JOURNAL OF CRUSTACEAN BIOLOGY, 29(1): 51–56, 2009
DIEL AND SEMI-LUNAR CYCLES IN THE SWIMMING ACTIVITY OF THE INTERTIDAL,
BENTHIC AMPHIPOD COROPHIUM VOLUTATOR IN THE UPPER BAY OF FUNDY, CANADA
David Drolet and Myriam A. Barbeau
(DD, corresponding author; and MAB) Department of Biology, University of New Brunswick, PO Box 440011,
Fredericton, New Brunswick, Canada, E3B 5A3 ([email protected] and [email protected])
ABSTRACT
Although movement of individuals has important consequences on population dynamics and various ecological interactions, it is often
difficult to quantify fully. We investigated the temporal variation in the number of the amphipod Corophium volutator swimming in the
water column during periods of immersion over an intertidal mudflat in the Bay of Fundy, Canada, in spring-summer 2006. Swimming is
an important mode of dispersal, since the number of swimming amphipods can peak at over 30,000 individuals within a 20-cm-diameter,
stationary plankton net over a period of immersion of ;4 h. Amphipods swim throughout spring-summer, but abundance in the water
column is less in May than in the other months. As well, amphipods swim during the day and night, but the number swimming shows
periodicity in relation to diel time of high tide, with peaks when high tides occur around 1:45 am. Finally, the number of amphipods
swimming shows periodicity in relation to lunar cycles, with peaks around the time of new moon and full moon. We developed a statistical
model describing the swimming activity of C. volutator based on month, diel time of high tide, and day of the lunar calendar. The model
accurately predicts the timing of peaks, but does not predict well the amplitude of the highest peaks. Overall, the model gives a very good
approximation of the number of swimmers (61% of the variation is explained) and provides a strong basis for future modeling of spatial
population dynamics of C. volutator.
KEY WORDS: Corophium volutator, dispersal, swimming, periodic regression, lunar cycle, diel cycle
DOI: 10.1651/08-3010.1
a maximum of 7% of the population swims at one time.
Swimming activity in C. volutator displays cyclic variations that are related to season, lunar phase and time of the
day (Holmstrom and Morgan, 1983; Hughes, 1988).
Usually, more amphipods are present in the water column
when high tides occur at night, around new moon or full
moon during the summer (Holmstrom and Morgan, 1983;
Hughes, 1988). However, there seems to be important
geographical variation in the swimming activity of C.
volutator. Hughes (1988) only recorded swimming activity
at high tides occurring at night in England, whereas
Holmstrom and Morgan (1983) and Hughes and Horsfall
(1990) reported swimming activity during the day for
different sites also in England. In contrast to other studies,
Lawrie and Rafaelli (1998b) reported peaks in swimming
activity occurring during the progression from spring to
neap tide in the Ythan Estuary (Scotland). All these studies
on swimming behaviour of C. volutator were conducted in
Europe and to our knowledge swimming behaviour of C.
volutator has never been studied on the coast of North
America.
Given the important geographical variation in swimming
of C. volutator, results from Europe would be poor
predictors of swimming of North American C. volutator
populations. The objectives of this study were (1) to
determine whether individuals in a population of C.
volutator in the upper Bay of Fundy (Canada) exhibit
swimming behaviour and (2) develop a statistical model
describing the numbers of C. volutator swimming in the
water column over time.
INTRODUCTION
Movement of individuals is an important component of
population dynamics (Turchin, 1998) that has consequences
on the spatial distribution of populations (Turchin, 1991), as
well as on the outcome of predator-prey, host-parasites and
competitive interactions (Tilman and Kareiva, 1997).
Movement in benthic, intertidal species can occur through
displacement on the substratum, re-suspension (Norkko et
al., 2001) or active swimming in the water column (Hughes,
1988). Diel and lunar cycles offer predictable environmental
cues. Many species are affected by light and consequently
show diel cycle in activity (Drolet et al., 2004). Lunar
cycles are known to influence the timing of spawning
(Gaudette et al., 2006), release of juveniles (McCurdy et al.,
2000) and activity schedule of animals (Meireles and
Queiroga, 2004).
The burrowing amphipod Corophium volutator (Pallas,
1776) is a dominant species in the intertidal soft-sediment
environments in North America and Europe (Watkin, 1942;
Peer et al., 1986). During periods of emersion and
immersion, amphipods, mostly large adult males, move
by crawling at the surface of the mud resulting in short
displacement; this behaviour is thought to be related to
mate finding (Boates and Smith, 1989; Boates et al., 1995;
Lawrie and Raffaelli, 1998a). During periods of immersion,
C. volutator individuals can leave their burrows and swim
in the water column (Hughes, 1988; Essink et al., 1989;
Lawrie and Rafaelli, 1998b) likely resulting in longer
dispersal distances. All size classes swim but small
juveniles and large adults are over-represented in the water
column (Hughes, 1988). Hughes (1988) estimated that
51
52
JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 29, NO. 1, 2009
MATERIALS AND METHODS
Study Site
The study was conducted on the intertidal mudflat of Pecks Cove in the
upper Bay of Fundy, Canada (see Peer et al. (1986) for maps of the region).
The area exhibits semi-diurnal tides with some of the highest amplitude in
the world (often higher than 13 m). The mudflat is about 3.5 km wide and
extends to a maximum of 1000 m from the line of high water. The entire
mudflat supports a dense population of Corophium volutator with densities
peaking at over 60,000 ind. m2. Data on diel time of high tide and tidal
amplitude were obtained from the website of the Department of Fisheries
and Oceans of Canada (www.waterlevels.gc.ca).
Sampling Methods
We sampled the water column using three stationary plankton nets that were
free to rotate with the net opening facing the water current. The nets were
deployed 1 m above the mudflat using metal posts sunk in the mud at
a distance of 150 m from shore, and separated from each other by 5 m. The
height of the nets was decided based on logistical constraints due to length
of the nets. Preliminary trials showed comparable numbers of amphipods
caught in nets set 1 m above the substrate compared to nets set 30 cm above
the substrate (unpublished data). The mesh size of the nets was 180 lm,
which is small enough to retain every size class of C. volutator (Crewe
et al., 2001). The nets were 45 cm long with a 20-cm diameter opening and
a 9-cm diameter funnel at the exit. Matter drifting in the water was collected
by a 500-ml plastic bottle that was fitted on the exit of the funnel. We cut
a slit on the side of the bottles to permit water to evacuate the bottle. The slit
was covered by 180-lm mesh to retain the matter.
Sampling was conducted as often as logistically possible over individual
periods of immersion between 18 May and 24 August 2006. Bottles were
set on the nets at low tide and recovered during the following low tide.
Material retained in the bottles was preserved in 95% ethanol. The number
of C. volutator was later quantified under dissecting microscope. When
a large quantity of matter was collected, we used a Folsom plankton splitter
to divide the samples until a reasonable amount of material was obtained.
The quantity sorted ranged from the full sample for small samples to 1/64
for especially large samples.
Based on intertidal height of the nets and predicted water levels, we
quantified the relationship between immersion time (h) and tidal amplitude
(m): immersion time ¼ 3.2 þ 3.2 ln (tidal amplitude) (non-linear
regression: F1,8 ¼ 234.43, P , 0.0001, R2 ¼ 0.97). During the course of
our study, immersion time varied from 2.2 to 4.9 h. We divided the number
of amphipods caught per net by immersion time to obtain a rate (amphipods
per hour). Although current speed may vary with tidal amplitude, we
unfortunately do not have these data for our site, and so cannot convert
number of amphipods caught to amphipod density (per m3). To check for
potential biases due to possibly varying water velocity during the lunar
cycle, we regressed the residuals of the final model against tidal amplitude.
If a larger amount of water filtered per unit time during spring tides is
causing an increase in number of amphipods caught (rather than actual
numbers swimming), then we should see an increase in residuals with
increasing tidal amplitude.
Statistical Analysis
We used multiple regression analysis to describe the temporal variation in
the number of swimming C. volutator caught per plankton net per hour of
immersion. The model included a term for the month at which samples were
collected, b1 Mt. The subscript t, used here and for other variables, refers to
the time at which a sample was collected. Mt was a ‘dummy’ variable
taking the value of 0 if the sample was collected in May and 1 for the rest of
the summer. We divided the spring-summer in these two periods because
juvenile amphipods were absent from the population in May, but present in
the other months. Furthermore, a preliminary analysis of variance indicated
that the number of amphipods varied significantly with month (ANOVA:
F3,81 ¼ 8.23, P , 0.0001). Tukey’s post-hoc test indicated that fewer
amphipods were caught in May compared to the other months, but that the
numbers did not differ significantly between June, July and August.
The model also included terms for time of day at which high tide
occurred; this variable was considered cyclical and was analyzed as in
a periodic regression (deBruyn and Meeuwig, 2001):
b2 sin ht þ b3 cos ht ;
where ht is the angular-transformed time of high tide (time/24 h 3 2p). A
significant cos ht regression represents a peak occurring at midnight or noon
(depending on the sign of the coefficient), whereas a significant sin ht
regression coefficient represents a peak occurring at 6:00 or 18:00
(depending on the sign). A combination of two significant coefficients
represents a peak whose timing depends on the sign and amplitude of the
coefficients.
Finally, the model included terms for day of the lunar calendar which
was also considered cyclical:
b4 sin rt þ b5 cos rt þ b6 sin 2rt þ b7 cos 2rt ;
where rt is the angular-transformed day of the lunar calendar (day/29 3
2p). A significant cos rt regression coefficient represents a peak occurring
at full moon or new moon (depending on the sign), whereas a significant sin
rt regression coefficient represents a peak at the first or last quarter
(depending on the sign). We included the terms sin 2rt and cos 2rt because
a preliminary analysis indicated that two peaks occurred during each lunar
cycle (Batschelet, 1981). A significant cos 2rt regression coefficient
represents two peaks occurring at the first and last quarter, or at full and new
moon (depending on the sign), whereas a significant sin 2rt regression
coefficient represents two peaks occurring between new moon and the first
quarter, or between full moon and the last quarter (depending on the sign).
Combinations of significant coefficients for the rt and 2rt terms permit
fine-tuning of when peaks occur and how high they are.
We decided not to include the interaction terms (i.e., Month Diel time,
Month Lunar phase, Diel time Lunar phase, and Month Diel time Lunar phase) since a preliminary analysis showed that they only slightly
increased the fit of the model, but made interpretation more difficult. Thus,
our full model was:
Yt ¼ l þ b1 Mt þ ðb2 sin ht þ b3 cos ht Þ
þ ðb4 sin rt þ b5 cos rt þ b6 sin 2rt þ b7 cos 2rt Þ;
where Yt is the number of C. volutator per hour of immersion at time t, l
is the intercept and b’s are the parameters to be estimated. The number
of amphipods per net per hour of immersion was transformed using
log10(datum þ 1) to correct for heterogeneity of variances and nonnormality of residuals. Graphical inspection of the residuals from the full
model, plotted over time (the 99 days of the study), did not reveal any serial
correlation. As well, there were no outliers: all residuals had an absolute
value less than 3.5 standard deviations (Draper and Smith, 1998).
Collinearity of independent variables was also not a problem, as indicated
by low variance inflation factors (VIF ’ 1; Draper and Smith, 1998). An
analysis with Akaike’s information criterion (AIC) was used to rank the
contribution of each term in the model (Burnham and Anderson, 2002).
RESULTS
The number of Corophium volutator caught in plankton nets
over one period of immersion was highly variable and
ranged between 0 6 0 (mean 6 SE) to 29963 6 2095 ind.
net1 (Figs. 1 and 2). We found obvious differences in the
number of C. volutator caught per net in relation to diel time
of high tide, with higher numbers caught when high tides
occurred at night; we thus plotted the values for night (high
tide occurring between 21:00 and 7:00) and day separately
(Figs. 1 and 2). There also seemed to be a semi-lunar cycle
in the number of C. volutator caught with peaks around time
of full moon and new moon.
All terms were included in the final model for the number
of C. volutator caught per net per hour of immersion, and
were significant at a ¼ 0.05, with the exception of the cos
(lunar phase) term (Table 1). Overall, the model accurately
predicted the number of swimming C. volutator per hour of
immersion (F7,234 ¼ 52.19, P , 0.0001, R2 ¼ 0.61, Fig. 3).
The number of swimming C. volutator was lower in May
compared to the rest of the spring-summer (June, July and
August). Using standard trigonometric identities and the
parameters estimated (Table 1), the model can be re-written
in terms of phase amplitude and phase shift of cosinusoidal
functions as:
53
DROLET AND BARBEAU: SWIMMING COROPHIUM VOLUTATOR IN BAY OF FUNDY
Fig. 1. Number of Corophium volutator (6 SE, n ¼ 3) caught in
stationary plankton nets during periods of immersion occurring at night
(21:00 – 07:00), in relation to lunar cycles (open circles are new moons and
full circles are full moons) and tidal amplitude (solid line) at Pecks Cove,
Bay of Fundy, Canada, in spring-summer 2006. Note the change in scale of
y-axis between the two panels.
Fig. 2. Number of Corophium volutator (6 SE, n ¼ 3) caught in
stationary plankton nets during periods of immersion occurring at daytime
(07:00 – 21:00), in relation to lunar cycles (open circles are new moons and
full circles are full moons) and tidal amplitude (solid line) at Pecks Cove,
Bay of Fundy, Canada, in spring-summer 2006. Note the change in scale of
y-axis between the two panels.
DISCUSSION
Yt ¼ 0:48 þ 1:28 Mt þ ½0:77 cosðht 0:45Þ
þ ½0:48 cosð2rt 0:62Þ þ ½0:16 cosðrt 1:11Þ;
where ht is the angular transformed diel time of high tide
and rt is the angular transformed day of lunar calendar. This
shows that log-transformed number of swimming C.
volutator per net per hour of immersion oscillated between
0.77 and 0.77 as a function of diurnal time of high tide
with a single peak at approximately 1:45 am (Fig. 4). The
log-transformed number of swimming C. volutator caught
per net per hour of immersion oscillated between 0.48 and
0.48 as a function of day of lunar calendar with two peaks
per cycle occurring one day after new moon and one day
after full moon (Fig. 5). Finally, the term for lunar cycle
with a single peak shows that the first peak (one day after
new moon) was slightly higher than the second peak
(Fig. 5). When the observed and predicted values are
transformed back to number of amphipods per net per hour
of immersion, rather than log10 (datum þ 1), we can
visualize the fit of the model (Fig. 6): the timing of the peaks
and the day/night variation are accurate. However, the
model under-predicts the amplitude of the highest peaks.
Based on AICs, we found that the factor that explained
most of the variation was Diel time of high tide, followed by
Month, Moon phase with two peaks per cycle, and finally
Moon phase with a single peak per cycle.
The residuals of the final model showed no relationship
with tidal amplitude (linear regression; F1,241 ¼ 0.51, P ¼
0.48; Fig. 7).
We described the temporal variation (between May and
August 2006) in the abundance of Corophium volutator
caught in stationary plankton nets set 1 m above the
substratum on the mudflat of Pecks Cove in the upper Bay
of Fundy, Canada. To our knowledge, this is the first study
confirming that populations of C. volutator on the coast of
North America swim during periods of immersion. During
peaks in activity, the number of C. volutator swimming is
very high (we captured about 30,000 individuals in 20-cm
diameter nets over one period of immersion). This has
profound implications on our understanding of the ecology
of C. volutator in the Bay of Fundy. The finding that large
numbers of C. volutator swim high above the substratum,
e.g., 1 m above the surface of the mudflat, reveals that largeTable 1. Parameter estimates of the regression model for the number of
Corophium volutator (Pallas, 1766) caught per plankton net per hour of
immersion as a function of month of sampling, diel time of high tide and
lunar phase. h is the angular-transformed diel time of high tide (time/24 3
2p) and r is the angular-transformed day of lunar calendar (day/29 3 2p).
The data were log10-transformed prior to the analysis.
Variable
Estimate (b’s)
SE
t-value
Intercept
Month
sin h
cos h
sin r
cos r
sin 2r
cos 2r
0.48
1.28
0.34
0.70
0.14
0.07
0.26
0.39
0.12
0.13
0.06
0.06
0.06
0.06
0.07
0.05
4.00
10.11
5.53
12.23
2.37
1.08
3.97
7.06
p
,
,
,
,
0.0001
0.0001
0.0001
0.0001
0.02
0.28
0.0001
, 0.0001
54
JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 29, NO. 1, 2009
Fig. 3. Plot of observed number of Corophium volutator caught in
stationary plankton nets per hour of immersion against numbers predicted
by the model, showing the fit of the model.
scale dispersal is an important process. This further suggests
that areas will be recolonized rapidly following small-scale
disturbance (although swimming might not be sufficient to
rapidly compensate for large-scale disturbance; Shepherd
and Boates, 1999).
Although we did not quantify water velocity directly, and
so did not calculate density of swimming amphipods, we are
confident that our rate (amphipods caught per hour of
immersion) provides a good estimation of density. This is
supported by the lack of relationship between the residuals
of our model and tidal amplitude (Fig. 7). Water velocity
thus would not explain any more variation not already
explained by the model. Also, although peaks in swimming
activity occurred during periods of new and full moon, the
Fig. 4. Effect of diel time of high tide on the number of swimming
Corophium volutator caught per net per hour of immersion. Residuals were
calculated by subtracting, from the observed numbers of amphipod caught,
the numbers predicted by a partial model that excluded the terms for diel
time of high tide. Dots represent observed residuals and solid line represents
prediction of the partial model including only terms for diel time of high
tide.
Fig. 5. Effect of lunar cycle on the number of swimming Corophium
volutator caught per net per hour of immersion. Residuals were calculated
by subtracting, from the observed numbers of amphipod caught, the
numbers predicted by a partial model that excluded the terms for lunar
cycle. Dots represent observed residuals and solid line represents prediction
of the partial model including only terms for lunar cycle.
highest tidal amplitude did not coincide with these periods
(Figs. 1 and 2). This again supports the idea that swimmers
are denser during new and full moon, and that the observed
pattern is not a consequence of stronger currents and larger
amounts of water filtered by the nets.
Fig. 6. Predicted (dashed line) and observed (dots, mean 6 SE, n ¼ 3)
number of swimming Corophium volutator caught per net per hour of
immersion in Pecks Cove, Bay of Fundy, Canada, in spring-summer 2006.
Note the change in scale of y-axis for the different panels.
DROLET AND BARBEAU: SWIMMING COROPHIUM VOLUTATOR IN BAY OF FUNDY
We found a significant reduction in the abundance of C.
volutator swimming in the water column in May compared
to June, July and August (Figs. 1 and 2, Table 1). This
compares well to Hughes (1988), who reported that
swimming only occurs during the summer months, and to
Holmstrom and Morgan (1983), who reported less well
defined cycles of swimming activity during the winter. The
reason for such seasonal variation is likely due to the
demography of C. volutator. Density of C. volutator is at
a minimum in May in the Bay of Fundy following suspected
intense winter mortality (Partridge, 2000). The first reproduction event usually occurs in mid to late May,
resulting in a significant increase in density in early June
and throughout the summer (Murdoch et al., 1986; Peer
et al., 1986). Furthermore, swimming behaviour appears
size-specific, with juveniles being over-represented in the
water column (Hughes, 1988; Lawrie and Rafaelli, 1998b).
In sum, the presence of high densities, skewed towards
juvenile size classes on the mudflat in the summer likely
explains the higher abundance of swimmers. We are
currently investing spatio-temporal variation in the size
structure of swimming C. volutator, as well as comparing
this size structure to that of burrowed amphipods.
Similar to European studies, swimming activity at our
study site is more important when high tides occur at night
(Figs. 1, 2 and 4). However, swimming also occurred during
the day, which is consistent with Holmstrom and Morgan
(1983) and Hughes and Horsfall (1990), but not Hughes
(1988). Therefore, there seems to be geographical differences on the diel cycle in activity of C. volutator. Previous
researchers suggested that swimming mostly at night was an
adaptation to reduce predation risk (Hughes, 1988; Lawrie
and Rafaelli, 1998b), as visual predators are less efficient
foragers at night (James and Heck, 1994). However, the
water in the upper Bay of Fundy is heavily loaded with
sediment resulting in Secchi depth of less than 10 cm.
Known predators of C. volutator in the Bay of Fundy either
are shorebirds that feed on exposed mud (Hicklin and Smith,
1984), benthic feeding fish that ingest C. volutator by
swallowing sediment (McCurdy et al., 2005) or pelagic fish
that rely on visual cues to locate prey (Gillmurray and
Daborn, 1981). So, it is unlikely that light intensity
influences predation risk encountered by swimming C.
volutator in the Bay of Fundy. Increased nocturnal
swimming in C. volutator might be related to aquatic
predators being less active at night or to some yet
unidentified reason.
On the mudflat of Pecks Cove, C. volutator swims mostly
around the time of full and new moons (Fig. 5). This
influence of the lunar cycle on the swimming activity of C.
volutator was previously reported in Europe (Holmstrom
and Morgan, 1983; Hughes, 1988; Lawrie and Rafaelli,
1998b). However, the timing of the cycle appears to differ
between different sites. Hughes (1988) reported peaks in
activity during spring tides, whereas Lawrie and Rafaelli
(1998b) reported peaks during the transition between spring
to neap tide. Increased tidal amplitude around new and full
moons might be the cue triggering swimming in C.
volutator as proposed by Morgan (1965).
55
Fig. 7. Residuals of the full model plotted against tidal amplitude,
showing the lack of extra explanatory power of tidal amplitude.
We clearly demonstrated that swimming of C. volutator
occurs in the Bay of Fundy. Based on month, diel time of
high tide and day of the lunar calendar, we developed
a statistical model that explains 61% of the total variation
(Figs. 3 and 6). This model will be useful in future attempts
to model the spatial population dynamics of C. volutator at
Pecks Cove. However, important issues remain unresolved,
including the spatial variation in abundance of swimmers,
the direction of swimming, the distance traveled by
swimmers, the relationship between density on the mudflat
and the abundance of swimmers, and finally the relationship
between abundance of swimmers and density of individuals
settling on the mudflats. We are currently investigating most
of these questions. Similar surveys at other mudflats are
required to generalize our results to the entire Bay of Fundy.
In addition, we need to determine whether swimming occurs
only over the mudflats or if it also occurs between different
mudflats. This would determine whether mudflats should be
viewed as open or closed populations of C. volutator.
ACKNOWLEDGEMENTS
We are grateful to A. Botta, M. Coffin, and J. Horsfall for their assistance in
the field and laboratory, to D. J. Hamilton for logistical help and use of
laboratory space at Mount Allison University, and to J. A. Kershaw and
T. R. Turner for statistical advice. This manuscript benefited from
comments by D. J. Hamilton, H. L. Hunt, J. Watmough, and an anonymous
reviewer. The research was founded by a National Science and Engineering
Research Council (NSERC) Discovery Grant to M. A. B., Science Horizons
Internships from Environment Canada, and the Work-Study Program at the
University of New Brunswick (UNB). D.D. received financial support from
a NSERC Postgraduate Scholarship, a Board of Governors Merit Award
(UNB) and Marguerite and Murray Vaughan Fellowships in Marine
Sciences (UNB).
REFERENCES
Batschelet, E. 1981. Circular statistics in biology. Academic Press, New
York.
Boates, J. S., M. R. Forbes, M. Zinck, and N. McNeil. 1995. Male
amphipods (Corophium volutator [Pallas]) show flexible behaviour in
relation to risk of predation by sandpipers. Ecoscience 2: 123-128.
———, and P. C. Smith. 1989. Crawling behaviour of the amphipod
Corophium volutator and foraging by semipalmated sandpipers, Calidris
pusilla. Canadian Journal of Zoology 67: 457-462.
56
JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 29, NO. 1, 2009
Burnham, K. P., and D. R. Anderson. 2002. Model selection and
multimodel inference, second edition. Springer-Verlag, New York.
Crewe, T. L., D. J. Hamilton, and A. W. Diamond. 2001. Effects of mesh
size on sieved samples of Corophium volutator. Estuarine, Coastal and
Shelf Science 53: 151-154.
deBruyn, A. M. H., and J. J. Meeuwig. 2001. Detecting lunar cycles in
marine ecology: periodic regression versus categorical ANOVA. Marine
Ecology Progress Series 214: 307-310.
Draper, N. R., and H. Smith. 1998. Applied regression analysis. John
Wiley-Interscience, New York.
Drolet, D., J. H. Himmelman, and R. Rochette. 2004. Effect of light and
substratum complexity on microhabitat selection and activity of the
ophiuroid Ophiopholis aculeata. Journal of Experimental Marine Biology and Ecology 313: 139-154.
Essink, K., H. F. Kleef, and W. Visser. 1989. On the pelagic occurrence and
dispersal of the benthic amphipod Corophium volutator. Journal of the
Marine Biological Association of the United Kingdom 69: 11-15.
Gaudette, J., R. A. Wahle, and J. H. Himmelman. 2006. Spawning events in
small and large populations of the green sea urchin Strongylocentrotus
droebachiensis as recorded using fertilization assays. Limnology and
Oceanography 51: 1485-1496.
Gillmurray, M. C., and G. R. Daborn. 1981. Feeding relations of the
Atlantic silverside Menidia menidia in the Minas Basin, Bay of
Fundy. Marine Ecology Progress Series 6: 231-235.
Hicklin, P. W., and P. C. Smith. 1984. Selection of foraging sites and
invertebrate prey by migrant semipalmated sandpipers, Calidris pusilla
(Pallas), in Minas Basin, Bay of Fundy. Canadian Journal of Zoology
62: 2201-2210.
Holmstrom, W. F., and E. Morgan. 1983. Variation in the naturally
occurring rhythm of the estuarine amphipod, Corophium volutator
(Pallas). Journal of the Marine Biological Association of the United
Kingdom 63: 833-850.
Hughes, R. G. 1988. Dispersal by benthic invertebrates: The in situ
swimming behaviour of the amphipod Corophium volutator. Journal of
the Marine Biological Association of the United Kingdom 68: 565-579.
———, and I. M. Horsfall. 1990. Differences in the swimming behaviour
of the amphipod Corophium volutator from different populations. Journal of the Marine Biological Association of the United Kingdom 70: 143148.
James, P. L., and K. L. Heck. 1994. The effects of habitat complexity and
light intensity on ambush predation within a simulated seagrass
bed. Journal of Experimental Marine Biology and Ecology 176: 187200.
Lawrie, S. M., and D. Rafaelli. 1998a. Activity and mobility of Corophium
volutator: a field study. Marine and Freshwater Behaviour and
Physiology 31: 39-53.
———, and ———. 1998b. In situ swimming behaviour of the amphipod
Corophium volutator (Pallas). Journal of Experimental Marine Biology
and Ecology 224: 237-251.
McCurdy, D. G., J. S. Boates, and M. R. Forbes. 2000. Reproductive
synchrony in the intertidal amphipod Corophium volutator. Oikos 88:
301-308.
———, M. R. Forbes, S. P. Logan, D. Lancaster, and S. I. Mautner. 2005.
Foraging and impacts by benthic fish on the intertidal amphipod
Corophium volutator. Journal of Crustacean Biology 25: 558-564.
Meireles, S., and H. Queiroga. 2004. Use of artificial collectors shows
semilunar rhythm of planktonic dispersal in juvenile Hydrobia ulvae
(Gastropoda: Prosobranchia). Journal of the Marine Biological Association of the United Kingdom 84: 761-766.
Morgan, E. 1965. The activity rhythm of the amphipod Corophium
volutator (Pallas) and its possible relationship to changes in hydrostatic
pressure associated with the tides. Journal of Animal Ecology 34: 731746.
Murdoch, M. H., F. Bärlocher, and M. L. Laltoo. 1986. Population
dynamics and nutrition of Corophium volutator (Pallas) in the Cumberland Basin (Bay of Fundy). Journal of Experimental Marine Biology and
Ecology 103: 235-249.
Norkko, A., V. J. Cummings, S. F. Thrush, J. E. Hewitt, and T. Hume.
2001. Local dispersal of juveniles bivalves: implications for sandflat
ecology. Marine Ecology Progress Series 212: 131-144.
Pallas, P. S. 1766. Elenchus Zoophytorum Sistens Generum Adumbrationes
Generaliores et Specierum Cognitarum Succinctas Descriptiones cum
Selectis Auctorum Synomymis- pp. (1-28), 1-451. Hagæ Comitum. (F.
Varrentrapp).
Partridge, V. A. 2000. Aspects of the winter ecology and spring
recolonization of the Windsor mudflat. M.Sc. thesis, Acadia University,
Wolfville, Nova Scotia, Canada.
Peer, D. L., L. E. Linkletter, and P. W. Hicklin. 1986. Life history and
reproductive biology of Corophium volutator (Crustacea: Amphipoda)
and the influence of shorebird predation on population structure in
Chignecto Bay, Bay of Fundy, Canada. Netherlands Journal of Sea
Research 20: 359-373.
Shepherd, P. C. F., and J. S. Boates. 1999. Effects of a commercial
baitworm harvest on semipalmated sandpipers and their prey in the Bay
of Fundy hemispheric shorebird reserve. Conservation Biology 13: 347356.
Tilman, D., and P. Kareiva. 1997. Spatial Ecology. The role of space in
population dynamics and interspecific interactions. Princeton University
Press, Princeton.
Turchin, P. 1991. Translating foraging movement in heterogeneous
environments into the spatial distribution of foragers. Ecology 72:
1253-1266.
———. 1998. Quantitative analysis of movement: Measuring and
modelling population redistribution in animals and plants. Sinauer
Associates, Sunderland.
Watkin, E. E. 1942. The yearly life-cycle of the amphipod Corophium
volutator. Journal of Animal Ecology 10: 77-93.
RECEIVED: 20 March 2008.
ACCEPTED: 17 July 2008.