Experimental manipulation of suspended culture socks: growth and behavior of juvenile mussels (Mytilus spp.).
KEY WORDS: mussel aquaculture, Mytilus, bivalve growth, competition, suspended culture, carrying capacity
High population densities characteristic of many suspension-feeding benthic invertebrates such as bivalves, tunicates, and hydrozoans may lead to intraspecific competition (Okamura 1986, Boromthanarat & Deslous-Paoli 1988, Frechette et al. 1992, Mueller 1996, Taylor et al. 1997). For example, it has been demonstrated in natural shellfish beds and in cultured populations that there is competition for space and food (Ceccherelli & Barboni 1983, Frechette & Lefaivre 1990, Frechette et al. 1992, Maximovich et al. 1996, Mueller 1996). Advantages to aggregation, common in benthic suspension-feeders, include protection from predators, reproductive success (Okamura 1986), and optimization of hydrodynamic regimes leading to higher seston fluxes (Gibbs et al. 1991). However, as density increases, crowded individuals may suffer negative effects such as the inability to effectively open valves for feeding and growth constraints caused by spatial limits (Bertness & Grosholz 1985, Frechette & Despland 1999). Aquaculture represents a special case of aggregation where the density of suspension-feeders is maximized to promote high commercial yields. Feeding and growth inhibition has been observed in cultured mussels, where high densities and complex webs of byssal threads restrict mobility (Cote & Jelnikar 1999, Frechette & Despland 1999).
In bivalves, crowding has various negative impacts on feeding, including restriction of shell gape and thus filtration behavior. Jorgensen et al. (1988) suggested that a reduction in the valve gape of mussels causes a retraction of the mantle edges that ultimately results in a decrease in the distance between gill filaments. This in turn affects the pressure pump and flow rate through the inhalant siphon (i.e., reduction in clearance rate and food uptake). Valve gape and siphon area are believed to be indicators of physiological activity such as pumping and filtration in mussels and other bivalves (Newell & Bayne 1980, Jorgensen et al. 1988, Dolmer 2000, Ward et al. 1992, Thorin et al. 1998, Thorin et al. 2001). Specifically, a strong positive relationship has been detected between blue mussel pumping rate and exhalant siphon area, as well as a relationship between valve gape, respiration, and feeding (Newell et al. 2001). Given the implied importance of shell gape to feeding, the impact of crowding on condition and growth should be indicated by reduced shell gape in cultured animals. There are, however, few corroborative observations of mussel arrangement in natural beds or culture conditions.
In addition to physical crowding, dense populations of suspension-feeders reduce water column seston concentrations, potentially resulting in food limitation (Smaal & van Stralen 1990, Gibbs et al. 1991, Lesser et al. 1992, Mueller 1996). The effects of seston depletion are known to occur at farm scales (Ibarra 2003), however, local depletion at the sock level and its implications for small-scale food availability are poorly understood (Alunno-Bruscia et al. 2000). At the ecosystem scale, the carrying capacity of a farm is determined by primary production and diffusion-advection, whereas locally it is determined primarily by physical renewal (Smaal et al. 1998). Acquisition and competition for food will affect the commercial yield (size and condition) of cultured mussels, where local-scale seston depletion could produce asymmetric competition resulting in a skewed size distribution caused by the mortality of smaller individuals (Lesser et al. 1992, Frechette & Despland 1999). Frechette and Lefaivre (1990) studied the interdependence of competition for food and space of mussels cultivated on bouchots using self-thinning relationships, and hypothesized that if competition for space produces lateral forces that reduce shell gaping, then the ensuing reduction in filtration would result in a decrease of food intake and growth. Alunno-Bruscia et al. (2000) tested this hypothesis in the laboratory and concluded that when crowding interferes with normal feeding, condition index will decrease inversely with mussel size.
There are mechanisms by which bivalves minimize physical and interference competition, such as density-dependent migration (McGrorty & Goss-Custard 1995), selection of orientation (Weissburg 2000), and extension of the mantle (Wildish & Kristmanson 1997). Studies of clumping behavior in sessile bivalves such as mussels have understandably been rare, because it is often assumed that bivalves display little postsettlement movement relative to spat (Cote & Jelnikar 1999). Yet, there is evidence that mussels, even as adults, are mobile (Mytilus californianus; Paine & Levin 1981), and this capability is well developed in juveniles. The foot provides locomotion by stretching, anchoring, and contracting, slowly pulling the mussel forward. Wildish and Kristmanson (1997) observed that juvenile mussels have the ability to cut their byssal threads, move, and reattach themselves in other locations. It is well known in mussel aquaculture that in the first week after stocking, mussels emerge and fix themselves on the sock mesh with byssal attachment.
Behavioral orientation has been observed in many aquatic organisms in response to predation, reproduction, and feeding (Stabell 1992, Weissburg & Zimmer-Faust 1994, Littorin & Gilek 1999, Weissburg 2000, de Vooys 2003). In the case of cultured mussels, it is proposed that they instinctively orient themselves towards their food source (i.e., into the current). In order for consistent orientation to occur, mussels must be stimulated by a quality and/or quantity signal and extract directional information from it (Sherman & Moore 2001). Directionality is important because the position of mussels on the rope can significantly affect their growth rate (Okamura 1986, Heasman et al. 1998). The importance of orientation is mediated via hydrodynamics, which play a major role in mussel feeding behavior (Frechette et al. 1989, Newell & Wildish 1997, Wildish & Kristmanson 1997, Newell et al. 2001). Mytilid feeding and growth has been shown to be positive at slower current speeds (Hildreth 1976, Wildish & Miyares 1990 and Newell et al. 2001), and inhibited at higher speeds (>25 cm [s.sup.-1]: Wildish & Miyares 1990, >20 cm [s.sup.-1]: Newell et al. 2001). Similar observations were made for deep-sea scallops (Placopecten magellanicus, Wildish et al. 1987, Wildish & Saulnier 1992), and for freshwater zebra mussels (Dreissena bugensis, Ackerman 1999).
Currents have a negative effect on exhalant siphon area (an indirect measurement of pumping rate) because of differential negative pressure when flow speeds exceed 20 cm [s.sup.-1] (Newell et al. 2001). Studies conducted on scallops suggest pressure differences between inhalant and exhalant siphons at high current speeds diminish the ability of the ciliary pump to capture particles efficiently (Wildish & Saulnier 1993). Highly mobile juvenile mussels orient themselves according to current speed and direction. At current speeds < 10 cm [s.sup.-1], mussels direct siphons towards the flow to increase particle flux, whereas at high flow speeds (>15 cm [s.sup.-1]) orient away from the flow to minimize negative pressure differences (Wildish & Miyares 1990).
Bivalve growth is also influenced by a variety of macro-environmental factors including temperature, salinity, and culture depth (Incze et al. 1980, Grant et al. 1993, Mueller 1996, Heasman et al. 1998, Archambault et al. 1999, Fuentes et al. 2000, Westerbom et al. 2002). Temperature has been observed to affect physiological condition and activity such as filtration (Incze et al. 1980), respiration (Widdows 1976) and reproduction in Mytilus sp. (Incze et al. 1980) with optima at 10[degrees]C to 20[degrees]C. Salinity has also been observed to be an important factor influencing growth rates. Although mussels can tolerate a wide range of salinity, shell growth is severely reduced under largely fluctuating salinity conditions (Incze et al. 1980) as well as in lower salinity conditions (Westerbom et al. 2002). Experiments examining the influence of depth on mussel production have yielded site-specific results yet they generally agree that ambient conditions rather than depth per se will affect production (Mueller 1996, Fuentes et al. 2000, Karayucel & Karayucel, 2000, Westerbom et al. 2002).
There are surprisingly few observations on the effects of stocking density on mussel growth in culture, despite its importance to the industry. Optimal stocking density is defined as the initial population density leading to maximum individual growth with minimum loss in density in the shortest time span (Frechette et al. 1996). Whereas we have emphasized overcrowding and food limitation, high densities also increase the risk of falloff from the socks and thus loss of commercial product. Previous research in Atlantic Canada, based on short-term experiments and biomass-density diagrams, suggested optimal stocking densities of juvenile mussels to be roughly 120-500 individuals per 30.5 cm (1 foot) of sleeve (Mallet & Myrand 1995, Frechette et al. 1996). Mueller (1996) determined those growth rates of Mytilus trossulus in raft culture varied significantly according to rope position within the raft but not rope quantity or depth. Fuentes et al. (2000) found that the stocking density of mussels (M. galloprovincialis) on rafts had no significant effect on the growth of the mussels. There has yet to be a detailed visual study on the effects of stocking density on the behavior and growth of M. edulis and M. trossulus juveniles in longline culture, especially relative to movement and orientation.
In the present study, our approach was to establish experimental culture socks identical to farm practices, and to observe mobility and growth of seeded spat over time. Considering the potential for crowding effects in cultured mussels, we hypothesize that (a) the positioning of juvenile mussels will evolve over time as individuals adjust their spatial configuration in response to crowding; mussels remaining in the interior of the sock will show reduced growth, (b) repositioning, orientation, and mobility will be density-dependent, (c) crowding will be reflected in the distribution and extent of shell gape and siphon area, (d) other factors such as temperature, current velocity, and culture depth that affect mussel feeding and growth will influence the above predictions.
MATERIALS AND METHODS
Mussels produced in Atlantic Canada are grown in suspended culture. The longline system consists of a rope (80-100 m) floated by buoys and anchored at either end. From this longline, mesh socks filled with mussels (see below) are suspended vertically in the water column until they reach commercial size (1.5-2 y; 5 cm shell length). Seed collection in Atlantic Canada is conducted from mid-May through September, and stocking continues until mid-December. At this time, longlines are submerged below the ice until April. The two species primarily cultivated in Atlantic Canada are the blue mussels, Mytilus edulis and M. trossulus. They were assumed to be the same species until the early 1990s, because it is difficult to distinguish them by external features (Mallet & Myrand 1995). Mytilus trossulus has a narrower, more delicate shell, and is more vulnerable to fracture during commercial processing (Mallet & Myrand 1995). Furthermore, it is believed to have lower annual meat yield and lower shell growth, thus further reducing its economic value (Mallet & Carver 1993).
The in situ culture experiment was conducted from September 2000 to August 2001 in Ship Harbour, an estuary on the Eastern Shore of Nova Scotia, Canada (Fig. 1). It is a long and narrow fjord (8 x 1 km) with the deepest water situated at the head of the system (Gregory et al. 1993). The farm, chiefly located in the inner basin, produces M. edulis and M. trossulus. At the upper extremity of the bay, Ship Harbour River discharges freshwater into a basin (15-25 m depth) whereas at the southern end of the estuary a shallow sill (7 m) is present. Tides are semidiurnal with a range of 1.4-2 m (Gregory et al. 1993).
The experiment consisted of observations of the effects of stocking density (high and low), culture species (M. edulis and M. trossulus) and culture site through repeated measurements (in situ images as well as direct sampling) of mussel shell length (SL), dry tissue weight (DW), condition, emergence, and orientation over a period of 11 mo. In addition, measurements were made on mussel valve gape and siphon area in relation to current speed and stocking density.
[FIGURE 1 OMITTED]
The two experimental sites were situated at the extremities of the farm culture area (Fig. 1). The first site was located just above the sill that separates the inner bay from the mouth of the estuary (Site 1-Outer). The second site was near the Ship Harbour River at the head of the estuary (Site 2-Inner). Sites were selected to represent different growing environments as well as being located outside the regular culture area to minimize interference from the farm.
Mussel seed purchased from a local farm in September 2000 were graded by size and socked as specified below before transplantation to the two sites. Italian tubular mesh socks used in the experiments have 40 mm openings, and are able to contain juvenile mussels of >25 mm in length. In each of the two sites, one longline of roughly 25 m was set up, suspending 12 socks (3 m in length) vertically in the water column. The longline was weighed down at both ends with cement blocks and buoys attached along the longline were used to keep it afloat. Each longline held triplicate socks 75 cm apart stocked at low density (~400 mussels/30.5 cm of sock; density used at Ship Harbour farm) and triplicates stocked at high density (~800 mussels/30.5 cm of sock) for both M. edulis and M. trossulus. The socks were deployed for approximately 1 y on September 18, 2000. Samples of 20 mussels from each sock were frozen to measure initial SL, DW, and condition.
A time series of in situ images of socks were obtained using a Kodak DC290 digital camera (resolution 1440 x 960 pixels) in a waterproof housing attached to a graduated pole lowering system. A Benthos underwater strobe was attached to the housing at a 45[degrees] angle from the camera lens direction, firing when the shutter was triggered. Photographs were taken at 1, 2, and 3 m sock depth, initially obtained twice daily for 5 days (September 18-22) and then once daily on September 27, October 5, October 27, and December 5. One sock of each design was retrieved and frozen at -20[degrees]C on December 5, 2000, March 22, and August 28, 2001 for laboratory analysis.
During the research period, experimental mussel lines were treated in the same manner as the farm's commercial lines. Occasionally, seastars (Asterias vulgaris) and fouling were manually removed from the socks. During the winter months (mid-December to mid-March) longlines were submerged to prevent ice damage.
During the first week of sampling, data were collected for temperature, salinity, current speed, and direction via InterOcean $4 current meters/CTD deployed at each site. Physical data were recorded every 30 min from September 18-22 at 4 m below the surface. Triplicate water samples for chlorophyll-a measurements were collected using a 5-L Niskin bottle at the three sampling depths at both sites on the morning and evening of September 20. One liter of seawater was filtered through Whatman 25 mm GF/C filters within 4 h of sampling. Chlorophyll-a concentrations were determined fluorometrically following acetone extraction (Strickland & Parsons 1972).
For each of the frozen socks obtained in December (month 3), June (month 9), and August (month 11) five mussels were randomly obtained from the interior and exterior of the sock at each of the sampling depths (1, 2, and 3 m). The "interior" was defined as mussels not having emerged from the mesh sock, whereas mussels were considered to be "exterior" or emerged if 2/3 or more of the shell was free of the culture sock. Shell length (cm) was measured for each sample mussel to the nearest 0.1 mm with calipers. Tissue was removed and weighed to the nearest 0.05 mg after drying 48 h at 60[degrees]C. A condition index was calculated as (dry weight [DW, [micro]g]/shell length [[SL, cm].sup.3]). Site 1-Outer had no data after December because of seaduck predation and the high density M. edulis culture sock was lost during winter at Site 2-Inner.
Images of mussels were analyzed with SigmaScan Pro v. 5 (SPSS Science, 2003) using known sock mesh dimensions as a reference scale. Image calibration was verified with direct measurements of SL. Variables measured were SL (cm), valve gape opening index, siphon area ([cm.sup.2]), emergence, and orientation of individuals outside of the sock. Shell length was measured on individual mussels from their umbo to the longest axis of the shell. Valve gape data were sorted into 5 categories: 0%, 25, 50%, 75%, and 100% shell opening (0% = completely closed: 100% = completely open). Siphon area was measured only on mussels whose angle allowed an adequate view and measurement of SL. A siphon index was calculated by dividing the siphon area by the SL of its corresponding mussel. Emergence was calculated as the number of mussels outside the culture sock in relation to all the mussels visible in an image, normalized to 0.5 m sock length.
We note that the images did not prove effective for tracking absolute mussel density on the socks. Because of the threedimensional aspects of the sock, two-dimensional images do not provide the necessary information to quantify changes in mussel numbers through time. Mussel density is also heterogeneous along the length of the sock (clumps or empty sections) so that even sections of sampled socks do not permit assessment of total sock population numbers.
Dependent variables measured in this study consist of two types, those collected from analysis of in situ images (mussel size, orientation, gape characteristics) and those obtained from sampling the experimental socks (shell and tissue metrics). The former variables represent a detailed time series related to the development of the sock population, and are described first. Data from the three sampling depths revealed statistically equivalent results and were combined for all further analysis. Image analysis data for individual mussels were averaged to obtain a single data point per image and within-sock variability was not considered. Analyses performed on these data considered triplicate sample socks as well as combined data obtained for a single day (>1 image per day early in the study). Preliminary analyses indicated that there were also no differences in shell length among replicate socks, allowing the minimum sample size for a single treatment combination to be 9 (3 depths * 3 replicated socks * # images per day). All data were tested for normality (Shapiro-Wilks), homogeneity of variance (Bartlett test), and transformed where necessary (Zar, 1999). The results of tests with no significant results are not presented in tables.
Two-way analysis of variance (ANOVA), with factors density and location, were performed separately for both species to compare mussel emergence over 0.5 m of culture sock on days 3 and 39. Because density was not a significant treatment in these two-way ANOVAs, density was combined in one-way ANOVAs using emergence as the dependent variable and species as a treatment for Site 1-Outer and Site 2-Inner separately. These analyses were conducted for both days 3 and 39.
Valve Gape and Siphon Index
Current speeds, obtained periodically during the first weeks, were matched with data on siphon index and valve gape for the same time period. The frequency distribution of current speeds was very distinct, allowing a post hoc grouping into 3 range categories (2: 0-4 cm [s.sup.-1]; 10:10-11 cm [s.sup.-1]; 15:14-15 cm [s.sup.-1]; see below). Because there was no significant species difference in valve gape and siphon index (t-tests: P > 0.05), the results for both species were pooled to obtain a larger sample size. Valve gape and siphon index data were then examined by two-way ANOVA against current speed and stocking density for Site 1-Outer only. Although shell gape was initially categorical data, means obtained from single images generated a continuous variable used in ANOVA.
Orientation of individual mussels was characterized by dividing them into positive Cartesian categories determined by drawing a line from the umbo to the longest part of the shell and then associating the resulting angle with its degree class. The data were classed into 22.5[degrees] intervals starting from 0[degrees] (mussel orientation pointing towards the sea surface), to 180[degrees] (mussels pointing towards the sea bottom). Orientation data were collected only on mussels considered emerged. Circular statistics were used to analyze orientation data by converting degrees to radians and calculating the resultant vector. A Rayleigh uniformity test was used to determine if data were uniformly distributed or concentrated around a mean direction. Watson-Williams F-tests, equivalent to ANOVAs for the mean vector of each sample, were performed to compare mean angles of orientation between treatments. One-way analyses were conducted separately for density and species at each location for the first week of submergence combined and at day 39. Circular statistics were performed using Oriana v. 2 (Kovach Computing Services, 2005; http://www.kovcomp.co.uk).
Size measurements based on samples from collected socks were used to determine SL, DW and condition after 3 mo of the sampling period. Although the experiment was carried out over a 1 y period, most of the data presented are from month 3 (December), the last samples collected with all treatments represented. Two types of analysis were conducted using these dependent variables. First, as with measurements from images, two-way ANOVAs were carried out using site and stocking density treatments. Next, two-way ANOVAs were executed using stocking density and emergence (interior, exterior) treatments. In the latter analyses, data for both species were combined since preliminary assessment indicated no difference in their response. A few of these cases did not achieve normality via transformation, but parametric ANOVAs were applied for consistency, given that ANOVA is robust in spite of these violations (Peckham & Sanders, 1972). Inflation of type I error through repeated analyses is noted, however two-way ANOVAs were deemed effective for the isolation of site and density effects. This approach also avoids interpretation problems of higher order ANOVAs or multivariate tests.
Observations of temperature, salinity, and current speeds revealed that the two sites were different from one another, with faster currents and warmer saltier water at Site 1-Outer. Mean salinity for Site 2-Inner near the Ship Harbour River averaged 29.2, with little variation and only a slow increase during the initial two weeks of the study. Salinity at Site 1-Outer averaged 29.6 with more variability and a slightly larger range than Site 2-Inner, likely fluctuating between river and ocean influence. The temperature differed between the two sites during the two-week sampling period by ~2[degrees]C with lower temperatures at Site 2-Inner, and slightly more variability at Site 1-Outer. Current meter records demonstrated higher flow speeds near the mouth of the estuary, ranging from 0.250.7 cm [s.sup.-1]. Currents follow the general axis of the estuary, dominated by north-northwest tidal flow, as well as southeast ebb, possibly affected by the sill. The bay narrows at the head but is considerably deeper with generally constant and relatively low current speeds (average = 2.5 cm [s.sup.-1]) during the sampling period. This area of the estuary has a predominantly southeast current, following the shoreline near the site. A lower velocity northern current was also observed.
The experiment began before the fall phytoplankton bloom. Chlorophyll-a levels at both sites were similar (mean = 1.84 [micro]g [L.sup.-1]) and pigment stratification between the three sampling depths was not observed. These data are comparable to previously gathered measurements at Ship Harbour (Keizer et al. 1996, Strain 2002).
Some insight into the range of environmental variation during the study period may be obtained from concurrent studies (Ibarra 2003). Data collected within the mussel lease near Site 1-Outer identified two storms with wind speeds exceeding 40 km [h.sup.-1] on October 6 and October 15. The first storm apparently induced an upwelling event within 3 days and was responsible for an increase in salinity of 1 and decrease in temperature of 6.8[degrees]C. The pycnocline also rose at from 6 m to 2 m below the surface. A four day bloom beginning on October 16 increased chlorophyll-a concentrations by a factor of three. Similar blooms have been recorded in October in Ship Harbour (Keizer et al. 1996).
Mussels emerged relatively slowly from within the sock, with 70% occupying the "outside" by 39 days (Fig. 2). Representative images from the time series indicated considerable movement in the socks during the first week of deployment. In the initial hours after submergence, mussels were entirely inside the sock and filtering (valves open). Within four days few mussels were completely emerged although some individuals were partly outside the sock (Fig. 2). At this time preferential orientation is evident with mussel valves pointing towards the exterior of the sock and umbos towards the center. After 9 days the sock is becoming obscured by emergence, a process that is largely complete by day 17. At 17 days, mussels are mostly emerged and settled, although some mussels remain inside the culture sock. As of 39 days, sock images show mussels to be completely emerged and feeding, with little change between this date and day 78. The mature sock demonstrates that the proportion of the population characterized as emerged far exceeds the proportion of interior mussels.
Mussel emergence through time based on image analysis of all experimental socks was described by a semilog relationship (Emergence = -7.5142 + 13.3838 * In Days, [R.sup.2] = 0.48). Two-way ANOVAs of emergence for each species as a function of density and site treatments indicate that for M. edulis there was significantly greater emergence at Site 2-Inner than Site 1-Outer at day 39 (Table 1). It is a notable result that higher density stocking did not invoke greater mussel emergence for either species. Mytilus trossulus initially emerged more rapidly than M. edulis at Site 1-Outer (one-way ANOVA, P = 0.0002), but at Site 2-Inner and later dates (day 39) there were no significant interspecific differences.
Valve Gape and Siphon Index
A frequency distribution of currents corresponding to the timing of sock images indicated three distinct categories, 0-4, 10-11, and 14-15 cm [s.sup.-1], with medians at 2, 10, and 15 cm [s.sup.-1], respectively. Shell gape measurements (species combined) revealed that regardless of the current speed, mussels generally did not close their shells completely (Fig. 3A). No trends or patterns in gape were apparent as a function of current speed category (P > 0.05). Similarly, valve gape displayed no differences between density treatments.
When siphon index was plotted as a function of flow speed category there was a slight increase at 10 cm [s.sup.-1] (Fig. 3B). Considerable variation was observed in siphon index, even within speed categories. Two-way ANOVA results (species combined) indicated a significant difference in siphon index among the three current speed categories, and SNK posthoc comparison tests verified that siphon index at the middle speed category was significantly greater than the low and high categories (Table 2). This suggests an optimal speed for feeding activity.
Density alone had no effect on siphon index, but a positive interaction occurred for density and current speed (Table 2). Mussels in high density socks exhibited slightly larger siphon indices relative to low density at the low current speeds (0-4 cm [s.sup.-1]) as well as at 14-15 cm [s.sup.-1] whereas the opposite was observed at 10-11 cm [s.sup.-1] (Fig. 3B).
Mussels displayed preferential orientation at both study sites (See Fig. 4 for example), with r-values for orientation ranging from 0.86-1.0 for both locations and times, and significant uniformity in orientation (Rayleigh P < 0.05, with the exception of low density Mytilus edulis for Site 2-Inner at day 39 because of a single data point). Initially the orientation of mussels approximated 80[degrees], with the exception of M. edulis low density treatment that had a mean orientation of ~87[degrees]. Over time, mussel orientation changed towards the surface with means ranging from 56-78[degrees]. There was an effect of density and species at Site 1-Outer for the first week of submergence, where M. edulis showed a greater preference towards horizontal orientation at low density compared with high density and compared with M. trossulus, within the low density treatment (Watson-Williams P = 0.015 and 0.007, respectively). There was no significant effect of density or species at Site 2-Inner for either the first week or day 39 (P > 0.05).
[FIGURE 2 OMITTED]
Mean shell length (SL) monitored from in situ images showed a positive relationship with time but no apparent growth until day 9 (SL = 2.445 + 0.017 Days, r = 0.97, Fig. 5).
Direct sampling of experimental socks for growth showed that SL was relatively uniform across treatment and species in December (month 3). Shell length in M. edulis displayed no differences as a function of density and site (Fig. 6A). Results for M. trossulus showed a significant effect of site on SL, but no density influence (Table 3). A significant interaction term indicated that the 20% increase in SL found at Site 1-Outer was detectable only at low density.
December results for dry tissue weight (DW) showed more obvious species and treatment effects than SL (Fig. 6B). For M. edulis, both site and density factors were significant (Table 4), with 22% greater tissue weight at Site 2-Inner than Site 1-Outer, and a 16% increase at high density compared with low density (Fig. 6B). Treatment effects were significant in M. trossulus as well (Table 4), with tissue weight greater at Site 2-Inner, by 25% (Fig. 6B). Density was also significant, although differences were in opposite directions at Site 1-Outer compared with Site 2-Inner. The interaction term was not significant however, because the results are strongly influenced by the doubling of DW in high compared with low density stocking at Site 1-Outer (Table 4).
[FIGURE 3 OMITTED]
Based on the contribution of SL and meat weight, calculation of condition provides a relative index of growth, with consistent results for both species. Mytilus edulis had significantly greater condition (17% increase) at Site 2-Inner compared with Site 1-Outer, and significantly greater condition at high density compared with low density at Site 1-Outer (Fig. 6C, Table 5). Mytilus trossulus displayed similar density effects (16% increase in condition at high density), but site differences were not significant (Fig. 6C, Table 5).
A comparison of growth between the two species indicated similar SL initially, but with M. trossulus becoming significantly larger by month 3 at both sites (one-way ANOVA for each site; P < 0.042). Mytilus edulis initially had higher condition when socks were filled (one-way ANOVA; P < 0.016), but no interspecific difference in condition persisted through time.
Analyses of the effects of density and emergence (interior, exterior) on SL showed that exterior sock mussels (species combined) were significantly larger in every case, and that the effects of density and time were consistent at both sites (Table 6), a result that was not obvious from analyses of each species in site/density treatments (Table 3). Namely, during month 3, high density culture socks produced significantly better shell growth at both sites (8% in both cases), and emerged mussels revealed significantly greater SL (11% and 13% for Site 1-Outer and Site 2-Inner, respectively, Table 6A). A significant interaction term at Site 1-Outer demonstrated that the high density factor was beneficial for mussels in the interior of the culture socks (13%), but did not influence SL of the emerged mussels. The opposite was true at Site 2-Inner, where some aspect of crowding of interior mussels had a negative influence on SL (11%). The same analysis completed at months 9 and 11, for Site 2-Inner, consistently revealed emergence as being an important factor with the exterior mussels presenting greater SL (Table 6B). Density was a negligible factor at these times.
[FIGURE 4 OMITTED]
Dry weight results were comparable to SL with significantly enhanced tissue weight for emergence at 3 mo (19.6% and 20.6% for the outer and inner sites, respectively) as well as at months 9 and 11 (65.2% and 71.2%, respectively [Table 7A, B]). The density factor was also significant in month 3 for both sites (42.6% and 12.6% for the outer and inner sites, respectively) with high density culture generating better meat weight.
Mussel condition as a function of these factors at month 3 indicated significant effects of density (11% increase in condition at high density), but not emergence at Site 1-Outer (Table 8). Contrary to these results, Site 2-Inner had no density effect yet significant effects of emergence on condition with interior mussels increased in condition by 13% (Table 8). An interaction revealed better conditions for exterior mussels in the high density culture socks (15%). The results obtained at months 9 and 11 showed no effects for either density or emergence.
[FIGURE 5 OMITTED]
A quantitative examination of the development of a socked culture population produced new insights into the early stages of mussel culture and intraspecific competition in these species. Following submergence of the experimental socks at Ship Harbour in September, a slight increase in shell length became perceptible after 9 days. The lack of production during the first week of the grow-out period may be attributable to transplantation shock, acclimation to the environment, and energy spent towards emergence and byssal attachment to the substrate. The presence of a fall phytoplankton bloom in mid-October likely contributed to the subsequent shell growth observed between October and December. Growth rates averaged for all treatments for the entire study period (11 mo) were 29.57 [+ or -] 0.42 SE mm in shell length and 73.3 [+ or -] 0.40 SE mg dry tissue weight, slightly higher than previously recorded for other Nova Scotia culture (Mallet & Carver 1989).
Tissue production was considerably higher at the inner site. Characterization of food sources was not extensive enough to compare sites throughout the study, but current speed is a significant aspect of intersite differences. Although the daily current speed averages were not likely high enough to generate a negative effect on mussel filtration rates (September, outer site 5.8 cm [s.sup.-1]; October, middle of basin 10 cm [s.sup.-1]; Ibarra 2003), the current at the outer site was variable and did exceed speeds of 20 cm [s.sup.-1] in "normal" weather conditions. At these times, the inner site did not exhibit spikes nor did current speeds ever exceed 9.9 cm [s.sup.-1] in September. Factors contributing to inferior production at the outer site may thus include higher, variable current speeds and their negative effect on mussel pumping (Wildish et al. 1987, Newell et al. 2001). This is discussed below in the specific context of shell gape and siphon area.
[FIGURE 6 OMITTED]
Compared with site, density is a less obvious determinant of early mussel growth in socks. The lack of any clear negative density effect on shell growth is counterintuitive because of crowding and space limitation expected at high density stocking (Bertness & Grosholz 1985, Okamura 1986). Experimental observations suggest that stocking densities have a more pronounced effect on mussel tissue growth than shell growth, but the nature of the results varied for Mytilus edulis and Mytilus trossulus. The results were somewhat as expected for M. trossulus, exhibiting lower tissue yields for high density stocking, but solely at the inner site. This finding concurs with previously recorded results indicating evidence for crowding at high density (Bertness & Grosholz 1985, Okamura 1986, Boromthanarat & Deslous-Paoli 1988). In contrast, M. edutis manifested noticeably better dry weights at high density stocking for both sites. The effects of stocking density on growth are difficult to interpret because effective density is believed to change with time. Movement and emergence of juveniles may serve to adjust initial densities to alleviate crowding and inhibition of gape and feeding. Simultaneously, consistent shell growth provides continuous pressure to make further adjustments such that almost all mussels are emerged after 3 mo. This process explains why there is no disadvantage of higher stocking densities, but cannot account for the further result of enhanced growth at higher density.
Negative density-dependent effects have been previously recorded for Mytilus sp. (Okamura 1986, Boromthanarat & Deslous-Paoli 1988, Alunno-Bruscia et al. 2000). However, studies performed on the arrangement of mussels in aggregate formations has revealed that position within groups is of consequence when considering density and that individuals at the edge of groups fare better animals in the center of groups (Okamura 1986, Newell 1990). Despite the overt appearance, there is not substantial horizontal layering of mussels on socks (personal observation); it could be argued that mussels growing on suspended culture socks are equivalent to edge mussels and do not suffer from overcrowding. In fact, studies performed on suspended culture revealed no benefit of thinning throughout the growth period (Ceccherelli & Barboni 1983). The densities used in the present study are larger than those used by Fuentes et al. (2000) who also found no negative effects of density on growth.
The disappearance of treatment effects through time has been previously noted for local density (Frechette et al. 1996), stock differences (Frechette et al. 1996), and rope densities on rafts (Heasman et al. 1998). It has been speculated that mussels will mitigate crowding through density-dependent self-thinning (Boromthanarat & Deslous-Paoli 1988, Frechette et al. 1992, Maximovich et al. 1996, Alunno-Bruscia et al. 2000). Because absolute numbers of mussels in the socks could not be followed via our methodology, it is not known if in additional to behavioral sorting, fall-off or mortality was important in our results.
It is noteworthy that mussels in this study achieved marketable sizes (>5 cm) within one year of culture, instead of 1.5-2 y expected in Atlantic Canada. The superior production observed in the experimental socks compared with those of the farm is of interest as the mussels were generally subject to similar conditions and handling. The most obvious differences of the experimental lines are their distance from other longlines and attendant seston depletion documented in Ship Harbour by Ibarra (2003). Experimental mussels were also able to take advantage of the fall bloom whereas new socks deployed as late as December in commercial culture would not benefit from this food event.
This study provides the first detailed examination of mussel emergence rates from culture socks. Juvenile mussels demonstrate considerable movement abilities after socking and submergence, with substantial foot activity, likely for chemoreception and/or positioning purposes. Mussels initially demonstrate preferential orientation with umbos towards the center of the sock and valve opening towards the outside, presumably to alleviate mechanical stress on valves and allow adequate filtration. Mussels closer to the mesh of the sock emerge first and appear to be propelled to the exterior by the pressure exerted by less favorably positioned individuals. The relative emergence of mussels at day 3 was 6% to 8.5% with no intersite differences. Thirty-nine days after deployment, emergence was 60% at the outer site and 81% at the inner site, a significant difference for M. edulis. The emergence and settling of mussels outside the sock was slower than indicated from a controlled laboratory environment, where they were observed to completely emerge in a few days (personal observation). This distinction may be explained by differences in the flow field around the sock.
Hydrodynamic characteristics of the two sites likely occupy a significant role in juvenile M. edulis emergence after deployment of the socks. The outer site characterized by stronger currents slowed emergence by ~30%. Mussels submerged in rougher conditions have a tendency to produce more byssal threads to secure their positions (Eyster & Pechenik 1987). Increased potential for detachment may reduce the rate of emergence, mytilus trossulus emergence however, was not affected by the conditions present at either site. These observations agree with those of Bates and Innes (1995) and Comesana et al. (1999) confirming that this species is frequently found in exposed sites.
No substantial effects of density were apparent on rate of juvenile emergence for either species, suggesting that intraspecific pressure does not force emergence. In addition, a denser array of overlapping mussels and byssus neither hinders nor benefits maneuverability of these species to the outside of the sock. Between the two species, M. trossulus had significantly faster emergence by 75% at day 3, possibly in association with its proclivity toward more extreme environments (Riginos & Cunningham 2005). Site differences in emergence do not appear to generate lasting effects on overall mussel production, because complete emergence occurs eventually even at high density.
The number of mussels remaining in the interior of the socks compared with the initial value was difficult to quantify because those that did not survive quickly decomposed, but small numbers (~20 per 30.5 cm) remained in the interior regardless of initial density. Live mussels collected in the interior of the socks demonstrated signs of mechanical stress such as shell thickening and distortion.
At all sampling dates, differences in shell length and tissue yields between interior and exterior mussels were apparent, favoring emergent mussels. High density culture socks generated superior shell and tissue production at both sites in month 3 of the experiment, with more pronounced effects of density on tissue yields than on shell length.
Although the meat yield was significantly lower for interior mussels, the corresponding lack of shell growth results in values of condition index comparable to exterior mussels. In addition to spatial limitation and shell deformation, food limitation is also likely caused by the minimal potential for shell opening. Although some interior mussels produced substantial meat yield, they were commercially unusable and thus considered a loss. Because the total quantity of mussels having been "lost" inside the culture sock could not be determined, an overall assessment of commercial losses could not be quantified. However, as we indicated above, emergence leaves relatively few mussels in the sock interior.
A drawback of this study is that total mussel numbers could not be tracked, and thus changes in density through the experiment were hard to assess. At higher densities, pressure from emerging mussels may increase falloff such that an equilibrium density occurs on the socks, and the high density treatment is not maintained. Falloff is not often documented in studies of mussel growth, because various cages or nets are used for containment rather than culture socks (Ceccherelli & Barboni 1983, Cote & Jelnikar 1999, Bayne 2000). Similarly, seastar predation may have reduced densities because emerged mussels and/or higher densities may be more vulnerable to predation. Previous studies of seastar predation on cultured mussels have demonstrated reductions in density (North pacific seastar, Asterias amurensis: 80 [m.sup.2] to 5-7 [m.sup.2]; Ross et al. 2003), however predation losses caused by the seastars or crabs have not been well quantified in North Atlantic cultures. Although sea duck predation was also observed in our study, there was no evidence that it was responsible for the partial thinning of the socks. These processes of mortality are a consequence of using a realistic culture method, representative of farm practices, but subject to less controlled experimental conditions.
Valve Gape and Siphon Area
It has been postulated that crowding in mussels culture leads to inhibited valve gape and reduced feeding and growth (Jorgensen et al. 1986, Boromthanarat & Deslous-Paoli 1988). For emerged mussels, density-dependent physical interference did not affect valve gape, a result consistent with growth results. Visual observation revealed that shells of the mussels were mostly clear of byssus in either density treatment. Although movement of juvenile mussels may provide a mechanism to reduce mechanical pressure of crowding, greater variability in shell gape was observed at high density stocking, suggesting that certain mussels may be subject to interference.
Current speeds ranging from 0-15 cm [s.sup.-1] produced no obvious effects on valve gape, confirming previous in situ studies for M. edulis, which indicate independence of valve gape and speed in a range from 10-30 cm [s.sup.-1] (Newell et al. 2001). Although it has been observed that gape and pumping rate are linked (Riisgard & Randlov 1981, Famine et al. 1986, Jorgensen et al. 1988), previous studies using obligate reduced gaping revealed that shell constraint had little or no effect on mussel growth (Frechette & Despland 1999). Furthermore, it has been observed that the highly variable shell openings of mussels are linked to circatidal and circadian rhythms (Ameyaw-Akumfi & Naylor 1987), which were not taken into consideration in the present study.
Siphon area may be a more specific indicator of feeding activity and condition than shell gape. Density and current speed both affected this variable. At low density, siphon area responded to flow speed in nonlinear fashion with a peak of siphon area at medium speeds. Previous studies performed in situ under constant seston concentrations, show an inverse relationship between siphon area and current speeds ranging from 6-25 cm [s.sup.-1] (Wildish & Miyares 1990) and 10-30 cm [s.sup.-1] (Newel1 et al. 2001). This same tendency was observed for cultured mussels in the field at the higher current categories (10-15 cm [s.sup.-1]). Reduced siphon areas may occur at increased flows where the adverse pressure differences between siphon openings causes a reduction in filtration (Wildish et al. 1987, Wildish & Saulnier 1992, Newell et al. 2001). A similar reduction in siphon area detected at low flow velocities (0-4 cm [s.sup.-1]) might reflect decreased particle flux. Previously, siphon opening has been linked with particle concentration, with mussels exhibiting increased feeding rates under better food conditions (Newell et al. 2001, Ward et al. 2003).
A similarity in patterns of valve gape and siphon area among current speeds was detectable at high density stocking suggesting that when crowding is substantial the response of these two variables may be related. Nonetheless, at low density stocking, besides the basic requirement of open valves for siphon extension, the two variables appear to be uncorrelated (Newell et al. 2001, Riisgard et al. 2003). Newell et al. (2001) suggested that valve gape and siphon area were not governed by the same factors and that siphon area is sensitive to ambient hydrodynamics. Siphon area has been shown to depend on particle quality (Smaal and van Stralen, 1990, Wildish et al. 1992) and quantity (Newell et al. 2001, Riisgard et al. 2003, Ward et al. 2003), as well as having an inverse relationship with ambient flow (Newell et al. 2001). The interaction between siphon area and density suggests that optimal hydrodynamic response is sensitive to crowding.
In the first week of the experiment, emerged mussels exhibited a preferred direction with orientation predominantly on a horizontal plane. Through time mussels remained similarly clustered around a mean direction, but with values shifted towards an increasingly surface pointing mean. Experiments carried out in a one-way flow channel demonstrated that mussels tended to aggregate in an upstream direction (Vooys 2003). Mytilus edulis has also been observed to actively crawl on a rope approximately 5 cm towards the incoming flow of an aquarium in a matter of days (personal observation). Although mussels react to chemical stimuli (de Vooys 2003), it is still unclear whether they predominantly respond to chemical cues, flow direction, proximity of individuals or an interaction of several variables. Surprisingly little is known about such an obvious part of their early life history.
In general, density had no effect on mussel orientation. Density-dependent motion of mussels has been documented as active movement of individuals onto cleared parts of a rope as well as passive redistribution (Littorin & Gilek 1999). The lack of density effects on valve gape and orientation are consistent with the conclusion that mussels alleviate the mechanical effects of crowding through behavioral mechanisms.
Comparison of Mytilus sp. orientation shows no differences at the inner site, but at the outer site a slight difference occurred with a mean direction of 79.4[degrees] for M. trossulus compared with 86.8[degrees] for M. edulis, within the low density treatments. Previous studies comparing chemotaxis of M. edulis and M. galloprovincialis revealed species-specific behavior despite minimal difference in genotypes (Schneider et al. 2005). Their study suggested that there were differences in movements between these two species that could account for species-specific differences in growth and viability. Although the importance of micromovements in sessile invertebrates is still uncertain, it is likely consequential for species competing intensely for space and resources.
Although we collected no 3-dimensional flow data, it is expected that currents are stronger in the horizontal than in the vertical, which may account for the horizontal orientation and consistency between sites. It should be noted that there are likely complex small-scale flow patterns caused by the structure of the shells and sock (Wildish & Kristmanson 1997). Predominance in orientation was apparent despite the potential for localized microflow environments. Similarly, the potential for flow inhibition at the outer site was reflected in siphon area but not in orientation, such that the former represents a "tuning mechanism" for optimizing horizontal orientation.
Both Mytilus species were of similar length at the beginning of the experiment but at 3 too, M. trossulus displayed an advantage in shell growth at both outer and inner sites. Previous studies of growth in M. edulis and M. trossulus indicate that the latter had lower shell growth and tissue yields and was prone to shell fracture (Freeman et al. 1994, Mallet & Carver 1999, Penney et al. 2002). Mallet and Carver (1995) have estimated the economic value of M. edulis to be 1.7 times that of M. trossulus in off-bottom cultures. Based on meat yields and shell growth, these results are not supported in this study, however, shell fragility may entail important losses during commercial processing. Previously observed morphological variations in these two species were also distinct in our study with M. trossulus possessing a more elongated form and lower shell height of (Mallet & Carver 1995, Innes & Bates 1999).
Growth of juvenile mussels in socks represents a potentially critical stage in the population dynamics. Although we have focused on an aquaculture context, our results are applicable to determinants of growth in natural mussel beds where gregarious behavior is also featured. In the present experiments, short-term measures of response to crowding including valve gape, siphon area, and emergence were coupled with integrated measures including shell and tissue growth. In general, crowding imposed by culture sock conditions had mixed effects, but there was little indication of negative density-dependence and even demonstration of positive benefits of higher numbers in initial stocking. The lack of density effects was displayed in behavioral variables such as emergence and gape, as well as integrated variables such as shell and tissue growth. As in previous studies, mussel mobility and behavior were species-specific; Mytilus trossulus is adapted to high current conditions and showed comparable movement abilities independent of site.
The most prominent environmental factor appeared to be current speed. Although high flow offers improved seston flux, it may also cause reduced pumping rates in mussels (Newell et al. 2001). Interestingly, the yields of mussels in this study were superior in the low flow conditions of the inner estuary. This study offers support to the hypothesis that there is a relationship between feeding and flow fields characterized by a unimodal response or optimum at a middle current range (Wildish & Kristmanson 1997). Siphon area was one variable that was affected by crowding in interaction with current speed.
In concurrence with previous studies, valve gape was shown to be independent of flow at the speeds recorded. In addition, conditions in the high density socks did not restrict mussel valve gape as was previously hypothesized. It is probable that individual mussels continuously adjust to the presence and growth of neighbors through small movements, which enable proper shell gape. The results of this field study coincide with previous in vitro findings asserting that mussel valve and siphons are governed by different factors. Preferential orientation of mussels is proposed to be in response to the predominant flow and thus food source.
Image analyses was deemed effective for measurement of multiple variables on culture socks, and allowed time series data to be collected in the natural environment without disturbance. However, this method is limited in that 2-dimensional images reveal incomplete information on absolute density and positioning, and total mussel numbers cannot be tracked. It is assumed that some self-thinning occurs via density-dependent mortality and falloff (Ceccherelli & Barboni 1983, Bayne 2000) and predation (Cote & Jelnikar 1999, Beadman et al. 2003).
Even where density is maintained at higher levels, mussels "solve" crowding by behavioral sorting. As a result, almost all individuals are emerged and able to feed and grow without overcrowding. There is likely a density at which the exterior positions are fully occupied and mussels are forced to reside in the interior. We have documented that when internal positioning is suboptimal, it will lead to higher mortality or poor product. Overstocking will ultimately be limited by the size of the sock or by falloff past maximal density. From a practical standpoint, it would be of interest to determine this value and ascertain whether density-dependent growth occurs when maximal packing is applied. Although there is concern that growth studies applied to cultured mussels are often deficient because they use cages or other containers rather than socks, this study suggests that conditions of the sock environment such as crowding can be less important than macroenvironmental factors (e.g., current speed) in determining growth.
The authors thank Bryan Schofield for help in designing and deploying the camera and AquaPrime Mussel Ranch for access to their site and help in setting up the experiments. Funding was provided by the Natural Sciences and Engineering Research Council of Canada (NSERC)
Ackerman, J. D. 1999. Effect of velocity on the filter feeding of dreissenid mussels (Dreissena polymorpha and D. bugensis): Implications for trophic dynamics. Can. J. Fish. Aquat. Sci. 56: 1551-1561.
Alunno-Bruscia, M., P. S. Petraitis, E. Bourget & M. Frechette. 2000. Body size-density relationship for Mytilus edulis in an experimental food-regulated situation. Oikos 90:28-42.
Ameyaw-Akumfi, C. & E. Naylor. 1987. Temporal patterns of shellgape in Mytilus edulis. Mar. Biol. 95:237-242.
Archambault, P., C. W. McKinsey & E. Bourget. 1999. Large-scale shoreline configuration influences phytoplankton concentration and mussel growth. Estuar. Coast. Shelf Sci. 49:193-208.
Bates, J. A. & D. J. Innes. 1995. Genetic variation among populations of Mytilus spp. in eastern Newfoundland. Mar. Biol. 124:417-424.
Bayne, B. L. 2000. Relations between variable rates of growth, metabolic costs and growth efficiencies in individual Sydney rock oysters (Saccostrea commercialis). J. Exp. Mar. Biol. Ecol. 251:185-203.
Beadman, H., R. Caldow, M. Kaiser & R. Willows. 2003. How to toughen up your mussels: using mussel shell morphology plasticity to reduce predation losses. Mar. Biol. 142:487-494.
Bertness, M. D. & E. Grosholz. 1985. Population dynamics of the ribbed mussel, Geukensia demissa: The costs and benefits of an aggregated distribution. Oecol. 67:192-204.
Boromthanarat, S. & J. M. Deslous-Paoli. 1988. Production of Mytilus edulis L. reared on bouchots in the Bay of Marennes-Oleron: Comparison between two methods of culture. Aquaculture 72:255-263.
Ceccherelli, V. U. & A. Barboni. 1983. Growth, survival and yield of Mytilus galloprovincialis Lamk, on fixed suspended culture in a bay of the Po River Delta. Aquaculture 34:101-114.
Comesana, A. S., J. E. Toro, D. J. Innes & R. J. Thompson. 1999. A molecular approach to the ecology of a mussel (Mytilus edulis--M. trossulus) hybrid zone on the east coast of Newfoundland, Canada. Mar. Biol. 133:213-221.
Cote, I. M. & E. Jelnikar. 1999. Predator-induced clumping behaviour in mussels (Mytilus edulis Linnaeus). J. Exp. Mar. Biol. Ecol. 235:201-211.
de Vooys, C. G. N. 2003. Effect of tripeptide on the aggregational behaviour of the blue mussel Mytilus edulis. Mar. Biol. 142:1119-1123.
Dolmer, P. 2000. Feeding activity of mussels Mytilus edulis related to near-bed currents and phytoplankton biomass. J. Sea Res. 44: 221-231.
Eyster, L. S. & J. A. Pechenik. 1987. Attachment of Mytilus edulis L. larvae on algal and byssal filaments is enhanced by water agitation. J. Exp. Mar. Biol. Ecol. 114:99-110.
Famme, D., H. U. Riisgard & C. B. Joergensen. 1986. On direct measurement of pumping rates in the mussel Mytilus edulis. Mar. Biol. 92:323-327.
Frechette, M. & E. Despland. 1999. Impaired shell gaping and food depletion as mechanisms of asymmetric competition in mussels. Ecoscience 6:1-11.
Frechette, M. & D. Lefaivre. 1990. Discriminating between food and space limitation in benthic suspension feeders using self-thinning relationships. Mar. Ecol. Prog. Ser. 65:15-23.
Frechette, M., A. E. Aitken & L. Page. 1992. Interdependence of food and space limitation of a benthic suspension feeder: Consequences for self-thinning relationships. Mar. Ecol. Prog. Ser. 83:55-62.
Frechette, M., P. Bergeron & P. Gagnon. 1996. On the use of self-thinning relationships in stocking experiments. Aquaculture 145:91-112.
Frechette, M., C. A. Butman & W. R. Geyer. 1989. The importance of boundary-layer flows in supplying phytoplankton to the benthic suspension feeder, Mytilus edulis L. Limnol. Oceanogr. 34:19-36.
Freeman, K. R., K. L. Perry, T. G. DiBacco & D. J. Scarratt. 1994. Observations on two mytilid species from a Nova Scotian mussel farm. Can. Tech. Rep. Fish. Aquat. Sci. 1969:47.
Fuentes, J., V. Gregorio, R. Giraldez & J. Molares. 2000. Within-raft variability of the growth rate of mussels, Mytilus galloprovincialis, cultivated in the Ria de Arousa (NW Spain). Aquaculture 189:39-52.
Gibbs, M. M., M. R. James, S. E. Pickmere, P. H. Woods, B. S. Shekespeare, R. W. Hickman & J. Illingworth. 1991. Hydrodynamics and water column properties at six stations associated with mussel forming in Pelorus Sound, 1984-85. NZ. J. Mar. Fresh. Res. 25:239-254.
Grant, J., C. W. Emerson & S. E. Shumway. 1993. Orientation, passive transport, and sediment erosion features of the sea scallop Placopecten magellanicus in the benthic boundary layer. Can. J. Zool. 71:953-959.
Gregory, D., B. Petrie, F. Jordan & P. Langille. 1993. Oceanographic, geographic and hydrological parameter of Scotia-Fundy and Southern Gulf of St. Lawrence inlets. Can. Tech. Rep. Hydrograph. Ocean Sci. 143:1-248.
Heasman, K. G., G. C. Pitcher, C. D. McQuaid & T. Hecht. 1998. Shellfish mariculture in the Benguela system: Raft culture of Mytilus galloprovincialis and the effect of rope spacing on food extraction, growth rate, production, and condition of mussels. J. Shellfish Res. 17:33-39.
Hildreth, D. I. 1976. The influence of water flow rate on pumping rate in Mytilus edulis using a refined direct measurement apparatus. J. Mar. Biol. Ass. UK56:311-319.
Ibarra, D. 2003. Estimation of seston depletion by cultured mussels (Mytilus spp.) using measurements of diffuse attenuation of solar irradiance from optical moorings. Msc. Thesis, Dalhousie University, Halifax, NS. pp. 127.
Incze, L. S., R. A. Lutz & L. Watling. 1980. Relationships between effects of environmental temperature and seston on growth and mortality of Mytilus edulis in a temperate northern estuary. Mar. Biol. 57:147-156.
Innes, D. J. & J. A. Bates. 1999. Morphological variation of Mytilus edulis and Mytilus trossulus in eastern Newfoundland. Mar. Biol. 133:691-699.
Jorgensen, C. B., P. S. Larsen, F. Mohlenberg & H. U. Riisgard. 1986. The bivalve pump. Mar. Ecol. Prog. Ser. 34:69-77.
Jorgensen, C. B., P. S. Larsen, F. Mohlenberg & H. U. Riisgard. 1988. The mussel pump: properties and modelling. Mar. Ecol. Prog. Ser. 45:205-216.
Karayucel, S. & I. Karayucel. 2000. The effect of environmental factors, depth and position on the growth and mortality of raft-cultured blue mussels (Mytilus edulis L.). Aquacult. Res. 31:893-899.
Keizer, P. D., T. G. Milligan, D. V. S. Rao, P. M. Strain & G. Bugden. 1996. Phytoplankton monitoring program: Nova Scotia component-1989 to 1994. Can. Tech. Rep. Fish. Aquat. Sci. 2136:74.
Lesser, M. P., S. E. Shumway, T. Cucci & J. Smith. 1992. Impact of fouling organisms on mussel rope culture: Interspecific competition for food among suspension-feeding invertebrates. J. Exp. Mar. Biol. Ecol. 165:91-102.
Littorin, B. & M. Gilek. 1999. A photographic study of the recolonization of cleared patches in a dense population of Mytilus edulis in the northern Baltic proper. Hydrobiologia 393:211-219.
Mallet, A. L. & C. E. Carver. 1989. Growth, mortality, and secondary production in natural populations of the blue mussel, Mytilus edulis. Can. J. Fish. Aquat. Sci. 46:1154-1159.
Mallet, A. L. & C. E. Carver. 1993. Temporal production patterns in various size groups of the blue mussel. J. Exp. Mar. Biol. Ecol. 170:75-89.
Mallet, A. L. & C. E. Carver. 1995. Comparative growth and survival patterns of Mytilus trossulus and Mytilus edulis in Atlantic Canada. Can. J. Fish. Aquat. Sci. 52:1873-1880.
Mallet, A. & C. Carver. 1999. Maritime distribution and commercial production performance of Mytilus edulis and Mytilus trossulus. Bull. Aquacult. Assoc. Can. 99:7-13.
Mallet, A. & B. Myrand. 1995. The culture of the blue mussel in Atlantic Canada. In: Boghen, A.D. (Ed.), Cold water aquaculture in Atlantic Canada. Canadian Institute for Research on Regional Development, Moncton, NB, pp. 255-296.
Maximovich, N. V., A. Sukhotin & Y. S. Minichev. 1996. Long-term dynamics of blue mussel (Mytilus edulis L.) culture settlements (the White Sea). Aquaculture 147:191-204.
McGrorty, S. & J. D. Goss-Custard. 1995. Population dynamics of Mytilus edulis along environmental gradients: Density-dependent changes in adult mussel numbers. Mar. Ecol. Prog. Ser. 129:197-213.
Mueller, K. W. 1996. A preliminary study of the spatial variation in growth of raft-cultured blue mussels Mytilus trossulus in northern Puget Sound, Washington. J. World Aquacult. Soc. 27:240-246.
Newell, C. R. 1990. The effects of mussel (Mytilus edulis, Linnaeus 1758) position in seeded bottom patches on growth at subtidal lease sites in Maine. J. Shellfish Res. 9:113-118.
Newell, R. I. E. & B. L. Baync. 1980. Seasonal changes in the physiology, reproductive condition and carbohydrate content of the cockle Cardium (= Cerastoderma) edule (Bivalvia: Cardiidae). Mar. Biol. 56:11-19.
Newell, C. R. & D. J. Wildish. 1997. The effects of current speed on exhalent siphon area and shell gape in blue mussels under constant seston regimes. J. Shellfish Res. 16:339.
Newell, C. R., D. J. Wildish & B. A. MacDonald. 2001. The effects of velocity and seston concentration on the exhalant siphon area, valve gape and filtration rate of the mussel Mytllus edulis. J. Exp. Mar. Biol. Ecol. 262:91-111.
Okamura, B. 1986. Group living and the effects of spatial position in aggregations of Mytilus edulis. Oecol. 69:341-347.
Paine, R. T. & S. A. Levin. 1981. Intertidal Landscapes: Disturbance and the Dynamics of Pattern. Ecol. Monogr. 51:145-178.
Peckham, P. P. & J. P. Sanders. 1972. Consequences of failure to meet assumptions underlying the fixed effects analysis of variance and covariance. Rev. Edu. Res. 42:237-293.
Penney, R. W., M. J. Hart & N. Templeman. 2002. Comparative growth of cultured blue mussels, Mytilus edulis, M. trossulus and their hybrids, in naturally occurring mixed-species stocks. Aquacult. Res. 33:693-702.
Riginos, C. & C. W. Cunningham. 2005. Local adaptation and species segregation in two mussel (Mytilus edulis and Mytilus trossulus) hybrid zones. Mol. Ecol. 14:381-400.
Riisgard, H. U. & A. Randlov. 1981. Energy budgets, growth and filtration rates in Mytilus edulis at different algal concentrations. Mar. Biol. 61:227-234.
Riisgard, H. U., C. Kittner & D. F. Seerup. 2003. Regulation of opening state and filtration rate in filter-feeding bivalves (Cardium edule, Mytilus edulis, Mya arenaria) in response to low algal concentration. J. Exp. Mar. Biol. Ecol. 284:105-127.
Ross, D. J., C. R. Johnson & C. L. Hewitt. 2003. Variability in the impact of an introduced predator (Asterias amurensis: Asteroidea) on soft sediment assemblages. J. Exp. Mar. Biol. Ecol. 288:257-278.
Schneider, K. R., D. S. Wethey, B. S. T. Helmuth & T. J. Hilbish. 2005. Implications of movement behavior on mussel dislodgement: exogenous selection in a Mytilus spp. hybrid zone. Mar. Biol. 146:333-344.
Sherman, M. L. & P. A. Moore. 2001. Chemical orientation of brown bullheads, Ameiurus nebulosus, under different flow conditions. J. Chem. Ecol. 27:2301-2318.
Smaal, A. & M. van Stralen. 1990. Average annual growth and condition of mussels as a function of food source. Hydrobiologia 195:179-188.
Smaal, A. C., T. C. Prins, N. Dankers & B. Ball. 1998. Minimum requirements for modelling bivalve carrying capacity. Aquat. Ecol. 31:423-428.
Stabell, O. B. 1992. Olfactory control of homing behaviour in Salmonids. In: T. J Hara, editor. Fish chemoreception. London: Chapman and Hall. pp. 249-270.
Strain, P. M. 2002. Nutrient dynamics in Ship Harbour, Nova-Scotia. Atmos. Ocean. 40:45-58.
Strickland, J. D. & T. R. Parsons. 1972. A practical handbook of seawater analysis. Bull. Fish. Res. Board. Can. 167:310.
Taylor, J. J., R. A. Rose & P. C. Southgate. 1997. Effects of stocking density on the growth and survival of juvenile silver-lip pearl oysters (Pinctada maxima, Jameson) in suspended and bottom culture. J. Shellfish Res. 16:569-572.
Thorin, S., H. Bourdages & B. Vincent. 1998. Study of siphon activity in Mya arenaria (L.) in the intertidal zone by means of an underwater video camera. J. Exp. Mar. Biol. Ecol. 224:205-224.
Thorin, S., T. Robinet, P. Lafaille & B. Vincent. 2001. Shape and surface variations of siphon openings during complete cycles in Mya arenaria in the intertidal zone. J. Mar. Biol. Ass. UK 81:505-515.
Ward, J. E., H. K. Cassell & B. A. MacDonald. 1992. Chemoreception in the sea scallop Plocopecten magellanicus (Gmelin). 1. Stimulatory effects of phytoplankton metabolites on clearance and ingestion rates. J. Exp. Mar. Biol. Ecol. 163:235-250.
Ward, J. E., J. S. Levinton & S. E. Shumway. 2003. Influence of diet on pre-ingestive particle processing in bivalves I: Transport velocities on the ctenidium. J. Exp. Mar. Biol. Ecol. 293:129-149.
Weissburg, M. J. 2000. The fluid dynamical context of chemosensory behavior. Biol. Bull. 198:188-202.
Weissburg, M. J. & R. K. Zimmer-Faust. 1994. Odor plumes and how blue crabs use them in finding prey. J. Exp. Biol. 197:349-375.
Westerbom, M., M. Kilpi & O. Mustonen. 2002. Blue mussels, Mytilus edulis, at the edge of the range: population structure, growth and biomass along a salinity gradient in the north-eastern Baltic Sea. Mar. Biol. 140:991-999.
Widdows, J. 1976. Physiological adaptation of Mytilus edulis to cyclic temperatures. J. Comp. Physiol. 105:115-128.
Wildish, D. & D. Kristmanson. 1997. Benthic suspension feeders and flow. New York: Cambridge University Press.
Wildish, D. J. & M. P. Miyares. 1990. Filtration rate of blue mussels as a function of flow velocity: preliminary experiments. J. Exp. Mar. Biol. Ecol. 142:213-219.
Wildish, D. J. & A. M. Saulnier. 1992. The effect of velocity and flow direction on the growth of juvenile and adult giant scallops. J. Exp. Mar. Biol. Ecol. 155:133-143.
Wildish, D. J. & A. M. Saulnier. 1993. Hydrodynamic control of filtration in Placopecten magellanicus. J. Exp. Mar. Biol. Ecol. 174:65-82.
Wildish, D. J., D. D. Kristmanson, R. L. Hoar, A. M. DeCoste, S. D. McCormick & A. W. White. 1987. Giant scallop feeding and growth responses to flow. J. Exp. Mar. Biol. Ecol. 113:207-220.
Wildish, D. J., D. D. Kristmanson & A. M. Saulnier. 1992. Interactive effect of velocity and seston concentration on giant scallop feeding inhibition. J. Exp. Mar. Biol. Ecol. 155:161-168.
Zar, J. H. 1999. Biostatistical analysis, 4th ed. Upper Saddle River, N J: Prentice-Hall Inc.
JUDITH SENECHAL, JON GRANT * AND MARIE-CLAUDE ARCHAMBAULT
Department of Oceanography, Dalhousie University, Halifax, Nova Scotia, B3H 4J1, Canada
* Corresponding author. E-mail: email@example.com
TABLE 1. Two-way ANOVA of mussel emergence from culture sock after submergence as a function of density (Low ~400/30.5 cm; High ~800/30.5 cm) and location (Site 1-Outer; Site 2-Inner) for Mytilus edulis at day 39. Sum of Mean Probability Source Term DF Squares Square F-Ratio Level A: Density 1 182.6044 182.6044 0.48 0.4952 B: Location 1 5,765.856 5,765.856 15.22 0.0008 * AB 1 13.60069 13.6007 0.04 0.8516 S 21 7,957.237 378.916 Total (Adjusted) 24 14,399.03 Total 25 * Term significant at alpha = 0.05. TABLE 2. Two-way ANOVA for siphon index as a function of current speed (cm [s.sup.-1]) and stocking densities (Low ~400/30.5 cm; High ~800/30.5 cm) from September 18-28, 2001, Site 1-Outer only, species combined. Sum of Mean Probability Source Term DF Squares Square F-Ratio Level A: Density 1 0.035458 0.035458 1.21 0.2719 [E.sup.02] [E.sup.02] B: Current 6 1.9299 0.3216 10.97 0.0000 * speed AB 6 1.458 0.2423 8.29 0.0000 * S 565 16.5622 0.029314 [E.sup.02] Total 578 19.7666 (Adjusted) Total 579 * Term significant at alpha = 0.05. TABLE 3. Two-way ANOVA comparing shell length of mussels three months after submergence as a function of density (Low ~400/30.5 cm; High ~800/30.5 cm) and location (Site 1-Outer; Site 2-Inner) for Mytilus trossulus. Sum of Mean Probability Source Term DF Squares Square F-Ratio Level A: Density 1 23.612 23.612 1.02 0.3173 B: Site 1 114.098 114.098 4.92 0.0306 * AB 1 482.233 482.233 20.79 0.0000 * S 56 1,298.784 23.192 Total (Adjusted) 59 1,918.729 Total 60 * Term significant at alpha = 0.05. TABLE 4. Two-way ANOVA comparing dry weight of mussels three months after submergence as a function of density (Low ~400/30.5 cm; High ~800/30.5 cm) and location (Site 1-Outer; Site 2-Inner) for Mytilus edulis and M. trossulus. Sum of Mean Source Term DF Squares Square Mytilus edulis A: Density 1 2.703[E.sup.-02] 2.703[E.sup.-02] B: Site 1 0.059 0.0595 AB 1 1.725[E.sup.-02] 1.725[E.sup.-02] S 56 0.280 5.017[E.sup.-03] Total (Adjusted) 59 0.384 Total 60 Mytilus trossulus A: Density 1 0.146 0.146 B: Site 1 0.102 0.102 AB 1 8.370[E.sup.-08] 8.370[E.sup.-08] S 56 0.439 7.844[E.sup.-03] Total (Adjusted) 59 0.687 Total 60 Probability Source Term F-Ratio Level Mytilus edulis A: Density 5.39 0.0239 * B: Site 11.87 0.0010 * AB 3.44 0.0689 S Total (Adjusted) Total Mytilus trossulus A: Density 18.64 0.0001 B: Site 13.05 0.0006 * AB 0.00 0.9974 S Total (Adjusted) Total * Term significant at alpha = 0.05. TABLE 5. Two-way ANOVA comparing the condition of mussels three months after submergence as a function of density (Low ~400/30.5 cm; High ~800/30.5 cm) and location (Site 1-Outer; Site 2-Inner) for Mytilus edulis and M. trossulus. Sum of Mean Probability Source Term DF Squares Square F-Ratio Level Mytilus edulis A: Density 1 30.116 30.116 16.30 0.0001 B: Site 1 17.075 17.075 9.24 0.0035 * AB 1 0.807 0.807 0.44 0.5114 S 56 103.498 1.848 Total (Adjusted) 59 151.498 Total 60 Mytilus trossulus A: Density 1 14.737 14.737 6.45 0.0138 * B: Site 1 5.478 5.478 2.40 0.1271 AB 1 0.286 0.286 0.13 0.7247 S 56 127.940 2.284 Total (Adjusted) 59 148.441 Total 60 * Term significant at alpha = 0.05. TABLE 6. Two-way ANOVA comparing shell length of mussels as a function of individuals having remained inside the sock and those having effectively emerged (Interior/Exterior), and as a function of density (Low ~400/30.5 cm; High ~800/30.5 cm) for months 3 (December), 9 (June) and 11 (August). Data from both sites are analyzed separately at month 3 (A) and data for months 9 and 11 represent Site 2-Inner only (B). All analyses are for both species combined. Sum of Mean Source Term DF Squares Square A Month 3, Site 1-Outer A: Density 1 293.9989 293.9989 B: Interior/Exterior 1 628.4012 628.4012 AB 1 80.8658 80.8658 S 151 2,755.157 18.2461 Total (Adjusted) 154 3,691.628 Total 155 Month 3, Site 2-Inner A: Density 1 0.0305 0.0305 B: Interior/Exterior 1 0.1187 0.1187 AB 1 1.2528[E.sup.-02] 1.2528[E.sup.-02] S 116 0.3502 3.0193[E.sup.-03] Total (Adjusted) 119 0.512 Total 120 B Month 9 A: Density 1 41.0411 41.0411 B: Interior/Exterior 1 4,971.861 4,971.861 AB 1 0.0111 0.0110 S 86 3,708.407 43.1210 Total (Adjusted) 89 9,348.361 Total 90 Month 11 A: Density 1 54.838 54.838 B: Interior/Exterior 1 8,335.706 8,335.706 AB 1 70.502 70.502 S 87 1,892.486 21.752 Total (Adjusted) 90 10,788.5 Total 91 Probability Source Term F-Ratio Level A Month 3, Site 1-Outer A: Density 16.11 0.0001 * B: Interior/Exterior 34.44 0.0000 * AB 4.43 0.0369 * S Total (Adjusted) Total Month 3, Site 2-Inner A: Density 10.10 0.0019 * B: Interior/Exterior 39.31 0.0000 * AB 4.15 0.0439 * S Total (Adjusted) Total B Month 9 A: Density 0.95 0.3320 B: Interior/Exterior 115.30 0.0000 * AB 0.00 0.9873 S Total (Adjusted) Total Month 11 A: Density 2.52 0.116 B: Interior/Exterior 383.20 0.0000 * AB 3.24 0.0753 S Total (Adjusted) Total * Term significant at alpha = 0.05. TABLE 7. Two-way ANOVA comparing dry weight of mussels as a function of individuals remaining inside the sock and those having effectively emerged (Interior/Exterior), and as a function of density (Low ~400/30.5 cm; High ~800/30.5 cm), for months 3 (December), 9 (June), and 11 (August). Data from both sites are analyzed separately at month 3 (A) and data for months 9 and 11 represent Site 2-Inner site only (B). All analyses are for both species combined. Sum of Mean Source Term DF Squares Square A Month 3, Site 1-Outer A: Density 1 0.3506 0.3506 B: Interior/Exterior 1 8.0759[E.sup.-02] 8.0759[E.sup.-02] AB 1 6.7298[E.sup.-03] 6.7298[E.sup.-03] S 116 0.8493 7.3215[E.sup.-02] Total (Adjusted) 119 1.2874 Total 120 Month 3, Site 2-Inner A: Density 1 0.1801 0.1801 B: Interior/Exterior 1 0.5156 0.5156 AB 1 2.9895[E.sup.-03] 2.9895[E.sup.-03] S 116 2.8994 2.4995[E.sup.-02] Total (Adjusted) 119 3.5982 Total 120 B Month 9 A: Density 1 9.63071[E.sup.-05] 9.6307[E.sup.-05] B: Interior/Exterior 1 6.7042 6.7042 AB 1 7.3655[E.sup.-04] 7.3655[E.sup.-04] S 85 3.2832 3.8626[E.sup.-02] Total (Adjusted) 88 10.7328 Total 89 Month 11 A: Density 1 0.004 0.004 B: Interior/Exterior 1 7.063 7.063 AB 1 70.502 0.055 S 86 0.055 1.955[E.sup.-02] Total (Adjusted) 89 9.224 Total 90 Probability Source Term F-Ratio Level A Month 3, Site 1-Outer A: Density 47.89 0.0000 * B: Interior/Exterior 11.03 0.0012 * AB 0.92 0.3397 S Total (Adjusted) Total Month 3, Site 2-Inner A: Density 7.21 0.0083 * B: Interior/Exterior 20.63 0.0000 * AB 0.12 0.7301 S Total (Adjusted) Total B Month 9 A: Density 0.00 0.9603 B: Interior/Exterior 173.57 0.0000 * AB 0.02 0.8905 S Total (Adjusted) Total Month 11 A: Density 0.23 0.6326 B: Interior/Exterior 361.26 0.0000 * AB 2.86 0.0946 S Total (Adjusted) Total * Term significant at alpha = 0.05. TABLE 8. Two-way ANOVA comparing condition of mussels as a function of individuals remaining inside the sock and those having effectively emerged (Interior/Exterior), and as a function of density (Low 400/30.5 cm; High -800/30.5 cm) for month 3 (December), both sites separately. All analyses are for both species combined. Sum of Mean Probability Source Term DF Squares Square F-Ratio Level Month 3, Site 1-Outer A: Density 1 36.654 36.654 16.71 0.0001 * B: Interior/Exterior 1 4.1582 4.1582 1.90 0.1713 AB 1 2.4158 2.4158 1.10 0.2962 S 116 254.5018 2.194 Total (Adjusted) 119 297.7297 Total 120 Month 3, Site 2-Inner A: Density 1 2.2073 2.2073 0.94 0.3355 B: Interior/Exterior 1 23.631 23.6310 10.01 0.002 * AB 1 16.7758 16.7758 7.11 0.0088 * S 116 273.8031 2.3604 Total (Adjusted) 119 316.4172 Total 120 * Term significant at [alpha] = 0.05.
|Printer friendly Cite/link Email Feedback|
|Author:||Senechal, Judith; Grant, Jon; Archambault, Marie-Claude|
|Publication:||Journal of Shellfish Research|
|Date:||Aug 1, 2008|
|Previous Article:||Genotypic diversity of culturable Vibrio species associated with the culture of oysters and clams in Galicia and screening of their pathogenic...|
|Next Article:||Genotype-dependent variability in somatic tissue and shell weights and its effect on meat yield in mixed species [Mytilus edulis L., M. trossulus...|