|
| |
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
First published online June 29, 2006
Journal of Experimental Biology 209, 2713-2725 (2006)
Published by The Company of Biologists 2006
doi: 10.1242/jeb.02315
Multidimensional analysis of suction feeding performance in fishes: fluid speed, acceleration, strike accuracy and the ingested volume of water
1 Section of Evolution and Ecology, University of California, One Shields
Avenue, Davis, CA 95616, USA
2 Department of Mechanical Engineering, Rochester Institute of Technology,
76 Lomb Memorial Drive, Rochester, NY 14623-5604, USA
* Author for correspondence (e-mail: tehigham{at}ucdavis.edu)
Accepted 3 May 2006
 |
Summary |
|---|
 TOP Summary IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Key words: volume, Centrarchidae, Lepomis, Micropterus, swimming, ram, kinematics, prey capture, feeding, DPIV, ingested volume, accuracy, suction feeding, performance
|
Introduction |
|---|
|
TOPSummaryIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
; Muller and Osse,
1984; Norton and Brainerd,
1993; Van Wassenbergh et al.,
2005; Wainwright et al.,
2001).
One commonly cited metric of suction feeding performance is the maximum
speed of the water entering the mouth cavity. Fluid speed is attractive as a
metric of suction feeding performance because the drag experienced by a prey
item in the suction flow will be proportional to the square of fluid speed
relative to the prey (i.e. difference between the speed of the prey and the
speed of the fluid surrounding the prey) and because a rapid flow presumably
limits the time for prey escape. Variation in fluid speed has been inferred
indirectly by tracking the prey movement towards the predator
(Norton and Brainerd, 1993;
Cook, 1996;
Van Leeuwen, 1984;
Wainwright et al., 2001;
Waltzek and Wainwright, 2003)
or by measuring the magnitude of subambient pressure in the buccal cavity
(Lauder, 1980;
Lauder, 1983;
Lauder et al., 1986;
Nemeth, 1997;
Sanford and Wainwright, 2002;
Carroll et al., 2004). A few
studies have directly quantified the flow speeds generated by suction feeders
(Muller and Osse, 1984;
Van Leeuwen, 1984;
Ferry-Graham et al., 2003;
Day et al., 2005;
Higham et al., 2005a), but to
date there has been no attempt to measure the peak fluid speeds that an
individual or species is capable of generating. Acceleration of the fluid may
also contribute to the effectiveness of suction feeding because of the
acceleration reaction force that it generates and exerts on the prey in the
flow. Although Drost et al. estimated the acceleration of fluid during suction
feeding in larval fish (Drost et al.,
1988), the potential importance of fluid acceleration has been
ignored in most recent considerations of suction feeding performance (e.g.
Carroll et al., 2004;
Van Wassenbergh et al., 2005;
Wainwright et al., 2001).
While the speed and acceleration of the fluid entering the mouth generate
forces and generally limit the time available to the prey for escape, there
are several features of the ingested volume of water that could influence
performance of the strike (Wainwright et
al., 2001). The larger the volume of water ingested by a fish
during the strike, the less chance the prey item has of escaping the flow.
While the flow of water entering the mouth can be modulated by swimming speed
(Muller and Osse, 1984;
Higham et al., 2005a) and how
fast the buccal cavity is expanded (Day et
al., 2005), little is known regarding the modulation of the shape,
size and flow of the ingested volume of water. For example, although it is
clear that an increase in swimming speed will result in a narrower and more
elongate ingested volume of water (Weihs,
1980; Higham et al.,
2005a), it is not clear how ram speed influences the total volume
of ingested water or the flow rate of this volume. Most studies have either
measured volume indirectly from morphology
(Cook, 1996;
De Visser and Barel, 1998;
Viladiu et al., 1999;
Van Wassenbergh et al., 2005)
or from geometric estimates of volume change of the head during the strike
(Van Leeuwen, 1984). With
fishes, these methods underestimate the total volume of water ingested during
the strike, because while fish ingest water through their buccal cavity they
expel water out a caudal valve located at the posterior end of the opercular
cavity. This ability to expel water while ingesting more enables fish to
circumvent the constraints of buccal cavity volume, which would otherwise
limit the ingested volume (Muller and
Osse, 1984; Van Leeuwen,
1984; Day et al.,
2005). In fact, fish species such as rainbow trout and bluegill
sunfish can ingest a volume of water much greater than the volume of their
buccal cavity (Van Leeuwen,
1984; Day et al.,
2005). In addition, fish are able to maintain a flow of water
until the mouth is almost closed (Day et
al., 2005; Higham et al.,
2005a) which likely decreases the chances of the prey escaping.
Understanding how fish can manipulate this ingested volume of water may give
new insights into suction feeding performance.
An understudied aspect of suction feeding performance is the ability to
accurately direct the flow of water that is generated external to the fish's
mouth. Accurate positioning of the mouth relative to the prey item is
essential for prey capture. Further, the suction flow is ephemeral, requiring
the fish to time its placement to maximize the effectiveness of the flow.
Several factors including approach speed, timing of mouth opening and mouth
size are likely important for determining the accuracy of prey capture. The
accuracy of a predator is typically quantified as the number of successful
attempts relative to the number of failed attempts (e.g.
Nyberg, 1971;
Drost, 1987;
McLaughlin et al., 2000). This
approach is designed to compare the ability of different species to feed on a
particular prey type. However, it would be useful to design a metric of
accuracy that relates to the ability of fish to position prey within the flow
of water that they generate during suction feeding.
In this study we adopt a multidimensional view of suction feeding performance in a comparative analysis of two species that have been the focus of a tremendous amount of research on feeding functional morphology and ecology, the largemouth bass, Micropterus salmoides and the bluegill sunfish, Lepomis macrochirus. Using digital particle image velocimetry (DPIV) we measure the peak fluid speeds and local acceleration generated by these species feeding on elusive prey in both the earthbound and fish's frames of reference. We measure the volume of water captured during feeding and the rate of volume flow entering the mouth. Finally, we introduce a new method for quantifying strike accuracy and use this method to compare the two species. Based on the fluid speeds calculated in the earthbound frame of reference, our results strongly confirm previous interpretations of feeding performance in these species. We find that bluegill generate higher fluid speeds and accelerations, and are more accurate with their strike. Largemouth bass ingest a larger volume of water during the strike and generate higher volumetric flow rates. Because of their higher swimming speed, largemouth bass actually generate higher fluid speeds than bluegill sunfish in the fish's frame of reference, allowing them to close more quickly on the prey item.
|
Materials and methods |
|---|
|
TOPSummaryIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
;
Collar et al., 2005). These
genera (Lepomis and Micropterus) are monophyletic and sister
taxa with a most recent common ancestor estimated about 24 million years ago
(Near et al., 2005). The fish
were collected in Yolo County, California, USA, brought back to the University
of California, Davis and housed individually in 100-liter aquaria at 22°C.
Fish were maintained on a diet of cut squid (Loligo sp.), goldfish
(Carassius auratus), ghost shrimp (Palaemonetes sp.), and/or
small annelid `tubifex' worms. All maintenance and experimental procedures
used in this research followed a protocol that was reviewed by the University
of California, Davis Institutional Animal Care and Use Committee. We analyzed
data from three bluegill sunfish with standard lengths of 15.3 cm, 15.0 cm and
15.4 cm, and from three largemouth bass with standard lengths of 16.6 cm, 17.7
cm and 18.0 cm. We emphasize that the comparisons we draw between the two
species are therefore restricted to fish in a narrow size range. It is likely
that many of the parameters we measured will change with body size in these
species (Carroll et al.,
2004).
Experimental protocol
Each fish was placed in the experimental tank and trained to feed in the
laser sheet (see below). At the onset of experiments, the individual was kept
at one end of the tank and restrained behind a door [see
fig. 1 in Higham et al.
(Higham et al., 2005a)]. A
tubifex worm (about 2 cm), ghost shrimp (about 2 cm) or goldfish (about 2-3
cm) was then introduced via plastic tubing or attached to a thin
wire. The prey was held within the laser light sheet and within the camera
field of view, and the door was lifted. Although the tubifex worms were not
attached to something, they were barely moving in the field of view.
|
;
Day et al., 2005;
Higham et al., 2005a). An
Innova-90 5 W argon-ion continuous wave laser (Coherent, Inc., Santa Clara,
CA, USA) was used in combination with a set of focusing lenses and mirrors to
produce a vertical laser sheet that was approximately 10 cm wide and 1 mm
thick in the aquarium. To visualize the flow of water, the aquarium was seeded
with 12 µm silvercoated, hollow glass spheres (Potter Industries, Inc.,
Carlstadt, NJ, USA) with a specific gravity of 1.05. Mirrors above and below
the tank were used to illuminate both above and below the head of the fish
during feeding. Lateral-view video sequences were recorded using a NAC
Memrecam ci digital system (Tokyo, Japan) operating at 500 images
s-1. The field of view ranged from 5.1x6.7 cm to 8.4x11
cm, depending on the species. Additionally, a Sony CCD camcorder (Tokyo,
Japan), operating at 30 images s-1, was used to capture anterior
view images for each sequence in order to determine the orientation and
position of the fish relative to the laser sheet. While we only analyzed
sequences recorded in lateral view in this study, we have found that the flow
pattern is approximately radially symmetric about the long axis of the fish
(Day et al., 2005).
An adaptive mesh cross correlation algorithm
(Scarano and Riethmuller,
1999) was used to calculate velocities from image pairs. The
distance that particles traveled between image pairs (2 ms interval) was
determined within interrogation windows of varying dimensions, depending on
the species (e.g. bluegill=0.9x0.9 mm, bass=0.13x0.13 mm), with
50% overlap between interrogation windows. The algorithm then returned a
two-dimensional grid of two components of measured velocity for each image
pair that was processed.
The instantaneous velocity measurement at a single measurement point
contains a random error that we measured using published methods
(Day and McDaniel, 2005).
Assuming adequate density of seed particles, uncertainty due to random errors
is a function of the interrogation window size and the diameter of an
individual particle's images. The conditions in our experiments lead to a
random error of approximately 5%.
A transect extending forward from the center of the fish's mouth was studied to measure the speed of the fluid entering the mouth. The closest position to the mouth where accurate measurements of velocity vectors were made in 100% of the sequences was at a distance away from the mouth aperture equal to one half of the peak gape diameter (PG) of the fish for the feeding sequence. The measurement at this position was validated in every trial. All velocities reported in this paper are at this distance and on the centerline, and we refer to the magnitude of these velocities as `fluid speeds'.
Frames of reference
The frame of reference is an important factor when quantifying fluid speeds
during suction feeding (Muller and Osse,
1984; Higham et al.,
2005a). An increase in ram velocity will increase the fluid speed
relative to the fish. Thus, we measured the fluid speeds in the earthbound
frame of reference directly and then calculated fluid speed in the fish's
frame of reference. To calculate the latter, we added the magnitude of the
forward velocity of the fish to the fluid speed generated by suction. The
forward velocity of the fish was always calculated in a direction towards the
prey item. We note that this method of changing frame of reference is
appropriate for fluid velocity measurements on the centerline, where all water
movement is in this axis, but it would not generalize to positions away from
the centerline where water movement also has a y and z
component. As explained below, certain variables are expressed in both frames
of reference while others are only calculated in one of them.
Data analysis
Only those sequences in which the laser sheet intersected the mid-sagittal
plane of the fish (verified with the anterior view camera) and in which the
fish were centered on the filming screen in lateral view were used for
analyses. Using IMAGE J version 1.33 (NIH, Washington, DC, USA), the
x and y coordinates of the tip of the upper and lower jaws
were digitized for each image (2 ms intervals) starting prior to the onset of
mouth opening and continuing until the mouth was closed. These points were
used to calculate gape distance as a function of time and to determine the
value of peak gape for each sequence. Time to peak gape (TTPG) was
measured as the time from 20% to 95% of maximum gape
(Sanford and Wainwright, 2002;
Day et al., 2005;
Higham et al., 2005a). This
method reduces errors that are related to a variable rate of early mouth
opening and the difficulty in clearly identifying the point where the peak
value is reached in an asymptotic relationship. TTPG was measured as
an indicator of the rate of buccal expansion
(Sanford and Wainwright,
2002). In order to determine the ram velocity during feeding, we
first digitized the x and y coordinates of the anterior
margin of the eye for each frame. Ram velocity was the first derivative of the
displacement of the eye. Ram velocity varied throughout the strike and the
speeds reported in this paper, with the exception of the volumetric flow rate
calculation (see below), are those measured at the time of 95% of maximum
gape, which approximates the time of maximum fluid speed. The temporal pattern
of kinematic events and fluid speeds was investigated as was done previously
for the bluegill sunfish (Day et al.,
2005). The relative timing of key kinematic events was determined
manually from the graphed profiles of each feeding and the mean and standard
deviation of timing of these was calculated as a percentage of
TTPG.
To determine the volume of the ingested parcel of water (V), we
manually tracked particles going into the mouth using IMAGE J. Any particle
that entered the mouth between mouth opening and mouth closing was considered
ingested. We then defined a boundary around the outer limit of particles (in
the frame at the onset of mouth opening) that entered the mouth and digitized
several points (>20) along this boundary
(Fig. 1). Assuming the flow
field was symmetric about the long axis of the fish
(Day et al., 2005), we
calculated the total volume of ingested water by integration of the
two-dimensional boundary.
To determine the rate of volume ingestion (dV/dt), we
initially employed the methods described above in a stepwise fashion to
determine the volume at several times throughout the gape cycle. For the same
sequences, we calculated dV/dt by multiplying the area of
the mouth aperture by the speed of the fluid (at the mouth aperture) entering
the mouth at intervals of 2 ms. Although we measured fluid speeds at a
distance equal to 
maximum gape away from the aperture, we converted
these values to fluid speed at the mouth aperture by multiplying by 3.6
(bluegill) and 4.6 (bass), which follows the methods employed in earlier work
(Day et al., 2005;
Higham et al., 2005a). The
volumetric flow rate was only calculated in the fish's frame of reference, so
the fluid speeds described above were added to the magnitude of the forward
velocity of the fish to determine the fluid speed for the calculation of
dV/dt relative to the mouth aperture. The forward velocity
of the fish was the average of the velocity at the onset of the strike and the
velocity at the time of maximum gape. From the relationship of
dV/dt versus time, we determined the maximum
dV/dt value and the time of this maximum for a comparison
with the timing of kinematic variables. The two methods for calculating
dV/dt generated similar results so we employed the latter
method for the remainder of the sequences.
The methods for determining the shapes of the ingested volume of water are
discussed in greater detail elsewhere
(Higham et al., 2005a). In
short, we measured the maximum height and the length of the boundary described
above and converted the measurements to an aspect ratio of the ingested volume
in lateral view. The ingested volume was more narrow and elongate as values of
this ratio decreased.
To determine the accuracy of the strike, we digitized points along the edge of the ingested volume and used these to determine the center of the parcel of water (COP; Fig. 1). For the shrimp and tubifex worms, the center of mass (COM) was the best-approximation of the center of the body. For the goldfish prey, the COM was located slightly dorsal and posterior to the pectoral fin insertion site on the body. All measurements of the COM of the prey were made at the onset of the strike prior to any movement. We then calculated the straight-line distance from the COP to the edge of the ingested volume, with the line passing through the estimated center of mass (COM) of the prey item. In order to determine where along this line the prey was, we calculated the straight-line distance from the COP to the COM of the prey. The accuracy index (AI) was defined as AI=1-(distance to prey/distance to volume boundary) such that higher values of AI indicate the prey item was closer to the COP. Values of 1 indicate the prey is located at the COP (i.e. the value within the parentheses above would be zero). If the prey were located on the boundary of the ingested volume, the AI would equal zero (i.e. the value within the parentheses above would be one).
In addition to this overall metric of accuracy, we measured the vertical (relative to the x-axis) and the horizontal (relative to the y-axis) accuracies for each strike (Fig. 1). We first calculated the distance between the COM of the prey item and either the x-axis (Ay) or the y-axis (Ax) (Fig. 1). We then calculated the straight-line distance from the boundary of the ingested parcel of water to the axis of interest (going through the COM of the prey). The accuracy relative to the axis of interest was defined as Ax or y=1-(distance to prey from axis/distance to boundary from axis). Values of 1 indicate the prey item is located on the axis of interest and values of zero indicate that the prey is located on the boundary of the ingested parcel of water.
Maximum fluid speed, measured at a distance of of maximum gape
away from the center of the mouth, was quantified for a large number (96) of
sequences for bluegill sunfish and largemouth bass. The purpose of this was to
determine the average maximum fluid speed for each species under a variety of
feeding situations.
In a separate analysis to characterize maximum suction feeding performance, we selected the maximum value for fluid speed and fluid acceleration for each individual using fluid speeds in the earthbound and fish's frames of reference. The acceleration we are measuring is the local acceleration of the fluid since our measurements are at a single location. Thus, acceleration in our study refers to the local acceleration of the fluid. We also selected the maximum value for dV/dt, which was only calculated in the fish's frame of reference. The local acceleration of the fluid in the earthbound frame of reference was calculated by dividing the maximum fluid speed by the time to peak fluid speed (TTPFS), which is the time from 20% of peak gape to the time of maximum fluid speed. To calculate local fluid acceleration in the fish's frame of reference, we added the average ram speed of the strike to the maximum fluid speed and then divided this by TTPFS. To characterize the accuracy for the maximum performance strikes, we first selected the three sequences per individual that exhibited the highest fluid speeds. For each individual, we then averaged the three values of accuracy for these strikes.
Statistical analyses
Prior to performing any statistical analyses, we log10
transformed all of the variables (with the exception of the accuracy index) in
order to normalize variances. For the accuracy index, we used an arcsin
transformation. For each species separately, we performed mixed-model multiple
regressions in order to determine the effects of TTPG, ram speed and
maximum gape on the following dependent variables: (1) total volume of
ingested water, (2) maximum rate of volume flow (dV/dt), and
(3) the accuracy index (AI). We performed analyses of variance
(ANOVAs) in order to determine the effects of species on several variables.
For these analyses, the independent variables were species (fixed) and
individual (nested within species; random). In order to correct for multiple
statistical tests, 
(0.05) was adjusted using a sequential Bonferroni
test (Rice, 1989). We used
SYSTAT version 9 (SPSS Inc., Chicago, IL, USA) for all statistical
analyses.
|
Results |
|---|
|
TOPSummaryIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
; Higham et al.,
2005a). Largemouth bass had an average maximum gape twice that of
bluegill sunfish (Table 1). The
average swimming velocity at the time of prey capture was substantially higher
in largemouth bass than bluegill sunfish
(Table 1). The average time to
peak gape (TTPG), however, did not differ significantly between the
two species. For both species, fluid speed FS (in both the earthbound
and fish's frame of reference) decreased as a function of TTPG,
although there was more variation in FS independent of TTPG
for bass than for bluegill (Fig.
2). Bluegill sunfish generated higher average peak fluid speeds
than largemouth bass in the earthbound frame of reference, but the opposite
result was observed for fluid speeds in the fish's frame of reference.
|
|
As was done for bluegill sunfish (Day et
al., 2005), a mean scaled velocity profile was found by fitting a
fourth order polynomial to 58 pooled feedings from all bass
(r2=0.985), as shown in
Fig. 3. The s.d. of residuals
of scaled fluid speeds SFS about the mean scaled velocity profile are
shown as error bars in the figure, SFSpooled-bluegill
=0.348x4-2.49x3+6.61x2-7.78x+
3.56 (see Day et al., 2005),
SFSpooled-bass
=0.986x4-5.80x3+12.53x2-12.07x+4.59.
|
Characteristics of the ingested volume
Largemouth bass ingested a significantly larger volume of water
(V) than bluegill sunfish (Table
1), and ingested the volume of water at a faster rate as indicated
by the rate of volume change (dV/dt;
Table 1;
Fig. 4). For both largemouth
bass and bluegill, approximately 50% of the total V was ingested by
the time of maximum gape (Fig.
4). Furthermore, maximum dV/dt occurred at
approximately the time of maximum gape. This is almost coincident with the
timing of peak fluid speed, as measured at a distance of peak gape
(Figs 4,
5). In general, the temporal
pattern of key kinematic and fluid mechanical events is very similar for the
two species (Fig. 5), with one
exception being that bass tend to ingest their prey relatively later in the
mouth opening sequence than bluegill sunfish. The time of peak fluid speed is
nearly simultaneous with the onset of peak gape for both species (Figs
4,
5).
|
|
For both species, larger values of peak gape resulted in a significantly greater volume of ingested water, and a faster rate of volume ingestion (Fig. 6). In contrast, neither species exhibited a significant relationship between time to peak gape (TTPG) and the volume of water ingested (Fig. 7A). For bluegill, and not largemouth bass, the rate of volume flow into the mouth was significantly higher when the TTPG was reduced (Fig. 7B). Finally, for both species, an increase in ram velocity resulted in a significantly greater ingested volume of water (Fig. 8).
|
|
|
The average height to length ratio of the ingested volume of water was
similar in the two species (Table
1; Fig. 9). The
height and length were, on average, approximately equal. With an increase in
ram velocity, the ingested volume of water became significantly more elongate
and narrow in largemouth bass (Fig.
9). This relationship was also found in bluegill sunfish, and is
presented elsewhere (Higham et al.,
2005a).
|
|
Suction feeding performance
From the sequences used for these analyses, bluegill sunfish generated
significantly higher fluid speeds and higher magnitudes of acceleration in the
earthbound frame of reference, and exhibited significantly greater strike
accuracy compared to largemouth bass (Table
2; Fig. 11).
However, largemouth bass generated higher, but not significantly different,
fluid speeds in the fish's frame of reference
(Table 2;
Fig. 11A). The maximum
volumetric flow rate was significantly higher for bass compared to bluegill
(Table 2;
Fig. 11B).
|
|
|
Discussion |
|---|
|
TOPSummaryIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
), a fact that has been interpreted as reflecting a greater
scope for fluid speed. The larger mouth and buccal cavity of bass has been
suggested to be well-suited to feeding on the larger and more elusive fishes,
crayfish and shrimp that dominate their diet
(Werner, 1977;
Norton and Brainerd,
1993).
Fluid speed and acceleration
The speed of the water entering the mouth has generally been considered the
most important measure of suction feeding performance
(Muller et al., 1982;
Van Leeuwen and Muller, 1984;
Ferry-Graham et al., 2003;
Higham et al., 2005a), but the
technical difficulty associated with measuring fluid speed directly has slowed
progress in this area. This study represents the first direct measurement and
comparison of maximum fluid speeds between species. Bluegill sunfish generated
higher fluid speeds in the earthbound frame of reference than largemouth bass
(Table 1), confirming the
interpretation made previously that the greater capacity for subambient buccal
pressure in bluegill indicated a greater scope for fluid speed
(Carroll et al., 2004).
However, in the fish's frame of reference, largemouth bass actually generated
higher fluid speeds than bluegill sunfish. Largemouth bass achieve this by
exhibiting higher ram speeds than bluegill
(Table 1). Thus, it seems that
bass would benefit from feeding in open areas where they could maximize their
approach speed and thus maximize the fluid speeds they generate during suction
feeding. In a habitat that precludes high attack speeds, bluegill will likely
be able to generate higher fluid speeds (relative to the fish).
Bluegill accelerate the fluid faster than largemouth bass regardless of the
frame of reference (Table 2).
In the case of a prey item moving as if it were a particle of water, greater
fluid speed and acceleration will minimize the time that prey have to initiate
an escape response. If prey are not moving along with the suction flow, such
as when prey are sucked off a holdfast, higher flow velocities will generate
larger drag forces and higher accelerations will generate a larger
acceleration reaction, which resists changes in velocity and thus prevents the
deceleration of fluid (Daniel,
1984). High velocity and high acceleration of the fluid should
both result in improved suction feeding performance.
Muller and Osse describe strategies of suction feeding that are defined, in
part, by the timing of the opening of the opercular valves
(Muller and Osse, 1984).
Largemouth bass and bluegill sunfish both open their opercular cavities after
the prey is captured and after the time of maximum gape (T. E. Higham,
unpublished). However, largemouth bass open this valve earlier in the strike
sequence than bluegill sunfish (approximately 3 ms after maximum gape for bass
versus 19 ms for bluegill). Muller and Osse suggest that it is
beneficial for a fish to combine early opening of the opercular valve with
swimming (Muller and Osse,
1984). Given that the ram velocities of largemouth bass are much
greater than bluegill sunfish (Table
1), it seems that the two species in our study follow this trend.
Assuming that ram speed does not affect mouth expansion, it is interesting to
note that for either species it does not seem that an increase in ram speed
results in earlier opercular opening, which contradicts the hypothesis by
Muller and Osse (Muller and Osse,
1984).
Ingested volume
For each species, the primary correlate of increases in the ingested volume
was increasing gape distance, measured here as an indication of overall buccal
expansion. Time to peak gape (TTPG) had little influence on the
ingested volume, despite its considerable effect on the peak fluid speed
entering the mouth in bluegill sunfish (Day
et al., 2005). Although fluid speed was faster with a shorter
TTPG, this was countered by a shorter duration of the suction flow
that resulted in little net change in volume. By increasing the ingested
volume of water, the distance from the prey (if centered in this volume) to
the edge of the parcel of fluid increases and thus the chance of the prey
escaping may be less likely.
Largemouth bass ingested a much larger volume of water than did bluegill
(Table 1). Bass are piscivorous
ram-suction feeders that eat primarily evasive prey including fish, crayfish
and penaeid shrimp (Nyberg,
1971; Huskey and Turingan,
2001). Thus, the larger ingested volume may be an adaptation to
feeding on evasive prey. Based on measured buccal volumes of these two species
(D. C. Collar, unpublished), bluegill can ingest up to 2.5 times the size of
the buccal volume (max. ingested=9510 mm3) while bass ingest up to
3.2 times the size of the buccal volume (max. ingested=61337 mm3).
Both species are able to maintain a unidirectional flow of water that is
expelled from the opercular cavity at the same time that more water is being
ingested. The ability of largemouth bass to ingest more water relative to
their buccal volume than bluegill could be related to the observation that
they swim faster during prey capture, which will generate a greater passive
flow of water through the buccal cavity
(Table 1).
Based on unpublished values of buccal lengths for largemouth bass and bluegill sunfish, the individuals in our study had buccal lengths of 2.29 cm (bluegill) and 2.81 cm (bass) (D. C. Collar, unpublished). If one assumes that the shape of the expanding buccal cavity is similar in the two species, then differences in buccal length should result in differences in elevation of the regressions of gape diameter on ingested volume (Fig. 6A). We found no difference between species in this scaling pattern, suggesting that minor differences in buccal shape may counter the effects of a difference in buccal length (analysis of covariance, Pspeciesxgape=0.42, Pspecies=0.06).
Volumetric flow rate
For both bluegill and largemouth bass the primary mechanism for modulating
the volumetric flow rate was to modulate the extent of buccal expansion during
the strike, as indicated by maximum gape
(Fig. 6). Shorter times to peak
gape (TTPG) resulted in higher volumetric flow rates in bluegill, but
did not affect the volumetric flow rate for largemouth bass
(Fig. 7). This difference
between the two species suggests that largemouth bass are modulating another
kinematic variable to regulate the flow of water through the buccal cavity in
a different way than bluegill. This could be accomplished with a greater
decoupling of jaw rotation and buccal expansion in bass, or by modulation of
the opercular opening. Interestingly, bass exhibit a greater amount of
variation in fluid speed for a given TTPG
(Fig. 2), supporting the idea
that other kinematic variables are being modulated by bass.
Accuracy during feeding
Our study presents a novel method for quantifying accuracy during the
feeding event by measuring the location of the prey relative to the center of
the ingested parcel of water (Fig.
1). We suggest that this may be an important, understudied aspect
of suction feeding performance. It is well known that proper timing of the
strike is essential to a successful outcome. A fish will not capture prey if
the strike occurs too early, when the prey is not in range, and strikes will
be unsuccessful when the mouth of the predator is too close to the prey item
prior to mouth opening (Nyberg,
1971; Webb and Skadsen,
1980; Coughlin,
1991). Our observations indicate that bluegill have a superior
ability to position the suction flow field on the prey item. This has not been
previously demonstrated and may reflect some level of compensation for the
fact that bluegill generate a smaller flow field than bass. Bluegill often
feed on midwater zooplankton, such as cladocera, as well as small benthic
chironomid larvae and vegetation dwelling insect larvae
(Keast, 1978;
Mittelbach, 1981;
Mittelbach, 1984;
Brown and Colgan, 1984;
Collar et al., 2005). Their
ability to generate small, well-directed regions of high-speed suction flow
may be a key factor in their well-documented ability to rapidly remove large
numbers of these small prey from a feeding arena
(Mittelbach, 1981).
One explanation for why bass were less accurate than bluegill could be that
the average ram speed of bass was higher than bluegill (48.0±5.7 cm
s-1 versus 8.1±2.0 cm s-1). It has been
shown for other species that accuracy decreases with an increase in attack
velocity (Webb and Skadsen,
1980). Whether largemouth bass are less accurate because they swim
at higher speeds or whether they swim faster because they rely less on
accuracy is not fully understood. Higham et al. found that bluegill sunfish
developed a narrower, more elongate, and more focused ingested volume of water
with an increase in ram speed (Higham et
al., 2005a). Having the area that is influenced by suction
generation directed more in front of the fish could make accuracy a more
important factor. It has been suggested that braking could increase the
accuracy of a suction-feeding fish (Lauder
and Drucker, 2004; Higham et
al., 2005a; Higham et al.,
2005b), and this might be more important for fish that ingest a
relatively small volume of water. Future studies that relate braking during
suction feeding to mouth size (indirect measure of ingested volume) would
provide further insight into this issue.
Interestingly, while the species were equally good at positioning the prey
horizontally, bluegill were better at positioning the prey vertically. With an
increased ram velocity, the height to length ratio of the ingested volume of
water decreased for both bluegill sunfish
(Higham et al., 2005a) and
largemouth bass (Fig. 9). Given
that bluegill ingest a smaller volume of water, the vertical dimension of this
volume at maximal ram speeds would be much smaller than that of bass at a
comparable ram speed. Thus, if the attack strategy involves swimming at a
speed that reduces the vertical height of the ingested volume of water, then
the vertical accuracy of bluegill would be especially significant. The length
of the ingested volume of water increases with an increase in swimming speed
(Higham et al., 2005a),
suggesting less of a constraint along the horizontal axis with changes in ram
speed.
In studies of larval fishes, accuracy was quantified by the perpendicular
uptake distance (PUD) between the centroid of the prey and the
longitudinal axis of the fish (extending outward from the fish to the location
of the prey) (Drost, 1987;
Coughlin, 1991). In these
cases, measurements were dependent on the orientation of the predator, both in
ventral and lateral view. Without measuring the area of water influenced by
suction, these studies lacked a scaleable metric that could be applied to a
variety of morphologically distinct species. Our measures differ since they
are relative to the x-axis of the ingested volume rather than the
central axis of the mouth. Additionally, our measures are scaled to the
dimensions of the ingested volume rather than just a distance from the axis of
interest.
Suction feeding performance
Successful prey capture using suction depends on several aspects of
predator behavior and the pattern of water flow that is generated during the
strike, including a fluid speed high enough to draw the prey towards the
predator, ingesting a volume of water great enough to entrain the prey, and
accurate positioning of the suction flow field. Given this multidimensional
nature of suction feeding performance it is interesting to ask how different
species are distributed in the suction feeding performance space. Without
taking ram velocity into account, our results suggest the possibility that a
trade-off exists between the ability to generate high fluid speeds and the
volumetric flow rate. Bluegill generate higher fluid speeds than bass, but
bass overcome a slower fluid speed and are able to move more water per time
into the mouth. It is known from previous work that bluegill are able to
generate greater suction pressure magnitudes than largemouth bass, and this
has been attributed to bluegill having a greater force capacity of the epaxial
musculature that elevates the cranium during the strike and a higher
mechanical advantage for the transfer of force from this muscle to the
expansion of the buccal cavity (Carroll et
al., 2004). At a given body size they also have a lower area of
the buccal cavity across which the expansive epaxial forces are distributed.
In comparison with bluegill, largemouth bass are modified to achieve greater
volume of expansion for a given amount of epaxial contraction.
Thus, the basic trade-off in biological musculoskeletal lever systems between transfer of force and displacement may underly a trade-off in design of the fish suction feeding mechanism that results in a contrast between species modified to generate high fluid speeds and accelerations (i.e. high force transfer) versus other species that generate high volume and volumetric flow rate (i.e. large displacements). As additional comparative data are generated it will be instructive to see how morphological variation maps onto the distribution of fish species in this performance space. We predict that a common pattern will be that species that generate high fluid speeds will commonly exhibit small mouths and a high capacity to deliver force to buccal expansion, while species that ingest a large volume and generate high values of dV/dt will have larger buccal cavities and more efficient buccal expansion.
The underlying basis of differences between species in strike accuracy may simply be that accuracy decays with increasing approach speed of the predator (Fig. 10). If this proves to be the case generally, then a second trade-off is identified that may influence the distribution of species in the suction feeding performance space.
In the fish's frame of reference, largemouth bass generate higher fluid speeds than bluegill sunfish. Although bluegill sunfish are capable of exerting greater drag forces on the prey during suction feeding (due to higher flow speeds in the earthbound frame), largemouth bass close the distance to the prey faster. This suggests that bluegill might be specialized for drawing relatively non-evasive prey into their mouths either from the water column or from substrate, whereas bass are adept at overtaking highly evasive prey. The strategy employed by largemouth bass requires that the prey not be close to substrate or the high swimming velocity might result in a collision, suggesting that a trade-off exists between feeding performance and the habitat in which they feed.
|
Acknowledgments |
|---|
|
References |
|---|
|
TOPSummaryIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Brown, J. A. and Colgan, P. W. (1984). The ontogeny of feeding behaviour in four species of centrarchid fish. Behav. Proc. 9,395 -411.[CrossRef]
Carroll, A. M., Wainwright, P. C., Huskey, S. H., Collar, D. C.
and Turingan, R. G. (2004). Morphology predicts suction
feeding performance in centrarchid fishes. J. Exp.
Biol. 207,3873
-3881.
Collar, D. C., Near, T. J. and Wainwright, P. C. (2005). Comparative analysis of morphological diversity: does disparity accumulate at the same rate in two lineages of centrarchid fishes? Evolution 59,1783 -1794.[CrossRef][Medline]
Cook, A. (1996). Ontogeny of feeding morphology and kinematics in juvenile fishes: a case study of the cottid fish Clinocottus analis. J. Exp. Biol. 199,1961 -1971.
Coughlin, D. J. (1991). Ontogeny of feeding behavior of first-feeding Atlantic salmon (Salmo salar). Can. J. Fish. Aquat. Sci. 48,1896 -1904.
Daniel, T. L. (1984). Unsteady aspects of aquatic locomotion. Am. Zool. 24,121 -134.
Day, S. W. and McDaniel, J. C. (2005). PIV measurements of flow in a centrifugal blood pump: steady flow. J. Biomech. Eng. 127,244 -253.[Medline]
Day, S. W., Higham, T. E., Cheer, A. Y. and Wainwright, P.
C. (2005). Spatial and temporal patterns of water flow
generated by suction feeding bluegill sunfish Lepomis macrochirus
resolved by Particle Image Velocimetry. J. Exp. Biol.
208,2661
-2671.
De Visser, J. and Barel, C. D. N. (1998). The expansion apparatus in fish heads, a 3-D kinetic deduction. Neth. J. Zool. 48,361 -395.
Drost, M. R. (1987). Relation between aiming and catch success in larval fishes. Can. J. Fish. Aquat. Sci. 44,304 -315.
Drost, M. R., Osse, J. W. M. and Muller, M. (1988). Prey capture by fish larvae, water flow patterns and the effect of escape movements of prey. Neth. J. Zool. 38, 23-45.
Ferry-Graham, L. A., Wainwright, P. C. and Lauder, G. V. (2003). Quantification of flow during suction feeding in bluegill sunfish. Zoology 106,159 -168.
Higham, T. E., Day, S. W. and Wainwright, P. C.
(2005a). Sucking while swimming: evaluating the effects of ram
speed on suction generation in bluegill sunfish Lepomis macrochirus
using digital particle image velocimetry. J. Exp.
Biol. 208,2653
-2660.
Higham, T. E., Malas, B., Jayne, B. C. and Lauder, G. V.
(2005b). Constraints on starting and stopping: behavior
compensates for reduced pectoral fin area during braking of the bluegill
sunfish Lepomis macrochirus. J. Exp. Biol.
208,4735
-4746.
Huskey, S. H. and Turingan, R. G. (2001). Variation in prey-resource utilization and oral jaw gape between two populations of largemouth bass, Micropterus salmoides. Environ. Biol. Fishes 61,185 -194.[CrossRef]
Keast, A. (1978). Trophic and spatial interrelationships in the fish species of an Ontario temperate lake. Environ. Biol. Fishes 3,7 -31.
Lauder, G. V. (1980). The suction feeding
mechanism in sunfishes (Lepomis): an experimental analysis.
J. Exp. Biol. 88,49
-72.
Lauder, G. V. (1983). Prey capture
hydrodynamics in fishes: experimental test of two models. J. Exp.
Biol. 104,1
-13.
Lauder, G. V. and Drucker, E. G. (2004). Morphology and experimental hydrodynamics of fish fin control surfaces. IEEE J. Oceanic Eng. 29,556 -571.[CrossRef]
Lauder, G. V., Wainwright, P. C. and Findeis, E. (1986). Physiological mechanisms of aquatic prey capture in sunfishes: functional determinants of buccal pressure changes. Comp. Biochem. Physiol. 84A,729 -734.[CrossRef]
McLaughlin, R. L., Grant, J. W. A. and Noakes, D. L. G. (2000). Living with failure: the prey capture success of young brook charr in streams. Ecol. Freshw. Fishes 9, 81-89.[CrossRef]
Mittelbach, G. G. (1981). Foraging efficiency and body size: a study of optimal diet and habitat use by bluegills. Ecology 62,1370 -1386.[CrossRef]
Mittelbach, G. G. (1984). Predation and resource partitioning in two sunfishes (Centrarchidae). Ecology 65,499 -513.[CrossRef]
Muller, M. and Osse, J. W. M. (1984). Hydrodynamics of suction feeding in fish. Trans. Zool. Soc. Lond. 37,51 -135.
Muller, M., Osse, J. W. M. and Verhagen, J. H. G. (1982). A quantitative hydrodynamical model of suction feeding in fish. J. Theor. Biol. 95, 49-79.[CrossRef]
Near, T. J., Bolnick, D. I. and Wainwright, P. C. (2005). Fossil calibrations and molecular divergence time estimates in centrarchid fishes (Teleostei: Centrarchidae). Evolution 59,1768 -1782.[CrossRef][Medline]
Nemeth, D. H. (1997). Modulation of buccal pressure during prey capture in Hexagrammos decagrammus (Teleostei: Hexagrammidae). J. Exp. Biol. 200,2145 -2154.[Abstract]
Norton, S. F. and Brainerd, E. L. (1993). Convergence in the feeding mechanics of ecomorphologically similar species in the Centrarchidae and Cichlidae. J. Exp. Biol. 176, 11-29.[Abstract]
Nyberg, D. W. (1971). Prey capture in the largemouth bass. Am. Midl. Nat.86,128 -144.[CrossRef]
Rice, W. R. (1989). Analyzing tables of statistical tests. Evolution 43,223 -225.[CrossRef]
Sanford, C. P. J. and Wainwright, P. C. (2002). Use of sonomicrometry demonstrates the link between prey capture kinematics and suction pressure in largemouth bass. J. Exp. Biol. 205,3445 -3457.
Scarano, F. and Riethmuller, M. L. (1999). Iterative multigrid approach in PIV image processing with discrete window offset. Exp. Fluids 26,513 -523.
Van Leeuwen, J. L. (1984). A quantitative study of flow in prey capture by rainbow trout, with general consideration of the actinopterygian feeding mechanism. Trans. Zool. Soc. Lond. 37,21 -77.
Van Leeuwen, J. L. and Muller, M. (1984). Optimum sucking techniques for predatory fish. Trans. Zool. Soc. Lond. 37,137 -169.
Van Wassenbergh, S., Aerts, P. and Herrel, A.
(2005). Scaling of suctionfeeding kinematics and dynamics in the
African catfish, Clarias gariepinus. J. Exp.
Biol. 208,2103
-2114.
Viladiu, C., Vandewalle, P., Osse, J. W. M. and Casinos, A. (1999). Suction feeding strategies of two species of Mediterranean Serranidae (Serranus cabrilla and Serranus scriba). Neth. J. Zool. 49, 81-95.[CrossRef]
Wainwright, P. C., Ferry-Graham, L. A., Waltzek, T. B., Carrol,
A. M., Hulsey, C. D. and Grubich, J. R. (2001). Evaluating
the use of ram and suction during prey capture by cichlid fishes.
J. Exp. Biol. 204,3039
-3051.
Waltzek, T. B. and Wainwright, P. C. (2003). Functional morphology of extreme jaw protrusion in neotropical cichlids. J. Morphol. 257,96 -106.[CrossRef][Medline]
Webb, P. W. and Skadsen, J. M. (1980). Strike tactics of Esox. Can. J. Zool. 58,1462 -1469.
Weihs, D. (1980). Hydrodynamics of sucti
