Abstract

Bryozoans are suspension feeding colonial animals that remain attached to the substratum or other surfaces. How well a bryozoan can feed in a particular flow regime could help determine the distribution and abundance of that bryozoan. I tested how velocity of flow affects feeding rate in four species of bryozoans in the laboratory and how these species perform in different flow regimes in the field. I found that one species, Membranipora membranacea, had a higher ingestion rate than did the other three species at all velocities of flow tested. Membranipora also had a higher rate of ingestion at intermediate velocities, while velocity did not have as strong an effect on ingestion rate in the other three species. As predicted from the feeding experiments, all four species generally had greater abundance, attained a larger size, grew faster, and survived longer in flow regimes in which feeding is higher. Also as predicted, Membranipora had greater abundance, attained a larger size, grew faster, and survived longer than did the other three species both in slower and faster flow regimes in the field. Understanding how flow affects feeding can help predict the distribution and abundance of bryozoans in the field. Because especially efficient filterers like Membranipora can grow faster and have higher survival under a wide range of conditions of flow, this species may be able to outcompete many other species or take advantage of ephemeral habitats, thereby becoming a potentially effective invasive species as has been seen in the Gulf of Maine.

Introduction

The vast majority of colonial animals such as corals, hydrozoans, ascidians, and bryozoans are benthic suspension feeders that depend on currents of water to supply nutrients, eliminate wastes, and disperse gametes and larvae. The effect of velocity of flow on the capture of food is especially important because acquisition of energy is essential to perform most other life processes, and therefore, feeding performance may be a good proxy for fitness. In this study, I focus on how water flow affects feeding in bryozoans.

A bryozoan colony is made of individual units called zooids that are replicated by asexual budding. A zooid consists of a protective housing, or zoecium, that encloses and protects the living tissues. Bryozoans actively generate their own feeding currents with a crown of ciliated tentacles called a lophophore. Not all zooids are capable of feeding, but the ones that do feed have a movable part, or polypide, that contains the digestive system and bears the lophophore. Zooids arranged in different ways create different colony shapes, and zooids within the colony can coordinate their feeding currents. The feeding currents of the colony depend on the arrangement of lophophores within the colony and the shape of the colony as a whole (Cook 1977; Winston 1978, 1979; Lidgard 1981; Okamura 1984, 1985; McKinney 1990).

Many studies have investigated the feeding of bryzoans; however, many unresolved issues and methodological problems remain. The majority of studies were conducted in still water (Cook 1977; Winston 1978, 1979; Riisgård and Goldson 1997; Nielsen and Riisgård 1998), yet bryozoans probably experience still water rarely, if ever. A few studies have investigated bryozoan feeding in flowing water, but, with one exception (Okamura 1992), these do not relate feeding to distribution and abundance. The influence of velocity on feeding in an erect, tree-shaped colony, Bugula stolonifera (Okamura 1984), and an encrusting sheet-shaped colony, Conopeum reticulum (Okamura 1985), was variable, appearing to depend on both shape and size of the colony. However, a generic problem with all previous studies is the assessment of feeding in only a small number of ambient velocities of flow and the assumption that the relationship between feeding and flow is linear. Clearly, if the capture of food is a nonlinear function of velocity, then feeding must be measured over a wider range of velocities.

Many factors can influence capture of particles by bryozoans, including characteristics of the food (e.g. type, size, concentration); shape and size of the colony; size, shape, number, location, and level of activity of zooids; and environmental conditions (e.g. temperature, water velocity, salinity, competition from neighbors). While a number of studies look at the effect of one or more of these factors on bryozoan feeding (see McKinney 1990 for a review) and interactions among them may be important, no studies hold all of these other factors constant and examine how feeding success varies under a relatively wide range of velocities and in multiple species of different colony shapes. Whether water velocity has a consistent effect on success of capturing particles, and whether the relationship between water velocity and that success has an effect on the abundance and distribution of bryozoans remain unclear.

I measured how feeding varies over a broad range of flows for bryozoans of different colony shape, and how flow affects abundance, growth, and survival in the field. I had three predictions: (1) feeding changes as a unimodal function of velocity, (2) bryozoans will have higher abundance, growth, and survival in flows where feeding is high, and (3) colonies in the form of encrusting sheets will perform better in high-flow velocities than those in the form of erect trees.

The first prediction, that feeding will show a unimodal function of velocity, is based on theory as well as on previous experiments with other suspension feeders. According to Wildish and Kristmanson (1997), the rate of capture of food particles by suspension feeders generally follows a “continuous unimodal function” with a peak in rate of capture of food at intermediate freestream velocities. If the concentration of particles is held constant, the continuous-unimodal-function theory suggests that the rate of capture of particles at first increases with increased velocity, most likely due to an increase in the rate at which particles are encountered. Eventually, a maximum is reached beyond which the rate at which particles are captured is unaffected by further increase in velocity. This maximum particle-capture rate may be determined by such things as the size of the filtration surface or limits on the rate of transport of particles to the mouth. At yet higher velocities, flow begins to suppress capture of particles because the particles pass by too quickly to be diverted into the feeding structure or the feeding structure itself is deformed by the high flow and is less effective at capturing particles. At very high velocities, capture of particles may stop altogether, either because no food can be captured or the animal ceases feeding. The filtration rates of most suspension feeders that have so far been adequately examined fit the unimodal-function theory and Wildish and Kristmanson (1997) claimed this to be a general principle of suspension feeding. Examples of animals that fit the unimodal-function theory include: cnidarians such as the sea pen Ptilosarcus gurneyi (Best 1988), the gorgonian corals Pseudopterogorgia acerosa, and P. americana (Sponaugle and Labarbera 1991), an alcyonacean soft coral (McFadden 1986), an annelid terebellid polychaete Lanice conchilega (Denis et al. 2007) and 10–12 species of marine and freshwater molluscan bivalves (Ackerman 1999), including the bay scallop Argopecten irradians (Kirby-Smith 1972; Wildish et al. 1987). Bryozoans have been inadequately studied for evaluating how feeding varies as a function of velocity, so it is unclear if the unimodal-function theory holds for all phyla that have suspension feeders.

The unimodal-function theory implies there is a velocity, or range of velocities, at which filtration rate is maximal. Does this help determine which habitats are most suitable for an organism? The second prediction is simply based on the idea that bryozoans should grow faster and survive longer and thus have greater abundance in environments in which there is greater capture of food. A few studies have related feeding success to the distribution and abundance of suspension feeders in the context of their flow environment. These include investigations of sea pens (Best 1988), octocorals (Sebens 1984; Patterson 1991a, 1991b), and sea anemones (Koehl 1977; Shick et al. 1979; Sebens 1981). One study on bryozoans showed that both feeding success and growth decreased with increasing velocity for an encrusting species, Electra pilosa (Okamura 1992). That study provided evidence that feeding patterns are reflected in growth, but more research is needed to see if this pattern is general.

While there may be a velocity at which feeding is optimized that accordingly predicts the most suitable habitats for bryozoans, encrusting colonies may be able to live in a wider range of flows. Erect colonies may be less successful in environments with high ambient velocities because of a higher risk of dislodgment or damage (Cheetham and Thomsen 1981; Cheetham 1986) and because high drag can cause deformation of the filtering surface and reduce the efficiency of filtration (Best 1988). Encrusting colonies are exposed to slower flows in the velocity gradient close to the substratum. Since an encrusting colony would experience slower local flow and has a much wider area attached to the substratum than does an erect colony, the encrusting colony would need less support to prevent dislodgement and would still be able to feed without the colony being deformed by the flow. This advantage of the encrusting form explains the third prediction that encrusting sheets should have higher ingestion rates than do erect trees at high ambient velocities of flow.

Methods

Feeding as a function of velocity

Collection and preparation of colonies

Ingestion rate was measured for four species of bryozoans: Membranipora membranacea, Celleporella hyalina, Bugula pacifica, and Schizoporella varians. Membranipora and Celleporella are encrusting sheets, while Bugula and Schizoporella are erect trees (Fig. 1). Small colonies of each species were collected from the floating docks at the Friday Harbor Laboratories in Friday Harbor, WA, USA. Colonies of the two erect species were each glued with cyanoacrylite gel onto a Plexiglas plate (5 cm×5 cm). Because the glue only touched a very small part of the colony, where no feeding zooids where located, and the colonies were given plenty of time to recover after gluing (generally weeks), it was assumed that the glue had no effect on bryozoans’ feeding rates. Colonies of the two encrusting species were carefully peeled from a red alga (Mazzealla splendens) and then placed on a glass slide (3 cm×1 cm) in a dish of seawater. The encrusting colonies were allowed to grow onto the slides until they were firmly attached (usually 1–2 days). Once colonies were firmly attached, the plates and slides were placed in racks and hung from the floating docks so the bryozoans could feed until needed for experiments. No difference in morphology was observed between colonies naturally growing in the field and colonies transferred and grown for experimentation.

Fig. 1

Photographs of the four bryozoan species used in this study. (A) M. membranacea, (B) C. hyalina, (C) S. varians, and (D) B. pacifica.

Fig. 1

Photographs of the four bryozoan species used in this study. (A) M. membranacea, (B) C. hyalina, (C) S. varians, and (D) B. pacifica.

Protocols for measuring ingestion rate

Feeding experiments were conducted in a recirculating flow-tank with a working section of 70 cm×10.2 cm×13 cm (Vogel and LaBarbera 1978). The flow-tank was filled with 0.45 μm filtered seawater and kept in a cold room at 12°C to maintain temperature similar to that of the bryozoans’ natural habitat. Blue-dyed polystyrene beads (mean diameter = 10.3 μm, SD = 0.94 μm, density = 1050 kg/m3) at a concentration of 1000 beads/ml were used as food particles. Bead concentrations were measured using a nanoplankton counting meter. Sizes and concentrations of particles were chosen to coincide with those encountered in the field, as in other studies (Okamura 1984, 1985). Polystyrene beads have been used in many studies of suspension feeders (Nygaard et al. 1988; Shimeta and Koehl 1997), including bryozoans (Okamura 1984, 1985; Pratt 2005), and bryozoans have been shown to ingest them in large quantities (Okamura 1984, 1985). None of the species in the present study were ever observed to reject beads. Polystyrene beads are particularly convenient because they are not digested. Colonies can therefore be preserved and feeding rates determined later by counting the number of beads ingested.

A plate or slide containing a single colony was positioned flush with the bottom of the flow-tank so only the bryozoan protruded into the flow. After feeding for 20 min, the colony was removed, fixed in formalin, rinsed in 70% EtOH, and cleared in 50% glycerol. Feeding rate was measured for 3–5 colonies of each species at each velocity. Colonies used were close to 1 cm in length (Bugula: 0.98 ± 0.16 cm, Celleporella: 0.84 ± 0.19 cm, Membranipora: 0.88 ± 0.19 cm, Scrupocellaria: 1.43 ± 0.23 cm).

Rate of ingestion was measured in five freestream velocities: 0, 0.4, 2.3, 4.3, and 7.0 cm/s (Table 1). Velocities were measured by tracking particles in video recordings over known distances, using a laser light sheet to illuminate a 2D view. Near the substratum, the velocity (U, cm/s) showed a linear relationship with height above the bottom (z, cm). The slope of this relationship (dU/dz) allowed calculation of the shear velocity (U*)  

(1)
formula
(Vogel 1994), where ν is the kinematic viscosity (in this case ν = 0.0128 cm2/s). Estimates of U* on the most exposed blades in a kelp bed range between 0.22 and 0.60 cm/s (Koehl and Alberte 1988); thus, the velocities used in this study (U* between 0–0.37 cm/s) are likely to be within the lower range of velocities experienced by bryozoans living on kelps in the field.

Table 1

Characteristics of the five velocities of flow used in the feeding experiments

Freestreama (cm s−1Linear Relationshipb 
Range (cm) Slope R2 U* 
NA NA NA NA 
0.4 ± 0.01 0–1 0.29 0.94 0.061 
2.3 ± 0.05 0–0.8 2.5 0.96 0.18 
4.3 ± 0.31 0–0.5 6.1 0.90 0.28 
7.0 ± 0.74 0–0.35 10.7 0.86 0.37 
Freestreama (cm s−1Linear Relationshipb 
Range (cm) Slope R2 U* 
NA NA NA NA 
0.4 ± 0.01 0–1 0.29 0.94 0.061 
2.3 ± 0.05 0–0.8 2.5 0.96 0.18 
4.3 ± 0.31 0–0.5 6.1 0.90 0.28 
7.0 ± 0.74 0–0.35 10.7 0.86 0.37 

aValues of freestream velocity are the means ± SD of velocities measured between 2.5 and 5 cm above the bottom.

bThere was a linear relationship between velocity (U) and height above the bottom (z). Range: range in heights above the bottom where the linear relationship was found. Slope: slope of the linear relationship (dU/dz). R2: coefficient of determination.

U*: calculated shear velocity.

The total number of zooids and the number of beads ingested by each zooid in the colony were counted using a compound microscope at 100×. Only autozooids (with or without polypides) were counted for each species. Autozooids have a polypide in them at some point and are capable of feeding. Some autozooids will lose their polypides and serve another purpose, such as support, in an erect-tree colony or in an excurrent space in an encrusting colony. Membranipora only has autozooids but the other three species also have heterozooids with other functions. Mature Bugula colonies have reproductive ovicells and defensive avicularia, Scrupocellaria have defensive vibracula, and Celleporella have small, nonfeeding male zooids. However, autozooids are substantially larger than these heterozooids and thus make up most of the area of the colonies. Therefore, only autozooids are considered in this study and the terms autozooid and zooid are used synonymously.

Zooid ingestion rate is defined here as the number of beads eaten by a zooid divided by the time spent eating them (beads/min). Only the zooids that ate one or more beads were included in calculating mean zooid ingestion rates. Despite removing the large number of zeros by only including the zooids that ate, the data was still skewed and could not be transformed to conform to a normal distribution. Therefore, zooid ingestion rates were analyzed with a generalized, linear, mixed model using a negative binomial distribution (Proc GLIMMIX, SAS 9.1, SAS Institute 2006) with species (Bugula, Celleporella, Membranipora, Scrupocellaria) and velocity (0, 0.4, 2.3, 4.3, and 7.0 cm/s) as fixed effects and colony (3–5 colonies per treatment) as a random blocking effect.

While zooid ingestion rate can be used to compare the average feeding success of zooids that ate, comparing feeding success at the colony level is also important. Colony ingestion rate is defined here as the number of beads eaten by an entire colony divided by the total number of zooids in the colony and the time they spent eating (beads/zooid/min). Colony ingestion rates were normally distributed so an ANOVA model was used (Proc GLM, SAS 9.1, Freund et al. 1986) with species and velocity as fixed effects. Not all of the zooids were capable of eating since they all did not contain polypides. To be able to better interpret colony ingestion rates, it is useful to know how many zooids had polypides as well as how many of those polypides actually fed. Thus, I compared the percent of polypides that fed (number of polypides that fed divided by the total number of polypides times 100) and the percent of zooids that fed (number of polypides that fed divided by the total number of zooids times 100) using a two-way ANOVA model with species and velocity as fixed effects.

Interactions were compared a priori as simple effects using the Tukey–Kramer adjustment for multiple comparisons (or T-method as in Hayter 1989). This approach allowed comparison of each velocity within a species and each species within a velocity. I also performed a priori orthogonal polynomial tests for linear, quadratic, and cubic trends in velocity.

Sizes of zooids and colonies

Because feeding can be influenced by the size of zooids and of colonies and because these can vary among species, I measured several aspects of the morphology of zooids and colonies. To compare zooid size among species, I determined the area of the zooid, the volume of the lophophore, the height of the zooid, and the number of tentacles per zooid. The area of the zooid was calculated by multiplying the zooid's maximum length times its width. The volume of the lophophore was calculated by:  

(2)
formula
in which V = volume of the lophophore, h = the height of the lophophore, D1 = the diameter of the base of the lophophore, and D2 = the diameter of the top of the lophophore. I calculated zooid height by adding the height of the introvert to the height of the lophophore.

Size of a colony was estimated by measuring the diameter (encrusting sheets) or height (erect trees) of the colony as well as the area. To measure the area, the outline of the colony was traced onto an acetate sheet. For encrusting sheets, the perimeter of the colony was easy to trace since they are essentially 2D sheets. For erect trees, the best nondestructive way to get a colony outline was to lay the colony on its side and trace it in side-view. Traced outlines were scanned into a computer and the area digitized using Scion Image. I also counted the number of branches in colonies of the two erect bryozoans by counting the number of tips of branches.

Analysis of covariance (ANCOVA) was used to compare how different variables scale among species (Proc GLM). First, the slopes were compared and if they were not significantly different (i.e. if P > 0.05), a common slope was used and the intercepts compared. All data for ANCOVA were transformed by natural logarithm.

Field experiments

Velocities at field sites

The two sides of the floating docks at the Friday Harbor Laboratories provided sites exposed to contrasting flow regimes. The “protected” site faced the shore, while the “exposed” site faced the harbor. Both sides of the dock have abundant sessile organisms that live on the tires or the abundant macroalgae that hang from the sides of the dock. Ambient flow was measured at each site by two methods. In the first, an electromagnetic current meter (Marsh–McBirney Model 511) measured free-stream velocity of the water. Maximum and minimum velocities were measured 15 cm from the dock and 30 cm below the surface of the water for each side of the dock over 1-min periods every 4 h over 24 h. The second method used solid-state dissolution modules (Pep-O-Mint Life Savers) to measure the relative ambient velocity near the substratum by using a substance that dissolves in water at a rate proportional to water velocity. Dissolution rate is measured by weighing the substance before and after a specified time. Blocks of gypsum have often been used for this purpose (Muus 1968; Doty 1971; but see Porter et al. 2000), but Pep-O-Mint Life Savers are also well suited for these measurements (Koehl and Alberte 1988). Ten replicate Life Savers were weighed, sewn onto kelp blades or tires, allowed to dissolve for 10 min, removed, and weighed again. Dissolution rate was calculated as the difference in weight before and after deployment, divided by the time deployed in the water (10 min).

One-tailed t-tests were used to compare the mean maximum and mean minimum velocity measured at the exposed site versus the protected site. A two-way fixed ANOVA model was used to compare the effect of site (exposed, protected), substrate (algae, tire), and the site-by-substrate interaction on the dissolution rate of Life Savers (Proc GLM).

Observations of abundance and size

Abundance was assessed by estimating by eye the percent cover of each bryozoan species in 5 cm×5 cm quadrats. Three quadrats were sampled on each of three substrata (tires, red algae, and brown algae) in five replicated blocks (tires) in each site. Blocks, substrata, and locations of quadrat sample were chosen at random. In addition to the measurements of percent cover, the length (diameter or height) of the two largest colonies of each species (if present) was measured in each block in each site. Abundance and size were measured four times over the course of late spring through the summer (5/25, 6/18, 7/29, 8/24) in the year 2000.

A four-way ANOVA model of percent cover with exposure, species, substratum, and time as fixed factors would not converge because of too many zero values. Therefore, I used a series of other models to make the comparisons of most interest. For the first model, I compared percent cover for each species separately with a three-way interaction of exposure (protected or exposed), substratum (tires, red algae, or brown algae), and time (0, 24, 65, or 91 days) as fixed effects and block as a random effect. I then compared the maximum percent cover over time for each species with a two-way interaction between exposure and substratum as fixed effects and block as a random effect. Finally, I found the maximum percent cover for each substratum-by-time interaction and then compared these values with a two-way interaction between exposure and species (Bugula, Celleporella, Membranipora, Scrupocellaria) as fixed effects and block as a random effect.

A three-way ANOVA model of maximum colony length with exposure, species, and time as fixed factors similarly would not converge because of too many zero values. Therefore, I compared the length of the largest colony for each species separately with exposure and time as fixed effects and block as a random effect. Then I compared the length of the largest colony for each time separately with exposure and species as fixed effects and block as a random effect.

Proc GLIMMIX with the assumption of a normal distribution was used for all the analyses of percent cover and colony length. The interactions were all compared a priori as simple effects using the Tukey–Kramer adjustment for multiple comparisons. I also performed a priori orthogonal polynomial tests for linear, quadratic, and cubic trends in time.

Experiments on survival and growth

Growth and survival of three of the species (Membranipora, Bugula, and Scrupocellaria) were measured over 1 month at the two sites. Celleporella was not included because there were not enough colonies available for analysis. Single colonies of each species were allowed to settle or were glued (using cyanoacrylite gel) onto 16 replicate Plexiglas plates (11 cm×7 cm) that were attached vertically to PVC racks suspended on each side of the dock. Each rack held eight plates of each species. Colonies of each species were randomly assigned to location. Growth and survival of Membranipora and Bugula colonies were measured in the summer of 2000, while Scrupocellaria was measured in the summer of 2001. Measurements were taken four times regularly over approximately 40 days. During measurements, plates were cleaned of everything except the one colony of interest so as to minimize competition. Survivorship of each colony was noted, and measurements of area were made as already described.

Survival rate was calculated by:  

(3)
formula
where l = the proportion of the original cohort surviving to day t, N(t) = number of colonies at time t, t = time (in days), N(t0) = the number of colonies at t0, and t0 = time zero. Relative growth rate was calculated by:  
(4)
formula
where RGR = relative growth rate, At = area after 12 days, At0 = initial area, t = 12 days. Even though area was measured four times over about 40 days, growth was calculated over 12 days because this period represents the longest time for which a relatively large sample size was available (many of the colonies did not survive to the next date of measurement).

Survival was analyzed using a Cox regression proportional-hazards model, and ties were handled using the exact method (Proc PHREG, SAS 9.1, Allison 1995). The covariates used in the full model were (1) two dummy variables for the different species (Bugula and Scrupocellaria with Membranipora as the withheld group), (2) a dummy variable for site (with the exposed site as the dummy variable and the protected site as the withheld group), and (3) initial area. Relative growth rate was compared with a two-way ANOVA model with species (Bugula, Membranipora, or Scrupocellaria) and exposure (protected or exposed) as fixed effects (Proc GLM). Interactions were compared a priori as simple effects using the Tukey–Kramer adjustment for multiple comparisons.

Results

Feeding as a function of velocity

Ingestion rates

Zooid ingestion rates were significantly different among species (P3,56 < 0.0001): Membranipora had greater zooid ingestion rates than did Bugula and Scrupocellaria (P < 0.0001) and these three species had more frequent zooid ingestion than did Celleporella (P ≤ 0.001) (Fig. 2). The random colony effect added significant variation to the data (P < 0.0001) but the velocity main effect and species-by-velocity interaction were not significant (velocity: P4,74 > 0.05, species×velocity: P12,55 > 0.05). All four species together had a significant quadratic trend in velocity (P < 0.05) but when each species was tested separately only Membranipora (P = 0.001) and Scrupocellaria (P < 0.05) had significant quadratic trends in velocity.

Fig. 2

Zooid (A) and colony (B) ingestion rates as a function of water velocity for four bryozoan species. Values are means ± SEM.

Fig. 2

Zooid (A) and colony (B) ingestion rates as a function of water velocity for four bryozoan species. Values are means ± SEM.

Colony ingestion rates were also significantly different among species (P3,54 < 0.0001): Membranipora had greater ingestion rates than did the other three species (P < 0.0001). The interaction between species and velocity was also significant (P12,54 < 0.01) in which Membranipora had significantly greater colony ingestion rates than did the other three species from 0.4 to 7 cm/s (P < 0.01) and no species differed at 0 cm/s (all P > 0.05). Velocity had an overall effect (P4,54 < 0.05): colony ingestion rates were higher for all species at 2.3 cm/s than at 7 cm/s (P < 0.05) and showed a quadratic trend (P < 0.001). When the simple interactive effects were compared, only Membranipora showed any significant effect among velocities; colony ingestion rate was greater at intermediate velocities (0.4, 2.3, and 4.3 cm/s > 0 and 7 cm/s; all P < 0.05) and there was a significant quadratic trend (P < 0.0001).

The overall main effect of velocity on percent feeding was not significant (percent of polypides: P4,54 > 0.05, percent of zooids: P4,54 > 0.05). However, there was a small, but significant, linear decrease in the percentage that fed as a function of velocity (percent of polypides: P < 0.05, percent of zooids: P < 0.05). The interaction between species and velocity was not significant (percent of polypides: P12,54 > 0.05, percent of zooids: P12,54 > 0.05). The percent of polypides and zooids that fed differed significantly among species (both P3,54 < 0.0001) (Fig. 3). Membranipora had a significantly greater percent of polypides and zooids that fed than did each of the three of the other species (each P < 0.01), and Bugula had significantly greater percent of polypides and zooids that fed than did Celleporella and Scrupocellaria (each P < 0.05).

Fig. 3

The percent of polypides and zooids that fed for each of the four bryozoan species. Values are means ± SEM.

Fig. 3

The percent of polypides and zooids that fed for each of the four bryozoan species. Values are means ± SEM.

Size of zooids and colonies

Membranipora zooids are generally larger in area and height and have greater volume of the lophophore than do the other three species (Table 2). Scrupocellaria zooids have the second largest lophophore volume and Celleporella has the smallest. Bugula, Celleporella, and Scrupocellaria had similar zooid areas.

Table 2

Mean (± one standard error of the mean) zooid area, zooid height, number of tentacles, and volume contained by the lophophore for four species of bryozoan zooids

Species Zooid area (mm2Zooid height (mm) No of tentacles Lophophore volume (mm3
Celleporella 0.1456 ± 0.0001 0.1294 ± 0.0002 13.0 ± 0.0 283.2 ± 0.004×10−4 
Bugula 0.1232 ± 0.0005 0.0550 ± 0.0015 15.2 ± 0.2 419.7 ± 0.061×10−4 
Scrupocellaria 0.1080 ± 0.0006 0.1023 ± 0.0007 16.0 ± 0.0 711.9 ± 0.149×10−4 
Membranipora 0.2622 ± 0.0016 0.2419 ± 0.0053 16.7 ± 0.4 842.2 ± 0.547×10−4 
Species Zooid area (mm2Zooid height (mm) No of tentacles Lophophore volume (mm3
Celleporella 0.1456 ± 0.0001 0.1294 ± 0.0002 13.0 ± 0.0 283.2 ± 0.004×10−4 
Bugula 0.1232 ± 0.0005 0.0550 ± 0.0015 15.2 ± 0.2 419.7 ± 0.061×10−4 
Scrupocellaria 0.1080 ± 0.0006 0.1023 ± 0.0007 16.0 ± 0.0 711.9 ± 0.149×10−4 
Membranipora 0.2622 ± 0.0016 0.2419 ± 0.0053 16.7 ± 0.4 842.2 ± 0.547×10−4 

The slopes of the equations comparing the number of polypides as a function of the number of zooids were significantly different among species (P < 0.05, Table 3); Membranipora had a significantly greater slope than did Bugula and Celleporella (P < 0.05). At a value of 250 zooids, Bugula and Membranipora had a greater number of polypides than did Celleporella (P < 0.05). At a value of 500 zooids, Bugula and Membranipora had a greater number of polypides than did Scrupocellaria and Celleporella (P < 0.05), and Scrupocellaria had more polypides than did Celleporella (P < 0.001). At a value of 1000 zooids, Celleporella had significantly fewer polypides than did the other three species (P < 0.01).

Table 3

Results of the ANCOVA tests to compare morphological scaling in four species of bryozoans

 P versus Z Br versus Z Z versus A Z versus L 
Species Slope Intercept Slope Intercept Slope Intercept Slope Intercept 
Celleporella 0.70 0.85   0.89 2.42 1.45 2.76 
Bugula 0.84 0.45 0.85 −1.19 1.00 2.69 1.45 1.70 
Scrupocellaria 0.90 −0.14 0.69 −0.77 0.78 2.96 1.45 2.64 
Membranipora 1.19 −1.68   0.66 3.09 1.45 2.53 
 P versus Z Br versus Z Z versus A Z versus L 
Species Slope Intercept Slope Intercept Slope Intercept Slope Intercept 
Celleporella 0.70 0.85   0.89 2.42 1.45 2.76 
Bugula 0.84 0.45 0.85 −1.19 1.00 2.69 1.45 1.70 
Scrupocellaria 0.90 −0.14 0.69 −0.77 0.78 2.96 1.45 2.64 
Membranipora 1.19 −1.68   0.66 3.09 1.45 2.53 

A: total colony area (mm2); Br: total number of branches in the colony; L: length (mm); P: total number of zooids with polypides; Z: total number of zooids. Values are from equations in the general form ln Y = α ln X + β where α is the slope estimate and β is the intercept estimate. If slopes were not found to be significantly different in the full model, then common slopes were used and intercepts compared (see text for details).

Bugula had a significantly greater slope than did Scrupocellaria when the number of colony branches was compared as a function of the number of zooids (P < 0.05; Table 3). Within the range of the data (from ∼250 to 1000 zooids), Bugula always had a significantly greater number of branches for a given number of zooids (P < 0.0001), meaning that Bugula colonies are bushier than are those of Scrupocellaria.

The slopes of the equations comparing the number of zooids as a function of the colony area were significantly different among species (P = 0.05, Table 3). Bugula and Celleporella had significantly greater slopes than did Membranipora (P < 0.05), and Bugula also had a significantly greater slope than did Scrupocellaria (P < 0.05). At an area of 25 mm2, Bugula had a significantly greater number of zooids than did the other three species (P < 0.001) and Scrupocellaria had a greater number of zooids than did Membranipora (P < 0.05). At areas in the range of 50 to 100 mm2, Bugula had a significantly greater number of zooids than did Celleporella (P < 0.001), and Scrupocellaria had a greater number of zooids than did Membranipora (P < 0.001).

The slopes of the equations comparing the number of zooids as a function of the length of the colony were not significantly different among species (P > 0.05), so they were set equal and the intercepts compared (Table 3). For a given length of colony, Bugula had a significantly greater number of zooids than did each of the three other species (P < 0.0001) and Celleporella had a significantly greater number of zooids than did Membranipora (P < 0.01).

Field experiments

Velocities at field sites

The mean maximum velocity at the exposed site (23.6 ± 6.6 cm/s) was significantly greater than at the protected site (8.6 ± 1.9 cm/s) (P12 < 0.05). The mean minimum velocity was also significantly greater at the exposed site (3.3 ± 0.5 cm/s) than at the protected site (1.3 ± 0.5 cm/s) (P12 < 0.01). Dissolution rates were similar between substrates within a site (protected, algae: 0.0024 ± 0.0001 g/s; protected, tires: 0.0024 ± 0.0001 g/s; exposed, algae: 0.0036 ± 0.0002 g/s; exposed, tires: 0.0036 ± 0.0001 g/s), but dissolution was significantly greater on the exposed side than on the protected side (site: P1,32 < 0.0001; substrate: P1,32 > 0.05; site×substrate: P1,32 > 0.05).

Observations of abundance and size

Membranipora and Celleporella were found on both sides of the dock, while Bugula was rarely and Scrupocellaria was never found on the exposed side. Scrupocellaria only occurred on tires, Bugula grew on tires and sometimes on brown algae, Celleporella grew on red algae and occasionally on brown algae, and Membranipora grew on red and brown algae. Percent cover increased as a linear function of time for Bugula (protected side, brown algae P < 0.05; protected side, tires P < 0.0001) and Membranipora (exposed side, brown algae P < 0.05; protected side, brown algae P < 0.0001; protected side, red algae P < 0.0001; but not exposed side, red algae where P > 0.05), but not for Celleporella (protected side, red algae P > 0.05) or Scrupocellaria (protected side, tires P > 0.05) (Fig. 4). The block random factor added significant variation to the percent cover of Bugula (P < 0.01) but not for any other species (each P > 0.05). When comparing the time with the maximum percent cover for each species, percent cover on the protected side was greater than on the exposed side of the dock (Bugula on tires P < 0.01; Membranipora on red algae P < 0.05; Scrupocellaria on tires P < 0.01), but there was no difference in some cases (Bugula on brown algae P > 0.05; Celleporella on red algae P > 0.05 and on brown algae P > 0.05; Membranipora on brown algae P > 0.05). In all cases where the comparison could be made, Membranipora had greater percent cover than did the other species (each P < 0.001).

Fig. 4

Percent cover as a function of time on three different substrata (brown algae, red algae, tires) and two flow environments (exposed or protected side of the docks at the Friday Harbor Laboratories) for four species of bryozoans: (A) M. membranacea, (B) C. hyalina, (C) B. pacifica, and (D) S. varians. Values are means ± SEM. Note that the scale is different for (A) but (B–D) have the same scale.

Fig. 4

Percent cover as a function of time on three different substrata (brown algae, red algae, tires) and two flow environments (exposed or protected side of the docks at the Friday Harbor Laboratories) for four species of bryozoans: (A) M. membranacea, (B) C. hyalina, (C) B. pacifica, and (D) S. varians. Values are means ± SEM. Note that the scale is different for (A) but (B–D) have the same scale.

Maximum length of the colony increased linearly with time on the protected side of the dock for all four species (each P < 0.05) (Fig. 5). It also increased linearly with time on the exposed side of the dock for Membranipora (P < 0.0001) but not for Celleporella (P > 0.05). Membranipora attained larger maximum size on the protected side than it did on the exposed side of the dock on days 24 and 65 (P < 0.0001), but Celleporella did not differ in maximum size between sides of the dock at any time (P < 0.05) and Bugula and Scrupocellaria were not found on the exposed side. Membranipora attained a larger maximum length than did any of the other species during days 24–91 (each P < 0.01), Bugula and Scrupocellaria had larger maximum lengths than did Celleporella on days 65 and 91 (each P < 0.001), and Bugula had a larger maximum length than did Scrupocellaria on day 91 (P < 0.001). The block random factor added significant variation to the percent cover of all species (each P < 0.05) except Celleporella (P > 0.05).

Fig. 5

Maximum length of colonies (diameter for encrusting colonies and height for erect colonies) as a function of time for four species of bryozoans in two flow environments (exposed or protected side of the docks at the Friday Harbor Laboratories). Values are means ± SEM.

Fig. 5

Maximum length of colonies (diameter for encrusting colonies and height for erect colonies) as a function of time for four species of bryozoans in two flow environments (exposed or protected side of the docks at the Friday Harbor Laboratories). Values are means ± SEM.

Experiments on survival and growth

Survival of none of the three species of bryozoans was affected by the initial area of the colony (P > 0.05). Membranipora had a significantly greater survival rate than did Bugula or Scrupocellaria overall as well as in each flow environment (P < 0.05): Membranipora was more than six times more likely to survive than was either Bugula or Scrupocellaria on the protected side and more than three times more likely to survive than either species on the exposed side of the dock (Fig. 6). Bugula and Scrupocellaria did not have significantly different survival rates (P > 0.50). Overall, all three species had significantly greater survival on the protected side than on the exposed side of the dock (P < 0.001). When comparing each species separately, Membranipora and Scrupocellaria both had greater survival on the protected side (each P < 0.05), while Bugula did not show a difference in survival between the protected and exposed sides of the dock (P > 0.05). Both Membranipora and Scrupocellaria were three times more likely to survive on the protected side of the dock than was Bugula.

Fig. 6

Survival as a function of time for three species of bryozoans in two flow environments (exposed or protected side of the docks at the Friday Harbor Laboratories).

Fig. 6

Survival as a function of time for three species of bryozoans in two flow environments (exposed or protected side of the docks at the Friday Harbor Laboratories).

Consistent with the data on survival, Membranipora had a significantly greater relative growth rate than did Bugula or Scrupocellaria overall as well as in each flow environment (P < 0.0001), but Bugula and Scrupocellaria did not have significantly different relative growth rates from each other (P > 0.05) (Fig. 7). Overall, relative growth rates were greater on the protected side of the dock (P < 0.01). When comparing each species separately Membranipora had a greater relative growth rate on the protected side (P < 0.01), but neither Bugula nor Scrupocellaria showed a difference in relative growth rate between the protected and exposed sides of the dock (P > 0.05).

Fig. 7

Relative growth rate for three species of bryozoans in two flow environments (exposed or protected side of the docks at the Friday Harbor Laboratories). Values are means ± SEM.

Fig. 7

Relative growth rate for three species of bryozoans in two flow environments (exposed or protected side of the docks at the Friday Harbor Laboratories). Values are means ± SEM.

Discussion

Feeding as a function of velocity

The zooid ingestion rates of Membranipora and Scrupocellaria as well as the colony ingestion rate of Membranipora followed a quadratic trend as a function of velocity as predicted; however, a quadratic trend was not found for the other species. The lack of this trend for the other two species, Celleporella hyalina and Bugula pacifica, suggests that the unimodal theory of Wildish and Kristmanson (1997) does not hold true for all suspension feeders. As found for some of the bryozoans in this study, studies on other suspension feeders also provide support for the unimodal function theory (Kirby-Smith 1972; McFadden 1986; Wildish et al. 1987; Best 1988; Sponaugle and Labarbera 1991; Wildish and Saulnier 1993). However, previous data on how water velocity affects the rates at which bryozoans feed generally show that feeding rate is either not affected by velocity or decreases with increased velocity (Okamura 1984, 1985, 1990, 1992). Such inconsistencies may be explained by measuring feeding in too narrow a range of velocities (Eckman and Duggins 1993) or not using the optimal type or concentration of particles (Okamura 1987b, 1990).

Feeding as a function of morphology

The general shape of the colony (encrusting versus erect) did not have a consistent effect on feeding. I expected that at higher velocities encrusting sheet colonies would have higher ingestion rates than would erect tree colonies. However, one encrusting species, Membranipora, had a higher feeding rate than did the erect species at all velocities while the other encrusting species, Celleporella, had the lowest feeding rate. This result suggests that the overall form of the colony is not the most important factor determining feeding success and few generalizations can be made. To properly assess the effect of colony form on feeding success, species differences need to be controlled. For example, it is important to take size of the colony into account since size does affect feeding in bryozoans (Pratt 2005). However, what measure of colony size to use in comparisons is not easy to determine because differences among species make it difficult to truly compare like with like. For example, only autozooids were considered in this study, while some species also produced heterozooids. Thus, total zooid numbers for Membranipora truly consider all the zooids in the colony since it only produces autozooids, but not for Bugula, Scrupocellaria, or Celleporella because, in addition, they produce some heterozooids. While area of a colony may be a useful measure of overall size, one species may have more zooids per unit area and more polypides as a function of zooid number, thereby giving a higher feeding capacity per unit area. Bugula had more zooids as a function of area, more zooids per unit length, more branches as a function of total number of zooids, and smaller zooids than did Scrupocellaria. The ingestion rates of zooids and of colonies of these two species were similar and yet Bugula had a higher percentage of polypides and zooids that fed. Membranipora had fewer zooids as a function of area and fewer per unit length than did the other species, but this is because it had larger zooids. Not only did Membranipora have more polypides as a function of the total number of zooids, but it had the greatest percentage of polypides and zooids that fed, as well as the greatest ingestion rates by zooids and colonies. Whether Membranipora's greater feeding rates are caused by having larger zooids, its encrusting shape, some other unmeasured factor, or a combination of morphology and/or behaviors is unclear.

Not only do morphological differences among the species make comparing them difficult, but differences among species cannot be effectively taken into account when using comparative methods to establish phylogenetic relationships (Harvey and Pagel 1991) because we currently lack an adequate hypothesis of phylogenetic relationships among bryozoan taxa. A previous study was able to avoid this problem by artificially creating an erect tree form of Membranipora. This erect form had a higher feeding rate than did the encrusting form at a moderate free-stream flow of 3 cm/s (Pratt 2005). Whether the erect shape continues to have higher feeding than the encrusting shape in flows of higher velocities is unknown.

While the effect of colony form on feeding success remains unclear, differences other than in the morphology of zooids or colonies can influence feeding. Different species may show increased efficiencies of retention or have preferences for particles of different sizes (Shumway et al. 1985; Okamura 1987a; Riisgård 1988; Newell et al. 1989; Riisgård and Manríquez 1997; Lisbjerg and Petersen 2001), and capture of particles of different sizes may be differentially affected by the magnitude of the ambient velocity of water (Okamura 1990; Shimeta 1996; Shimeta and Koehl 1997). Species also may vary in their preference for food of specific quality or surface chemistry (“taste”) (Labarbera 1978; Rassoulzadegan et al. 1984; Sierszen and Frost 1992; Pedrotti 1995). More research is needed to determine if any of these factors can help explain the differences in feeding found among the species in this study.

The relatively high-feeding rate of Membranipora may result from larger zooids with greater water-pumping capacity. Larger lophophores have been shown to have faster incurrent water velocities (Best and Thorpe 1986), which should increase particle flux and therefore ingestion rate. However, zooid size cannot be the only factor that determines feeding success because the larger Scrupocellaria zooids have greater water pumping capacity than do Bugula zooids, yet Scrupocellaria did not always have greater feeding success than Bugula.

A further explanation for the high-feeding success of Membranipora relates to the arrangement of the zooids within its colonies. Membranipora zooids are packed closely together in a sheet and this increases the capture of food with that of single zooids or of zooids that are more widely spaced (Eckman and Okamura 1998; Pratt 2004). Additionally, the results of the current study showed that Membranipora had a higher percent of polypides or zooids that captured food compared with the other three species, which supports the idea that Membranipora is especially good at filtering. If the heterozooids of the other species were taken into consideration, they would have an even lower percentage of zooids that fed, which suggests that Membranipora invests heavily in having a high-feeding capacity, and the higher percent of polypides that fed suggests that the polypides that are there are very effective at capturing food. Other unmeasured factors may also contribute to the high-feeding rate of Membranipora. For example, it is possible that Membranipora is especially efficient at filtering particles that are 10 µ in size. The effect of particle size on feeding in these four bryozoan species would have to be measured to test this hypothesis.

Performance in the field

Since the velocities measured on the protected side of the floating docks were in the range at which bryozoan feeding was high, all the species were expected to be more abundant, to grow faster, and to survive longer at the protected site. For the most part, this prediction was supported. All four species were either found exclusively on the protected side of the dock or had a greater percent cover on the protected side. Since Membranipora had higher ingestion rates at higher velocities than did the other three species, it makes sense that Membranipora was the only species found in abundance on the exposed side of the dock. Membranipora grew to a larger maximum size than did the other three species, and attained the largest colony size on the protected side. However, by the last sampling date the percent cover and the maximum size for Membranipora was similar between the protected and exposed sides. This could be a result of predation by nudibranchs on the protected side. While predator densities and predation was not measured, predation was considerably more evident on the protected side of the dock (personal observation). The transplant experiment added further evidence that distribution and abundance is correlated with differences in growth and survival: all three species tested had higher survival and faster growth on the protected side, and Membranipora showed higher survival and faster growth than did the other species. Part of the reason that Membranipora grows faster and attains a larger size may be due to its larger zooids and to the lower energetic investment because of only making autozooids.

Overall, the results are consistent with the hypothesis that bryozoans would have greater abundance, faster growth rates, and higher survival in habitats dominated by water velocities where feeding success is high. Previous research on the effect of velocity on growth in bryozoans show mixed results. For example, many studies have investigated the effect of flow velocity on growth in Celleporella hyalina but the results are not consistent. Celleporella can have more rapid growth in slower flow (Cancino and Hughes 1987), more rapid growth with increasing flow (Hughes and Hughes 1986), or no effect of flow on growth (Hermansen et al. 2001). Some of this variation may be due to a stronger effect of season than of flow velocity on growth rate (Cancino and Hughes 1987; Hughes 1992). It should also be noted that none of these studies relate growth rate to feeding success. The only previous study to do that similarly found a correlation between the two attributes (Okamura 1992).

Ecological implications

Feeding varies with water velocity for bryozoans, and this relationship is different for different species. Bryozoan distribution and abundance at least in part may be due to how water flow affects feeding, growth, and survival. Some bryozoans, such as Membranipora, are especially effective at capturing particles even at higher water velocities. Thus, species such as Membranipora may be found in a wider range of flows in nature. Another advantage of being an especially good filterer is that rapid growth can be achieved. Membranipora is most often found growing on kelps and other macroalgae (Bernstein and Jung 1979; Yoshioka 1982). While kelps are perennial, the blades can be more ephemeral and generally live <6 months in Macrocystis pyrifera (North 1961) and Laminaria longicruris (Chapman 1986). Membranipora can utilize this kind of habitat through high recruitment and rapid growth (McKinney and Jackson 1991). Not only does rapid growth give Membranipora a competitive advantage on such ephemeral substrata, but it also might contribute to its effectiveness as an invasive species. Membranipora was first seen in the Gulf of Maine in 1987, and within a few years it became the dominant organism living on kelps (Berman et al. 1992). Membranipora has a negative effect on the kelps (Lambert et al. 1992; Scheibling et al. 1999; Chavanich and Harris 2000; Levin et al. 2002; Saier and Chapman 2004) and as a result, substantial areas of the rocky subtidal along the Atlantic coast of Nova Scotia and the Gulf of Maine have shifted from domination by kelp to domination by the invasive macroalga Codium fragile ssp. tomentosoides (Harris and Tyrrell 2001; Levin et al. 2002; Mathieson et al. 2003; Scheibling and Gagnon 2006; Theriault et al. 2006). Understanding how water flow affects feeding in bryozoans not only provides insights about their distribution and abundance, but may also allow prediction of which bryozoans are likely to become problematic invasive species.

Acknowledgements

Thanks to G. Rivera and R. Blob for the invitation to participate in the symposium. I am grateful for comments made by S. Vogel, D. McShea, C. Cunningham, W. Kier, and E. Shaughnessy on early drafts of this manuscript as well as comments by two anonymous reviewers. I would like to thank A. O. D. Willows and staff for providing facilities and support at the Friday Harbor Labs. I was supported by an American Dissertation Fellowship from the American Association of University Women, a Wainwright Fellowship from the Friday Harbor Laboratories, and a Doctoral Dissertation Improvement Grant (IBN-0206457) from the National Science Foundation during the course of this research.

References

Ackerman
JD
Effect of velocity on the filter feeding of dreissenid mussels (Dreissena polymorpha and Dreissena bugensis): implications for trophic dynamics
Can J Fish Aquat Sci
 , 
1999
, vol. 
56
 (pg. 
1551
-
61
)
Allison
PD
Survival analysis using the SAS system: a practical guide
 , 
1995
Cary, NC
SAS Institute
Berman
J
Harris
L
Lambert
W
Buttrick
M
Dufresne
M
Recent invasions of the Gulf of Maine: three contrasting ecological histories
Conserv Biol
 , 
1992
, vol. 
6
 (pg. 
435
-
41
)
Bernstein
BB
Jung
N
Selective pressures and co-evolution in a kelp canopy community in Southern California, USA
Ecol Monogr
 , 
1979
, vol. 
49
 (pg. 
335
-
55
)
Best
BA
Passive suspension feeding in a sea pen – effects of ambient flow on volume flow-rate and filtering efficiency
Biol Bull
 , 
1988
, vol. 
175
 (pg. 
332
-
42
)
Best
MA
Thorpe
JP
Effects of food particle concentration of feeding current velocity in six species of marine Bryozoa
Mar Biol
 , 
1986
, vol. 
93
 (pg. 
255
-
62
)
Cancino
JM
Hughes
RN
The effect of water flow on growth and reproduction of Celleporella hyalina (L.) (Bryozoa: Cheilostomata)
J Exp Mar Biol Ecol
 , 
1987
, vol. 
112
 (pg. 
109
-
30
)
Chapman
ARO
Age versus stage: an analysis of age- and size-specific mortality and reproduction in a population of Laminaria longicruris Pyl
J Exp Mar Biol Ecol
 , 
1986
, vol. 
97
 (pg. 
113
-
22
)
Chavanich
S
Harris
LG
Pederson
J
Potential impact of the introduced bryozoan, Membranipora membranacea, on the subtidal snail, Lacuna vincta, in the Gulf of Maine
Marine bioinvasions
 , 
2000
Cambridge, MA
MIT Sea Grant College Program
(pg. 
157
-
63
)
Cheetham
AH
Branching, biomechanics and bryozoan evolution
Proce R Soc Lond B, Biol Sci
 , 
1986
, vol. 
228
 (pg. 
151
-
72
)
Cheetham
AH
Thomsen
E
Functional morphology of arborescent animals: strength and design of cheilostome bryozoan skeletons
Paleobiology
 , 
1981
, vol. 
7
 (pg. 
355
-
83
)
Cook
PL
Colony-wide water currents in living Bryozoa
Cah Biol Mar
 , 
1977
, vol. 
18
 (pg. 
31
-
47
)
Denis
L
Desroy
N
Ropert
M
Ambient flow velocity and resulting clearance rates of the terebellid polychaete Lanice conchilega (Pallas, 1766)
J Sea Res
 , 
2007
, vol. 
58
 (pg. 
209
-
19
)
Doty
MS
Measurement of water movement in reference to benthic algal growth
Botanica Marina
 , 
1971
, vol. 
14
 (pg. 
4
-
7
)
Eckman
JE
Duggins
DO
Effects of flow speed on growth of benthic suspension feeders
Bio Bull
 , 
1993
, vol. 
185
 (pg. 
28
-
41
)
Eckman
JE
Okamura
B
A model of particle capture by bryozoans in turbulent flow: significance of colony form
Am Nat
 , 
1998
, vol. 
152
 (pg. 
861
-
80
)
Freund
RJ
Littell
RC
Spector
PC
SAS system for linear models
 , 
1986
Cary, NC
SAS Institute, Inc
Harris
LG
Tyrrell
MC
Changing community states in the Gulf of Maine: synergism between invaders, overfishing and climate change
Biol Invasions
 , 
2001
, vol. 
3
 (pg. 
9
-
21
)
Harvey
PH
Pagel
MD
The comparative method in evolutionary biology
 , 
1991
Oxford
Oxford Universty Press
Hayter
AJ
Pairwise comparisons of generally correlated means
J Am Stat Assoc
 , 
1989
, vol. 
84
 (pg. 
208
-
13
)
Hermansen
P
Larsen
PS
Riisgård
HU
Colony growth rate of encrusting marine bryozoans (Electra pilosa and Celleporella hyalina)
J Exp Mar Biol Ecol
 , 
2001
, vol. 
263
 (pg. 
1
-
23
)
Hughes
DJ
Genotype-environment interactions and relative clonal fitness in a marine bryozoan
J Anim Ecol
 , 
1992
, vol. 
61
 (pg. 
291
-
306
)
Hughes
DJ
Hughes
RN
Life history variation in Celleporella hyalina (Bryozoa)
Proc R Soc Lond B
 , 
1986
, vol. 
228
 (pg. 
127
-
32
)
Kirby-Smith
WW
Growth of the bay scallop: the influence of experimental water currents
J Exp Mar Biol Ecol
 , 
1972
, vol. 
8
 (pg. 
7
-
18
)
Koehl
MAR
Effects of sea anemones on the flow forces they encounter
J Exp Biol
 , 
1977
, vol. 
69
 (pg. 
87
-
105
)
Koehl
MAR
Alberte
RS
Flow, flapping, and photosynthesis of Nereocystis luetkeana: a functional comparison of undulate and flat blade morphologies
Mar Biol
 , 
1988
, vol. 
99
 (pg. 
435
-
44
)
Labarbera
M
Particle capture by a Pacific brittle star – experimental test of aerosol suspension feeding model
Science
 , 
1978
, vol. 
201
 (pg. 
1147
-
9
)
Lambert
WJ
Levin
PS
Berman
J
Changes in the structure of a New England (USA) kelp bed: the effects of an introduced species?
Mar Ecol Prog Ser
 , 
1992
, vol. 
88
 (pg. 
303
-
7
)
Levin
PS
Coyer
JA
Petrik
R
Good
TP
Community-wide effects of nonindigenous species on temperate rocky reefs
Ecology
 , 
2002
, vol. 
83
 (pg. 
3182
-
93
)
Lidgard
S
Larwood
GP
Nielsen
C
Water flow, feeding, and colony form in an encrusting cheilostome
Recent and fossil bryozoa
 , 
1981
Fredensborg, Denmark
Olsen & Olsen
(pg. 
135
-
43
)
Lisbjerg
D
Petersen
JK
Feeding activity, retention efficiency, and effects of temperature and particle concentration on clearance rate in the marine bryozoan Electra crustulenta
Mar Ecol Prog Ser
 , 
2001
, vol. 
215
 (pg. 
133
-
41
)
Mathieson
AC
Dawes
CJ
Harris
LG
Hehre
EJ
Expansion of the Asiatic green alga Codium fragile subsp tomentosoides in the Gulf of Maine
Rhodora
 , 
2003
, vol. 
105
 (pg. 
1
-
53
)
McFadden
CS
Colony fission increases particle capture rates of a soft coral: advantages of being a small colony
J Exp Mar Biol Ecol
 , 
1986
, vol. 
1986
 (pg. 
1
-
20
)
McKinney
FK
Feeding and associated colony morphology in marine bryozoans
Rev Aquat Sci
 , 
1990
, vol. 
2
 (pg. 
255
-
80
)
McKinney
FK
Jackson
JBC
Bryozoan evolution
 , 
1991
Chicago, IL
University of Chicago Press
Muus
B
A field method for measuring “exposure” by means of plaster balls. A preliminary account
Sarsia
 , 
1968
, vol. 
34
 (pg. 
61
-
8
)
Newell
CR
Shumway
SE
Cucci
TL
Selvin
R
The effects of natural seston particle size and type on feeding rates, feeding selectivity and food resource availability for the mussel Mytilus edulis Linnaeus, 1758 at bottom culture sites in Maine (USA)
J Shellfish Res
 , 
1989
, vol. 
8
 (pg. 
187
-
96
)
Nielsen
C
Riisgård
HU
Tentacle structure and filter-feeding in Crisia eburnea and other cyclostomatous bryozoans, with a review of upstream-collecting mechanisms
Mar Ecol Prog Ser
 , 
1998
, vol. 
168
 (pg. 
163
-
86
)
North
WJ
Life-span of fronds of giant kelp, Macrocystis pyrifera
Nature
 , 
1961
, vol. 
190
 (pg. 
1214
-
5
)
Nygaard
K
Borsheim
KY
Thingstad
TF
Grazing rates on bacteria by marine heterotrphic microflagellates compared to uptake rates of bacterial-sized monodisperse flourescent latex beads
Mar Ecol Prog Ser
 , 
1988
, vol. 
44
 (pg. 
159
-
65
)
Okamura
B
The effects of ambient flow velocity, colony size and upstream colonies on the feeding success of Bryozoa: 1. Bugula stolonifera, an arborescent species
J Exp Mar Biol Ecol
 , 
1984
, vol. 
83
 (pg. 
179
-
94
)
Okamura
B
The effects of ambient flow velocity, colony size and upstream colonies on the feeding success of Bryozoa: 2. Conopeum reticulum, an encrusting species
J Exp Mar Biol Ecol
 , 
1985
, vol. 
89
 (pg. 
69
-
80
)
Okamura
B
Particle size and flow velocity induce an inferred switch in bryozoan suspension-feeding behavior
Biol Bull
 , 
1987
, vol. 
173
 (pg. 
222
-
9
)
Okamura
B
Ross
JRP
Seasonal changes in zooid size and feeding activity in epifaunal colonies of Electra pilosa
Bryozoa: past and present
 , 
1987
Bellingham, WA
Western Washington University Press
(pg. 
197
-
203
)
Okamura
B
Particle size, flow velocity, and suspension-feeding by the erect bryozoans Bugula neritina and Bugula stolonifera
Mar Biol
 , 
1990
, vol. 
105
 (pg. 
33
-
8
)
Okamura
B
Microhabitat variation and patterns of colony growth and feeding in a marine bryozoan
Ecology
 , 
1992
, vol. 
73
 (pg. 
1502
-
13
)
Patterson
MR
The effects of flow on polyp-level prey capture in an octocoral, Alcyonium siderium
Biol Bull
 , 
1991
, vol. 
180
 (pg. 
93
-
102
)
Patterson
MR
Passive suspension feeding by an octocoral in plankton patches: empirical test of a mathematical model
Biol Bull
 , 
1991
, vol. 
180
 (pg. 
81
-
92
)
Pedrotti
ML
Food selection (size and flavor) during development of echinoderm larvae
Invertebr Reprod Dev
 , 
1995
, vol. 
27
 (pg. 
29
-
39
)
Porter
ET
Sanford
LP
Suttles
SE
Gypsum dissolution is not a universal integrator of ‘water motion’
Limnol Oceanogr
 , 
2000
, vol. 
45
 (pg. 
145
-
58
)
Pratt
MC
The effect of zooid spacing on bryozoan feeding success: is competition or facilitation more important?
Biol Bull
 , 
2004
, vol. 
207
 (pg. 
17
-
27
)
Pratt
MC
Consequences of coloniality: influence of colony form and size on feeding success in the bryozoan Membranipora membranacea
Mar Ecol Prog Ser
 , 
2005
, vol. 
303
 (pg. 
153
-
65
)
Rassoulzadegan
F
Fenaux
L
Strathmann
RR
Effect of flavor and size on selection of food by suspension-feeding plutei
Limnol Oceanogr
 , 
1984
, vol. 
29
 (pg. 
357
-
61
)
Riisgård
HU
Feeding rates in hard clam (Mercenaria mercenaria) veliger larvae as a function of algal (Isochrysis galbana) concentration
J Shellfish Res
 , 
1988
, vol. 
7
 (pg. 
377
-
80
)
Riisgård
HU
Goldson
A
Minimal scaling of the lophophore filter-pump in ectoprocts (Bryozoa) excludes physiological regulation of filtration rate to nutritional needs. Test of hypothesis
Mar Ecol Prog Ser
 , 
1997
, vol. 
156
 (pg. 
109
-
20
)
Riisgård
HU
Manríquez
P
Filter-feeding in fifteen marine ectoprocts (Bryozoa): particle capture and water pumping
Mar Ecol Prog Ser
 , 
1997
, vol. 
154
 (pg. 
223
-
39
)
Saier
B
Chapman
AS
Crusts of the alien bryozoan Membranipora membranacea can negatively impact spore output from native kelps (Laminaria longicruris)
Botanica Marina
 , 
2004
, vol. 
47
 (pg. 
265
-
71
)
SAS Institute
The GLIMMIX procedure
2006
 
Cary, NC: SAS Institute, Inc
Scheibling
RE
Gagnon
P
Competitive interactions between the invasive green alga Codium fragile ssp. tomentosoides and native canopy-forming seaweeds in Nova Scotia (Canada)
Mar Ecol Prog Ser
 , 
2006
, vol. 
325
 (pg. 
1
-
14
)
Scheibling
RE
Hennigar
AW
Balch
T
Destructive grazing, epiphytism, and disease: the dynamics of sea urchin - kelp interactions in Nova Scotia
Can J Fish Aquat Sci
 , 
1999
, vol. 
56
 (pg. 
2300
-
14
)
Sebens
KP
The allometry of feeding, energetics, and body size in three sea anemone species
Biol Bull
 , 
1981
, vol. 
161
 (pg. 
152
-
71
)
Sebens
KP
Water flow and coral colony size: interhabitat comparisons of the octocoral Alcyonium siderium
Proc Natl Acad Sci USA
 , 
1984
, vol. 
81
 (pg. 
5473
-
7
)
Shick
JM
Hoffman
RJ
Lamb
AN
Asexual reproduction, population structure, and genotype-environment interactions in sea anemones
Am Zool
 , 
1979
, vol. 
19
 (pg. 
699
-
713
)
Shimeta
J
Particle-size selection by Pseudopolydora paucibranchiata (Polychaeta: Spionidae) in suspension feeding and in deposit feedings: influences of ontogeny and flow speed
Mar Biol
 , 
1996
, vol. 
126
 (pg. 
479
-
88
)
Shimeta
J
Koehl
MAR
Mechanisms of particle selection by tentaculate suspension feeders during encounter, retention, and handling
J Exp Mar Biol Ecol
 , 
1997
, vol. 
209
 (pg. 
47
-
73
)
Shumway
SE
Cucci
TL
Newell
RC
Yentsch
CM
Particle selection, ingestion, and absorption in filter-feeding bivalves
J Exp Mar Biol Ecol
 , 
1985
, vol. 
91
 (pg. 
77
-
92
)
Sierszen
ME
Frost
TM
Selectivity in suspension feeders - food quality and the cost of being selective
Arch Hydrobiol
 , 
1992
, vol. 
123
 (pg. 
257
-
73
)
Sponaugle
S
Labarbera
M
Drag-induced deformation: a functional feeding strategy in two species of gorgonians
J Exp Mar Biol Ecol
 , 
1991
, vol. 
148
 (pg. 
121
-
34
)
Theriault
C
Scheibling
R
Hatcher
B
Jones
W
Mapping the distribution of an invasive marine alga (Codium fragile spp. tomentosoides) in optically shallow coastal waters using the compact airborne spectrographic imager (CASI)
Can J Remote Sens
 , 
2006
, vol. 
32
 (pg. 
315
-
29
)
Vogel
S
Life in moving fluids: the physical biology of flow
 , 
1994
Princeton, NJ
Princeton University Press
Vogel
S
LaBarbera
M
Simple flow tanks for research and teaching
BioScience
 , 
1978
, vol. 
28
 (pg. 
638
-
43
)
Wildish
D
Kristmanson
DD
Benthic suspension feeders and flow
 , 
1997
, vol. 
xiii
 
Cambridge, New York
Cambridge University Press
pg. 
409
 
Wildish
DJ
Kristmanson
DD
Hoar
RL
Decoste
AM
McCormick
SD
White
AW
Giant scallop feeding and growth responses to flow
J Exp Mar Biol Ecol
 , 
1987
, vol. 
113
 (pg. 
207
-
20
)
Wildish
DJ
Saulnier
AM
Hydrodynamic control of filtration in Placopecten maggellanicus
J Exp Mar Biol Ecol
 , 
1993
, vol. 
174
 (pg. 
65
-
82
)
Winston
JE
Polypide morphology and feeding behavior in marine ectoprocts
Bull Mar Sci
 , 
1978
, vol. 
28
 (pg. 
1
-
31
)
Winston
JE
Larwood
GP
Abbott
MB
Current-related morphology and behaviour in some Pacific coast bryozoans
Advances in bryozoology
 , 
1979
New York
Academic Press Inc
(pg. 
247
-
68
)
Yoshioka
PM
Role of planktonic and benthic factors in the population dynamics of the bryozoan Membranipora membranacea
Ecology
 , 
1982
, vol. 
63
 (pg. 
457
-
68
)

Author notes

From the symposium “Going with the Flow: Ecomorphological Variation across Aquatic Flow Regimes” presented at the annual meeting of the Society for Integrative and Comparative Biology, January 2–6, 2008, at San Antonio, Texas.