Acquisition and progression of Perkinsus marinus infections by specific-pathogen-free juvenile oysters (Crassostrea virginica Gmelin) in a mesohaline Chesapeake Bay tributary.

ABSTRACT We conducted a 36-mo study to determine the effects of environmental variables and proximity to infected resident oysters on Perkinsus marinus infection acquisition and progression in specific-pathogen-free (SPF) juvenile oysters in the Patuxent River, MD, a mesohaline tributary of Chesapeake Bay. Multiple deployments of experimental SPF oysters were made at four sites along the river's salinity gradient, with three sites adjacent to infected resident oyster populations and one site isolated by at least 5 km from known oyster populations. Experimental cohorts were analyzed for dermo disease at deployment (time 0), approximately 2 and 4 wk, and at subsequent monthly intervals, using quantitative whole-body-burden alternative Ray's fluid thioglycollate medium (ARFTM) assays. Some oysters in some experimental groups acquired P. marinus infections as early as 10 d after placement in dermo disease-enzootic waters, and infections were detected in all groups by 62 d postdeployment. In several cohorts, first infections were simultaneously detected among >62% of experimental oysters, reflecting epizootics within 8 wk of their exposures to local infection pressures. Threshold values of 12%o for salinity and 24[degrees]C for water temperature influenced infection acquisition rates. Isolation of SPF juvenile oysters from known populations of infected resident oysters did not prevent or delay acquisition of dermo disease during drought conditions. Once initial infections occurred, infection prevalences and mean intensities increased over time in all experimental oyster cohorts.

KEY WORDS: dermo disease, SPF oyster, Patuxent River, disease transmission

INTRODUCTION

Maryland oyster landings declined dramatically from 1986 (1.5 million bushels) to 1994 (80,000 bushels), largely because of mortalities from diseases caused by Perkinsus marinus (dermo) and Haplosporidium nelsoni (MSX) (MDNR 1996). In contrast to H. nelsoni, which is restricted to waters with salinities consistently above 10 [per thousand] (Ford 1985), P. marinus infects oysters in waters of a wide salinity range (Ragone & Burreson 1993). Because of its wide salinity tolerance, P. marinus has become enzootic since the mid-1980s in most of the available oyster habitat in Chesapeake Bay (Burreson & Ragone Cairo 1996).

Several mechanisms may effect dermo disease transmission. Primarily, the disease spreads between oysters by infectious, waterborne parasite cells released from feces and decomposing tissues of infected oysters (Ray 1954a, Andrews 1988, Bushek et al. 2002). Maximum dermo disease transmission to experimental Chesapeake Bay oysters coincides with maximum abundances of waterborne P. marinus cells, and maximum mortality rates among infected neighboring oysters (Ragone Calvo et al. 2003). Moreover, other bivalve molluscs harbor Perkinsus sp. infections in Chesapeake Bay (Andrews 1954, Valiulis & Mackin 1969, Dungan et al. 2002), and may serve as additional dispersal sources of infectious parasite cells. Vector-mediated dermo disease transmission by ectoparasitic gastropods has been described (White et al. 1987), as has potential dispersal of P. marinus cells by oyster predators and scavengers (Hoese 1962). However, strong epizootiological significances for either of these P. marinus dispersal and delivery mechanisms are not confirmed for Chesapeake Bay waters.

The timing and severity of P. marinus infection acquisition and progression are promoted by warm winters and high summer temperatures (Ford 1996), salinities above 8%,, to 10%o (Krantz & Jordan 1996), and the synergistic effects of both physical variables (Chu et al. 1994). A seasonal pattern of regularly fluctuating infection prevalences and intensities occurs throughout the parasite's range, and is strongly influenced by water temperatures in the mid-Atlantic region and northward. Existing infections become latent during winter and spring (Andrews 1954, Crosby & Roberts 1990) and are then often undetectable by standard Ray's fluid thioglycollate medium (RFTM) assays of tissue subsamples (Ray 1952). Subsequently, infection prevalences and intensities increase from late spring through mid fall in Chesapeake Bay (Andrews & Hewatt 1957, Krantz & Jordan 1996), largely caused by proliferation of parasites in latent infections present among 70% to 100% of oysters surviving from the previous fall (Ragone Calvo & Burreson 1994), and also caused by the acquisition of new infections as waterborne parasites become available and invade additional hosts.

Various strategies for avoiding P. marinus infections, or for delaying mortalities of infected oysters, have been implemented to protect and enhance feral oyster populations and planted hatchery production lots in Chesapeake Bay. Examples include transferring juvenile oysters from high-salinity recruitment sites to low-salinity grow-out sites (Andrews 1988, Krantz & Jordan 1996), planting uninfected juvenile oysters on fallowed beds that are free of infected oysters, and planting oysters in areas isolated by distance from infected oysters (Andrews 1988, Andrews & Ray 1988). However, results of several investigations show that during high-salinity drought periods in Chesapeake Bay, juvenile oysters planted on fallowed or isolated grounds can rapidly acquire and die of P. marinus infections that are transmitted over wide distances (Andrews 1988, Andrews & Ray 1988, Krantz & Jordan 1996). Some current enhancement strategies attempt to use the reported low susceptibility of juvenile oysters to P. marinus infections (Ray 1954b) by planting uninfected juvenile hatchery oysters at sites typically characterized by lower-salinity waters. However, although very young juveniles in limited Gulf of Mexico experiments were less susceptible to dermo disease than yearlings or adults (Mackin 1951, Ray & Chandler 1955), those oysters nevertheless acquired P. marinus infections that intensified over time (Ray 1954b).

We report results of experiments that tested the effectiveness of planting P. marinus-free (specific-pathogen-free, SPF) oyster spat in a mesohaline Chesapeake Bay tributary, both adjacent to and at least 5 km distant from P. marinus-infected native oysters. Alternative Ray's fluid thioglycollate medium (ARFTM) assays of whole oysters were used to quantitatively assess P. marinus infection prevalences and intensities among experimental oyster cohorts that were deployed during different seasons and years along the salinity gradient in the tributary. We determined the maximum exposure times elapsed until infections were first detected, documented the long-term progression of P. marinus prevalence and mean infection intensity in selected groups, and compared infection acquisition and progression between cohorts deployed adjacent to and at distance from infected feral oysters. We also characterized infected oyster sizes when infections were first detected, determined environmental conditions associated with infection acquisition, and correlated environmental conditions associated with dermo disease progression.

MATERIALS AND METHODS

Field Methods

Specific-pathogen-free (SPF) experimental oysters were hatchery-produced for this study (Albright et al. 2007). Cohorts of oysters >3 mm mean shell height (ca. 2-3 wk postmetamorphosis) were caged for deployment 0.1 m off-bottom on anchored racks that were placed on natural oyster bars located along the Patuxent River salinity gradient. A sample of 30+ oysters was collected from each hatchery production lot for Perkinsus marinus infection analysis at the time deployments were made (time 0). Deployments were made in September 2000, May 2001, August 2001, May 2002, September 2002, and June 2003. During all years, deployment sites were located at Town Creek (TC, down-estuary, highest salinity), Gatton (GA, mid-estuary), and Holland Point (HP, up-estuary, lowest salinity). In both 2002 deployments, an additional cohort was placed at Sandgates (SA, mid-estuary), a site isolated by at least 5 km from any known oyster population but with a similar salinity and temperature regimen to that at Gatton (Fig. 1).

Environmental conditions were measured near each site by a Hydrolab Datasonde 3 Multiparameter Water Quality Datalogger that recorded water temperature ([+ or -] 0.15[degrees]C) and salinity ([+ or -]0.2[per thousand]) at 2-h intervals. Thirty feral oysters from natural oyster bars adjacent to the deployment sites were also collected and examined monthly, using the RFTM rectal tissue assay (Ray 1952) to estimate infection prevalences and relative intensities of local infection pressures generated from pathogen reservoir oyster populations (Albright et al. submitted 2006). Infection intensities were expressed as categorical ranks (0-7).

Disease Samples and Assays

Samples from each experimental cohort were collected at approximately 2, 4, and 8 wk post-deployment, and 30 oysters were arbitrarily selected from each sample to be assayed for the presence of Perkhlsus marinus. Sampled oysters were purged for 48 h at a minimum of 25[degrees]C in recirculating artificial seawater that was isotonic with deployment site waters, to clear any ingested parasite cells in passive gut transit. Assays continued at monthly intervals, at least until P. marinus was detected in three consecutive samples from each cohort.

An enhanced ARFTM whole-body burden assay was used for P. marinus isolation, detection, and enumeration (Bushek et al. 1994, La Peyre et al. 2003). We assessed the entire oyster to ensure detection of all parasites present, used a medium optimized to enlarge parasites and reduce parasite clumping to facilitate counting, and selectively removed oyster tissues by chemical hydrolysis (Fisher & Oliver 1996, La Peyre et al. 2003). The ARFTM medium was prepared with 995 mL of sterile alternative thioglycollate medium (Sigma A0465) containing 16 g [L.sup.-1] of artificial sea salts (MarineMix CS-150B), 209 [micro]g [mL.sup.-1] of lipid concentrate (Sigma L5146), and 50 [micro]g [mL.sup.-1] of chloramphenicol (Sigma C0378) to inhibit bacterial proliferation.

Three sizes of sterile centrifuge tubes (2-mL, 15-mL, 50-mL) were used, based on oyster shell heights. Tubes were preloaded with 0.5, 1.0, or 5.0 mL of sterile ARFTM, respectively, and tared. Small oysters (<1 g wet meat weight) were shucked directly into tared 2-mL centrifuge tubes, weighed, and homogenized. Larger oysters were shucked, diced into small pieces in the cupped oyster valve with flame-sterilized scissors, and all tissues and shell liquor were transferred directly from the shell into tared 15- or 50-mL centrifuge tubes. Tissues were then weighed and homogenized. A disposable sterile polypropylene pestle, a sterile Teflon[R] pestle, or a motorized homogenizer (Tissue Tearor[TM]) was used to homogenize oyster tissues. In all cases, homogenizer surfaces were rinsed into sample tubes with additional sterile ARFTM to recover any adherent material. The motorized homogenizer generator was cleaned between samples by immersing and operating it successively in 70% ethanol, 200 ppm sodium hypochlorite, and distilled water. To inhibit fungal proliferation, 200-400 [micro]L of nystatin (Sigma N1639) was added to the surface of the homogenate in each tube. Homogenized tissues in ARFTM were incubated at 28[degrees]C for 7 d.

After incubation, tissue suspensions were centrifuged at 1500 x gravity (g) for 10 min, supernates were aspirated to waste, and pellets were resuspended in 1-35 mL of 2 M sodium hydroxide, depending on pellet volume. The samples were hydrolyzed at 60[degrees]C for 1 6 h with periodic vortexing until oyster tissues dissolved. Tubes were again centrifuged at 1,500 x g for 10 min, supernates were aspirated to waste, and the pellets were resuspended in distilled water. Pellets were washed twice with 0.1 M sodium phosphate buffer (pH 7.2) containing 0.5 mg [mL.sup.-1] of bovine serum albumin (Sigma A9418) (PBS-BSA) to prevent clumping and surface adsorption of parasite cells. Final resuspension of intact P. marinus hypnospores was in 1 mL or 10 mL of PBS-BSA diluent, based on oyster size or visible pellet volume.

[FIGURE 1 OMITTED]

Where pellets were resuspended in 1 mL of PBS-BSA, the entire pellet was placed in a single well of a 24-well plate for microscopic counting. Fifty-microliters aliquots from pellets resuspended in 10 mL of diluent were placed in wells of a 96-well plate for counting. Aliquots that were too numerous to count were serially diluted until countable dilutions (100-300 hypnospores [well.sup.-1]) were obtained. Where aliquots contained <100 parasite cells, additional aliquots were counted until either a minimum count of 100 cells was reached or the entire pellet was enumerated. Parasites were stained with Lugol's iodine. The total number of parasites was recorded for each oyster, and P. marinus cells [g.sup.-1] oyster tissue wet weight was calculated.

For each sample, mean sample body burden (MSBB) and mean infected body burden (MIBB) were calculated. Mean sample body burden was calculated as the sum of P. marinus [g.sup.-1] tissue of all oysters in a sample divided by the sample size (n = 30), and is synonymous with the terms "mean abundance" (Bush et al. 1997) and weighted prevalence (Ragone Calvo & Burreson 1994). Mean infected body burden was calculated as the sum of P. marinus [g.sup.-1] tissue of infected individuals in a sample divided by the number of infected individuals in that sample, and is synonymous with the term 'mean intensity' (Margolis et al. 1982).

Immunoassay Confirmation of Selected Infections

Histological immunoassays were performed to confirm infections by visualizing in situ lesions in oysters from a selected sample. Additional oysters were chosen from a sample that showed a high prevalence when early infections were first detected by ARFTM assays. Those oysters were shucked and fixed whole in Davidson's fixative for 48 h. Fixed tissues were subdivided transversely into three pieces, dehydrated and infiltrated, oriented to yield transverse sections, and embedded in paraffin wax. Systematic sections were cut at 5-[micro]m thickness from several depths of each tissue block, immunofluorescence-stained with anti-P, marinus antibodies (Dungan & Roberson 1993), and examined microscopically for early P. marinus lesions.

Data Analyses

We defined first (initial) infections to have occurred when at least one parasite cell was detected from at least one host oyster in a sample from a cohort. All cohorts analyzed for infection acquisition were P. marinus-free when deployed. Because our sampling schedule was discontinuous, initial infections may have occurred at any time during the intersample period before first detection. Therefore, because the actual number of exposure days elapsed before first infections were acquired was unknown, we defined categorical infection acquisition rates relative to exposure periods preceding the sampling iterations at which infections were first detected: "rapid" (2 wk), "moderate" (4 wk), and "slow" (8 wk). Deployments were grouped by season: May-June deployments were classified as "Spring," and August-September deployments were classified as "Fall."

We analyzed for correlations between rate of infection acquisition, and both disease-condition and environmental cofactors. Infection prevalence, mean sample body burden, mean infected body burden, and oyster size were the disease-condition cofactors. Salinity, water temperature, and environmental infection pressure (from feral oysters), were the environmental cofactors. Samples in which infections were first detected were pooled to analyze oyster size relationships within and among cohorts, and a Student's t-test was used to compare mean sizes of infected and uninfected oysters. Associated mean salinity and temperature values were calculated from daily values for the days from deployment to first detection of dermo disease. These means were compared qualitatively to reported threshold values of temperature (25[degrees]C) and salinity (12 [per thousand]) that affect P. marinus pathogenicity (Ford & Tripp 1996). Environmental infection pressure was inferred from mean infection intensity values from rectal RFTM assays of feral oysters sampled simultaneously from existing populations adjacent to experimental cohorts in which initial infections were detected (Albright et al. 2007).

We examined infection progression by comparing prevalences, mean sample body burdens, and mean infected body burdens over time between pairs of cohorts within deployments. Comparisons of disease progression were statistically evaluated using the Kolmogorov-Smirnov chi-square test (Chakravarti et al., 1967). The effects of isolation by distance from infected oysters on dermo disease progression were examined by comparing infection prevalences and intensities between mid-estuary cohorts at GA and SA from the May and September 2002 deployments. Comparisons between deployments could not be made because deployments were not of equivalent duration. Linear correlation was used to quantify the relationship between oyster size and measures of P. marinus infection, expressed as prevalence, mean sample body burden, and mean infected body burden.

RESULTS

Rate of Infection Acquisition

There were 19 samples showing first infections because the Fall 2000 cohort at HP was lost to heavy predation by the flatworm Stylochus ellipticus. Some oysters in all remaining cohorts acquired P. marinus infections by their eighth week post-deployment (Fig. 2). All cohorts within a deployment generally had the same infection acquisition rates, and isolation from infected feral oysters did not delay infection acquisition. There was one apparent anomaly in data from the Spring 2002 deployment. In that deployment, SPF oysters at TC, SA, and HP acquired infections by 4 wk (13%, 7%, and 3% prevalence, respectively), but no infection was detected at GA until 8 wks (Fig. 2). We believe this apparent delay in infection acquisition at GA resulted from sampling error, because the 8-wk prevalence was relatively high (63%), indicating that infections were relatively well established in that cohort. Therefore, for analytical purposes, the Spring 2002 GA cohort was assigned the same 4-wk first infection acquisition rate ("moderate") seen in the other cohorts of that deployment.

[FIGURE 2 OMITTED]

Immunoassay Confirmation of Selected Infections

Infections were confirmed in oysters from the Fall 2001 GA cohort using antibody staining for Perkinsus sp. in fixed tissues. This cohort was selected for staining based on a 100% (30/30) prevalence of light initial infections (7-107 Perkinsus sp. cells [oyster.sup.-l]) that were detected at 4 wk post-deployment by whole-body ARFTM assays. Fluorescence immunostaining and examination of multiple systematic histological sections from tissues of ten oysters from that cohort revealed dividing P. marinus trophozoites in stomach epithelium (Fig. 3), gill, or mantle tissues of two assayed oysters (20%).

Disease Severity

Prevalence values for initial infections ranged from 3% to 73% of oysters in experimental cohorts, and fell into low- and high-prevalence groupings (Table 1). Most initial infection prevalences (17/19, 89%) were [less than or equal to] 20% (low), and only 2/19 (11%) samples had prevalences [greater than or equal to] 60% (high) (Table 2). All occurrences of high-prevalence initial infections were found in cohorts from GA. In the GA Fall 2001 cohort, infections were first detected at 2 wk, with 22/30 (73%) oysters infected. All other Fall 2001 cohorts also showed first infections at 2 wk, with 1-2/30 (3% to 7%) infected oysters. In the GA Spring 2002 cohort, infections were first detected at 8 wk with 19/30 (63%) infected oysters. In contrast, all other Spring 02 cohorts showed first infections at 4 wk with 1-4 (3% to 13%) infected oysters. Mean sample body burden for this cohort was higher than for all other Spring 02 cohorts, as was the mean number of P. marinus cells infected [oyster.sup.-1]. Therefore delayed detection was inferred, and this cohort was assigned a 4-wk infection rate (moderate).

[FIGURE 3 OMITTED]

Initial infections exhibited widely variable intensities (Table 2). Tissue weight-normalized body burdens of infected individual oysters ranged from 0.25-1000 parasites [g.sup.-1] of oyster tissue, but actual total P. marinus counts among infected individuals ranged only from 1-23 parasites per oyster. All but two samples showed mean sample body burden values below 12 parasites [g.sup.-1]; one had 23 parasites [g.sup.-1]; and one sample had 273.5 parasites [g.sup.-1]. Mean infected body burden values were also grouped, with 14 values between 1 and 60 parasites [g.sup.-1], 2 values in the 100s, and 3 values in the 300s (Table 1). Mean sample body burden values either mimicked mean infected body burden values (GA Fall 01 : mean sample body burden = 273.5 parasites [g.sup.-1], mean infected body burden = 373 parasites [g.sup.-1]) or radically differed (GA Fall 00: mean sample body burden = 10.0 parasites [g.sup.-1], mean infected body burden = 300 parasites [g.sup.-1]). Whereas mean sample body burden was poorly correlated with the rate of infection (P = -0.29), mean infected body burden was strongly correlated (P = -0.66) and was used for all analyses.

Interactive Effects of Mean Infected Body Burden and Prevalence on Rate of Infection

There were four combinations of mean infected body burden level and prevalence level: heavy/low, heavy/high, light/low, and light/high (Table 1). All but one of the samples with heavy mean infected body burdens showed "rapid" acquisition of infections. All samples with low mean infected body burdens showed "moderate" or "slow" acquisition of infections.

Oyster Size at Infection Acquisition

Oysters from initial-infection samples ranged in size from 3-42 mm shell height, with a mean of 16 mm; infected individuals ranged from 6-35 mm shell height, with a mean of 17 ram. Mean shell heights of infected oysters were not significantly different from their cohort mean shell heights (P = 0.52), indicating that individual susceptibility to infection was not a function of oyster size within the size range analyzed.

Drought Versus Freshet Conditions

Rainfall conditions during the study period consisted of three years of drought followed by a freshet year. A prevailing regional drought (beginning in 1999) elevated water salinities at all experimental sites above their long-term averages during 2000 through 2002. In contrast, high freshwater inflows (freshet conditions) during 2003 depressed salinities at all sites to values similar to long-term averages (http://mddnr.chesapeakebay.net/ bay_cond/station_select.cfm). Analysis of variance (P < 0.001) and Duncan's test showed increasing seasonal (spring and fall) mean salinities (all sites combined) through 2001 and 2002 (10.8, 12.9, 13.7, 17.0 [per thousand], respectively) with a sharply decreased mean salinity for Spring 2003 (6.5 [per thousand]). Mean spring salinities at each deployment site showed the same relative longitudinal gradient during all years, but variations in freshwater inflow conditions caused locations of salinity isoclines to shift in both up- and downstream directions (Table 3). Salinity values at all sites during 2002 (drought) were approximately 3 [per thousand] higher than corresponding 2001 values, and 2003 (freshet) values were approximately 3 [per thousand] lower than corresponding 2001 values.

Interactions of Seasonal and Environmental Effects on Infection Acquisition

There was a strong seasonal effect on infection acquisition rates and environmental conditions. The environmental conditions examined were salinity, temperature, and inferred infection pressure from adjacent feral oysters.

Clear salinity and water temperature threshold relationships to infection acquisition rate were observed. Because the ranges of mean salinities and water temperatures overlapped for Spring and Fall deployments, overall correlations of salinity and temperature with infection acquisition rate were weak. However, all "rapid" infection acquisitions occurred among Fall deployments and were associated with mean water temperatures above 24[degrees]C. All "slow" infection acquisitions occurred among Spring deployments and were associated with mean salinities below 12 [per thousand]. Cohorts that demonstrated "moderate" infection acquisition rates experienced environmental conditions above this salinity threshold and below this temperature threshold.

Presumptive infection pressure, expressed as mean rank infection intensity from adjacent feral oyster populations, was associated with the seasonal division of "slow" and "rapid" infection acquisition rates. Mean infection pressures on Spring deployments were low (0.5-2.1) and mean infection pressures on Fall deployments were consistently high (3.1-4.6). Because the ranges of infection pressure did not overlap between Spring and Fall deployments, infection pressure showed a strong overall correlation with infection acquisition rate (r = -0.75). However, because cohorts that acquired infections at "moderate" rates were distributed throughout the range of estimated infection pressures from feral oysters, within-season correlations between environmental infection pressure and infection acquisition rate were weak (Spring: r = -0.46, Fall: r = -0.06). Therefore, the correlation between infection pressure and infection rate provided no predictive value at any point in time. Hence, all subsequent analyses were performed on samples grouped by season or deployment.

Disease Progression and Associated Conditions

Like initial-infection rates, dermo disease progression in Fall 2000 and Spring 2002 deployments showed distinct seasonal influences, with prevalences and mean intensities following previously described seasonal patterns of maxima during summer and fall, and minima during winter and spring. Mean sample body burdens of infected samples were highly variable, but mean infection intensities showed regular seasonal variations that corresponded to seasonal variations in infection prevalences. Whereas oysters in cohorts from Fall deployments acquired infections more quickly than those from Spring deployments, infection prevalences among cohorts deployed during Fall 2000 remained low, and P. marinus was frequently undetectable through the subsequent winter and early spring months. As water temperatures increased during their first spring post-deployment, prevalences and mean intensities rapidly increased. During the second winter post-deployment, prevalences and mean intensities among cohorts at all sites declined. During the second spring, prevalences and mean intensities increased to, and sustained, higher values than during the previous spring and summer (Fig. 4). Mean infection intensities between 103 and [10.sup.6] P. marinus cells g t of oyster tissues, which respectively represent infection intensities associated with initial and lethal pathologies (Bushek et al. 1994), commonly occurred during the first summer and fall post-deployment, among juvenile oyster cohorts deployed during the previous fall (Fig. 4b). In Spring 2002 deployments, once oysters acquired infections, prevalences increased more rapidly than in Fall deployments. Within 3 mo of deployment, (with the exception of the cohort at TC) infection prevalences exceeded 90%. Whereas the prevalence at TC was only 10% at 3 mo, it had reached 53% by 4 mo. After reaching summer maxima, prevalences declined to near-zero values through the first fall and winter postdeployment. Infections were again detected during the second summer, with prevalences increasing during the summer and declining during the fall (Fig. 5). In some cohorts deployed during spring, mean infection intensity levels between [10.sup.3] and 106 P. marinus cells g I of oyster tissues, which are capable of causing pathological conditions, occurred during both the first and second summer and fall seasons post-deployment (Fig. 5b).

[FIGURE 4 OMITTED]

Strong increases in freshwater inflow, and corresponding decreases in estuarine salinities, influenced the magnitude of mean infection intensities. The 1999-2002 drought broke in late 2002. Summer 2003 maximum mean infection intensities for cohorts from the Fall 2002 deployment were orders of magnitude lower than those seen during summer for cohorts from the Fall 2000 deployment (Fig. 4, Fig 5).

[FIGURE 5 OMITTED]

Although there were strong differences in progression of dermo disease between deployments, results of Kolmogorov-Smirnov chi-square tests comparing distributions of prevalences or mean sample body burdens indicated that there were no significant differences between cohorts in any tested deployment. Although there were six deployments, there were insufficient samples from the Spring 2001, Fall 2001, and Spring 2003 deployments for comparison. All pairwise comparisons of prevalence in the Fall 2000, Fall 2002, and Spring 2002 deployments showed no statistically significant differences between cohorts (P values 0.513-1.000). Additionally, all pairwise comparisons of mean sample body burden showed that differences between cohorts were not significantly different (P values 0.098-0.841).

Comparison of disease progression between cohorts deployed at GA and SA during 2002 provided a measure of the effect of isolation from diseased feral populations. Although there were three deployments of these paired cohorts (Spring and Fall of 2002, and Spring 2003), there were insufficient samples from the Spring 2003 deployment for analysis. All comparisons using prevalence, mean sample body burden, and mean infected body burden showed P values >0.700, indicating no significant differences in disease progression between cohorts deployed adjacent to or >5 km distant from known populations of infected feral oysters.

Linear correlation was used to quantify the relationship between oyster size and disease severity (prevalence, individual body burden, mean sample body burden and mean infected body burden). As expected, based on generally longer exposure times for larger (older) oysters, the relationship between sample mean size and prevalence among all samples showed a significant correlation (r = 0.45 for wet weight, 0.51 for shell height). There was a very weak correlation between sample mean size and mean sample body burden (r = 0.13 for wet weight, 0.19 for shell height). There was no significant correlation between size and body burden (r = 0.03 for wet weight, 0.07 for shell height) among all individual animals (n = 7373), but the log relationship produced a fairly strong correlation (r = 0.30 for wet weight, 0.35 for shell height). Among infected samples only, there was also a very weak correlation between mean size of infected individuals and mean infected body burden (r = 0.11 for wet weight, 0.18 for shell height). Among infected individuals only (n = 1972), there was no significant correlation between size and body burden (r < 0.01 for wet weight, r = 0.07 for shell height), and a very weak correlation in the log relationship (r = 0.15 for wet weight, 0.21 for shell height).

DISCUSSION

Age and Size Effects

Our results show that juvenile oysters in mesohaline Chesapeake Bay regions can become infected by P. marinus in 8 wk or less. Because it was impossible to determine the exact day in a given sampling interval when the first infection was actually acquired, our estimates of time to first P. marinus infections in experimental juvenile oysters are conservative, maximum estimates (Fig. 2). Some oysters in each one of our experimental cohorts acquired infections within 8 wk ([less than or equal to] 67 d). Our results are consistent with those of other studies that examined very young oysters. As early as 1954, Ray (1954b) described new infections in two groups of juvenile oysters deployed adjacent to infected adults in Bayou Rigaud, LA, using RFTM assays of small tissue samples. Initial infections were detected at 6-8 wk of age in one group, and at 9 10 wk of age in a second group. In surveys of juvenile oysters in Galveston Bay, Ray (1987) found infections in 14% to 100% of sampled juvenile oysters. More recently in Chesapeake Bay, Paynter and Burreson (1991) detected infections at 45 d post-deployment among hatchery-reared juvenile oysters planted at 10-25 mm shell height. Ford et al. (2001) detected P. marinus in pooled samples from 8- to 10-mm oysters that had been exposed to Delaware Bay waters for 5-7 wk during the summer. Although there are additional studies of dermo disease in juvenile oysters, these have not generated disease data until the oysters were at least 8 mo old (Andrews & Hewatt 1957, Volety et al. 2000, Encomio et al. 2005). Our study assayed whole juvenile oysters for presence of P. marinus earlier and more frequently than others, and used a more sensitive ARFTM assay capable of routinely detecting parasite numbers orders of magnitude lower than those used by all but Ford et al. (2001). We detected Perkinsus marinus in all cohorts by 67 d post-deployment, at 6-35 mm shell height with a mean of 16 mm, and confirmed in situ infections in oysters from one cohort.

Our results indicate that initial infections that were detected by high-frequency sampling were usually characterized by low total parasite burdens, and that the proportion of infected oysters was typically low. Infection prevalences among our first-infected cohorts ranged from 3% to 73%, with 17/19 (89%) samples showing prevalences [less than or equal to] 17%. Total parasites enumerated from tissues of initially infected experimental oysters ranged from 1-23 P. marinus cells [oyster.sup.-1].

The less common instances (2/19, 11%) in which we detected initial infections at high prevalences (63% and 73%) also suggest that there were large differences in temporal and spatial levels of environmental infection pressures that produced initial infections in our experimental oysters. Both of these cohorts were located at GA, suggesting that some local characteristics of that site may have contributed to the concentration of infective parasite cells to levels that exposed larger numbers of experimental oysters to infective doses. Those characteristics may include large pulses of infective cells from nearby live and moribund feral oysters, as well as local hydrological properties that may act to import, retain, and concentrate parasite cells. Initial infections in these two cohorts occurred at high prevalences that created ostensible, nearly instantaneous epizootics in those experimental oyster populations within 8 wk of their deployments as SPF juveniles.

Salinity, Water Temperature, and Season Effects

This study tested the effects of salinity on P. marinus infection acquisition and long-term progression under natural environmental conditions. In Chesapeake Bay, differences in oyster habitats are characterized mainly by salinity, which is dominantly controlled by freshwater inflows. The drought that began in 1999 and continued through 2002 altered the usual salinity regime in the Patuxent River for four years. New salinity regimes were established that maintained a gradient along the river length, but salinities at all of our experimental sites were elevated above long-term averages. In contrast, the 2003 freshet depressed salinities below long-term averages at all sites, producing low-salinity conditions that would occur only at the farthest upstream site in a year with normal rainfall. Because of this natural temporal variability in freshwater inflows, our experimental sites could not be consistently defined by specific salinity ranges, and our infection acquisition results reflect the consequences of variable salinities. Cohorts deployed under freshet conditions generally acquired infections later than cohorts deployed during the drought, regardless of location. The exception was for cohorts deployed during Spring 2001 at GA and HP (lower salinity than TC), which also showed slow infection acquisition rates. Threshold effects of salinity on dermo disease transmission appear to have been operating in these cohorts.

We found that seasonal effects on infection acquisition rates appear to be the result of threshold effects of salinity and water temperature. The Spring 2003 deployment demonstrated that salinities below reported threshold values (Ford & Tripp 1996) overwhelmed any moderating temperature effects. Similarly, the Fall 2000 deployment demonstrated that temperatures near and above the reported threshold value of 25[degrees]C (Ford & Tripp 1996) overwhelmed inhibitory salinity effects. Because of the variable nature of salinity and water temperature, it is difficult to use intermittent field measures of these variables to accurately predict the disease status and potential for disease acquisition of local oyster populations. Cumulative field measures of temperature and salinity, expressed as means, combined with laboratory-determined threshold values, can be used to better predict the effects of environmental variables on P. marinus infection acquisition.

We found strong seasonal effects on infection acquisition rates. All "rapid" infection acquisition rates occurred among Fall deployments, all "slow" rates occurred among Spring deployments, and "moderate" rates occurred among both. Delayed placement of juvenile oysters into P. marinus enzootic waters until fall months to exploit declining water temperatures less favorable to P. marinus proliferation did not decrease the rate of disease acquisition in a biologically meaningful way. Juvenile oysters deployed during fall months are more likely to encounter waterborne P. marinus cells disseminated from heavily infected, dead, and dying feral oysters, than those deployed during spring months (Ragone Calvo et al. 2003). Even though infection prevalences and intensities were lower in early samples from Fall deployments, infections were still acquired rapidly, apparently overwintered, and thus provided established infections that intensified as soon as temperatures became suitable for P. marinus proliferation during the following spring.

Infection Pressure and Isolation Effects

Our results indicate that isolation of SPF juvenile oysters by >5 km from any known populations of infected feral oysters provided no protection from rapid acquisition of P. marinus infections during either drought or freshet years, at least under the salinity conditions (8 [per thousand] to 18 [per thousand]) extant at our paired mid-estuary experimental sites. Infections were detected earlier or simultaneously at the isolated site (SA) than they were at GA, which was adjacent to infected feral oysters. The primary mechanism of dermo disease spread is host-to-host transmission through the water column by parasite cells disseminated from dead and decomposing infected oysters (Ray 1954a, Andrews 1988, Ragone Calvo et al. 2003), and from parasites shed in feces of live infected oysters (Bushek et al. 2002). Large pulses in abundances of infective parasite cells in the water column coincide with mortality peaks of nearby infected oysters, and minor Perkinsus sp. abundance peaks found during periods of relatively low oyster mortalities suggest that other sources of infective cells may produce relatively low, continuous infection pressures (Ragone Calvo et al. 2003). Disease transmission may be influenced by proximity of infected and noninfected oysters, because the parasite must disperse to reach new hosts. Host proximity may therefore modulate dosage by altering dilution effects, and exposures exceeding a threshold infectious dose may govern acquisitions of new infections (Mackin 1962, Chu & Volety 1997). Although undetected patches of infected oysters might exist near our isolated site, our results demonstrate that P. marinus was transmitted through the water column for considerable distances from known populations of infected feral oysters, in abundances that were high enough to deliver infectious doses to our experimental oysters.

These results are consistent with those of previous studies in Louisiana and Chesapeake Bay. In early studies Mackin (1962) described cases where isolated populations of oysters in Louisiana waters acquired infections over distances as far as 33 km from a known source of infected oysters. In Chesapeake Bay, Andrews (1965, 1976, 1988) performed a number of proximity experiments with P. marinus-free adult oysters. In some cases disease transmission was suppressed, but not necessarily delayed, by distances of 15 m and 100 m from known infected oysters. In other experiments, separation did not prevent dermo disease from developing in disease-free oysters. Andrews (1988) cautioned that infected feral oysters consistently remain on, or are recruited to, cleared and fallowed oyster habitats as persistent sources of infectious P. marinus cells, and that complete removal of both infected oysters and cultch promoting feral oyster recruitment to oyster growing sites is nearly impossible. Andrews and Ray (1988) recommend isolation of disease-free juvenile oyster plantings by 0.4 km from infected populations, but caution that during a 2-y drought such a strategy was ineffective in preventing dermo disease transmission in Virginia waters. Burreson and Andrews (1988) also found that spatial isolation of oysters in Virginia waters did not prevent infections during drought conditions of 1985 to 1987. Therefore, under environmental conditions favorable to dissemination of infectious P. marinus cells, even oysters at significant distance from known sources of infective parasite cells may become infected at the same rate as oysters that are adjacent to infected oyster populations.

Disease Progression

We examined mean sample body burdens and mean individual body burdens over time to calculate rates of progression of infections; however sample variances were wide, and there were no consistent trends. Parasite loads in individual oysters varied widely, especially in the early stages of experimental cohort infection acquisition when infection prevalences and typical infection intensities were low. This may reflect variability in pathogen virulence and variability in individual oyster susceptibility. Data transformations were required for statistical analyses because of the scale of the body burden data and the relatively high sample variances. The latter resulted primarily from values that, under different circumstances, could be referred to as outliers. However, a few, heavily infected individuals within a given sample (thereby causing high sample variance) cannot be viewed as outliers (i.e., extreme random events); but instead may represent an integral functional component of dermo disease epizootiology. Low numbers of high-intensity infections in oyster populations may precipitate local dermo disease epizootics. Such occurrences of low prevalences of high intensity infections were observed by Ray (1954b) in early studies, and more recently by Ford et al. (1999).

According to Mackin (1962), once a nucleus of dermo disease is established in an oyster population, the density of that population (i.e., the proximity of hosts to one another) then controls the progress of the subsequent epizootic. Our results corroborate that observation, although the fact that our experimental oysters were contained at high densities may have promoted dermo disease transmission and progression, once initial infections occurred. Individual oysters that develop early heavy infections may serve as vectors for disseminating dermo disease to their neighbors, via fecal deposits or even early death and decomposition if conditions are favorable for rapid parasite proliferation. Infection intensities associated with early and overwintering infections were frequently below [10.sup.3] parasites [gram.sup.-1] oyster tissue, the detection limit of typical RFTM assays using tissue subsamples (Choiet al. 1989, Bushek et al. 1994), and prevalences were also low. During late summer and early fall, as conditions became favorable for P. marinus proliferation, infection intensities approached and reached potentially lethal levels of [10.sup.6] parasites [gram.sup.-1] oyster tissue (Choi et al. 1989, Bushek et al. 1994), with corresponding increases in prevalences resulting from inferred increases in numbers of parasites available in the immediate water column (Ragone-Calvo et al. 2003) (Fig. 4, Fig. 5).

The key difference between our Spring and Fall deployments was the speed at which infection intensities increased during the first months post-deployment. Because of exposure to increasing water temperatures, P. marinus infection prevalences and intensities increased more rapidly, and were sustained at higher values, among cohorts from Spring deployments than among those from Fall deployments. Although cohorts from Fall deployments acquired infections more quickly than those from Spring deployments, infection prevalences and intensities in cohorts deployed during fall remained low because of subsequent declining water temperatures. During drought conditions, P. marinus maintained overwintering infections in all cohorts. Low infection prevalences and intensities measured during winter demonstrated slight increases and decreases that were biologically insignificant. Where freshet conditions occurred during the first winter post-deployment, P. marinus was not detected in overwintering cohorts, as previously documented by Ragone Calvo and Burreson (1994).

Although there were statistically significant differences in speed of infection acquisition, and observed differences in early disease progression between Spring and Fall deployments, these did not result in differences in seasonal patterns of infection prevalence and intensity once the oysters entered their second summer. These patterns were similar to those described by Crosby and Roberts (1990) in a 2-y study of adult oysters in North Carolina waters.

Detection of Early and Low-intensity Infections

The whole body burden RFTM assay effectively detected low-intensity infections in our samples. Detection of initial P. marinus infections requires the ability to detect as few as one parasite cell, and only single cells were detected in some initial infections in our experimental cohorts (Table 2). Individual parasite cells stained blue-black with Lugol's iodine, were spherical, and measured 40-140 [micro]m in diameter. All initial infection intensities were below the detection limit of [10.sup.3] parasites [gram.sup.-1] tissue wet weight typical of RFTM assays that use solid tissue or hemolymph subsamples (Choi et al. 1989, Bushek et al. 1994). Ray (1952) concludes that the most sensitive RFTM assay results are obtained by staining parasites left behind after removal of oyster tissue, and Mackin (1962) notes that typical RFTM assay methods may not detect extremely light infections. Such methods frequently misdiagnose light infections as negative, and thus, standard RFTM methods used in earlier studies may not have been sensitive enough to detect early, low-intensity infections, even in very small oysters.

The whole body burden assay is crucial for detecting initial infections that are usually low-intensity by nature, because it ensures collection, and potential enlargement and detection, of any parasites present in sampled oysters. We found typical declines in prevalences and infection intensities in all winter and early spring samples, although infections continued to be detected sporadically during those seasons. Those results demonstrate that P. marinus is not completely eliminated from all oysters in affected populations, even though it may be absent or undetected in some samples or individuals.

Management Implications

We have demonstrated that under both drought and freshet conditions juvenile oysters in mesohaline Chesapeake Bay waters acquire P. marinus infections rapidly. Manipulating the deployment season does not result in a biologically meaningful change in time to acquisition of first infections. Specific-pathogen-free status does not prevent infection acquisition when juvenile oysters are planted in disease-enzootic waters. The high cost of SPF seed oysters, compared with their "natural" counterparts, may not be justified, given the short duration of their disease-free status once planted. Because very young oysters rapidly acquire enduring P. marinus infections, nursery propagation of hatchery seed in dermo disease-enzootic waters may quickly compromise the disease-free status of such oysters and destroy their value for subsequent introduction into waters requiring SPF seed. Isolation of SPF seed oysters by >5 km from infected oyster populations conferred no long-term benefits in delay of dermo disease acquisition or progression. Like similar results from higher salinity Virginia waters (Andrews & Ray 1988), this finding casts doubt on the efficacy of removing existing infected oysters from seed planting sites, or of simply planting seed oysters in areas distant from infected oyster populations.

Within estuarine systems characterized by inherently variable physical habitat conditions, geographic location is a weak predictor of water temperatures and salinities that drive dermo disease pathology and transmission dynamics. Therefore, geographic location cannot be used as a simple means to dependably identify or select oyster habitats with temperature and salinity characteristics adequate for oyster survival and growth, yet consistently protected from dermo disease impacts. Ranges of relevant environmental conditions also cannot be confidently predicted on the geographic and time scales used in this study. During times of severe or extended drought, even our most upriver sites ultimately exhibited salinities and disease prevalences typical of those at downriver sites during average freshwater inflow years (Albright et al. 2007). Temperature ranges also varied on the order of years. Thus, there are no locations that are consistently safe from conditions favorable to dermo disease, except perhaps the farthest upriver areas, where habitat is marginally adequate for oyster growth and survival. Our results may help refine management strategies for restoration and enhancement of Eastern oyster populations in Chesapeake Bay. Oyster management strategies must acknowledge that no practice thus far guarantees success in preventing the transmission of dermo disease. Management goals may be refined or redefined so that conditional successes in enhancing reproduction or fishery recruitment in the face of enzootic dermo disease can be maximized.

ACKNOWLEDGMENTS

The authors express their gratitude for the efforts of their late co-author Brian Albright for his contribution to the design and development of this study, construction of much of the equipment, and field sampling (including serving as research vessel captain) and lab work during the first year of this study. The authors are also grateful to Jud Blazek, Lee Hamilton, and Kristi Stevens who homogenized and digested thousands of oysters. Lee Hamilton also performed immunofluorescence stains. They also thank Dr. Don Merrit and Stephanie Tobash at the University of Maryland's Horn Point hatchery for providing the millions of oyster larvae used during this study, Dr. Mark Homer for early statistical guidance and analyses, and Heath Kelsey of NOAA for map graphics. The comments of anonymous reviewers improved the manuscript. Funding for this study was provided by the Sea Grant Oyster Disease Research Program (awards NA06RG0101 and NA96RG048), the Academy of Natural Sciences, and the Maryland Department of Natural Resources.

LITERATURE CITED

Albright, B. W., G. R. Abbe, C. B. McCollough, L. S. Barker & C. F. Dungan. 2007. Growth and mortality of dermo-disease-free juvenile oysters (Crassostrea virginica) at three salinity regimes in an enzootic area of Chesapeake Bay. J. Shellfish Res. 26:451-463.

Andrews, J. D. 1954. Notes on fungus parasites of marine bivalve mollusks in Chesapeake Bay. Proc. Natl. Shellfish. Assn. 45:157-163.

Andrews, J. D. 1965. Infection experiments in nature with Dermocystidium marinum in Chesapeake Bay. Ches. Sci. 6:60-67.

Andrews, J. D. 1976. Epizootiology of Dermocystidium marinum (=Labyrinthomyxa marina) in oysters. Proceedings of the First International Colloquium on Invertebrate Pathology. 172-174.

Andrews, J. D. 1988. Epizootiology of the disease caused by the oyster pathogen Perkinsus marinus and its effects on the oyster industry. Ant. Fish. Soc. Spec. Publ. 18:47-63.

Andrews, J. D. & W. G. Hewatt. 1957. Oyster mortality studies in Virginia. II. The fungus disease caused by Dermocystidium marinum in oysters of the Chesapeake Bay. Ecol. Monogr. 27:1-25.

Andrews, J. D. & S. M. Ray. 1988. Management strategies to control the disease caused by Perkinsus marinus. Am. Fish. Soc. Spec. Publ. 18:257-264.

Burreson, E. M. & J. D. Andrews. 1988. Unusual intensification of Chesapeake Bay oyster diseases during recent drought conditions. Proc. Oceans '85 Conf. pp. 799-802.

Burreson, E. M. & L. M. Ragone Calvo. 1996. Epizootiology of Perkinsus marinus disease of oysters in Chesapeake Bay. with emphasis on data since 1985. J. Shellfish Res. 15:17-34.

Bush, A. O., K. D. Lafferty, J. M. Lotz & A. W. Shostak. 1997. Parasitology meets ecology on its own terms: Margolis et al. revisited. J. Parasitol. 83:575-583.

Bushek, D., S. E. Ford & S. K. Allen. 1994. Evaluation of methods using Ray's fluid thioglycollate medium for diagnosis of Perkinsus marinus infections in the eastern oyster Crassostrea virginica. Ann. Rev. Fish Dis. 4:201-217.

Bushek, D., S. E. Ford & M. M. Chintala. 2002. Comparison of in vitro-cultured and wild type Perkinsus marinus, III. Fecal elimination and its role in transmission. Dis. Aquat. Org. 51:217-225.

Chakravarti, I., R. Laha & J. Roy. 1967. Handbook of methods of applied statistics, vol. 1. New York, NY: John Wiley and Sons,.

Choi, K. S., E. A. Wilson, D. H. Lewis, E. N. Powell & S. M. Ray. 1989. The energetic cost of Perkinsus marinus parasitism in oysters: quantification of the thioglycollate method. J. Shellfish Res. 8: 125-131.

Chu, F.-L. & A. K. Volety. 1997. Disease processes of the parasite Perkinsus marinus in eastern oyster Crassostrea virginica: minimum dose for infection initiation, and interaction of temperature, salinity and infective cell dose. Dis. Aquat. Org. 28:61-68.

Chu, F.-L., A. K. Volety & G. Constantin. 1994. Synergistic effects of temperature and salinity on the response of oysters (Crassostrea virginica) to the pathogen, Perkinsus marinus. J. Shellfish Res. 13:293.

Crosby, M. P. & C. F. Roberts. 1990. Seasonal infection intensity cycle of the parasite Perkinsus marinus (and an absence of Haplosporidium spp.) in oysters from a South Carolina salt marsh. Dis. Aquat. Org. 9:149-155.

Dungan, C. F., R. M. Hamilton, K. L. Hudson, C. B. McCollough & K. S. Reece. 2002. Two epizootic diseases in Chesapeake Bay commercial clams, Mya arenaria and Tagelus plebeius. Dis. Aquat. Org. 50:67-78.

Dungan, C. F. & B. S. Roberson. 1993. Binding specificities of mono-and polyclonal antibodies to the protozoan oyster pathogen Perkinsus marinus. Dis. Aquat. Org. 15:9-22.

Encomio, V. G., S. M. Stickler, S. K. Allen & F.-L. Chu. 2005. Performance of "natural dermo-resistant" oyster stocks survival, disease, growth, condition and energy reserves. J. Shellfish Res. 24:143-155.

Fisher, W. S. & L. M. Oliver. 1996. A whole oyster procedure for the diagnosis of Perkinsus marinus disease using Ray's fluid thioglycollate culture medium. J. Shellfish Res. 15:109-117.

Ford, S. E. 1985. Effects of salinity on survival of the MSX parasite Haplosporidium nelsoni (Haskin, Stauber, and Mackin) in oysters. J. Shellfish Res. 5:85-90.

Ford, S. E. 1996. Range extension by the oyster parasite Perkinsus marinus into northeastern United States: Response to climate change? J. Shellfish Res. 15:45-56.

Ford, S. E., A. Schotthoefer & C. Spruck. 1999. In vivo dynamics of the microparasite Perkinsus marinus during progression and regression of infections in eastern oysters. J. Parasitol. 85:273-282.

Ford, S. E. & M. R. Tripp. 1996. Diseases and defense mechanisms. In: Kennedy, V. S., R. 1. E. Newell, & A. F. Eble, Editors. The Eastern Oyster Crassostrea virginica. Maryland Sea Grant College, College Park. MD. pp. 581-660.

Ford, S. E., Z. Xu & G. Debrosse. 2001. Use of particle filtration and UV irradiation to prevent infection by Haplosporidium nelsoni (MSX) and Perkinsus marinus (Dermo) in hatchery-reared larval and juvenile oysters. Aquaculture 194:37-49.

Gieseker, C. 2001. Year 2000 Maryland Oyster Disease Status Report. Technical Report FS-SCOL-01-1, Maryland Department of Natural Resources Fisheries Service, Annapolis, MD. 27 pp.

Hewatt, W. G. & J. D. Andrews. 1954. Oyster mortality studies in Virginia. 1. Mortalities of oysters in trays at Gloucester Point, York River. Tex. J. Sci. 1954:121-133.

Hoese, H. D. 1962. Studies on oyster scavengers and their relation to the fungus Dermocystidium marhmm. Proc. Natl. Shellfish. Assn. 53:161-174.

Krantz, G. E. & S. J. Jordan. 1996. Management alternatives for protecting Crassostrea virginica fisheries in Perkinsus marinus enzootic and epizootic areas. J. Shellfish Res. 15:167-176.

La Peyre, M. K., A. D. Nickens, A. K. Volety, G. S. Tolley & J. F. La Peyre. 2003. Environmental significance of freshets in reducing Perkinsus marinus infection in eastern oysters Crassostrea virginica: potential management applications. Mar. Ecol. Prog. Ser. 248:165-176.

Mackin, J. G. 1951. Histopathology of infections of Crassostrea virginica (Gmelin) by Dermocystidium marinum Mackin, Owen, and Collier. Bull. Mar. Sci. Gulf and Caribbean 1:72-87.

Mackin, J. G. 1962. Oyster disease caused by Dermocystidium marimtm and other microorganisms in Louisiana. Pub. Inst. Mar. Sei 7:132-229.

Margolis, L., G. W. Esch, J. C. Holmes, A. M. Kuris & G. A. Schad. 1982. The use of ecological terms in parasitology (Report of an ad hoc committee of the American Society of Parasitologists). J. Parasitol. 68:131-133.

MDNR. 1996. Maryland Oyster Population Status Report: 1995 Fall Survey. Report No. MDDNRSP1-96, Maryland Department of Natural Resources Fisheries Service, Annapolis, MD. 16 pp.

Paynter, K. T. & E. M. Burreson. 1991. Effects of Perkinsus marinus infection in the eastern oyster, Crassostrea virginica: II. Disease development and impact on growth rate at different salinities. J. Shellfish Res. 10:425-431.

Ragone, L. M. & E. M. Burreson. 1993. Effect of salinity on infection progression and pathogenicity of Perkinsus marinus in the eastern oyster, Crassostrea virginica (Gmelin). J. Shellfish Res. 12:1-7.

Ragone Calvo. L. M. & E. M. Burreson. 1994. Characterization of overwintering infections of Perkinsus marinus (Apicomplexa) in Chesapeake Bay oysters. J. Shellfish Res. 13:23-130.

Ragone Calvo, L. M., C. F. Dungan, B. S. Roberson & E. M. Burreson. 2003. Systematic evaluation of factors controlling Perkbtsus marinus transmission dynamics in lower Chesapeake Bay. Dis. Aquat. Org. 56:75-86.

Ray, S. M. 1952. A culture technique for the diagnosis of infections with Dermocystidium marinum Mackin, Owen, and Collier in oysters. Science 116:360-361.

Ray, S. M. 1954a. Experimental studies on the transmission and pathogenicity of Dermocystidium marinum, a fungus parasite of oysters. J. Parasitol. 40:235.

Ray, S. M. 1954b. Studies on the occurrence of Dermocystidium marinum in young oysters. Proc. Natl. Shellfish. Assn. 1953:80-92.

Ray, S. M. 1987. Salinity requirements of the American oyster, Crassostrea virginica. In: Mueller, A. J., & G. A. Matthews, editors. Freshwater inflow needs of the Matagorda Bay System with focus on penaeid shrimp. NOAA Technical Memorandum NMFCSEFC-189, pp. E1-E28.

Ray, S. M. & A. C. Chandler. 1955. Dermocystidium marinum, a parasite of oysters. Parasitol. Reviews 4:172-200.

Valiulis, G. A. & J. G. Mackin. 1969. Formation of sporangia and zoospores by Labyrinthomyxa sp. parasitic in the clam Macoma balthica. J. Invertebr. Pathol. 14:268-270.

Volety, A. K., F. O. Perkins, R. Mann & P. R. Hershberg. 2000. Progression of diseases caused by the oyster parasites, Perkinsus marinus and Haplosporidium nelsoni, in Crassostrea virginica on constructed intertidal reefs. J. Shellfish Res. 19:341-347.

White, M. E., E. N. Powell, S. M. Ray & E. A. Wilson. 1987. Host-to-host transmission of Perkinsus marinus in oyster (Crassostrea virginica) populations by the ectoparasitic snail Boonea impressa (Pyramidellidae). J. Shellfish Res. 6:1-5.

CAROL B. MCCOLLOUGH, (1) * BRIAN W. ALBRIGHT, (2) GEORGE R. ABBE, (2) LINDA S. BARKER (3) AND CHRISTOPHER F. DUNGAN (1)

(1) Maryland Department of Natural Resources, Cooperative Oxford Laboratory, 904 S. Morris Street, Oxford, Maryland 21654; (2) Academy of Natural Sciences Estuarine Research Center, 10545 Mackall Road, St. Leonard, Maryland 20685; (3) Maryland Department of Natural Resources, 580 Taylor Avenue, Annapolis, Maryland 21401

* Corresponding author. E-mail: [e-mail]

TABLE 1.
Observed infection acquisition rates and corresponding measures
of infection severity. Fast = <2 wk, Moderate = 2-4 wk,
Slow = 4-8 wk. MIBB = mean infected body burden.
low 1-60, high >60. Prevalence - low <20%, high >60%.

                      Infection Acquisition
                              Rate
         Number of                            MIBB    Prevalence
          Cohorts    Fast   Moderate   Slow   Level     Level

             4        3        1              high       low
             1        1                       high       high
             1                          1     low        high
            13        1        7        5     low        low
Totals      19        5        8        6

TABLE 2.
Infection prevalence and intensity in the initially infected sample
from each cohort, and mean parasite counts from individual infected
oysters from these samples. % Prevalence = n infected/sample n. MSBB =
mean sample body burden, MIBB = mean infected body burden.

                                                  P. marinus
                                                cells infected
                                 Prevalence    [oyster.sup.-1]
Deployment       Cohort             (%)       (mean [+ or -] SD)

September 2000   Town Creek           3               1
                 Gatton               3               3
May 2001         Town Creek          10         1 [+ or -] 0
                 Gatton              17       1.2 [+ or -] 0.4
                 Holland Point        3               3
August 2001      Town Creek           7       1.5 [+ or -] 0.7
                 Gatton              73       4.5 [+ or -] 5.4
                 Holland Point        3               1
May 2002         Town Creek           7         1 [+ or -] 0
                 Gatton              63       3.5 [+ or -] 4.2
                 Sandgates           13         1 [+ or -] 0
                 Holland Point        3               1
September 2002   Town Creek           7       1.5 [+ or -] 0.7
                 Gatton              17         3 [+ or -] 1.6
                 Sandgates           17               1
                 Holland Point        7         1 [+ or -] 0
June 2003        Town Creek           3               1
                 Gatton               3               1
                 Holland Point        7       2.5 [+ or -] 0.7

                                           MSBB
Deployment       Cohort             (mean [+ or -] SD)

September 2000   Town Creek          2.8 [+ or -] 15.2
                 Gatton             10.0 [+ or -] 54.8
May 2001         Town Creek         10.6 [+ or -] 33.4
                 Gatton              0.8 [+ or -] 2.4
                 Holland Point       0.3 [+ or -] 1.5
August 2001      Town Creek         22.6 [+ or -] 97.1
                 Gatton            264.7 [+ or -] 290.1
                 Holland Point      0.01 [+ or -] 0.0
May 2002         Town Creek          2.2 [+ or -] 8.6
                 Gatton             11.7 [+ or -] 20.9
                 Sandgates           7.5 [+ or -] 23.2
                 Holland Point       0.7 [+ or -] 4.1
September 2002   Town Creek          3.1 [+ or -] 12.4
                 Gatton             10.0 [+ or -] 25.6
                 Sandgates           3.4 [+ or -] 9.1
                 Holland Point       3.2 [+ or -] 12.3
June 2003        Town Creek         0.04 [+ or -] 0.2
                 Gatton             0.05 [+ or -] 0.3
                 Holland Point       1.0 [+ or -] 4.3

                                           MIBB
Deployment       Cohort            (mean [+ or -] SD)

September 2000   Town Creek                83.3
                 Gatton                   300.0
May 2001         Town Creek        105.7 [+ or -] 32.4
                 Gatton              4.7 [+ or -] 4.5
                 Holland Point              8.3
August 2001      Town Creek        338.7 [+ or -] 212.1
                 Gatton            361.0 [+ or -] 278.1
                 Holland Point              0.2
May 2002         Town Creek         32.5 [+ or -] 13.0
                 Gatton             18.5 [+ or -] 24.3
                 Sandgates          56.3 [+ or -] 39.2
                 Holland Point             22.2
September 2002   Town Creek         45.7 [+ or -] 23.7
                 Gatton             60.2 [+ or -] 31. 1
                 Sandgates          20.6 [+ or -] 12.4
                 Holland Point      48.0 [+ or -] 6.5
June 2003        Town Creek                 1.3
                 Gatton                     1.6
                 Holland Point      15.3 [+ or -] 9.7

TABLE 3.
Long-term (1984-2004) mean spring salinities compared with
2001-2003 mean spring salinities for all deployment sites.
Site salinities during drought years 2001 and 2002 contrast
with those during freshet year 2003, when spring salinities
were driven down to 2001

    Long-term
      means              2001                2002

          [per                [per                [per
Site    thousand]   Site    thousand]   Site    thousand]

           16                           TC        16.1
           15                           GA/SA     15.3
           14       TC        13.6
           13       GA        13.0
           12                           HP        12.2
           11
TC         10
            9
GA/SA       8       HP         8.1
            7
HP          6

    Long-term
      means              2003

          [per                [per
Site    thousand]   Site    thousand]

           16
           15
           14
           13
           12
           11        TC       10.5
TC         10
            9       GA/SA      9.1
GA/SA       8
            7
HP          6        HP        5.5