Notes
GOLDMAN J. C. 1988. Spatial and temporal discontinuities of
biological processes in pelagic surface waters, p. 273-296.
1n J. Rothschild [ed.], Towards a theory on biological-physical interactions in the world ocean. Kluwer.
GRAF, G. 1989. Benthic-pelagic coupling in a deep-sea benthic
community. Nature 341: 437-439.
GUST, G., R. H. BYRNE, R. E. BERNSTEIN, P. R. BETZER, AND
W. BOWLES. 1992. Particle fluxes and moving fluids: Experience from synchronous trap collections in the Sargasso
Sea. Deep-Sea Res. 39: 1071-1083.
HONJO, S. 1978. Sedimentation of materials in the Sargasso
Sea at a 5367 m deep station. J. Mar. Res. 36: 469-492.
-.
1982. Seasonality and interaction of biogenic and lithogenic particulate flux at the Panama basin. Science 218:
883-884.
---,
S. MANGANINI, AND J. J. COLE. 1982. Sedimentation
of biogenic matter in the deep ocean. Deep-Sea Res. 29:
609-625.
-,
AND M. R. ROMAN. 1978. Marine copepod fecal pellets: Production, preservation and sedimentation. J. Mar.
Res. 36: 45-57.
LAMPITT, R. S., W. R. HILLER, AND P. G. CHALLENOR. 1993.
Seasonal and diel variation in the open ocean concentration
of marine aggregates. Nature 362: 737-739.
NEWTON-DOWNS, J., AND C. J. LORENZEN. 1985. Carbon:
pheopigment ratios of zooplankton fecal pellets as an index
of herbivorous feeding. Limnol Oceanogr. 30: 1024-l 036.
Llmnoi
Oceanogr , 4 l(6),
( 1996. bl the Amencan
1996. 1 359-l 366
Soclet) of Llmnolog?
and Oceanograph\
1359
RAMASWAMI, V.,R. R. NAIR,S.MANGANINI,
B. HAAKE, AND V.
ITTEKOTT. 1991. Lithogenic fluxes to the deep Arabian
Sea measured by sediment traps. Deep-Sea Res. 38: 169184.
SMALL, L. F., S. W. FOWLER, AND M. Y. UNLU. 1979. Sinking
rates of natural copepod feacal pellets. Mar. Biol. 51: 233241.
SMETACEK, V. S. 1985. Role of sinking in diatom life history
cycles: Ecological, evolutionary and geological significance.
Mar. Biol. 84: 239-25 1.
STROM, S. L. 1993. Production of pheopigments by marine
protozoa: Results of laboratory experiments analysed by
HPLC. Deep-Sea Res. 40: 57-80.
UTERM~HL, H. 1958. Zur Vervollkommnung
der quantitativen Phytoplankton-Methodik.
Mitt. Int. Ver. Theor. Angew.
Limnol. 9. 39 p.
URBAN, J. L., C. H. MCKENZIE, AND D. DEIBEL. 1993. Nanoplankton found in fecal pellets of macrozooplankton
in
coastal Newfoundland waters. Bot. Mar. 36: 267-28 1.
WEF-ER,G. 1989. Particle flux in the ocean: Effects of episodic
production, p. 139-l 54. Zn W. H. Berger et al. [eds.]. Productivity of the oceans: Present and past. Wiley.
Submitted: 11 January I995
-Accepted:7 Februaq? 1996
.4mended: 24 *May 1996
Inc
Does flow speed also have a direct effect on growth of
active suspension-feeders: An experimental
test on oysters
*dbstracf-The
direct effects of flow velocity on growth
of juvenile American oysters Crassostrca l,irginica (Gmelin) were tested without confounding by indirect effects of
vertical mixing by measuring growth inside pipes, varying
average flow over five levels (0, 0.5, 2.5, 4. and 7 cm SK’)
at each of two food concentrations. Growth increased with
food concentration and increased monotonically with flow
velocity over all flows tested for both food treatments. In
this experiment, food concentration and flow velocity acted
independently in their effects on growth, with a multiplicative term (flux) not making any significant contribution
beyond the independent, additive contributions
of food
and flow. Thus, the apparent positive response of growth
in this active suspension-feeder to external flux of food is
best interpreted as a coincidental by-product of the action
of two separate responses to the terms that comprise flux.
Benthic
suspension-feeders
harvest particulate food
materials from the surrounding water and thereby provide an important coupling between the pelagic and benthic environments, especially in shallow waters (Dame et
al. 1980; Officer et al. 1982; Peterson and Black 1987).
Although individual growth rates, production, and energy
transfers vary strongly with seasonally changing water
temperatures, metabolic rates. and reproductive
status
(Vahl 1980; Bayne and Newell 1983) growth of suspension-feeders over finer temporal and spatial scales seems
to be controlled by complex interactions among feeding
physiology, food availability,
and hydrodynamics
in the
region of the benthic boundary layer (e.g. Vogel 198 1;
Jorgensen 1983; Frechette et al. 1989).
Suspension-feeders can be divided into two basic feeding modes: passive and active. Passive suspension-feeders
intercept particles that advect past the organism and thus
are directly responsive to food flux- the product of concentration and flow speed (Patterson 1980; Muschenheim
1987); growth increases with increasing food concentration and flow speed (Muschenheim
1987; Sanford et al.
1994), although excessively high flow speeds threaten
damage to the food capture organs and can inhibit feeding
and growth (e.g. Okamura 1984). Active suspension-feeders, such as bivalve molluscs, pump water from the external environment to internal cavities thereby creating
their own internal feeding currents (Jorgensen 1983; Newell and Langdon 1995). Although it seems clear that growth
of active suspension-feeders should depend on food con-
1360
Notes
centration, a point that has been confirmed (e.g. Widdows
et al. 1979), it is not so obvious why growth of an active
suspension-feeder should be affected by flow speed in its
external environment (Eckman et al. 1989). Yet, several
previous studies have shown growth of active suspensionfeeders to respond to external flow velocity, although in
complex ways (Kirby-Smith
1972; Grizzle and Morin
1989; Grizzle et al. 1992).
Processes that might relate current velocity to growth
rate in active suspension-feeders can be categorized as
indirect or direct. Growth of active (and also passive)
suspension-feeders can be enhanced at higher flow speeds
because of enhancement of turbulent vertical mixing. In
faster flows, greater vertical mixing reduces the intensity
of vertical gradients in food concentration
created by
feeding of benthic suspension-feeders and minimizes the
resultant depletion of foods near the sea floor (Wildish
and Kristmanson
1979; Frechette et al. 1989; Skilleter
and Peterson 1994). This enhancement of vertical mixing
by flow increases represents an indirect effect of flow velocity on growth, acting through the mechanism of enhancing food concentrations around feeding animals. Increased bottom shear stress associated with higher flow
velocity also can influence growth indirectly by resuspending deposited materials and thus altering food quantity and quality (Bricelj and Malouf 1984; Emerson 1990;
Turner and Miller 1991).
Alternatively,
flow velocity might also act directly on
the feeding rate of active suspension-feeders, with the
most compelling mechanism involving the influence of
hydrostatic pressure generated by the external flow field
on the rate of movement (pumping velocity) of foodbearing water through the organism. This process is most
clearly suggested in studies of scallops, especially studies
demonstrating that the influence of external flow varies
with the orientation of the feeding animal to the direction
of flow. When the inhalant opening to the mantle cavity
is directed into the flow, growth rates of scallops are enhanced with increasing flows over a range of relatively
slow current velocities, whereas if the exhalant opening
is pointed into the flow, the unfavorable pressure gradient
inhibits feeding and reduces growth (Wildish et al. 1987;
Eckman et al. 1989; Wildish and Saulnier 1993). Despite
this evidence of direct effects of flow on feeding and growth
of scallops, general models of feeding by active suspension-feeders do not incorporate any direct effect of flow
speed on feeding rate (Frechette et al. 1989; Monismith
et al. 1990; O’Riordan et al. 1993), and Frechette et al.
(1994) have argued that the apparent relationship
of
growth in active suspension-feeders to flow and food flux
is illusory (an “epiphenomenon”),
driven only by the
indirect effects of flow on food concentration.
We designed our experiment as an explicit test of the
role of direct effects of flow and external flux of food
particles on growth of an active suspension-feeder. We
chose to test responses inside pipes with a Kirby-Smith
(1972) flow-tube apparatus. Our intent in using pipes was
not to mimic natural flow but rather to separate the direct
effects of flow from the indirect effects acting on food
concentrations. Such confounding of the effects of flow
speed and food concentration accompanies virtually any
field or laboratory experiment with an overlying water
column (e.g. Frechette et al. 1989). We selected the American oyster Crassostrea virginica (Gmelin) so as to extend
tests of the direct effects of flow on active suspensionfeeders beyond scallops and mussels. Furthermore, the
information collected in this study is of immediate value
to restoration and management of a species whose decline
has probably greatly affected the partitioning
of energy
between the pelagic and benthic realms and thereby the
ecology of entire estuaries (Newell 1988). Proper oyster
management is likely to include restoring and maintaining the now badly degraded reef habitat, which should
provide the topography to elevate the oysters above the
estuarine sea floor and thereby enhance flow velocities
and particle fluxes around them (Rothschild et al. 1994).
We designed and conducted an experiment measuring
growth of juvenile oysters inside pipes such that food
concentration (two levels) and flow velocity (five levels)
were varied in a factorial design. Juvenile oysters, provided by a shellfish hatchery to ensure genetic and historical identity among individuals
(10 weeks old; shell
width, 6-8 mm; shell height, 10-l 2 mm; total wet wt,
0.3-0.5 g), were exposed to current speeds of 0, 0.5, 2.5,
4, and 7 cm s-l for 45 d (1 September-l 5 October 1993)
in a Kirby-Smith
(1972) flow-tube apparatus virtually
identical to that used by Eckman et al. (1989). The flowtube apparatus consisted of 12 4-m-long PVC pipes (5.1cm i.d.) extending from the base of a rectangular head
box (0.75 x 0.75 x 2 m long). Water was pumped continuously from nearby Bogue Sound into the head boxes to
keep them overflowing and always full; water flowed out
through the pipes into a surrounding seawater pond, in
which the pipes were submerged at 3-cm depth. Two
replicates of each of the five flow velocities were produced
by fitting pipes with stoppers containing holes of different
sizes. A zero flow rate was produced by using stoppers
without holes. This zero flow treatment experienced some
sluggish, circulating flow within the pipes and the head
box, as indicated by both survival and some growth of
oysters. Average flow rate in pipes was calculated by dividing the measured volumetric discharge rate by the pipe
cross-sectional area. Flow speeds at the centerline of each
pipe where oysters were positioned were in theory twice
the mean flow rate calculated for each pipe (Vogel 198 1).
Pipes were cleaned of deposits weekly by removing the
stopper and allowing rapid flow for a brief time. Pipes
were shaded with opaque cloth to minimize algal growth
within them and reduce heat stress to oysters.
Twenty-four oysters were placed in each replicate pipe
by gluing animals individually
onto 1.7-mm-diameter
glass rods. Oysters were positioned at the centerline of
each pipe by fixing the glass rods to rubber disks and
inserting them into a hole in the pipe wall. The rubber
disks sat flush with the inside wall of the pipe. This design
(see Eckman et al. 1989) minimized induction of turbulence within pipes. The left valve of each oyster was glued
to the glass rod so that the umbo faced downward and
the anterior, inhalant side of the mantle faced upstream.
Oysters were spaced at lo-cm intervals with the first oys-
Notes
ter positioned 1.6 m downstream of the entrance of the
pipe.
To evaluate the direct influence of flow velocity under
differing food conditions and to test the influence of food
flux on growth by using different combinations of food
concentration and flow velocity, we used two head boxes
with different food concentrations-ambient
and reduced. Reduced food concentration was produced by filtering the inflowing water of one box through an 80-pm
mesh, thereby reducing mean chlorophyll a concentration
by 57% (2.36kO.02 vs. 1.02*0.01 pg Chl a liter-l; n =
2), measured by fluorometry (Parsons et al. 1984) as a
food quantity index at both the beginning and end of the
experiment (ANOVA, effect of filtration: Fl,,.,= 1.15 x 104,
P < 0.001; effect of time: F1,4 = 1.25 x 10-5, P >> 0.05;
interaction of filtration and time: F1,4= 1.25 x 10m5, P >
0.05). Two replicates of 10 food fluxes were thus created:
0, 1.2,5.9,9.4, and 16.5 pg Chl a cm-* s-l in the ambient
concentration treatment and 0, 0.5, 2.6, 4.1, and 7.2 pg
Chl a cm-* s-l in the reduced concentration treatment.
Filters were replaced every 8 h. Although filtration through
80 pm selectively eliminates chainlike and colonial phytoplankters, as well as blue-greens and large detrital particles, concentrations of POM (particulate organic matter)
(0.11 kO.03 g liter-l; n = 2) and PIM (particulate inorganic matter) (0.0 1 kO.005 g liter-l; n = 2) were the same
in both food concentration treatments and did not change
between the beginning and end of the experiment (ANOVA for POM, difference between food concentration
treatments: F1,4= 0.62; time: F1,4= 1.10; interaction: F1,4
= 0.00; all with P * 0.05; ANOVA for PIM, concentration: Fl,4 = 0.00; time: Fl,4 = 1.10; interaction: Fl,4 =
1.10; all with P > 0.05). Consequently, although filtration doubtless altered food quality as well as quantity,
the bulk of the particulate matter must have been <80
pm.
The response variable in this experiment was growth
of the oysters. Growth was measured for each individual
oyster by comparing starting and ending shell heights and
widths and total wet weights. Linear sizes were measured
to 0.1 mm by vernier calipers and wet weights were calculated on an electronic balance to 0.01 g after towel
drying the oysters. We report here only the response in
shell width for the 45-d period because all three response
variables revealed the same response patterns and growth
in shell width proved to be measured with least error.
Initial shell width did not vary significantly among treatments (ANOVA; F19,354= 1.02; P > 0.05), so there was
no need to use a size covariate in any analysis. To test
whether oysters within a pipe were exposed to the same
food conditions or whether, alternatively, foods were depleted by upstream oysters (cf. Kirby-Smith
1972; Wildish and Kristmanson 1985), we performed separate linear regressions of individual
oyster growth on position
(distance from the beginning of each pipe). Subsequently,
a two-way, crossed model 1 ANOVA with five levels of
current speed and two levels of food concentration was
conducted on untransformed growth data to test whether
average growth in shell width varied with average external
current speed, food concentration,
or their interaction.
1361
Cochran’s test detected no heterogeneity of variances at
cy = 0.05. Our design was pseudo-replicated at the level
of concentration because we had only one head tank for
each concentration,
but this feature does not influence
interpretation
of the test of flow velocity, which is the
primary focus of our study. Post hoc multiple comparisons were made with SNK tests. Finally, the growth results were further explored by multiple linear regression
to evaluate whether food flux (the product of flow speed
and food concentration) contributed significantly to explaining the growth responses after separate and independent effects of both flow speed and food concentration
were already taken into account.
The regressions of individual growth on position along
the pipe failed to detect a significant effect of position in
any of the 20 pipes (Fig. 1). In the pipe with the best fit,
position along the pipe could not explain even as much
as 10% of the variability
in growth among replicate individuals, and position, on average, explained only 2.3%
of this variation (r*-values given in Fig. 1). The plots of
individual growth along the pipes do not suggest reduced
growth with distance nor do they indicate a stronger effect
of distance at the lowest flow speeds, as would be predicted if depletion were occurring.
Both main factors, flow speed and food concentration,
exhibited highly significant effects (P < 0.001) on growth
in the two-factor ANOVA, but there was no significant
interaction between these factors (Table 1). For each food
concentration treatment, average growth of oysters increased monotonically
with average pipe flow speed over
the range tested (Fig. 2). The post hoc SNK tests at (x =
0.05 showed that growth was significantly greater at the
two highest average flow speeds, 4 and 7 cm s-l, than at
the two lowest flow speeds, 0 and 0.5 cm s- l. Growth at
2.5 cm s-l was also significantly greater than at 0 cm s-l.
The effect of reducing food concentrations by filtration
was to reduce growth of juvenile oysters at every flow
velocity (Fig. 2).
The multiple linear regression analyses revealed that
in a full model containing all four terms, food concentration, flow velocity, flow*, and flow x concentration,
only food concentration and flow velocity were significant
in partial F-tests (Table 2). Together, food concentration
and flow velocity accounted for 67% of the variance in
average growth among pipes and inclusion of the two
higher order terms explained only 6% more variance and
depressed the P-value relative to a reduced model containing only the two main factors (Table 2). We included
the quadratic (squared) term for flow in this full model
to account for the apparent asymptotic reduction in growth
at higher flows (Fig. 1) and the product of flow velocity
and food concentration so as to have a direct measure of
food flux in the model. A plot of average growth of juvenile oysters with changing food flux in this experiment
does reveal that growth increased with flux of food (as
must be true if growth increased with increases in both
component factors) such that a plateau appears to be
attained above m 10 pg Chl a cm-* s-l (Fig. 3). However,
inclusion of the direct measure of flux could not explain
significantly more variance in growth than is explained
Notes
1362
Reduced
2.
0 cm s-l
concentration
P= 0.31
Ambient
concentration
r>n - 0 cm s-l
/-2 = 0.07
P= 0.37
r2 = 0.06
0
10 4
2. ,0.5 cm s-l
. G
lo-~~'
0
.
l
P= 0.81
l .0
l
0
0.
r2 = 0.00
a --
0
.e
IIIIIIIIIIIIII1111111111
2. ,0.5 cm s-l
2.5 cm s-1
P= 0.71
r2 = 0.05
p= 0.42
r2 = 0.00
w
*O-4
cm4
160
200
P= 0.86
240
280
r2=
320
0.01
360
160
200
240
280
320
360
Distance from pipe opening (cm)
Fig. 1. Growth in shell width for juvenile oysters inside pipes as a function of position
along the pipe in duplicate pipes for each combination of two food concentrations and five
average flow-speed treatments. Lines drawn are the best-fit linear regressions; r2 represents
the proportion of variance in individual oyster growth explained by position along the pipe
and P is the statistical significance of the test of whether the slope equals zero.
by the independent, additive combination of food concentration and flow velocity.
Our primary motivation here was to answer two questions to help resolve a fundamental debate about active
suspension feeding. First, do observed relationships between growth in active suspension-feeders and either flow
velocity or external flux of food particles represent a direct
effect of velocity or of flux (e.g. Grizzle et al. 1992)? Al-
Notes
1363
Table 1. Results of a two-factor ANOVA testing whether
growth of juvenile oysters varied in response to food concentration, flow speed, and their interaction. Data are given in Fig.
1. ANOVA was conducted on untransformed data because variances were homogeneous (Cochran’s test, Ctitid = 0.62; C =
0.20; df = 10,l; P > 0.05).
1
I
=’
Source of variation
Food concn
Flow speed
Food concn x flow speed
Residual
df
MS
F-ratio
P
1
4
4
10
37.26
16.18
0.88
1.30
28.66
12.45
0.68
0.0003
0.001
0.62
tematively, does this relationship between growth and
flow velocity exist only because of indirect effects of flow
velocity acting on food concentration by enhancing vertical mixing and reducing near-bed depletion of foods (e.g.
Frechette et al. 1994)? There is little doubt that over some
range of concentrations from zero up to some rather high
concentration, growth and production of active suspension-feeders will generally increase with increasing food
concentrations. This positive relationship between food
concentration
and growth of active suspension-feeders
makes sense arguing from first principles and is supported
by observational and experimental evidence (e.g. Widdows et al. 1979). Beds of suspension-feeders clearly can
deplete food concentrations near the sea floor sufficiently
to reduce their own growth (Peterson 1982; Peterson and
Black 1987) and turbulent mixing induced by increased
flow speeds acts to break down the vertical gradient in
food concentration (Frechette et al. 1989; O’Riordan et
al. 1993; Skilleter and Peterson 1994). Thus, there is an
excellent basis for believing that indirect effects of flow
velocity are important contributors to the transfer of energy and materials from the water column to active and
passive suspension-feeders in the benthos.
The presence of potentially important indirect effects
of flow velocity on growth of active suspension-feeders,
however, does not imply the absence of additional direct
effects of flow velocity. In field conditions and even in
laboratory flumes with a proper vertical dimension to the
overlying water column, the indirect effect of flow on
vertical mixing interferes with detection of any possible
direct effect of flow on feeding. This confounding of direct
and indirect effects of flow on suspension-feeder growth
in nature and in flumes motivated us to establish our test
of oyster feeding response inside pipes, where despite
complex flow regimes steady flows can be maintained and
flow character is generally predictable. Furthermore, the
absence of a vertical dimension to the water column in
pipes prevents establishment of vertical concentration
gradients (Kirby-Smith
1972; Wildish and Kristmanson
1985; Eckman et al. 1989). Our intent was not to duplicate
natural boundary-layer flows, but rather to artificially isolate the direct effects of flow on feeding and growth from
the indirect effects that are typically confounded in nature
so that food depletion would not interfere with a test of
this one component of response to flow. Indeed, detailed
examination of growth of individual
oysters by location
8
T
2.5
Flow speed (cm s-1)
0
reduced
concentration
q
ambient
concentration
Fig. 2. Mean growth in shell width of juvenile oysters inside
pipes over 45 d as a function of average flow speed through the
pipes at each of two food concentrations (ambient at 2.36 and
reduced at 1.02 pg Chl a liter-l). Bars indicate 1 SE around the
means of 16-2 1 surviving oysters out of an initial 24 placed in
each pipe. A two-factor ANOVA (results in Table 1) demonstrated highly significant (P < 0.001) effects of both food concentration and flow speed with no significant interaction between the factors. Results of SNK contrasts of means are indicated by lines connecting flow treatments that did not differ
at (Y = 0.05.
along the pipes revealed a convincing lack of any horizontal food depletion in our experimental system (Fig.
1). Consequently, our test of the effects of flow velocity
under two separate food concentrations appears not to
confound indirect effects with direct effects of flow.
Analysis of the growth of juvenile oysters inside pipes
showed an unambiguous, direct effect of flow velocity,
which enhanced growth monotonically over the full range
of flows tested under both sets of food conditions (Fig.
2). This result is consistent with those of Wildish and
Kristmanson (1985) for blue mussels and of Wildish et
al. (1987), Eckman et al. (1989), and Wildish and Saulnier
(1993) for two species of scallops. Among these studies,
direct effects of flow were detectable, although the pattern
of response of growth to increasing flow differed. Here,
we tested responses to rather slow average pipe flows up
to only 7 cm s-l, which represent realistic conditions for
oysters in nature (Lenihan unpubl. data), but we did not
explore the potential inhibition
of feeding at high flows.
Our study thus extends previous tests of the direct effects
of flow velocity from mussels and scallops to oysters.
More importantly, the factorial design and analysis of the
tests of whether flow and flux have a direct effect on
growth of an active suspension-feeder permit enhanced
understanding of the first-principles
processes affecting
feeding and growth of active suspension-feeders.
Notes
1364
Table 2. Results of multiple regressions performed on oyster growth data. Model 1 is the
full model including all four terms. Model 2 is the reduced model presenting the regression
with the best fit; all factors in model 1 with insignificant effects on growth were removed.
Terms in the upper portion of each model are for the complete model. Terms in the lower
portion are for the partial F-tests performed on the separate factors of the model.
Source
df
MS
F-ratio
Partial
F-ratio
P
r2
Model 1
Width = 0.96 + 2.52(food concn) + 1.72(flow) - O.lO(flo~)~ - O.lB(food concn x flow)
25.42
10.20
0.0003
0.73
Regression
4
2.49
Residual
15
Total
19
Food concn
Flow
Flow2
Flow x food concn
10.30
8.36
2.47
0.87
Model 2
Width = 2.41 + 1.98(food concn) + 0.67(flow)
46.68
17.36
2
2.69
17
19
Regression
Residual
Total
13.03
21.69
Food concn
Flow
We did not investigate the specific mechanistic basis
by which flow velocity affected oyster growth in our experiments, but we can propose two possible hypotheses
to explain the response. Oysters are active suspensionfeeders that rely on internally generated flow to remove
particulate foods from the water (Newell and Langdon
1995). Oysters utilize steep shear gradients between internal water currents produced by cilia located along their
gill lamellae to position particles close to other cilia, which
then capture and transport them to the mouth. It seems
g- 121
-5
f
10 -1
E
Flow weeds
0
o reduced concentration
A ambient concentration
ABCDE-
2
10
4
6
8
0 cm
0.5 cm
2.5 cm
4.0 cm
7.0 cm
s-1
s-1
s-1
s-1
s-1
12
14
16
18
Food flux (pg Chl a cm-* s-l)
Fig. 3. Mean growth in shell width of juvenile oysters inside
pipes plotted as a function of mean food (Chl a) flux through
the pipes. Error bars represent f 1 SE around the mean of two
rep&&e pipes each containing 16-21 surviving oysters. Although this plot reveals an apparent relationship between growth
of this active suspension-feeder and external -flux of food, the
relationship is misleading because analyses (Tables 1, 2) imply
that it derives from a coincidental consequence of separate,
independent responses to the two components of flux-food
concentration and flow speed.
0.006
0.01
0.14
0.37
0.000 1
0.67
0.002
0.0002
reasonable to hypothesize that application of the low flows
used in our study could serve as a passive enhancement
of the oyster’s ability to move water through the mantle
cavity and thereby improve the efficiency of its own ciliary pump (cf. Vogel 198 1). This explanation is consistent
with the observation that direct effects of flow velocity
on scallop feeding and growth vary with orientation of
the inhalant opening to the flow direction, with growth
enhanced when flow strikes the inhalant opening and depressed when it strikes the exhalant opening (Wildish et
al. 1987; Eckman et al. 1989). Alternatively,
increasing
flow velocity may have acted to enhance oyster growth
because of complex fine-scale hydrodynamics.
It is conceivable that at very low flows, individual oysters create
localized envelopes of depleted water, thereby depressing
growth (a process also proposed by Wildish and Saulnier
1993). Increased flow could inhibit formation of these
depletion envelopes.
Because oyster growth increased with both increasing
food concentration and also with increasing flow velocity
in our study, there necessarily exists an apparent relationship between growth and food flux (Fig. 3) because
flux is the product of concentration and flow velocity.
This relationship appears from our analyses to be a consequence of the independent influences of food concentration and flow velocity, each acting in different ways.
We make this argument based on results of both the
ANOVA and the multiple regressions. The ANOVA revealed that the two main factors (food concentration and
flow velocity) each had positive effects on oyster growth
over the ranges tested, but the absence of any significant
interaction between the main effects shows that these contributions were independent (Table 1). If growth were
Notes
responding directly to changes in flux, the main effects
should interact, given that flux is the product not the sum
of these two variables. This conclusion that the variables
contribute separately and independently is further supported by the regression results, showing that the best
model to explain the variance in average growth of oysters
among treatments includes only the two separate main
effects and not the explicit flux term (Table 2). Consequently, we conclude that the apparent relationship between oyster growth and food flux (Fig. 3) is fundamentally misleading and coincidental.
The danger in perpetuating the notion that growth of
active suspension-feeders responds to flux (e.g. Grizzle
and Lutz 1989) as opposed to responding separately to
its individual components is twofold. First, because this
relationship is coincidental, it is likely to fall apart as the
range in variables is increased or as another species is
examined. This contention that the monotonically
increasing relationship between flux of food and growth in
active suspension-feeders is limited to a narrow range of
conditions is especially true for increasing flow velocities,
which seem unlikely to cause further direct enhancement
of feeding and are likely to inhibit feeding when above
- 10 cm s-l (Grizzle et al. 1992). Second, relating growth
of active suspension-feeders to food flux instead of the
two apparently causative factors inhibits understanding
of the underlying mechanisms, which are doubtless based
on feeding physiology and fine-scale hydrodynamics.
The development of a mechanistic understanding of
the role of flow velocity in affecting growth of oysters and
other active suspension-feeders has practical significance
to habitat restoration, conservation, and management in
estuaries. Oysters have been greatly depleted in the estuaries of the mid-Atlantic
coast, where they previously
transferred a large percentage of pelagic primary production to the benthos (Newell 1988; Baird and Ulanowicz
1989). This loss of the ecological vehicle for energy transfer between the water column and the benthos has been
blamed for a fundamental shift in the food webs of the
Chesapeake Bay ecosystem, favoring the development of
pelagic consumers including the noxious sea nettle (Newell 1988). Removal of a dominant suspension-feeding bivalve may conceivably open an ecological niche that could
promote successful establishment of exotic ecological analogs. The ongoing alteration of the San Francisco Bay
ecosystem by expansion and cascading effects of the introduced suspension-feeding
bivalve, Potamocorbula
(Nichols et al. 1990), may be a partial consequence of the
rarity of the native west coast oyster. The success of invasions of the asiatic clam Corbicula and the zebra mussel
Dreissena in freshwater aquatic systems may also relate
to declines in native suspension-feeding mussels within
those water bodies.
Along the Atlantic and Gulf Coasts, where estuarine
fisheries production is such a valuable ecosystem service,
protection and restoration of oyster reef habitat and dependent populations is a priority of environmental
management. Knowledge of the complete relationship of oyster growth and production to flow velocity will be of
practical value. Oyster reefs project above the sea floor,
1365
thereby influencing the local flow environment and potential growth and production of oysters (Rothschild et
al. 1994). Our results in combination with studies of the
larger scale hydrodynamic effects of altering reef size and
shape will help direct reef restoration and management
based upon a mechanistic understanding of the consequences of flow alteration.
Hunter 23.Lenihan
Charles H. Peterson
Jennifer M. Allen
University of North Carolina
Institute of Marine Sciences
Morehead City 28557
at Chapel Hill
References
BAIRD, D., AND R. E. ULANOWICZ. 1989. The seasonal dynamics of the Chesapeake Bay ecosystem. Ecol. Monogr.
59: 329-364.
BAYNE, B. L., AND R. C. NEWELL. 1983. Physiological energetics ofmarine molluscs, p. 625-639. In A. S. M. Saleuddin
and K. M. Wilbur [eds.], The Molluscs, V. 4. Academic.
BRKELJ, V. M., AND R. E. MALOUF. 1984. Influence of algal
and suspended sediment concentrations on the feeding
physiology of the hard clam, Mercenaria mercenaria.Mar.
Biol. 84: 155-165.
DAME, R. F., R. ZINGMARK, H. STEVENSON, AND D. NELSON.
1980. Filter feeder coupling between the estuarine water
column and benthic subsystem, p. 521-526. In V. S. Kennedy [ed.], Estuarine perspectives. Academic.
E~KMAN, J. E., C. H. PETERSON, AND J. A. CAHALAN. 1989.
Effects of flow speed, turbulence, and orientation on growth
of juvenile bay scallops, Argopectenirradians concentricus
(Say). J. Exp. Mar. Biol. Ecol. 132: 123-140.
EMERSON, C. W. 1990. Influence of sediment disturbance and
water flow on the growth of the soft-shell clam, Mya arenaria L. Can. J. Fish. Aquat. Sci. 47: 1655-1663.
F&~HETTE, M., C. A. BUTMAN, AND 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.
-,
D. LEFAIVRE,AND~. A.BUTMAN. 1994. Bivalve feeding and the benthic boundary layer, p. 325-369. In Bivalve
filter feeders. NATO AS1 Ser. 5. G 33. Springer.
GRIZZLE, R. E., R. LANGDON, AND W. H. HOWELL. 1992.
Growth responses of suspension-feeding bivalve molluscs
to changes in water flow: Differences between siphonate
and non-siphonate taxa. J. Exp. Mar. Biol. Ecol. 162: 2 13228.
AND R. A. LUTZ. 1989. A statistical model relating
hdrizontal seston fluxes and bottom sediment charactexis-
Acknowledgments
We thank Adrian Whichard for help in modifying the pipeflow apparatus designed by Braxton Tesh. We thank Hal Summerson, David Colby, Eric Holm, and Greg Skilleter for advice
and help with statistical analyses. We thank Fiorenza Micheli
and Mark Luckenbach for several helpful theoretical considerations and Fiorenza Micheli for help with the manuscript.
This study was supported by a research and shellfish restoration grant from the North Carolina General Assembly through
CIFO (Cooperative Institute of Fisheries Oceanography).
1366
Notes
tics to growth of Mercenaria mercenaria. Mar. Biol. 102:
95-105.
-,
AND R. E. MORIN. 1989. Effect of tidal currents, seston,
and bottom sediments on growth of Mercenaria mercenaria: Results of a field experiment. Mar. Biol. 102: 85-93.
JDRGENSEN, C. B. 1983. Fluid mechanical aspects of suspension feeding. Mar. Ecol. Prog. Ser. 11: 89-103.
KIRBY-SMITH, W. W. 1972. Growth of the bay scallop: The
influence of experimental water currents. J. Exp. Mar. Biol.
Ecol. 8: 7-18.
MONISMITH, S. G., J. R. KOSEFF, J. K. THOMPSON, C. A. O’RIORDAN, AND H. M. NEPF. 1990. A study of model bivalve
siphonal currents. Limnol. Oceanogr. 35: 680-696.
MUSCHJZNHEIM,D. K. 1987. The dynamics of near-bed seston
flux and suspension-feeding benthos. J. Mar. Res. 45: 473496.
NEWELL, R. I. E. 1988. Ecological changes in Chesapeake Bay:
Are they the result of overharvesting the American oyster,
p. 536-546. In Understanding
the estuary: Advances in
Chesapeake Bay research. Chesapeake Res. Consort. Publ.
129. CBP/TRS 24/88.
AND C. J. LANGDON. 1995. Mechanisms and physiol;gy of larval and adult feeding, p. 109-139. In A. Eble
et al. [eds.], The eastern oyster, Crassostrea virginica. Maryland Sea Grant Publ.
NICHOLS, F. H., J. K. THOMPSON, AND L. E. SCHEMEL. 1990.
Remarkable invasion of San Francisco Bay (California,
USA) by the Asian clam Potamocorbula amurensis. 2. Displacement of a former community. Mar. Ecol. Prog. Ser.
66: 95-101.
OFFICER, C. B., T. J. SMAYDA, AND R. MANN. 1982. Benthic
filter feeding: A natural eutrophication
control. Mar. Ecol.
Prog. Ser. 9: 203-2 10.
OKAMURA, B. 1984. The effects of ambient flow velocity, colony size, and upstream colonies on the feeding success of
Bryozoa. 1. Conopeum stolonifera Ryland, an arborescent
species. J. Exp. Mar. Biol. Ecol. 83: 179-193.
O’RIORDAN, C. A., S. G. MONISMITH, AND J. R. KOSEFF. 1993.
A study of concentration boundary-layer
formation over a
bed of model bivalves. Limnol. Oceanogr. 38: 17 12-l 729.
PARSONS,T. R., Y. MAITA, AND C. M. LALLI. 1984. A manual
of chemical and biological methods for seawater analysis.
Pergamon.
PATTERSON, M. R. 1980. Hydromechanical
adaptations in AZcyonium siderium (Octocorallia),
p. 183-200. In D. J.
Schneck [ed.], Biofluid mechanics. Plenum.
PETERSON,C. H. 1982. The importance of predation and intraand interspecific competition in the population biology of
two infaunal suspension-feeding bivalves, Protothaca staminea and Chione undatella. Ecol. Monogr. 52: 437475.
AND R. BLACK. 198 7. Resource depletion by active
suipension feeders on tidal flats: Influence of local density
and tidal elevation. Limnol. Oceanogr. 32: 143-166.
ROTHSCHILD, B. J., J. S. AULT, P. GOLJLLETQUER,AND M. HERAL.
1994. Decline in Chesapeake Bay oyster population: A
century of habitat destruction and overfishing. Mar. Ecol.
Prog. Ser. 111: 29-39.
SANFORD, E., D. BERMUDEZ, M. D. BERTNESS,AND S. D. GAINES.
1994. Flow, food supply, and acorn barnacle population
dynamics. Mar. Ecol. Prog. Ser. 104: 49-62.
SKILLETER, G. A., AND C. H. PETERSON. 1994. Control of foraging behavior of individuals within an ecosystem context:
The clam Macoma balthica and interactions between competition and siphon cropping. Oecologia 100: 268-278.
TURNER, E. J., AND D. C. MILLER. 199 1. Behavior and growth
of Mercenaria mercenaria during simulated storm events.
Mar. Biol. 111: 55-64.
VAHL, 0. 1980. Seasonal variations in seston and the growth
rate of the Iceland scallop Chlamys islandica (0-F. Miiller)
from Balsfjord, 70” N. J. Exp. Mar. Biol. Ecol. 48: 195204.
VOGEL, S. 198 1. Life in moving fluids. Princeton.
WIDDOWS, J., P. FIETH, AND C. M. WORRALL. 1979. Relationships between seston, available food, and feeding activity in the common mussel Mytilus edulis. Mar. Biol. 50:
195-207.
WILDISH, D. J., AND D. D. KRISTMANSON. 1979. Tidal energy
and sublittoral macrobenthic animals in estuaries. J. Fish.
Res. Bd. Can. 36: 1197-1206.
-,AND. 1985. Control of suspension feeding bivalve production by current speed. Helgol. Meersunters.
39: 237-243.
AND OTHERS. 1987. Giant scallop feeding and growth
reiponses to flow. J. Exp. Mar. Biol. Ecol. 113: 207-220.
-,
AND A. M. SAULNIER. 1993. Hydrodynamic
control
of filtration in Placopecten magellanicus. J. Exp. Mar. Biol.
Ecol. 174: 65-82.
Submitted: 27 June 1995
Accepted: 31 October 1995
Amended: 2 June 1996