constraints of shellfish aquaculture in Alaska 147 BIVALVE FEEDING AND NUTRITION Francisco J. Borrero and V. Monica Bricelj Vertical gradients in growth of juvenile bay scallops, Argopecten irradians, in relation to flow and seston characteristics in eelgrass meadows 148 Christopher J. Langdon Microcapsules and suspension-feeders — an update 148 Roger I. E. Newell, J. Evan Ward, Bruce A. Macdonald and J. Raymond Thompson Mechanisms of particle transport and ingestion in the eastern oyster, Crassosirea virf;inica 149 National Shellfisheries Association, Portland, Oregon Absiracis. 1993 Annual Meeting, May 31-June 3, 1993 123 Eric N. Powell, E. Wilson-Ormond, E. Hoffman and J. M. Klinck Phytoplankton stocks and the future of the Galveston Bay oyster fishery 149 J. Evan Ward and Bruce MacDonald In situ measurements of bivalve suspension-feeding: comparison between rates of scallops and mussels 149 E. Wilson-Ormond, E. N. Powell, E. E. Hofmann and J. M. Klinck Food availability to natural oyster populations: food, flow and flux 149 GENETICS AND BREEDING John W. Crenshaw Jr., Peter B. Heffernan and Randal L. Walker Effects of growout density on heritability of growth rate in the northern quahog, Mercenaria mercenaria ( Linnaeus, 1758) 1 50 Gregory A. DeBrosse and Standish K. Allen Jr. The suitability of land based evaluations of Crassosirea gigas as an indicator of performance in the field 150 Arnold G. Eversole and Peter B. Heffernan Gonadal neoplasia in Mercenaria mercenaria. M. campechiensis and their hybrids 150 Ximing Guo and Standish K. Allen Jr. Assessing reproductive sterility of triploids: aneuploid larvae produced from crosses between triploid and diploid Crassosirea gigas 151 Peter B. Heffernan and Randal L. Walker Second heritability estimate of growth rate in the southern bay scallop, Argopecten irradians concentricus (Say, 1822) 151 Ami E. Wilbur and Patrick M. Gaffney The effect of parental relatedness on progeny growth and viability in the bay scallop, Argopecten irradians 151 WEST COAST AQUACULTURE Dwight W. Herren The effectiveness of predator exclusion tubes for growout of the geoduck clam, Panopea abrupta 152 Thomas B. McCormick Abalone cultivation techniques 152 Walter Y. Rhee Hatchery techniques for the rock scallop (Crassadoma gigantea) larvae in the Puget Sound 152 Anja M. Robinson and Christopher J. Langdon The Suminoe oyster — candidate for the half-shell trade? 152 Douglas S. Thompson and Walt S. Cook Substrate additive studies for development of hardshell clam habitat 152 POSTER SESSION Brian F. Beal Overwintering hatchery-reared individuals oi Mya arenaria: a field test of site, clam size and intraspecific density 153 Fred S. Conte, Michael N. Oliver, and Heidi A. Johnson The effects of airlift circulation on the spacial distribution of Crassostrea gigas larvae set on strung cultch in circular tanks 153 Matthew S. Ellis, Jung Song and Eric N. Powell Status and trends analysis of oyster reef habitat in Galveston Bay, Texas 154 David W. Foltz and Shane K. Sarver Genetic structure of brackish water clams (Rangia spp.) 154 Susan E. Ford and Kathryn Alcox A comparison of methods for identifying molluscan hemocytes 154 Jean Gaudreault and Bruno Myrand Identification of a summer mortality-resistant population of blue mussels in the Magdalen Islands (Quebec, Canada) .. 154 124 Abstracts. 1993 Annual Meeting, May 31-June 3, 1993 National Shellfisheries Association, Portland, Oregon M. Giguere, G. Cliche, and S. Brulotte Reproduction of sea scallops {Placopecien magellanicus) and Iceland scallops (Chlamys islandica) in the Magdalen Islands 155 Dale S. Mulholland and Frank E. Friedl Potential of hemocytes taken from various body locations of the eastern oyster to interact with foreign materials 155 F. X. O'Beirn, P. B. Heffernan and R. L. Walker Ecosystem monitoring studies in coastal Georgia 155 James R. Postel and Rita A. Horner Toxic diatoms in western Washington waters 155 Elizabeth T. Rice Clam production in Ireland 156 Bob S. Roberson, Tong Li and Christopher F. Dungan Flow cytometric analysis of histozoic Perkinsus marinus cells 156 Nancy A. Stokes and Eugene M. Burreson Comparison of 1 6S-iike rDN A of Crassoslrea virginica and Haplosporidium nelsoni 156 R. L. Walker and P. B. Heffernan Age, growth rate, and size of the southern surfclam. Spisiila solidissima similis (Say, 1822) 157 J. Evan Ward, P. G. Beninger, B. A. Macdonald and R. J. Thompson Suspension-feeding mechanisms in bivalves: resolution of current controversies using endoscopy 157 Sheree J. Watson and Nicole M. Apelian Production of domoic acid by Pseudonitzschia ausiralis isolated from the southwestern Oregon coast following an ASP outbreak in Fall 1991 157 National Shellfisheries Association. Portland, Orcizon Abstracts. 1993 Annual Meeting, May 31-June 3. 1993 125 PARASITES AND DISEASES I TRIPLOIDS FOR FIELD TESTS? THE GOOD, THE BAD, AND THE UGLY. Standish K. Allen. Jr., Haskin Shellfish Research Laboraton,', Institute of Marine and Coastal Sciences. Rutgers University. Port Norris. NJ 08349. Interest and controversy surround the "proposal" to introduce Crassostrea gi^as to the east coast, putatively. to bolster the ailing oyster industry. Yet there is no empirical data on how C. gigas would perform here. Key is whether or not C. gigas are resistant to Dermo. or MSX-disease, or both. For the latter two questions, field exposure seems necessary. Even for ecological issues, the reliability of data extrapolated from land-based experiments is questionable. The GOOD: Triploids. because they are reproduc- tively incapacitated, provides a way to "safely" test C. gigas with little or no risk of reproduction. Use of F,. or greater, progeny reduces the risk of disease. Data show that triploids produce ga- mete types that vary little among individuals and that crosses using these gametes behave in predictable ways, all suggesting that the risk is estimable. The BAD: Recent evidence also suggests that there may be some spontaneous chromosome loss in triploids as they age. This surprising result means that analysis of individuals before field planting will be essential, perhaps yearly. And indi- vidual testing means a relatively small sample size, precluding pilot scale tests. The UGLY: There is no clear consensus on whether field tests using triploids should be approved; guidelines for approval of such tests are vague and variable; it is diftlcult to establish the distinction between an introduction for research pur- poses and a full scale release. This paper considers these points in view of the present crisis on the east coast oyster fishery. COMPARATIVE PHYSIOLOGY OF CRASSOSTREA VIR- GINICA AND C. GIGAS: GROWTH, MORTALITY, AND INFECTION BY PERKINSUS MARINUS. Bruce J. Barber,* Dept. of Animal, Veterinary &. Aquatic Sciences. University of Maine. Orono. ME 04469; R. Mann, Virginia Institute of Marine Science, College of William and Mary, Gloucester Point, VA 23062. Hatchery-produced oysters (the eastern oyster. Crassostrea vir- ginica. and the Pacific oyster. C. gigas), of the same age were held in quarantined flumes which received raw water from the York River. VA. From July 1991 to December 1993. growth and mortality were compared for experimental (dosed with Perkinsus marinus) and control (undosed) groups of both species. Both prevalence and intensity of P. marinus infections were greater in C. virginica than in C. gigas. The experimental C. virginica group had 100% prevalence (with heavy infections) by August 1992; maximum prevalence in the experimental C. gigas group was 80%. and only I heavy infection was found the entire study. Overall mortality of C. gigas (76%) was greater than that of C. virginica (45%); however, only mortality of C. virginica was related to infection by P . marinus. In December 1992 (at age 20 months), mean shell height of C. gigas (55 mm) was significantly greater (P s; 0.05) than that of C. virginica (41 mm). Shell height was lower in the experimental group compared to the control group of C. virginica but not of C. gigas. Thus C. gigas is more tolerant of P. marinus and grows faster than C. virginica. but may be less well adapted to environmental conditions prevailing in lower Chesapeake Bay. THE EFFECT OF WINTER TEMPERATURE AND SPRING SALINITY ON PERKINSUS MARINUS PREVALENCE AND INTENSITY: A LABORATORY EXPERIMENT. Eugene M. Burreson* and Lisa M. Ragone Calvo, Virginia Institute of Ma- rine Science. College of William and Mary. Gloucester Point. VA 23062. The role of low temperature and low salinity in controlling P. marinus was investigated under laboratory conditions which sim- ulated typical and extreme winter and spring environmental con- ditions. Oysters {Crassostrea virginica) infected with P. marinus were collected from the upper James River. VA in December 1991. individually marked and analyzed for P. marinus by he- molymph assay. The oysters were then subjected to a sequential treatment of various temperature and salinity combinations. In the first phase oysters were placed in recirculating seawater systems at 10 ppt and low temperature (IT and 4°C). Half of the oysters were treated at each temperature for 3 weeks and the other half were held for 6 weeks. In the second phase the oysters were gradually wanned to 12°C, adjusted to one of three salinities (3,6, and 15 ppt). and held for 2 weeks. Finally, all oysters were grad- ually adjusted to 25°C and 20 ppt and maintained for 4 weeks to determine if any observed declines in prevalence or intensity re- sulting from prior treatment were permanent. At the end of each phase P. marinus prevalence and intensity was assessed using hemolymph assay. Control oysters were maintained at I5°C and 15 ppt during treatment phase 1 and 2 and adjusted to 25°C and 20 ppt in phase 3. Low temperature exposure, alone, did not significantly effect P. marinus prevalence or infection intensity. However, declines in prevalence and intensity, relative to initial levels were observed after 2 weeks at 12°C and 3. 6, and 15 ppt. Perkinsus marinus prevalence and intensity in control oysters significantly increased as the experiment progressed. These results suggest that low win- ter temperatures have little effect on the annual abundance of P. marinus within an estuary, while springtime depressions in salinity are very important. OVERWINTERING INFECTIONS OF PERKINSUS MARI- NUS IN CHESAPEAKE BAY OYSTERS. Eugene M. Burre- son and Lisa M. Ragone Calvo,* Virginia Institute of Marine Science, College of William and Mary. Gloucester Point, VA 23062. The scarcity of overwintering infections of Perkinsus marinus in Chesapeake Bay oysters has long puzzled investigators. Typi- 126 Abstracts. 1993 Annual Meeting, May 31-June 3. 1993 National Shellfisheries Association, Portland, Oregon cally, prevalence of the pathogen declines in winter and infections are not easily disclosed by routine diagnosis using tissue cultured in thioglycollate medium (FTM). It is unknown whether cryptic stages of the parasite are harbored in the oyster during winter or whether elimination occurs; hence, the actual abundance and rel- ative contribution of overwintering infections to subsequent sum- mer prevalences is unclear. The objective of this investigation was to determine the nature and abundance of overwintering P. morinus infections. Infected oysters were placed in a tray and suspended from a pier in the lower York River, VA in November 1991. Every six weeks from November 1991 through May 1992 oysters (n = 25) were re- moved from the tray, examined for P. marinus by hemolymph analysis, gradually warmed in individual containers to 25°C and held for one month. After the incubation period, which permitted the development of very light and/or cryptic parasite stages to detectable levels, the oysters were reanalyzed for P. marinus by both hemolymph and tissue cultures in FTM. A second group of 25 oysters was sacrificed on each date, diagnosed using tissue FTM cultures, and examined for cryptic stages using immunoas- says. Prevalence of P. marinus gradually declined from 100% in November 1991 to 32% in April 1992. Incubation of oysters at 25°C always resulted in an increase of P. marinus prevalence and intensity, suggesting that the parasite was more abundant than FTM cultures indicated. Immunoassay did not reveal the presence of cryptic stages, although it was generally more sensitive than FTM diagnosis. Perkinsus marinus appears to overwinter at very light intensities in a high proportion of oysters. These infections are likely to be an important cause of summer mortalities. PERKINSUS MARINUS AS A SOURCE OF OYSTER MOR- TALITY IN COASTAL LAGOONS IN TABASCO, MEX- ICO. Eugene M. Burreson,* Virginia Institute of Marine Sci- ence, College of William and Mary, Gloucester Point, VA 23062; Victor Vidal-Martinez and Raul Sima-Alvarez, Centro de In- vestigacion de Estudios Avanzados del IPN Unidad Merida, C. P. Merida, Yucatan, Mexico. Periodic oyster mortality in coastal lagoons in Tabasco, Mex- ico in the southern Gulf of Mexico was attributed to the Mexican oil industry because of a previous small-scale oil spill near Mecoacan Lagoon. In an attempt to identify the cause of the oyster mortality the Mexican oil company Petroleos Mexicanos (PEMEX) funded CINVESTAV-IPN Unidad Merida to conduct a study that included pathology, effects of various pollutants and other water quality studies. This situation is very reminiscent of the sequence of events in Texas in the late 1940s that led to the discovery of P. marinus. As part of the PEMEX-funded study a survey of oyster beds was conducted in Mecoacan and Carmen y Machona lagoons in October, 1992. Subsequent thioglycollate culture diagnosis re- vealed the presence of Perkinsus in all beds sampled. Prevalence ranged from 60% to 100% and weighted prevalence ranged from 0.5 to 3.1. Previous samples from July, 1992 processed only for paraffin histology revealed prevalences of at least 50% and the presence of one extremely high Perkinsus infection. Immunoassay analysis of the Mexican samples using an anti-P. marinus antibody were positive. These results suggest that at least some of the oyster mortality in Mexico could be attributed to P. marinus. but more intensive areal and temporal surveys are necessary before the ef- fect of this pathogen can be determined with certainty. EVALUATION OF PERKINSUS MARINUS QUANTIFICA- TION TECHNIQUES USING FLUID THIOGLYCOLLATE MEDIA. David Bushek, Haskin Shellfish Research Laboratory. Institute of Marine and Coastal Sciences, Rutgers University, Port Non-is, NJ 08349. Accurate quantification of parasite burden is critical for com- paring host resistance, especially when resistance is a matter of degree. Perkinsus marinus loads can be estimated by culturing tissue or hemolymph samples in fluid thioglycollate media. Infec- tions are ranked in tissue samples whereas hemolymph samples are enumerated. The accuracy and sensitivity of these methods was checked against total body burden across seasons. Oysters were collected throughout the year from Delaware Bay beginning in July, 1992. Perkinsus marinus levels estimated with both tech- niques were regressed on total body burden. Correlations im- proved as parasitism peaked in the population, but variability was high; r" = 0.63 with tissue, 0.4 with hemolymph. Higher corre- lations with tissue apparently resulted from lumping infections into categories. Neither technique is sufficient when total body burden esti- mates must be determined accurately or when infection levels are low. Tissue samples are recommended for routine diagnostics be- cause they are quick, easy and moderately accurate. Hemolymph samples are only recommended when the oyster cannot be sacri- ficed. Contribution # K-32100-1-93 NJAES. CHEMOTHERAPY OF PERKINSUS MA/f/A't/S-INFECTED OYSTERS: A TWO WEEK BATH TREATMENT EXPERI- MENT WITH AMPROLIUM, CYCLOHEXIMIDE, MALA- CHITE GREEN. AND SULFADIMETHOXINE. Gustavo W. Calvo* and Eugene M. Burreson, Virginia Institute of Marine Science. College of William and Mary. Gloucester Point. VA 23062. A repeated measures design was used to determine the effect of chemical baths on reducing P. marinus infections in oysters. To that end, 300 oysters were collected from Pt. of Shoals in the James River in September and maintained in a static renewal tank with 1 (xm filtered York River water (temperature 20°C, salinity 20 ppt) at VIMS for one week. During that time, oysters were labelled and screened individually for P. marinus using the he- molymph technique. Then, 180 oysters with known infection in- National Shellfisheries Association. Portland. Oregon Ahsiracts. 1993 Annual Meeting, May 31-Junc 3. 1993 127 tensities were selected and assigned to 10 L aquaria in groups of 10. There were 8 chemical treatments (amprolium and sul- fadimethoxine as 100 mg/L and 10 mg/L baths, and cycloheximide and malachite green as 10 mgL and 1 mg/L baths) plus 1 untreated control treatment all run in duplicate. Chemicals were mixed with microalgae and added to aquaria, at the time of water renewal, everv' other day for 2 weeks. Dilution water consisted of 1 p-ni filtered York River water (warmed and maintained at 20°C, and 20 ppt). Oysters were fed ever>' day. After the 2 week treatment, P. marinus diagnosis was performed on a second hemolymph sample and on a combined rectum, gill and mantle sample taken from each oyster. Pre-treatment and post-treatment infection intensities in he- molymph samples were compared by Wilcoxon's signed rank test. Only oysters exposed to 10 mg/L of cycloheximide showed a significant decrease in infection levels. Tissue samples also re- vealed a higher proportion of oysters with lower infection inten- sities in the group exposed to 10 mg/L of cycloheximide than in the control or any other group. These results suggest that cyclo- heximide is effective in reducing P. marinus infections in oysters. Use of cycloheximide, however, is mostly restricted to laboratory applications. PERKINSUS MARI^US SUSCEPTIBILITY IN EASTERN (CRASSOSTREA VIRGINICA) AND PACIFIC (CRASSOS- TREA GIGAS) OYSTERS: TEMPERATURE AND SALIN- ITY EFFECTS. Fu-Lin E. Chu,* Carrie S. Burreson, Aswan! Volety, and Georgeta Constantin, Virginia Institute of Marine Science, School of Marine Science, College of William & Mary, Gloucester Point, VA 23062. Susceptibility of Crassostrea virginica to Perkinsus marinus was compared with diploid and triploid (2N and 3N) C. gigas at 10, 15, and 25°C in the first experiment and at 3 salinities, 3, 10, and 20 ppt, in the second experiment. In both experiments, oysters were challenged twice with P. marinus trophozoites. The temper- amre effect experiment was terminated 68 days after 1st challenge and 27 days after 2nd challenge by P. marinus. The salinity effect experiment was terminated 50 days after 1st challenge and 34 days after 2nd challenge by P. marinus. Results revealed that at 15 and 20°C, infection prevalence was higher in challenged C. virginica than in challenged 2N and 3N C. gigas. But at 10°C, challenged 3N C. gigas had a prevalence higher than challenged 2N C. gigas and C. virginica. In all salinity treatments, prevalence was higher in challenged C. virginica than challenged 2N and 3N C. gigas. Weighted prevalence increased with temperature and salinity and was highest in C. virginica groups. Since, in both experiments, much higher infection prevalence and intensity were found in non- challenged C. virginica than in non-challenged 2N and 3N C. gigas, part of the recorded prevalence and intensity in C. virginica may be attributed to the hidden infection from the field. High mortality occurred in both 2N and 3N C. gigas during temperature and salinity acclimation and at the 25°C and 3 ppt treatments. THE QUANTITATIVE EFFECTS OF PERKINSUS MARI- NUS ON REPRODUCTION AND CONDITION IN THE EASTERN OYSTER, CRASSOSTREA VIRGINICA. Dawn E. Dittman,* Haskin Shellfish Research Laboratory, Rutgers Uni- versity. Port Norris. NJ 08349. Dermo disease, caused by Perkinsus marinus. is responsible for high oyster mortality in many areas along the East Coast. The evidence that P. marinus causes a decrease in reproduction before death has been ambiguous. This study examines the effect of in- fection by P. marinus on reproduction and condition index of live oysters. Known-age susceptible oysters were exposed to P. marinus in 1990 and 1991 . Sixteen samples of 40 to 50 animals were taken at two to four week intervals. The animals were prepared for histo- logical analysis and cultured for P. marinus using standard tech- niques. P. marinus infection level was assigned to three catego- ries; none, light, and advanced. Condition index was calculated and percent gonad area was measured using an image analysis system. The data were analyzed using a Multivariate analysis of variance model. There was no significant effect on reproduction in the first year when infections were light dunng the reproductive period. In I99I the percent gonad area of the individuals with advanced infections was significantly lower than that of individuals with no infections and in most cases was lower than in individuals with light infec- tions. In all of the samples the condition index of oysters with advanced infections was lower than that of the uninfected oysters, and in most cases lower than that of the individuals with light infections. The results show that infection by P. marinus has a significant negative impact on the reproduction and the condition index the oyster before death. This is NJAES Publication No. 3250 1-K- 1-93. PHYSIOLOGICAL AND IMMUNOLOGICAL MEASURES OF APALACHICOLA BAY OYSTERS DURING A ONE- YEAR PERIOD. William S. Fisher* and James T. Winstead. U.S. Environmental Protection Agency, Center for Marine and Estuarine Disease Research, Environmental Research Laboratory, Gulf Breeze, FL 32561; Leah M. Oliver, Technical Resources Inc., Gulf Breeze, FL 32561; Patrice Edwards, Center for En- vironmental Diagnostics and Bioremediation, University of West Florida, Gulf Breeze, FL 32561. Most physiological and immunological measures of oyster health are influenced by changes in salinity and temperature. To apply such measures in assessment of oyster health requires knowledge of variations introduced by temperature and salinity patterns. A year-long study was performed on oysters (Crassos- trea virginica) from two subtidal, unpolluted, commercially- harvested sites in Apalachicola Bay, Florida. Oysters were col- lected monthly and multiple endpoints measured for each organ- ism. Physiological measures included gonadal index and state of 128 Abstracts. 1993 Annual Meeting, May 31-June 3, 1993 National Shellfisheries Association, Portland, Oregon maturation, condition index, tissue structure indices and he- molymph protein levels. Immunological measures included hemo- cyte morphology, mobility, phagocytic capacity and superoxide production as well as hemolymph lectin and lysozyme content. Parasite burdens and infection levels of Perkinsus marinus were quantified. Results demonstrated high variability for most endpoints, with seasonal (temperature) cycles in evidence and relatively rapid re- sponses to salinity events. Correlations among certain immuno- logical endpoints support current hypotheses of immunological fitness. It is concluded that assessment of oyster health requires a continuous monitoring scheme for each site under consideration to reduce potential misinterpretation of results. SPORES OF HAPLOSPORIDIUM NELSOM (MSX): FIND- INGS AND SPECULATIONS. Susan E. Ford* and Robert D. Barber, Rutgers University. Institute of Marine and Coastal Sci- ences. Haskin Shellfish Research Laboratory. Box B-8. Port Nor- ris. NJ 08349. The apparent rarity of spores produced in oysters infected with Haplosporidium nelsoni, cause of MSX disease, led to hypotheses that another host is involved in the life cycle. In contrast to pre- vious studies, which found spores in <1% of infected adult oys- ters, we report that infected spat have a high probability (>50%) of producing the spore stage. Advanced infections nearly always result in sporulation. In 1988, 30-35% of spat in lower Delaware Bay produced spores, whereas, that the figure has been only 5% in the last 4 years (1989-92). Up to 1.5 x 10** mature spores have been found in a single spat. We have also found spores morphological identical (by light microscopy) to those of W. nelsoni. ingested by oysters throughout Delaware Bay. Their presence in oyster guts during the summer coincides with the infective period for H. nelsoni. We estimate that the concentration of spores in the water processed by oysters must be several hundred per liter to account for their numbers in the digestive tract. Although annual spat sets are temporally and spatially variable, data from 35 years of sampling in Delaware Bay lead us to esti- mate that spat density is about 100 m " in an "average" year (I0'"-10'' total in the Bay). If the ingested spores are H. nelsoni. 10'^ to 10'" spat would be required, each producing 10'' spores, to yield estimated concentrations in Delaware Bay during summer. Five percent of the total estimated spat in the Bay would somewhat exceed this number. We do not know how long spores remain viable, how long they are present in the water column, and our estimates have not taken into account potential loss of spores from the estuary in current outflow, loss from the water column through biodeposition. or destruction by microbes in the sediment. The calculations suggest that spat could produce enough spores to serve as a primary host; nevertheless, the possibility of an alternate host still cannot be excluded. IN VITRO CONTINUOUS CULTURE OF PERKINSUS MARINUS TROPHOZOITES: OPTIMIZATION OF THE METHODOLOGY. Julie D. Gauthier* and Gerardo R. Vasta, Center of Marine Biotechnology. University of Maryland. Balti- more, MD 21202. A continuous pure culture of the oyster parasite Perkinsus. marinus was accomplished in a variety of DMEM (Dulbecco's Modified Essential Medium, currently used in our laboratory for hybridoma culture) based media at 26 ± 2°C with no added CO,. DMEM was dissolved in 23 ppt artificial sea water and 15 mM HEPES (final pH 7.4) and 100 U/ml each Penicillin-G and Strep- tomycin sulfate added. The effect of supplements including oyster serum (0.1-50.0%). fetal bovine serum (FBS) (0.1-20%) and HAM'S F-12 Nutrient Mixture (l;l or 1;2 DMEM:HAM's F-12) was investigated and the formulations optimized. Oyster hemo- cytes harboring large numbers of P. marinus trophozoites were washed in a high antibiotic sea water solution (4,000 U/ml Peni- cillin -I- 5000 ug/ml Streptomycin sulfate) and plated at equal density in different media formulations. Growth was determined by direct counting and 'H-thymidine incorporation. Further opti- mization of culture conditions (supplement additions, seeding den- sity and frequency of medium changes) was accomplished by adapting image analysis methodology. Optimal conditions at present time include the addition of 5% oyster serum to the three following formulations: 20% FBS/DMEM, 10% FBS/1:1 DMEM: HAM'S F-12 or 1:2 DMEM;HAM's F-12 (Serum-free). The cul- tured parasite proliferates by multiple fission and/or budding at an estimated doubling time of 24 hrs within the first 72 hrs. Light and electron microscopy and serology demonstrate that the cultured forms are morphologically and biochemically identical to the freshly isolated ones. The cultured trophozoite enlarges in thio- glycollate medium and stains dark blue in Lugol's solution, both diagnostic