Journal of Plankton Research Vol.22 no.9 pp.1779–1800, 2000
Transport of molluscan larvae through a shallow estuary
G.Curtis Roegner
Department of Oceanography, Dalhousie University, Halifax, Nova Scotia,
Canada B3H 4J1
Present address: School of Fisheries, Box 355020, University of Washington,
Seattle, WA 98195, USA
Abstract. Dispersal of invertebrate larvae is determined by larval swimming behavior, the length of
planktonic development and the hydrodynamic regime. Larvae of estuarine invertebrates must
refrain from export or invade an estuary after development in the ocean. This study investigates
retention patterns of estuarine molluscs by measuring time series of larval abundance in relation to
hydrodynamic processes. Previous investigations of larval dynamics have generally focused on larger
estuarine systems that are often stratified and have relatively long hydraulic residence times. The
estuary studied in this investigation supports dense populations of infaunal clams yet has a water
depth to tidal amplitude ratio near unity. To access processes affecting larval retention, the circulation
patterns of the estuary were measured with time series of salinity, temperature, pressure and horizontal velocity. Transport rates of larvae between ocean and estuary, and within the estuary proper,
were calculated from velocity and larval concentration time series. The daily residence time of the
estuary was determined for the summer spawning period. The results demonstrate that molluscan
larvae were routinely transported between the estuary and nearshore zone in tidal flows. Based on
the magnitude of the horizontal current velocities, passive transport of larvae predominates during
most of the tidal cycle in the estuary. Residence time calculations suggest that the ability of larvae to
remain in the estuary through larval development is unlikely, and there was no evidence of selective
retention of mature bivalve larvae in the estuary. Rather, larvae are exported rapidly from the estuary
and undergo development in the coastal ocean. Mesoscale physical processes in the coastal ocean
probably control variation in the delivery of larvae back to estuarine systems. Recruitment to this and
similar estuaries must therefore be dependent on invasion.
Introduction
For benthic invertebrates that reproduce via planktonic larvae, variation in the
delivery of larvae to benthic habitats is a fundamental determinant of recruitment
rates and population structure (Thorson, 1966; Connell, 1985; Young, 1987;
Rodriguez et al., 1993; Ólafsson et al., 1994). Dispersal of meroplankton is dependent largely upon the interaction of larval swimming behavior, the length of the
planktonic development period and the prevailing hydrodynamic regime. Failure
to return to a suitable site results in death or, if settlement occurs too far from
neighbors for successful reproduction, in a loss from the gene pool. Estuarydependent organisms are constrained by further complexity; larvae must either
refrain from export to the sea or re-invade the estuarine habitat after development in the coastal ocean (Weinstein, 1988).
The veliger larva is the primary dispersal stage for most estuarine bivalve and
gastropod molluscs. Mollusc veligers are relatively poor swimmers, with net horizontal swimming velocities in the order of 10–4 m s–1 (Chia et al., 1984). They are
incapable of significant directed motion in the horizontal plane since horizontal
current velocities in most environments greatly exceed this value. Veligers are
free to maneuver in the horizontal plane only during low flow periods or in the
© Oxford University Press 2000
1779
G.Curtis Roegner
benthic boundary layer, where friction reduces the mean flow velocity. In
contrast, mollusc larvae are more proficient at vertical swimming, and it is
hypothesized that cued vertical migration may allow for retention within estuarine systems (Pritchard, 1950; Bousfield, 1955; Hill, 1995). Vertical current velocities are generally low compared with vertical swimming velocities, and mollusc
larvae are certainly capable of traversing significant vertical distances during a
tidal period (Mileikovsky, 1973; Mann, 1985). However, they are always transported horizontally near the velocity of the ambient flow.
Many estuarine systems are relatively shallow (<5 m) and are well flushed by
tidal action and/or freshwater discharge (Geyer and Signell, 1992). How are
populations of benthic molluscs maintained in such energetic estuarine systems
where there are often weak vertical gradients of physical variables and a net
seaward flow of water? If the recruitment dynamics are controlled by the delivery of larvae to benthic habitats [rather than postlarval transport (Roegner,
1996)], the two contrasting scenarios for larval dispersal are development within
the natal system (retention) versus import to an estuary after development in the
coastal ocean (invasion). Without cued behavioral responses, export from the
estuarine system is likely when development time exceeds hydraulic residence
time. Simple scaling comparisons indicate that retention of larvae is enhanced by
fast development times and in larger estuarine systems (Ketchum, 1954; Platt et
al., 1972). Intuitively, dispersal and loss from the parental stock is likely in small,
highly flushed systems compared with larger estuaries with long residence times.
However, established populations of molluscs can be extremely dense in wellflushed systems.
The objective of the present study was to examine the time series of abundance of molluscan larvae in a tidally-dominated estuarine system in relation to
physical oceanographic parameters. The study system, the Eel River estuary,
Nova Scotia, Canada, typifies many of the small, narrow and shallow estuaries
of large tidal range to depth ratio found in this region. The Eel River supports
dense populations of the infaunal clams Mya arenaria L. and Macoma balthica
L., and also the epifaunal blue mussel Mytilus edulis L. (Roegner, 1996), each of
which reproduces via planktonic larvae. The results demonstrate a strongly
advective system where molluscan larvae were passively transported in tidal
flows. There were sizable exchanges of larvae between estuary and ocean, and
the circulation patterns and residence time calculations indicated little likelihood for autochthonous larval development. Bivalve populations within this and
similar estuarine systems probably depend on invasions after development in the
coastal ocean.
Method
Study site
The Eel River estuary is located on the Atlantic coast of Nova Scotia, Canada
(44°38.5N, 63°23.5E), and forms part of the larger Lawrencetown River estuary
(Figure 1). Exchange between coastal water and estuary occurs exclusively
through a tidal inlet (TI) which enters the Lawrencetown River ~0.6 km from the
1780
Molluscan larvae transport through a shallow estuary
Fig. 1. Study site. (a) Location of Eel River, Nova Scotia; (b) Eel River estuary. Stations TI, Tidal
Inlet; BS, Benthic Site; ME, Middle Estuary; WM, position of culvert. Shaded areas denote intertidal
sandflats.
Atlantic Ocean. The tidal inlet is ~200 m long, with a cross-section composed of
a 22 m wide main channel (maximum depth 2.6 m) bounded on the north by a
26 m wide intertidal sandflat. The relatively straight geometry of the inlet forces
bidirectional flows past the sampling site. High turbulence, standing waves and
mesoscale eddies were often observed, indicating the well-mixed nature of the
inlet water.
The Eel River estuary consists of three main sections: two basins drained by
the Eel River (Figure 1). The Eel River is a 1.025 km long channel with a surface
area of 2.4 105 m2. The seaward boundary is the tidal inlet. Depths at low tide
vary from intertidal to ~1.0 m, and the tidal amplitude ranges from 0.4 to 1.3 m.
The seaward end of the Eel River consists of flood-delta deposits which grade
from high intertidal islands dominated by cordgrass, Spartina spp., through intertidal sandflats, to subtidal channels of sand and eelgrass, Zostera marina (L).
1781
G.Curtis Roegner
Populations of the bivalves Mya, Macoma and Gemma gemma Totten exceeded
combined densities of 105 ind. m–2 in the sandflat and channel environments.
Landward of the flood-delta deposits, the mainly subtidal Eel River is characterized by silty sediments dominated by Zostera, where the density of bivalves was
reduced.
The second section is a shallow peripheral basin called Back Marsh (surface
area of approximately 0.7 105 m2), which drained into Eel River (Figure 1).
Back Marsh is composed of a complex of anastomosing tidal creeks with adjacent
intertidal sandflats in the seaward end, and shallow subtidal Zostera beds in the
landward end. One creek–sandflat area (Benthic Site, BS) is the site of a 3 year
mollusc recruitment study (Roegner, 1996). The tidal creek channel at Benthic
Site is 15 m wide and 100 m long, and has a low tide depth of 0.4–0.6 m at the
point of measurement.
The final section of the Eel River system is the 9.9 105 m2 West Marsh (WM)
that connects to the Eel River by a narrow, 1.5 m wide man-made culvert (Figure
1b). The culvert greatly restricts the exchange of water between the two sections.
However, freshwater input into the estuary is primarily to West Marsh, and the
basin acted as a freshwater and algal ‘reservoir’ (Roegner, 1998). The remainder
of this paper concerns the larval dynamics of the lower estuary from the tidal inlet
to the connecting culvert at West Marsh.
Data collection
The hydrodynamic patterns of the Eel River estuary were investigated over
several spatial and temporal scales relevant to larval mollusc transport. Time
series of pressure and temperature were measured on a seasonal time scale and
used to calculate the hydraulic residence time of the estuary. Time series of salinity, temperature, pressure and algal pigment concentration were made on the
estuarine spatial scale (among stations within the estuary) over 5–10 h periods;
these latter data are published separately (Roegner, 1998). Finally, meroplankton concentration, current velocity, salinity, temperature and pressure time series
were collected over two to three tidal cycles (24–36 h) at the Tidal Inlet station,
and over 5–7 h periods at the Benthic Site station during 1993 and 1994. These
investigations allowed the long-term flushing of the estuary to be compared with
variation of larval concentrations over tidal time scales.
Volume transport
Current velocity measurements made at Tidal Inlet and Benthic Site were used
to calculate volume transport between the estuary and nearshore zone and within
the estuarine creek system, respectively (Figure 1). At Tidal Inlet, measurements
were acquired on four dates spanning the expected periods of bivalve spawning.
Horizontal water velocity (U) was measured with Marsh–McBirney electromagnetic current meters secured on an aluminum pole at positions 0.15 and 0.50 m
above the bottom. The mean and standard deviation were calculated for 10 min
measurement intervals and logged. Pressure and temperature were recorded at
1782
Molluscan larvae transport through a shallow estuary
30 min intervals, and salinity values were measured periodically with an electronic conductivity–temperature meter. At Benthic Site, the current meters were
deployed separately 0.15 m above the bottom in accordance with the reduced
depths at low tide. At both sites, the cross-sectional area of the channel was
measured by standard surveying techniques.
Volume transport of water Q (m3 s–1) was calculated with velocity and tidal
height data in conjunction with channel topography (Hume and Bell, 1993).
Volume transport through the cross-section was determined at each sampling
interval t (s) by multiplying the depth-averaged horizontal velocity U (m s–1) by
the cross-sectional channel area A (m2),
Q = U A.
Channel area varied as a function of the tidal height, h. The relation was determined using least squares regression at both Tidal Inlet (r2 = 0.998, P < 0.001, n
= 15) and Benthic Site (r2 = 0.924, P < 0.001, n = 8) stations. The calculations of
Q assume that the vertically-averaged velocity profiles were horizontally homogeneous [i.e. dU / dy = 0 where y is the cross-channel direction (Hume and Bell,
1993)]. Velocity and transport in this study are positive in the landward direction
(import) and negative in the seaward direction (export).
The total volume exchanged during a tidal period, the tidal prism (m3), was
estimated by numerical integration for ebb and flood periods as
= 0T Q dt
where T is the duration in seconds of the ebb or flood tide (Fisher et al., 1979).
The tidal prisms can be used to compare the steady state volume balance of the
estuary as
E = F + VFW
where E is the ebb tidal prism, F is the flood tidal prism and VFW is the volume
of freshwater, all in units of m3 (Fisher et al., 1979; Hume and Bell, 1993).
Plankton transport
Plankton samples were collected with a submersible, battery-powered sump pump
which had an intake rate of ~0.25 l s–1. The intake hose was positioned 0.15 m
above the channel bottom and arranged with openings parallel with the bidirectional channel flow. Collections were made at 1–3 h intervals, and abundance
was estimated with three replicate 15 l water samples that were filtered through
80 µm mesh screens and preserved in a buffered formalin and Rose Bengal solution. Larvae were sorted under a dissection microscope to major taxonomic group.
Shell height of bivalve veligers was measured and the larvae were grossly categorized as immature (<200 µm) or competent for settlement (>200 µm).
The instantaneous tidal transport of mollusc larvae, QL (ind. s–1 s), was
1783
G.Curtis Roegner
calculated for each sample by multiplying mean larval concentration C (ind. m–3)
by the corresponding instantaneous volume transport Q,
QL = C Q,
where it is assumed that dU/dy = dC/dz = dC/dy = 0. The results are presented as
time series. Total transport during a tidal period, N (ind.), was estimated for
bivalves and gastropods by
N = 0T QL dt
where T is as above. Values of N were found by digitizing the area under plots of
spline-interpolated mean QL by elapsed time. For sample occasions when the
current meter was unavailable or malfunctioned, plankton concentrations are
plotted as times series.
Hydraulic residence times
Pressure and temperature records were collected every 30 min at Benthic Site
during summer 1994 (6 June to 30 August), which encompassed the expected
spawning period of the estuarine bivalves (Roegner, 1996). The pressure readings were used to calculate the tidally-driven exchange of water in the estuary by
use of the tidal exchange ratio, r (Ketchum, 1954). This was determined for each
tide as
rt = 1– (ht / zt–1),
where ht is the low tide water level at time t, and zt–1 is the previous high tide level.
The exchange ratio is the proportion of water exported from the estuary per tidal
cycle. Hydraulic residence times (TR, d) were calculated as
TR = Ttide / rt
where Ttide (d) is the time between successive low tides (which were unequal).
This technique was expected to give a lower boundary on residence times (Geyer
and Signell, 1992).
Results
Volume transport at Tidal Inlet
At the tidal inlet, time series of salinity, temperature and velocity varied with tidal
stage and exhibited consistent trends between sample dates. The time series for
volume transport closely followed that of velocity but was modified by low crosssectional channel areas during ebb. The following description for 27 June 1994
illustrated in Figure 2 summarizes the observed patterns.
Tidal flushing typically resulted in extensive exchange of water (Figure 2a).
1784
Molluscan larvae transport through a shallow estuary
Fig. 2. Measured and derived physical variables at the tidal inlet. (a) Channel cross-sectional area as
a function of tidal height. The cross-section was oriented on a north-south direction where 0 m is
north. The arrow shows the position of the instrument mooring. (b) Time series of water level ()
and velocity (U, O). (c) Time series of salinity () and temperature (Temp, ). (d) Time series of
volume transport (Q). F, flood tidal prism; E, ebb tidal prism. For velocity and volume transport,
positive values indicate import into the estuary and negative values denote export.
Flood currents lagged 1–2 h past the beginning of rising water until equalization
of pressure gradient forces between rising flood water and ebbing estuarine water
(Figure 2b). With increasing tidal height, the ebb velocity steadily declined to 0,
with a slack water period of <15 min. The flooding coastal water entered the
estuary as a pronounced tidal front, denoted by rapid changes in the salinity and
temperature traces, which propagated in excess of ~0.1 m s–1 and completely
mixed with and/or displaced the ebbing estuarine water (Figure 2c). After the
passage of the flood tide front, the flood velocity steadily increased to a maximum
of 0.5–0.6 m s–1. Total flood period lasted only 3.5–5 h, and peak discharge coincided with maximum velocities. Flood tidal prisms F ranged from 0.46 to
1.43 105 m3 (Table I, Figure 2d).
At the end of flood, current velocity steadily declined to 0 and again reversed
1785
G.Curtis Roegner
Table I. Transport of molluscan larvae per tidal prism for each of three sample dates in 1994. Bivalve
larvae are divided by size into immature (<200 µm) and competent for settlement (>200 µm) groups.
∑ designates net transport over the tidal cycles shown, with positive numbers indicating import and
negative export from the estuary. Estimated abundance of the two size factions does not sum to total
abundance due to variations in the spline-interpolated areas
Date
27 June
21 July
17 Aug
Tide
Flood1
Ebb1
Flood2
Ebb2
∑
Flood1
Ebb1
Flood2
Ebb2
∑
Flood1
Ebb1
Flood2
Ebb2
∑
Tidal prism
(105 m3)
Gastropods
(109 ind.)
Bivalves (107 ind.)
<200 µm
>200 µm
Total
1.16
2.64
1.57
1.44
–1.36
1.18
2.78
0.87
1.68
–2.41
0.94
2.44
0.53
1.75
–2.73
0.15
–0.19
0.23
–0.05
0.14
0.15
–2.05
0.11
–1.03
–2.82
0.04
–1.64
0.11
–0.24
–1.74
3.22
–2.87
5.77
–0.66
5.46
8.45
–2.57
1.85
–1.95
5.78
1.51
–0.72
0.17
–0.25
0.71
5.43
–3.44
6.51
–0.74
7.74
8.88
–3.11
1.89
–1.94
5.83
1.78
–0.94
0.28
–0.43
0.71
2.21
–0.57
1.28
–0.15
2.77
0.71
–0.86
<104
<104
–0.15
0.36
–0.11
0.02
–0.18
0.09
with only a short period of slack water (Figure 2b). The velocity of the ebbing
currents maximized 1–2 h past flood and remained relatively constant at 0.6–0.7
m s–1 for the next 4–6 h. However, discharge peaked near the beginning of ebb
and declined with decreasing cross-sectional channel area (Figure 2d). The ebb
period was generally 1.5 times longer than the flood, and total discharge during
ebb was 1.5–2 times the flood tidal prism (Table I; Figure 2d). During both ebb
and flood, water in the channel was well mixed.
The time series of temperature and salinity were reliable indicators of the
origin of the water masses moving past the sample station. Temperature during
ebb generally remained constant and salinity generally decreased until the
passage of the flood tidal front, when there was a rapid rise in salinity and drop
in temperature to coastal values (Figure 2c). As the front passed, salinity fluctuations could exceed 20 psu h–1, with concurrent temperature declines >10°C
(Figure 2c). Throughout the remainder of the flood period, salinity and temperature values remained relatively constant. During the initial period of ebb, corresponding to the period of high discharge, the salinity trace remained at coastal
values, indicating the exit of undiluted sea water from the previous flood tide.
Subsequently, salinity declined and temperature increased as water of mixed
salinity flushed from the estuary. The interface between the ocean and estuarine
water masses constituted an estuarine front.
These data indicate the estuary was well flushed, but water was not simply
sloshing back and forth between estuary and nearshore. The salinity signal shows
the estuarine water was not re-imported on the following flood tide. Instead, a
new parcel of nearshore water was entrained into the estuary during each flood
tide.
1786
Molluscan larvae transport through a shallow estuary
Fig. 3. Time series of gastropod transport and physical variables at Tidal Inlet during three dates in
1994. (a) 27 June; (b) 21 July; (c) 17 August. (), gastropod transport (QL ± S.D.); (), salinity (S, mg
g–1); thin solid line, volume transport (Q, m3 s–1). Black bar represents period of darkness.
Plankton transport at Tidal Inlet
There were large differences in plankton transport through the tidal inlet
between ebb and flood tides and among dates sampled. Striking differences were
found between the transport of mollusc veligers (Figures 3 and 4). In June, gastropod larvae were found in relatively low abundance with import and export being
approximately equal at 1.5–2.3 108 ind. (Table I; Figure 3). Over the two tidal
periods sampled in June, there was a net import of 1.4 108 ind. In contrast, the
July and August samples exhibited large exports of gastropod larvae from estuary
to nearshore with total N ranging from 0.2 to 20.0 108 ind. Import rates
remained at June levels. Net export was 2.8 109 ind. over the two tidal periods
sampled in July and 1.7 109 in August. Transport from the estuary peaked
several hours into ebb, usually in water masses associated with the estuarine
1787
G.Curtis Roegner
Fig. 4. Time series of biological transport and physical variables at Tidal Inlet during three dates in
1994. (a) 27 June; (b) 21 July; (c) 17 August. (), bivalve transport (QL ± S.D.); (), salinity (S, mg
g–1); thin solid line, volume transport (Q, m3 s–1). Black bar represents period of darkness. Note
change in scale of y-axis.
front. Clearly, in the latter months, export greatly exceeded import and the
estuary was a source of gastropod larvae for the nearshore.
Bivalve transport was one to two orders of magnitude less than that of
gastropods (2.8 106–8.9 107 ind.), and veligers were in relatively high abundance in the June and July samples compared with August (Table I; Figure 4).
Import of bivalves exceeded export in four of six tidal cycles, and a net import
was measured during each of the three sample dates. Larvae were imported
throughout the flood period, but exported larvae were found almost exclusively
in the high salinity water during early ebb. These patterns were also observed in
a 36 h study initiated on 6 August 1993 (Figure 5). Bivalve larvae were rarely
observed in the estuarine water (Figures 4 and 5). The nearshore zone was thus
a source of bivalve larvae for the estuary.
There was no consistent pattern of bivalve transport relative to development
stage (Table I, Figure 6). The majority of the bivalves (58–>99%) in a tidal prism
1788
Molluscan larvae transport through a shallow estuary
Fig. 5. Time series of biological and physical variables at Tidal Inlet during 6 August 1993. (a) Mean
water level; (b) (), salinity and , temperature (Temp); (c) concentration (± S.D.) of gastropod ()
and bivalve () larvae. Black bar represents period of darkness.
were immature larvae <200 µm in length. Most of these were actually ‘D’ larvae,
the early developmental stage. The peak abundance of larger larvae was observed
in June, and these animals were mainly blue mussels (Mytilus edulis). The sizefrequency structure of the flood and ebb larval populations was examined to
determine if selective retention of competent larvae was occurring (Table I,
Figure 6). The size-frequency patterns did not show consistent patterns of larger
larvae remaining in the estuary; in fact, large individuals were often enriched in
the ebb water. Note that individuals ≥ 300 µm were mostly juveniles being transported as bedload or by thread drifting (Roegner et al., 1995). The evidence does
not support an enhanced retention of larger larvae in the estuary.
Volume transport at Benthic Site
Similar time series of larval abundance and oceanographic variables were made
at Benthic Site to examine larval transport within the estuary. The following
description for 30 June 1994 summarizes the observed patterns (Figure 7). This
5 h period encompassed the end of ebb through most of the subsequent flood,
and measurements were made 1 day after the June sampling at the Tidal Inlet.
Volume transport during the latter half of ebb flow was greatly reduced due to
the small cross-sectional channel area, and the ebbing water was characterized by
constant high temperature and gradually rising salinity (Figure 7a and b). The
change in direction of volume transport from ebb to flood coincided closely with
the rise in water level. During the beginning of flood, the tide increased 0.20 m
1789
G.Curtis Roegner
Fig. 6. Size-frequency histograms of bivalve larvae (shell length, 0.05 mm intervals) partitioned into
flood (F) and ebb (E) prism pairs during three dates in 1994. (a–b) 27 June; (c–d) 21 July; (e–f) 17
August.
in height with little variation in salinity or temperature, after which there was a
precipitous drop in salinity while temperature remained constant. This is
evidence of horizontal mixing due to tidal trapping (Fisher et al., 1979; Roegner,
1998). The flood tide front followed the transit of this parcel of Eel River water.
The salinity and temperature characteristics indicate coastal water penetrated to
Benthic Site as a relatively unmixed water mass. Variation in tide, salinity and
temperature measured on other sample dates show similar patterns (Figure 8).
Further discussion of algal transport at Benthic Site can be found in Roegner
(Roegner, 1998).
Plankton transport at the Benthic Site
Transport of bivalve and gastropod veligers through the Benthic Site tidal creek
varied strongly with tidal stage. During the 30 June 1994 sample, the transport of
mollusc larvae was low during the measured ebb and early flood periods (Figure
8c). A pulse of gastropods moved through the creek 1 h after flood in the tidallytrapped water identified above. There was a concurrent peak in chlorophyll a
1790
Molluscan larvae transport through a shallow estuary
Fig. 7. Time series of biological and physical variables at Benthic Site 30 June 1994. (a) Water level
and volume transport (Q, solid line); (b) (), salinity and (), temperature; (c) transport (QL ± S.D.)
of gastropod () and bivalve () larvae.
Fig. 8. Time series of physical and biological variables at Benthic Site during three dates. (a–c) 10
September 1993; (d–f) 14 September 1993; (g–i) 25 May 1994. , water level; (), salinity; (), temperature; (), gastropod larvae concentration (± S.D.); (), bivalve larvae concentration (± S.D.). Note
differing scales of x- and y-axes.
1791
G.Curtis Roegner
(Roegner, 1998) but continued low numbers of bivalve veligers until the intrusion of the nearshore water, when bivalve transport reached ~4.0 103 ind. s–1.
Gastropods maintained transport of about 1.0 103 ind. s–1 during flood. Thus,
there was a clear import of bivalves to the benthic site in the high salinity flood
water, but no evidence of organisms exiting the system during the end of the
previous ebb period. Gastropod larvae first appeared in water originating in the
Eel River, and were subsequently imported to Benthic Site with coastal water.
Further examples of temporal variation of mollusc larvae in tidal creeks are
shown in Figure 8, for which velocity measurements are lacking. These results are
consistent with the conclusion that bivalve abundance peaked in ocean water and
gastropods in estuarine water.
Hydraulic residence time
Tidal elevation during summer 1994 ranged from <0.3 to 1.1 m (Figure 9a). Tidal
exchange ratios ranged from 0.2 to 0.9, with on average 50.7% (±12.3% S.D.) of
the water in the lower estuary being replaced per tide (Figure 9b). The estuary
was well ventilated. The temporal variation in the exchange ratios was due to tidal
inequality and the spring–neap cycle, and exchange was maximized during
periods of lower low water. Hydraulic residence times TR were <2 days throughout the summer (Figure 9c). Temperature varied on tidal, diurnal, lunar and
seasonal periodicity (Figure 9d). Rapid heating of estuarine water occurred when
low tide corresponded with days of high insolation, while cooler seawater
temperatures prevailed during flood. Temperature fluctuations in excess of 10°C
in <1 h were common during the day, while the temperature variation between
ebb and flood was generally ~5°C at night. Ocean water was clearly discernible
from estuarine water by temperature.
Discussion
Estuarine circulation patterns
Dispersal of larval molluscs results from the interaction of swimming behavior,
planktonic development rate and physical transport mechanisms. The relative
importance of these factors for determining larval retention and recruitment
intensities is contentious [see reviews by (Mann, 1986a; Scheltema, 1986; Stancyk
and Feller, 1986)]. It is clear that the instantaneous horizontal position of larvae
in an estuary is determined by advective transport (Butman, 1986, 1987). Net
horizontal swimming velocities of bivalves, at <10–4 m s–1 (Cragg and Gruffydd,
1975; Mann, 1986b), are insufficient for directed motion in estuarine currents that
can exceed 1 m s–1. Over time scales on the order of a tidal excursion, larvae will
be transported as near-passive particles with the water parcel in which they
reside. Extensive landward or seaward displacements often result; for example,
the tidal transport of veligers in James River (Andrews, 1983) or Delaware Bay
(Jacobsen, 1990; Jacobsen et al., 1990) can exceed 10 km tide–1, while Tremblay
and Sinclair (Tremblay and Sinclair, 1988) estimated larval scallop transport in
the Bay of Fundy to be 22–44 km day–1.
1792
Molluscan larvae transport through a shallow estuary
Fig. 9. Time series of water level (a); exchange ratio, r (b); hydraulic residence times TR (c); and
temperature (d) during Summer 1994.
The Eel River typifies shallow, well-flushed estuaries found on the northwest
Atlantic coast. Water exchange is intensified by a low depth to tidal amplitude
ratio. This generates rapid tidal currents, which in Eel River exceeded 0.6 m s–1
nearly every tide. Circulation patterns exhibited steep time-varying horizontal
gradients in salinity and temperature. Salinity time series (Roegner, 1998) and
cross-correlations of pressure sensor data (not shown) indicate the tidal excursion length ranged between the Middle Estuary and West Marsh stations. A
typical flood tide completely replaced water in this 7.5 104 m2 area and forced
the estuarine water mass back towards West Marsh. At the height of flood tide,
this process formed a region between Middle Estuary and West Marsh of strong
horizontal salinity gradient [greater than 15 psu in ~500 m (Roegner, 1998)].
During ebb, first the recently imported coastal water, then the mixed estuarine
water and finally, fresher water from West Marsh flushed from the estuary. On
average, ~50% of the water in the estuary was exchanged every tide, and
hydraulic residence times ranged between 0.5 and 2 days (Figure 9). The seaward
half of the lower estuary (where clam populations were most abundant) was well
flushed by every tide and experienced wide salinity and temperature fluctuations.
The Back Marsh embayment containing the Benthic Site station had a low
drainage area and generally maintained higher salinity than at Middle Estuary
(Roegner, 1998). Imported ocean water rapidly replaced estuarine water during
flood, but the ocean water was usually preceded by a tidally trapped slug of low
salinity water originating from Eel River (Figures 7 and 8). This process was an
important mechanism for material exchange between Back Marsh and Eel River
1793
G.Curtis Roegner
and indeed, a larger scale occurrence of the same phenomenon was noted
between Eel River and Lawrencetown River (Figure 2). It is interesting that
systems such as Back Marsh and Eel River, that were so closely connected physically, maintained such disparate water masses. Set-up by the tide was responsible for the retention of relatively large volumes of water in Back Marsh. Set-up
occurs during unequal tidal amplitudes when a small ebb tide follows a large flood
tide. Water entrained in the system during the flood tide is not completely
removed during the subsequent ebb, thus increasing retention time. In Eel River,
however, the retention time was normally only a single tidal period (Figure 8).
Plankton transport through the Eel River estuary
Large numbers of mollusc veligers were imported into the estuary during each
flood tide (Table I, Figures 3 to 5). In 1994, between 106 and 108 bivalve, and 107
and 109 gastropod larvae entered the estuary per flood tide. Larval concentrations in the coastal water ranged between 101 and 104 ind. m–3. These values are
in the range of previous studies of shelf and nearshore environments [i.e.
(Harding et al., 1986; Mann, 1986b; Scrope-Howe and Jones, 1986; Tremblay and
Sinclair, 1988, 1990a, b; Raby et al., 1994)]. There were also large imports of holoplanktonic dinoflagellates (Ceratium spp.), and crustaceans such as calanoid
copepods, with the coastal water. Meroplankonic echinoderm pluteii, annelid
larvae and crustacean nauplii and zoeae were found in varying abundance.
Advected plankton penetrated well into the estuary during a typical tidal excursion.
During ebb, there was extensive export of material from estuary to sea. In July
and August, 109 gastropods were transported to the nearshore in a single ebb tide
(Figure 3). These numbers were 10–100 times the number of gastropods
imported. Gastropod larvae were abundant in estuarine water but were concentrated near the interface between ocean and estuarine water. Peak gastropod
transport occurred in or near this estuarine front. Gastropods clearly had an estuarine origin. Similarly, chlorophyll a derived from autochthonous algal blooms in
West Marsh was exported to the coastal water (Roegner, 1998).
These transport patterns contrast the time series for bivalve larvae. Peak
bivalve export generally coincided with peak ebb volume transport, which
occurred within 1.5 h of the change in tide (Figure 4). This water was invariably
of high salinity, indicating its coastal origin, and demonstrates the passive export
of veligers that had entered on a previous flood tide. Bivalves were rarely
observed in the estuarine water (Figures 4 and 5).
There were net imports of bivalve veligers on each sampling date. Most were
immature larvae. A comparison of the size-frequency histograms of bivalves in
the flood and ebb prisms does not support the selective retention of mature
larvae. Further, net import during a flood–ebb pair was of the same order as the
maximum tidal transport of that date (Table I). This suggests tidal set-up and
short-term storage of water in the estuary caused the observed retention of immature larvae. However, active retention in the shallow eel grass beds, or mortality,
1794
Molluscan larvae transport through a shallow estuary
cannot be excluded. The sampling frequency was not continuous and gaps in the
time series created uncertainties in the flux and transport calculations.
When could behavior influence retention in the Eel River?
The net import of bivalves into the estuary coupled with the low concentrations of
bivalves in the tidal channels (in contrast to the gastropod pattern) leads one to
wonder if bivalve swimming behavior is influencing larval distributions. Larvae can
affect their horizontal position if they can select water column strata of differing
velocities. Such ‘active’ retention is hypothesized to occur by vertical migration
cued to estuarine circulation, whereby interacting with shear zones (such as the
benthic boundary layer), larvae can elicit some control over their net horizontal
position (Pritchard, 1952; Bousefield, 1955; Butman, 1986; Hill, 1995). This process
relies on residual circulation patterns that are manifest over many tidal cycles.
Laboratory studies indicate vertical swimming and sinking speeds of many larval
forms are adequate to transverse more than 10 m of water within a tidal period
(Mileikovsky, 1973; Mann, 1988), and evidence exists that environmental cues such
as salinity (Haskin, 1964; Hidu and Haskin, 1978; Mann, 1988; Mann et al., 1991),
pressure (Bayne, 1963; Cragg and Gruffydd, 1975; Mann and Wolfe, 1983) and light
(Bayne, 1964; Hidu and Haskin, 1978) can initiate a directed swimming response.
However, vertical migration patterns must be precisely linked with circulation
patterns and development time for retention to be effected (Hill, 1995).
In tidally-dominated estuaries where vertical stratification is weak or nonexistent, retention is likely to be dependent on refugia from strong horizontal
velocities. Larval swimming behavior in laboratory flume flow has revealed that
free-stream velocities above ~0.15 m s–1 result in passive transport of bivalve
larvae (Butman, 1986; Jonsson et al., 1991; Grassle et al., 1992). In Eel River, the
velocity time series indicate larvae were subjected to rapid advection during most
of the tidal period. At Tidal Inlet, free-stream velocities greater than 0.15 m s–1
predominated the flow regime, and maximum velocities exceeded 0.6 m s–1. These
conditions virtually eliminate any possibility of behaviorally-mediated control of
ingress or export. Similarly, the velocity regime measured in the main tidal
channel and tidal creeks limited the periods when vertical swimming behavior
could strongly affect tidal transport. During flood, near-bottom flows <0.15 m s–1
occurred only for a short period during the beginning and end of the flood period.
Maximum currents peaked at ~0.3 m s–1 in the tidal creeks and commonly
resulted in the erosion of bottom sediments (Roegner et al., 1995; Roegner, 1996).
Regardless of vertical position, larvae would be transported significant distances
into the estuary as near-passive particles during most of the flood tide cycle.
Ebb velocities within the estuary varied more widely and if entrained into a
tidal creek, larval transport distance would depend on stage of the ebb tide. At
Benthic Site, the initial ebb flow had velocities of ~0.2 m s–1 and due to the high
water level, it was during this period that most volume transport occurred. Subsequently, the channel water level drained to between 0.1 and 0.3 m and velocities
were reduced to between 0.01 and 0.10 m s–1 for more than 5 h (Figures 7 and 8).
1795
G.Curtis Roegner
In this low velocity shallow water, larvae would have ample opportunity to
maneuver. These appear to be optimal conditions for mature larvae to select a
settlement site. However, larvae remaining in the free-stream position would
traverse completely through the Back Marsh creek system (>900 m) and into the
higher flows of the Eel River during a single ebb period. Ebb velocities in the Eel
River consistently exceeded 0.3 m s–1, and larvae would again be subjected to
passive transport. Once near the inlet, ebb velocities increased to ~0.6 m s–1, and
larvae would have no opportunity for behavioral regulation of their horizontal
position. Larvae were thus afforded few refuges from transport in the channel
environments during a tidal cycle.
Longer term refugia from transport appear to exist only in the shallow subtidal
portions of Back Marsh or between the Middle Estuary and West Marsh stations.
Reduced currents in these sections would allow vertical swimming to influence
position. The benthos of the basins was dominated by extensive beds of eelgrass
(emergent at low tide), in which larvae could find refuge from currents. These
regions were not adequately sampled in this study, but movement into such
boundary layer conditions would likely reduce transport (Eckman, 1983). While
water in the basins was completely exchanged every few tidal periods, these are
also areas experiencing tidal set-up and increased hydraulic residence time. Such
periods of short-term retention on the order of days may be important for settlement of mature larvae imported from the nearshore zone.
Comparison of molluscan groups
It is noteworthy that the large export signal measured for gastropods was not
observed for bivalves, despite the similarity in locomotor abilities and physical
attributes (i.e. density) of the veliger forms. This is surprising given the high
biomass of reproductively-active bivalves within the estuary (Roegner, 1996),
which are capable of a large larval production (Brousseau, 1978). The discrepancy between the export signals of larval bivalves and gastropods is likely
explained by differences in reproductive modes of the molluscs and the sampling
periodicity I used to sample them. The numerically dominant gastropod was the
small prosobranch, Hydrobia minuta (Totten), which lays attached eggs that
hatch after ~1 week into a free-swimming veliger larvae (Barnes, 1988, 1990).
Variation in rates of egg laying and hatching contribute to a relatively long period
of larval presence in the estuarine waters. The common periwinkle, Littorina
littorea (L.), was also extremely abundant and local populations exhibit two
spawnings per year (Chase and Thomas, 1995), again increasing the temporal
window of larval presence. In contrast, the bivalves within the system Mya,
Macoma and Mytilus, are broadcast spawners with planktonic development times
of 2–5 weeks. A synchronous bivalve reproductive effort and rapid exportation
of the weakly-swimming early larval forms within the relatively coarse 3–4 week
sampling intervals suggests the peak abundance of bivalve larvae in the estuary
was probably not sampled. Additionally, since it was not possible to sample the
plankton continuously, net transport of bivalves out of the estuary may have been
underestimated. Short-term retention may be important for the settlement of
1796
Molluscan larvae transport through a shallow estuary
mature larvae, but larvae spawned within the Eel River are probably removed
from the system within a few tidal periods.
Export and retention in estuarine systems
Investigations of bivalve larval dynamics made in a variety of estuarine systems
supports the importance of tidal advection for controlling larval abundance.
Many systems exhibit evidence of larval invasion. Carriker examined the larval
dynamics of oysters, Crassostrea virginica (Gmelin), and hard clams, Mercenaria
mercenaria L., in the highly flushed Home Pond in Gardiners Bay, NY (Carriker,
1951). Relatively high concentrations (>1.3 104 ind. m–3) of oyster larvae were
endogenously produced, but were rapidly reduced by tidal flushing. Few larvae
developed to maturity in situ. Ebbed water reportedly dispersed quickly in the
nearshore (depending on wind stress), thus transporting propagules to Long
Island Sound. Concentrations up to 1.2 104 ind. m–3 were imported through the
tidal inlet, and it was concluded that recruitment depended on invasion. Larval
Mercenaria were also passively transported, but because development time was
only 7 days, clams had a greater likelihood of retention.
In the larger Little Egg Harbor, NJ, Carriker (Carriker, 1961) found export of
Mercenaria to be dependent on proximity to the tidal inlet. Individual cohorts of
clam larvae were identified and followed; swarms of newly-spawned veligers initially maintained spatial coherence, but horizontal dispersion eventually resulted
in more uniform distributions. Spawning events and larval retention were
maximal during neap tides, while complete loss of cohorts occurred during high
flushing events during spring tides. Large horizontal gradients in larval concentration were also noted, with the highest abundance in central lower bay (>6.7 105 ind. m–3), and low concentrations near the inlet.
Christy and Stancyk computed net larval fluxes over 25 h periods through a tidal
inlet in South Carolina (Christy and Stancyk, 1982). They found a high larval
exchange between estuary and nearshore, but few net fluxes differed significantly
from zero over a given sample period. Bivalve larvae had a net export in only 1
month (July), during which time export was greater during neap (1.4 1010 ind.)
than spring (1.1 1010 ind.) tide samples. There were net imports in all other
months, ranging from 4.0 107 (November) to 4.7 109 (May) ind. Net gastropod
transport varied more (8.0 105 to 2.9 109 ind.) and export dominated in three
of eight sample periods. There was no consistent pattern of transport in relation to
tidal stage. Larval behavior was invoked to explain these findings (however, the
153 µm mesh size used undersampled the abundance of smaller mollusc larvae).
Carlson et al. measured large concentrations (>2.4 104 ind. m–3) of ‘D’ stage
bivalve larvae entering a Maine coastal embayment during flood, and the
depleted abundance of veligers and phytoplankton exiting during ebb flow was
attributed to benthic grazers (Carlson et al., 1984). Booth and Sephton monitored
oyster larvae (Crassostrea) abundance and the physical oceanography of Caraquet Bay, New Brunswick, Canada (Booth and Sephton, 1993). They found
patchy distributions and rapid changes of all larval size classes with time, until a
strong intrusive event emanating from the adjacent Baie de Chaleurs flushed the
1797
G.Curtis Roegner
larvae from the estuary. A week later, larvae re-entered the system at settlement
size. Detailed vertical and horizontal sampling revealed little evidence of retention or diurnal vertical migration.
Conceptual model of larval recruitment patterns
The physical oceanography of systems like the Eel River appears to preclude
autochthonous recruitment of invertebrate larvae with medium length planktonic
development periods. Larvae exported from these well-flushed estuaries must
undergo development in the coastal ocean. The temporal persistence of populations in the estuary is therefore dependent on larval invasion. During the tidal
periods sampled in Eel River, a significant exchange of bivalve larvae was
measured through the estuarine system. A net import of bivalve larvae was estimated for the periods sampled. However, few of these larvae were developmentally capable of settlement. Recruitment rates depend on the import of
physiologically-mature larvae. These animals can react rapidly to chemical cues
associated with specific settlement substrates (Turner et al., 1994). With active
searching and settlement behaviors, larvae can easily settle from the water
column within a tidal period.
An important unresolved question is the origin of the imported larvae. It is
unlikely that larvae exported from the Eel River system could remain very long
near the tidal inlet. Water imported into the Eel River was of coastal salinity and
temperature, indicating the estuarine plume was not ‘resampled’. Rather, prevailing longshore currents and wind-generated cross-shelf currents would result in
advection away from the estuary. However, this does not mean exported larvae
are lost to the metapopulation. If the larvae remain in proximity to the coast, or
are transported landward from further offshore, they can re-enter one of the
many estuarine systems in this area. Probably, all the estuaries along the coast
contribute to a larval pool that resides in the nearshore zone. Large-scale advective processes affecting the distribution of the larval pool likely determine rates
of larval supply to estuaries and set controls on recruitment rates. In the deeper,
stratified coastal waters, larval swimming behavior may be important for moderating transport rates and ingress into estuaries (Harding et al., 1986; Raby et al.,
1994). Physical and biological variables in the coastal boundary layer and near
estuarine plumes must be more extensively sampled in order to improve our
understanding of estuarine recruitment of benthic invertebrates.
Acknowledgements
I thank C.André, G.MacIntyre, R.Marsh, T.Sutherland and P.Simard for their
Herculean Labors in the field and my Dissertation committee, T.Bowen, J.Grant,
B.Hargrave and T.Rowell, for insightful discussions. This research was supported
by Canadian DFO/Subvention grants and a Dalhousie University Graduate
Fellowship.
1798
Molluscan larvae transport through a shallow estuary
References
Andrews,J.D. (1983) Transport of bivalve larvae in James River, Virginia. J. Shellfish Res., 3, 29–40.
Barnes,R.S.K. (1988) On reproductive strategies in adjacent lagoonal and intertidal-marine coastal
populations of the gastropod Hydrobia ulvae. J. Mar. Biol. Ecol., 68, 365–375.
Barnes,R.S.K. (1990) Reproductive strategies in contrasting populations of the coastal gastropod
Hydrobia ulvae. II. Longevity and life-time egg production. J. Exp. Mar. Biol. Ecol., 138, 183–200.
Bayne,B.L. (1963) Responses of Mytilus edulis larvae to increases in hydrostatic pressure. Nature, 198,
406–407.
Bayne,B.L. (1964) The biology of mussel larvae. In Bayne,B.L. (ed.), Marine Mussels, Their Ecology
and Physiology. Cambridge University Press, pp. 81–120.
Booth,D.A. and Sephton,T.W. (1993) Flushing of oyster (Crassostrea virginica) larvae from a small
tidal bay. Can. Tech. Rep. Fish. Aquat. Sci., 1945.
Bousfield,E.L. (1955) Ecological control of the occurrence of barnacles in the Miramichi Estuary. Nat.
Mus. Can. Bull., 137, Biol. Ser., 46, 1–69.
Brousseau,D.J. (1978) Spawning cycle, fecundity and recruitment in a population of soft-shell clam,
Mya arenaria, from Cape Ann, Massachusetts. Fish. Bull., 76, 155–166.
Butman,C.A. (1986) Larval settlement of soft-sediment invertebrates, some predictions based on an
analysis of near-bottom velocity profiles. In Nihoul,J.C.J. (ed.), Marine Interfaces Ecohydrodynamics. Elsevier Oceanogr. Ser. Vol. 42, Elsevier, Amsterdam, pp. 487–513.
Butman,C.A. (1987) Larval settlement of soft-sediment invertebrates, the spatial scales of pattern
explained by active habitat selection and the emerging role of hydrodynamical processes. Oceanogr.
Mar. Biol. Ann. Rev., 25, 113–165.
Carlson,D.J., Townsend,D.W., Hilyard,A.L. and Eaton,J.F. (1984) Effect of an intertidal mudflat on
plankton of the overlying water column. Can. J. Fish. Aquat. Sci., 41, 1523–1528.
Carriker,M.C. (1951) Ecological observations on the distribution of oyster larvae in New Jersey
estuaries. Ecol. Monogr., 21, 19–38.
Carriker,M.R. (1961) Interrelation of functional morphology, behavior and autecology in early stages
of the bivalve Mercenaria mercenaria. J. Elisha Mitchell Sci. Soc., 77, 168–241.
Chase,M.E. and Thomas,M.L. (1995) Evidence for double recruitment in the common periwinkle,
Littorina littorea (Linnaeus, 1758), on Pendleton Island, New Brunswick, Canada. J. Shellfish Res.,
14, 153–158
Chia,F., Buckland-Nicks,J. and Young,C.M. (1984) Locomotion of marine invertebrate larvae, a
review. Can. J. Zool., 62, 1205–1222.
Christy,J.H.and Stancyk,S.E. (1982) Timing of larval production and flux of invertebrate larvae in a
well-mixed estuary. In Kennedy,V.S. (ed.), Estuarine Comparisons. Academic Press, NY, pp.
489–504.
Connell,J.H. (1985) The consequences of variation in initial vs post-settlement mortality in rocky
intertidal communities. J. Exp. Mar. Biol. Ecol., 93, 11–45.
Cragg,S.M. and Gruffydd,L.D. (1975) Swimming behavior and the pressure responses of the veliconcha larvae of Ostrea edulis L. In Bayne,H. (ed.), Proceedings of the 9th European Marine Biology
Symposium. Aberdeen University Press, pp. 43–57.
Eckman,J.E. (1983) Hydrodynamic processes affecting benthic recruitment. Limnol. Oceanogr., 28,
241–257.
Fisher,H.B., List,E.J., Koh,R.C.Y., Imberger,J. and Brooks,N.H. (1979) Mixing in Land and Coastal
Waters. Academic Press, NY.
Geyer,W.R. and Signell,R.P. (1992) A reassessment of the role of tidal dispersion in estuaries and
bays. Estuaries, 15, 97–108.
Grassle,J.P., Snelgrove,P.V.R. and Butman,C.A. (1992) Larval habitat choice in still water and flume
flows in the opportunistic bivalve Mulinia lateralis. Neth. J. Sea Res., 30, 33–44.
Harding,G.C., Vass,W.P., Hargrave,B.T. and Pearre,S.Jr (1986) Diel vertical movements and feeding
activity of zooplankton in St. Georges Bay, N.S., using net tows and a newly developed passive trap.
Can. J. Fish. Aquat. Sci., 43, 952–967.
Haskin,H.H. (1964) The distribution of oyster larvae. Exp. Mar. Ecol. Oc. Publ. Narragansett Mar.
Lab., 2, 76–80.
Hidu,H. and Haskin,H.H.. (1978) Swimming speeds of oyster larvae Crassostrea virginica in different
salinities and temperatures. Estuaries, 1, 252–255.
Hill,A.E. (1995) The kinematical principles governing horizontal transport induced by vertical
migration in tidal flows. J. Mar. Biol. Assoc. UK, 75, 3–13.
Hume,T.M. and Bell,R.G. (1993) Methods for determining tidal flows and material fluxes in estuarine
cross-sections. N. Z. Water Quality Pub., 22.
1799
G.Curtis Roegner
Jacobsen,T.R. (1990)Observational and model studies of physical processes affecting benthic larval
recruitment in Delaware Bay. J. Geophy. Res., 95, 20, 331–320, 345.
Jacobsen,T.T., Milutinovic,J.D. and Miller,J.R. (1990) Recruitment in estuarine benthic communities,
the role of physical processes. In Chang,R.T. (ed.), Coastal and Estuarine Studies. Vol. 38, pp. 513–525.
Jonsson,P.R., André,C. and Lindegarth,M. (1991) Swimming behavior of marine bivalve larvae in
a flume boundary-layer flow, evidence for near-bottom confinement. Mar. Ecol. Prog. Ser., 79,
67–76.
Ketchum,B.H. (1954) Relation between circulation and planktonic populations in estuaries. Ecology,
35, 191–200.
Mann,R. (1985) Seasonal changes in the depth-distribution of bivalve larvae on the southern New
England shelf. J. Shellfish Res., 5, 57–64.
Mann,R. (1986a) Sampling of bivalve larvae. In Jameson,G.C. and Bourne,N. (eds), North Pacific
Workshop on Stock Assessment and Management of Invertebrates. Can. J. Fish. Aquat. Sci., 92,
107–116.
Mann,R. (1986b) Arctica islandica (Linne) larvae, Active depth regulators or passive particles? Am.
Malacol. Bull., 3, 51–77.
Mann,R. (1988) Distribution of bivalve larvae at a frontal system in the James River, Virginia. Mar.
Ecol. Prog. Ser., 50, 29–44.
Mann,R. and Wolfe,C.C. (1983) Swimming behaviour of larvae of the ocean quahog Arctica islandica
in response to pressure and temperature. Mar. Ecol. Prog. Ser., 13, 211–218.
Mann,R., Campos,B.M. and Luckenbach, MW. (1991) Swimming rate and responses of larvae of three
mactrid bivalves to salinity discontinuities. Mar. Ecol. Prog. Ser., 68, 257–269.
Mileikovsky,S.A. (1973) Speed of active movement of pelagic larvae of marine bottom invertebrates
and their ability to regulate their vertical position. Mar. Biol., 23, 11–17.
Ólafsson,E.B., Peterson,C.H. and Ambrose,W.G.Jr (1994) Does recruitment limitation structure
populations and communities of macro-invertebrates in marine soft sediments, The relative significance of pre- and post-settlement processes. Oceanogr. Mar. Biol. Annu. Rev., 32, 65–109.
Platt,T., Prakash,A. and Irwin,B. (1972) Phytoplankton nutrients and flushing of inlets on the coast
of Nova Scotia. Le Naturliste Canadien, 99, 253–261.
Pritchard,D.W. (1950) Distribution of oyster larvae in relation to hydrographic conditions. Proc. Gulf
Carib. Fish. Inst., 5, 123–132.
Raby,D., Lagadeuc,Y., Dodson,J. and Mingelbier,M. (1994) Relationship between feeding and vertical
distribution of bivalve larvae in stratified and mixed waters. Mar. Ecol. Prog. Ser., 103, 275–284.
Rodriguez,S.R., Ojeda,F.P. and Inestrosa,N.C. (1993) Settlement of benthic marine invertebrates.
Mar. Ecol. Prog. Ser., 97, 193–207.
Roegner,G.C. (1998) Hydrodynamic control of the supply of suspended chlorophyll a to infaunal
estuarine bivalves. Estuarine Coastal Shelf Sci., 47, 369–384.
Roegner,G.C. (1996) Hydrodynamic Factors Affecting the Recruitment of Bivalve Molluscs in a
Tidally-dominated Estuary. Ph.D. Dissertation, Dalhousie University, Halifax, Nova Scotia,
Canada.
Roegner,G.C., André,C. Lindegarth,M., Eckman,J.E. and Grant,J. (1995) Transport of recently
settled soft-shell clams (Mya arenaria L.) in laboratory flume flow. J. Exp. Mar. Biol. Ecol., 187,
13–26.
Scheltema,R.S. (1986) On dispersal of planktonic larvae of benthic invertebrates, an eclectic overview
and summary of problems. Bull. Mar. Sci., 39, 290–322.
Scrope-Howe,S. and Jones,D.A. (1986) The vertical distribution of zooplankton in the western Irish
Sea. Estuarine Coastal Shelf Sci., 22, 785–802.
Stancyk,S.E and Feller,R.J. (1986) Transport of non-decapod invertebrate larvae, An overview. Bull.
Mar. Sci., 39, 257–268.
Thorson,G. (1966) Some factors influencing the recruitment and establishment of marine benthic
communities. Neth. J. Sea Res., 3, 267–293.
Tremblay,M.J. and Sinclair,M.M. (1988) The vertical and horizontal distribution of sea scallop
(Placopecten magellanicus) larvae in the Bay of Fundy in 1984 and 1985. J. Northw. Atlant. Fish. Sci.,
8, 43–53.
Tremblay,M.J. and Sinclair,M. (1990a) Sea scallop larvae Placopecten magelanicus on Georges Bank,
Vertical distribution in relation to water column stratification and food. Mar. Ecol. Prog. Ser., 61,
1–15.
Tremblay,M.J. and Sinclair,M. (1990b) Diel vertical migration of sea scallop larvae Placopecten magellanicus in a shallow embayment. Mar. Ecol. Prog. Ser., 67, 19–25.
Turner,E.J., Zimmer-Faust,R.K. Palmer,M.A., Luckenbach,M. and Pentcheff,N.D. (1994) Settlement
of oyster (Crassostrea virginica) larvae, effects of water flow and a water-soluable chemical cue.
1800