University of South Florida
Scholar Commons
Graduate Theses and Dissertations
Graduate School
January 2012
The biomechanics of tongue projection in the frog
_Rana pipiens_: dynamics and temperature effects
Paula Sandusky
University of South Florida, [email protected]
Follow this and additional works at: http://scholarcommons.usf.edu/etd
Part of the American Studies Commons, Biomechanics Commons, Other Education Commons,
and the Physiology Commons
Scholar Commons Citation
Sandusky, Paula, "The biomechanics of tongue projection in the frog _Rana pipiens_: dynamics and temperature effects" (2012).
Graduate Theses and Dissertations.
http://scholarcommons.usf.edu/etd/4219
This Thesis is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in Graduate
Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected].
The Biomechanics of Tongue Projection in the Frog Rana pipiens:
Dynamics and Temperature Effects
by
Paula E. Sandusky
A thesis submitted in partial fulfillment
of the requirements for the degree of
Master of Science
Department of Integrative Biology
College of Arts and Sciences
University of South Florida
Major Professor: Stephen M. Deban, Ph.D.
Philip J. Motta, Ph.D.
Henry Mushinsky, Ph.D.
Date of Approval:
April 29, 2012
Keywords: Anurans, Inertial elongation, Elastic mechanism, Q10, Feeding
Copyright © 2012, Paula E. Sandusky
Table of Contents
List of Tables ................................................................................................................... iii
List of Figures ................................................................................................................... iv
Abstract .............................................................................................................................. v
Introduction ........................................................................................................................ 1
Materials and Methods ........................................................................................................5
Feeding Experiments ...............................................................................................5
Anatomy ..................................................................................................................7
Kinematic and Dynamic Analyses ..........................................................................8
Statistical Analysis ................................................................................................11
Results ...............................................................................................................................13
Prey Capture Kinematics ......................................................................................13
Prey Capture Dynamics ........................................................................................14
Lunge Kinematics and Dynamics .........................................................................15
Anatomy ................................................................................................................15
Discussion .........................................................................................................................17
Prey Capture Kinematics ......................................................................................17
Prey Capture Dynamics ........................................................................................19
Lunge Kinematics and Dynamics .........................................................................20
Anatomy and Convergence with Bufo ...................................................................21
Conclusion ........................................................................................................................24
References .........................................................................................................................26
Appendix: Figures and Tables ...........................................................................................30
Figure 1. Anatomical features and digitized points ...............................................31
Figure 2. Scatterplots of ballistic mouth opening kinematic and dynamic results at
experimental temperatures .....................................................................32
Figure 3. Scatterplots of tongue projection kinematic and dynamic results at
experimental temperatures ......................................................................32
i
Figure 4. Scatterplots of prey transport and tongue retraction kinematic and
dynamic results at experimental temperatures ........................................33
Figure 5. Scatterplots of lunge kinematic and dynamic results at experimental
temperatures ............................................................................................34
Figure 6. Montage comparing feedings at 10°C and 25°C ...................................35
Table 1. Minimum and maximum values for variables measured in the
Rana pipiens feeding experiments .........................................................36
Table 2. ANCOVA results for kinematic and dynamic variables examined .........37
ii
List of Tables
Table 1. Minimum and maximum values for variables measured in the Rana
pipiens feeding experiments ..................................................................................36
Table 2. ANCOVA results for kinematic and dynamic variables examined .....................37
iii
List of Figures
Figure 1. Anatomical features and digitized points. ..........................................................30
Figure 2. Scatterplots of ballistic mouth opening kinematic and dynamic results at
experimental temperatures .................................................................................31
Figure 3. Scatterplots of tongue projection kinematic and dynamic results at experimental
temperatures ........................................................................................................32
Figure 4. Scatterplots of prey transport and tongue retraction kinematic and dynamic
results at experimental temperatures ...................................................................33
Figure 5. Scatterplots of lunge kinematic and dynamic results at experimental
temperatures ........................................................................................................34
Figure 6. Montage comparing feedings at 10°C and 25°C ...............................................35
iv
Abstract
Ectotherms such as frogs must either function within environments with
temperatures amenable to their physiological needs, or find means to reduce the impact of
temperature on their activities. Recent studies on reptile and amphibian feeding have
shown convergent use of elastic recoil to drive feeding movements, thereby decoupling
temperature’s effects on muscle from movement and allowing the animals to feed over
broader temperature ranges. Rana pipiens specimens (n=5) were exposed to three
ambient temperatures (10º, 15º, and 25º C) at which feeding behavior was imaged at 6000
Hz. The image sequences yielded detailed kinematic and dynamic information for jaw,
tongue, and body movements, including velocity, acceleration, power, duration, and
excursion. Previously published studies have examined feeding in ranid frogs; however,
those studies employed slower frame rates that did not permit analysis of instantaneous
accelerations and velocities, and depressor and jaw-tongue complex mass specific power
outputs in Rana feeding have not yet been established. Specimens were dissected for
morphological measurement and calculation of mass-specific dynamics relative to the m.
depressor mandibulae and the center of mass of the jaw-tongue complex. Previous
studies on tongue projection in Bufo terrestris have shown that the rapid jaw depression
that inertially elongates the tongue relies on elasticity in the depressor muscles. Further,
because this movement is elastically driven, it is less sensitive to temperature than a
v
completely muscle-driven movement would be. Because Rana also feeds through inertial
elongation of the tongue (as does Bufo, in which the mechanism is convergent), I
hypothesized that Rana would demonstrate thermal insensitivity in its feeding kinematics
and dynamics in a pattern similar to that documented in Bufo. Experimental results
indicated that portions of the feeding cycle related to the initial, ballistic phase were at
least moderately thermally insensitive. At all temperatures studied, Rana reached
approximately half of the depressor mass-specific power of Bufo, demonstrating that
Bufo’s depressor mass-specific power output is not the minimum value necessary for
inertial elongation. I further hypothesize that thermal independence and power output in
excess of that achievable by muscle alone during the initial, ballistic mouth opening
phase of feeding suggests the involvement of an elastic mechanism convergent with that
of Bufo terrestris.
vi
Introduction
Ectotherms’ physiologic function relies on a thermally variable environment.
Adapting these functions to these environments requires either limiting activity periods
and/or ranges to those times and places with amenable temperatures, or physiologically
specializing to reduce any functional disruption temperature may cause. Temperature’s
influence is seen in the changes in physiological rates as an organism experiences varying
temperatures. The change in a physiological rate over ten degrees celsius, or Q10, is a
quantitative representation of how sensitive a particular process is to changes in
temperature (Hill, 1951; Rao, 1954; Belehradek, 1957; Bennett, 1984; Bennett, 1985).
Muscular responses to temperature changes have been well documented, particularly in
ectotherms (Putnam and Bennett, 1982; Rome and Kushmerick, 1983; Bennett and JohnAlder, 1984; Bennett, 1984; Hirano and Rome, 1984; Renaud and Stevens, 1984;
Bennett, 1985; Else and Bennett, 1987; John-Alder et al., 1989; Rome and Sosnicki,
1990; Altringham and Block, 1997; Wilson et al., 2000). Within the 20° - 30°C range, for
example, muscle is prone to dramatic changes in performance rates such as contraction
velocity (Q10 = 2.0 - 2.5), rate of tension development (Q10 = 2.36), and maximal power
output (Q10 = 2.42) (Bennett, 1985). Heart rate can also increase alongside temperature,
as observed in yellowfin tuna (Q10 = 2.37 over 18° - 28°C range) (Korsmeyer et al.,
1997). Depending upon the variable and the temperature, an increase in rate may provide
an advantage, such as greater blood perfusion; however, without some compensatory
1
mechanism, the necessary decrease in rate at lower temperatures can be detrimental to
organismal performance.
Many species thermoregulate behaviorally, seeking temperatures that will meet
their needs by basking or traveling between thermally differing areas to warm or cool
their bodies (Dreisig, 1984; Bennett, 2004). Other animals will avoid temperature issues
almost entirely by using thermally insensitive means, such as elastic recoil-based
movements, to reduce the effects of temperature on their physiology. Jumping insects use
resilin pads and “snapping” portions of their exoskeletons, rather than direct muscle
contraction, to propel themselves (Sutton and Burrows, 2008; Burrows, 2011).
Chameleons exploit an elastic recoil mechanism in their tongues to allow them to launch
(and therefore feed) at consistent velocities across wide temperature ranges (de Groot and
van Leeuwen, 2004; Anderson and Deban, 2010), and plethodontid salamanders have
convergently evolved thermally insensitive tongue projection that also makes use of
elastic recoil (Deban and Richardson, 2011). Recent research in ectotherm biology has
shown a prominent role for elastic recoil in powering feeding mechanisms, with an added
benefit of broadened capacity for feeding in thermally variable environments or
conditions (Huey and Hertz, 1984; de Groot and van Leeuwen, 2004; Anderson and
Deban, 2010; Deban and Lappin, 2011; Deban and Richardson, 2011). These elastic
mechanisms generally provide amplification of mechanical power, which enhances
performance (Roberts and Azizi, 2011).
Bufonid anurans use elastic recoil as a mechanism to depress their jaws rapidly,
which drives their inertial tongue projection. Elastic recoil in the m. depressor
2
mandibulae, which wraps from the posterior portion of the mandible to the top of the
head, lifts the retroarticular process of the mandible onto which the depressor inserts.
This elevation applies a torque to the jaw joint and drops the front of the mandible, and
the flap-like tongue, which is attached only at its rostral end, extends inertially
(Nishikawa, 2000; Lappin et al., 2006). Portions of this feeding behavior, such as the
velocity of the jaw during the initial mouth opening that propels the tongue, have been
found not to be significantly affected by temperature in Bufo terrestris (Deban and
Lappin, 2011). The elastic elements contributing to depressor recoil in Bufo terrestris are
believed to provide thermal insensitivity because elastic rate properties are significantly
less temperature sensitive than muscular ones (Alexander, 1966; Bennett, 1984; Bennett,
1985; Barnes and Ingalls, 1991; Denny and Miller, 2006).
Inertial elongation is not isolated to the Bufonidae. The Ranidae and
Dendrobatidae, among others, employ this mechanism convergently (Nishikawa, 2000;
Hoegg et al., 2004; van der Meijden, 2006). While performance and morphological
observations have established the convergence, whether or to what degree ranids and
bufonids use the same elastic recoil system to drive their tongue projection has yet to be
established. Outwardly the feeding behavior in Rana and Bufo appears different, even
before kinematics and dynamics are compared. Whereas Bufo will closely approach prey
and whip its tongue in a tight, low-profile trajectory, Rana throws its body forward and
gapes as its tongue lifts out of the mouth and flips. Both approaches, however, result in
the tongue inertially elongating to catch, and then pull back, a prey item (Nishikawa,
2000).
3
While Rana pipiens is convergent with Bufo for inertial elongation, convergence
in the elastic recoil mechanism that drives Bufo’s jaw depression has not been
established. The thermal sensitivity of the steps of the feeding gape cycle in Rana, as
determined by the Q10 values for rate properties relevant to inertial elongation, would be
a useful tool for uncovering such a convergence. For instance, Bufo’s initial mouth
opening, termed “ballistic” for its rapid achievement of peak velocity after the start of
movement (Lappin et al., 2006; Deban and Lappin, 2011), has been shown to be more
thermally neutral than the primarily muscle-driven final closing of the mouth after the
prey item has been pulled back into the buccal cavity (Deban and Lappin, 2011). Q10
trends can be used to compare other phases in the gape cycle for convergence in
performance and physiology. Movements that rely more on muscle will have higher Q10
values for their excursions, velocities, and accelerations and higher inverse Q10 (1/Q10)
values for their durations, because durations can be expressed as rates through their
reciprocals. Conversely, movements that rely primarily on elastic recoil will have lower
Q10 values for their rate properties, reflecting less of a performance difference across the
range of experimental temperatures (Hill, 1951; Bennett, 1984; Burrows and Sutton,
2008; Anderson and Deban, 2010; Burrows, 2011; Deban and Lappin, 2011; Deban and
Richardson, 2011). In those cases in which there is no difference between the measured
rates at the compared temperatures, the Q10 would be expected to be 1.
I hypothesize that the thermal sensitivity of tongue projection in Rana will show
that an elastic recoil mechanism is involved, and that elastically amplified power in Rana
will be similar to that of Bufo.
4
Methods and Materials
Five Rana pipiens Schreber 1782 were purchased commercially and housed as a
group in a plastic bin with ad libitum access to water deep enough for swimming. The
animals were maintained at approximately 22°C and were fed a diet of crickets, beetles
(Dermestes maculatus), and beetle larvae in random combination. All procedures in this
study were approved by the University of South Florida Institutional Animal Care and
Use Committee (protocol W3620).
Feeding Experiments
Frogs’ feedings were imaged at 6000 Hz (1/12,000 shutter speed) with a Fastcam
SA4 or Fastcam 1024 PCI camera (Photron USA, San Diego, California, USA). The
image frame was calibrated to a centimeter scale before each imaging session.
The animals were imaged individually, with the right side of the frog always
facing the camera on a plastic foam stage covered with wet paper towels and set against a
dark background. Crickets were presented at haphazard distances in front of the frog
(2.345 - 10.026 cm (5.305 ± 0.287 cm)). Frogs were presented with one cricket for each
feeding event, and the animals were permitted to continue feeding until they no longer
demonstrated an interest in the prey. Feedings continued with five to ten minute intervals
between a feeding and the next presentation of prey.
5
A total of 103 feeding events were recorded, and these were evaluated for
visibility of digitizing landmarks (see below) and correct orientation of the animal within
the frame (± ~15° relative to the imaging plane) to form a subset of 46 feeding events
comprised of three or four for each animal that would be analyzed to provide data for
every individual at each of the three experimental temperatures. Events were included in
the analysis if valid kinematic data could be extracted from their image sequences.
Frogs fed at three different ambient temperatures (10, 15, and 25°C) to form the
basis of the comparisons in the study. While attempts were made to include a broader
biologically relevant temperature range by including 5 and 30°C in the study, eliciting
feedings via inertial elongation proved difficult at those temperatures; thus, the 10-25°C
range was employed instead 1.
Imaging took place in a programmable environmental chamber (Environmental
Growth Chambers, Chagrin Falls, Ohio, USA) with additional halogen lighting (Source
Four PAR MCM, Electronic Theatre Controls, Middleton, Wisconsin, USA) to improve
image quality. This lighting was turned off after each successfully imaged feeding to
prevent elevating the animal’s body temperature artificially. The chamber’s set
temperature and the animal’s body temperature were documented for every feeding
recorded. Frog body temperatures were measured with a handheld infrared thermometer
(Sixth Sense LT300, Williston, Vermont, USA; ±1°C accuracy) held directly either above
the head or facing the tympanum to ensure measurement of temperature in the animal’s
body rather than from an environmental surface. Body and chamber temperature were
At 5°C, the frogs fed by jaw prehension, rather than tongue prehension. At 30°C, the
frogs were extremely active and did not reliably stay in the analysis frame.
6
1
within 0.38±0.18°C. Individuals experienced haphazardly determined temperature
sequences over the duration of the experiment. Animals were acclimated to chamber
temperatures for one to two hours prior to each imaging session, until body temperature
reached chamber temperature.
Anatomy
Following completion of imaging, animals were euthanized via overdoses of
MS-222. Specimens were stored in ice prior to dissection. Snout-vent lengths and body
masses were recorded for each individual upon thawing. The muscle fibers of the m.
depressor mandibulae were removed as completely as possible from both sides and the
top of each frog’s head. The left and right depressor muscles were soaked in amphibian
Ringer’s solution for 15-30 minutes, lightly toweled to reduce extraneous fluid mass, and
weighed on a balance (Virtual Measurements and Control model VB-302A, Santa Rosa,
California, USA, ±0.001g accuracy).
The jaw and tongue complex (including skin, buccal musculature, the majority of
the hyoid plate, and the retroarticular processes) was removed from each specimen and
soaked in amphibian Ringer’s solution, then the mass measured as above. The distance
along a rostrocaudal axis of the jaw-tongue complex from the center of the jaw joint to
the mandible tips and the distance from the joint center to the tip of the retroarticular
process were measured with calipers (Mitutoyo Corporation, Kanagawa, Japan; ±0.1 mm
accuracy). These measurements represented in-lever and out-lever portions of the
mandible. The location of the center of mass relative to the point of jaw articulation was
7
determined through the use of a balance and a ruler using published methods (Deban and
Lappin, 2011). Mean values were calculated for combined depressor mass, jaw-tongue
complex mass, and jaw length for use in the dynamic analyses.
Kinematic and Dynamic Analyses
Image sequences were imported into NIH ImageJ (Abramoff et al., 2004) on a
MacBook Pro computer (Apple, Inc., Cupertino, California, USA) for digitization of
dynamically relevant points and for measurement of durations and excursions. The x,y
positions of four anatomical landmarks were recorded in each frame of the feeding
sequences. The landmarks were (1) the tip of the upper jaw, (2) the tip of the mandible,
(3) the jaw joint, visible as a bulge just ventral to the caudal edge of the tympanum, and
(4) the anatomic tongue tip. Digitizing for each sequence began 100-200 frames before
the frog began to lunge and ended once the mouth closed.
To determine excursions, durations, average and instantaneous velocities,
accelerations, and power, image sequences were analyzed in phases with reference to the
inertial elongation gape cycle (Nishikawa, 2000; Deban and Lappin, 2011). First, the
animal starts a full-body lunge toward its prey. Before finishing the lunge, the animal
opens its mouth in an initially rapid, ballistic opening phase, and the tongue moves
forward to begin exiting the mouth. The gape stabilizes briefly while the tongue
continues moving and makes contact with its target, and then the gape widens further for
prey transport as the tongue retracts to bring the prey into the buccal cavity. Once the
8
prey has passed through the gape, the mouth closes. The animal then returns to its
crouched resting posture.
To standardize measurement of durations, the gape cycle was treated as a series of
timestamped events. The first frame in which the mouth was seen to open (as in ballistic
mouth opening) was event 1. The body of the tongue first crossing the gape was event 2.
The earliest observed gape stabilization at the end of ballistic opening was event 3. The
tongue tip reaching the substrate was event 4. The first observed widening of the gape
was event 5. The onset of tongue shortening was event 6. Achievement of maximum gape
was event 7. The return of the tongue tip to the buccal cavity and disappearing into the
pharynx was event 8. Final mouth closing was event 9. Durations were thus calculated
through subtraction of timestamp values. Ballistic mouth opening was event 3 minus
event 1. Transport mouth opening was event 7 minus event 5. Mouth closing was event 9
minus event 7. Tongue projection was event 4 minus event 2. Tongue retraction was event
8 minus event 6.
Excursions were measured with reference to the scale frame (see above) in
ImageJ as linear distances between landmarks at given frames of interest, and these
frames were the same as the start/end frames for the durations measured. Initial prey
distance was the distance between the frog’s upper jaw tip and the nearest edge of the
prey item, measured in the first digitized frame of the feeding sequence (and therefore
before the animal began its lunge). The gape at the end of ballistic mouth opening was the
distance between the jaw tips in the final frame of ballistic mouth opening. Maximum
gape, similarly, was the distance between the jaw tips at the conclusion of transport
9
mouth opening. Maximum tongue reach was the distance between the tip of the lower
jaw and the anatomical tongue tip, measured at the end of tongue projection (i.e., once
the tongue tip had made contact with the substrate).
Average velocities of ballistic opening, tongue projection, transport opening,
tongue retraction, and mouth closing were calculated by dividing the applicable gape or
tongue excursions by the durations over which they were achieved.
Peak velocity, acceleration, and power values for ballistic mouth opening were
calculated for all feeding sequences relative to the mean placement of the center of mass
of the jaw-tongue complex. These dynamics variables were calculated with a custom
script in R statistical software (R Development Core Team, 2009) designed to determine
the gape distance based upon the length of the arc traveled through mouth opening and
centered at the jaw joint. Arc radii extended to the tips of the upper jaw and mandible for
accurate gape arc measurement. A quintic spline (using the PSpline package, available
through CRAN) was fit to this instantaneous gape distance relative to time, which was
calculated from the frame rate of the image sequence. The script evaluated the location of
the center of mass of the jaw-tongue complex as a percentage of the total jaw length from
the jaw joint, which it then “followed” through its arc trajectory to prepare a
displacement function to be rendered along with its first and second derivatives. The first
and second derivatives of the displacement function represented the instantaneous
velocity and acceleration, respectively, of the jaw-tongue center of mass. Mass specific
power was calculated as the product of the instantaneous velocity and acceleration
values. Peak values for all three dynamic variables were reported.
10
Absolute power was calculated by multiplying the peak power by the mean mass
of the jaw-tongue complex. Depressor muscle mass-specific power was calculated by
dividing the absolute power by the mean mass of the paired mm. depressores mandibulae
muscles.
Lunge distance, duration, and mean velocity were calculated in a similar fashion,
based upon a quintic spline applied to the location of the jaw joint within the image frame
over the course of the sequence.
Statistical Analysis
All kinematic and dynamic variables (Table 1) were log10 transformed in Excel®
prior to statistical analysis. Statistical analyses were performed in JMP 5.1 software (SAS
Institute, Cary, North Carolina, USA) on Apple MacBook Pro and iMac computers.
Analyses of covariance (ANCOVAs) were performed to examine the effects of individual
animal, measured body temperature, and prey distance on the log10-transformed
kinematic and dynamic values. The interaction of individual and temperature was also
tested for all variables, but when the results of these interaction analyses proved not
statistically significant for all but one variable, the remaining variables were tested
without the interaction terms to conserve statistical power. A total of 17 variables were
tested and assessed for significance using a Bonferroni corrected α of 0.0029 (0.05 / 17).
Q10 values were calculated based upon the partial regression coefficient
estimating the slope of the linear relationship between log10 transformed response
variables versus the frogs’ measured body temperatures, including the influence of initial
11
prey distance and individual animal. The partial regression coefficient was then
multiplied by 10 to form the exponent of a base-10 antilogarithm as follows:
Q10 = 10(PRC x 10).
(1)
Excursions and dynamic values were evaluated for thermal sensitivity by their Q10
values; in particular, the statistical tests evaluated the variation from an expected Q10 of 1
(Table 2). Temporal properties, such as durations, were evaluated as rates, and inverse
Q10 values (1/Q10) reflected those temperature effects.
12
Results
Prey Capture Kinematics
Over all 46 feedings, spanning frog body temperatures of 10.1 - 26.7°C, frogs
spent 9 - 24 msec for ballistic mouth opening, ending with a maximum gape of 10 - 16
mm. The tongue took 17 - 92 msec after first passing through the gape to reach its full
extent, which was 9 - 38 mm. Transport opening of the mouth required 55.3 - 180.5 msec,
and the maximum gape was 14 - 31 mm. Tongue retraction ended (and the tongue
returned to the buccal cavity) within 58 - 172 msec, and the mouth closed within 48 - 258
msec after the animal reached its maximum gape (Table 1).
Temperature variation did not produce significantly different results for the
duration of ballistic mouth opening, gape width at the end of ballistic mouth opening, or
the width of the gape at the end of mouth opening for prey transport. Temperature did
produce significantly different results for the maximum tongue extension after projection,
the durations of tongue projection and transport mouth opening, and the durations of
tongue retraction and final mouth closing (Table 2).
The duration of ballistic mouth opening was not statistically significant for any of
the covariates (1/Q10 = 1.16). The gape at the end of ballistic mouth opening was
significant for individual (p < 0.0001). Maximum tongue projection length was
significant for individual (p < 0.0001) and increased with temperature (p < 0.0001, Q10 =
1.24). Duration of tongue projection was significant for individual (p < 0.0001) and
13
interaction of individual and temperature (p = 0.0006) and decreased with temperature (p
< 0.0001, 1/Q10 = 1.43). Duration of transport mouth opening decreased with temperature
(p < 0.0001, 1/Q10 = 1.35). Final gape at the end of transport mouth opening was
significant for individual (p < 0.0001). Duration of tongue retraction decreased with
temperature (p < 0.0001, 1/Q10 = 1.33). Duration of mouth closing was significant for
temperature (p < 0.0001, 1/Q10 = 2.08) and initial prey distance (p = 0.0020) (Table 2).
Prey Capture Dynamics
Across all feedings, the mean velocity of ballistic mouth opening (as measured at
the jaw tips) was 0.37 - 1.58 m sec-1, while the maximum instantaneous velocity of the
mandibular center of mass was 0.25 - 0.80 m sec-1 with a peak acceleration of
31.3 - 99.3 m sec-1 sec-1. Maximum mass-specific power of the m. depressor mandibulae
was 115.10 - 1783.23 W kg-1 of depressor muscle mass. The mean velocity of tongue
projection was 0.14 - 2.10 m sec-1, and the mean velocity of tongue retraction was 0.06 0.54 m sec-1. The mean velocity of transport mouth opening was 0.10 - 0.54 m sec-1. The
mean velocity of mouth closing was 0.06 - 0.46 m sec-1 (Table 1).
Temperature variation produced statistically significant results for all prey capture
dynamic variables. The mean velocities of tongue projection, tongue retraction, transport
mouth opening, and mouth closing showed strongly significant p-values (p = < 0.0001)
(Table 2).
All dynamic variables showed significance for effects of temperature. Mean
velocity of ballistic opening increased at warmer temperatures (Q10 = 1.20, p = 0.0011).
14
Maximum instantaneous velocity of ballistic opening increased at cooler temperatures
(Q10 = 1.19, p = 0.0016), as did peak acceleration (Q10 = 1.33, p = 0.0007). Maximum
mass-specific power of ballistic opening increased considerably at warmer temperatures
(Q10 = 1.57, p = 0.0007) (Figure 2). The mean velocities of tongue projection and
retraction had Q10 values of 1.66 and 1.65, respectively (p < 0.0001 for both), and the
mean velocity of transport opening had a Q10 value of 1.23 (p < 0.0001) (Figures 2 and
3). The mean velocity of mouth closing had a Q10 value of 2.17 (p < 0.0001) (Figure 3).
All three increased significantly with warmer temperatures.
Lunge Kinematics and Dynamics
The average feeding lunge was 2.2 - 8.7 cm in length and lasted 77 - 351 msec, as
measured over 41 feedings. The average lunge velocity was 0.01 - 0.36 m sec-1, as
measured over 40 feedings (Table 1).
Only lunge duration had any statistical significance for temperature (p = 0.0021,
1/Q10 = 1.31), and it decreased with increasing temperature. Lunge distance
(p < 0.0001), lunge duration (p < 0.0001), and average lunge velocity (p = 0.0021) were
significant for initial prey distance. Average lunge velocity increased with increasing
temperature (Q10 = 1.64, p = 0.0118) (Figure 5).
Anatomy
The mean snout-vent length of the frogs was 8.42 cm (± 0.15 cm). The mean body
mass was 35.51 gm, although body mass was not recorded for one gravid female (SVL
15
8.81 cm). The mean jaw-tongue complex mass was 0.90 gm (± 0.049 gm). The mean
location of the center of mass of the jaw-tongue complex relative to the articulation point
of the jaw, expressed as a percentage of jaw length, was 37.86% (± 4.12%). The mean
mass of the left m. depressor mandibulae was 0.02 gms (± 0.0041 gms), and the mean
mass of the right m. depressor mandibulae was 0.02 gms (± 0.0033 gms). The combined
mean depressor mass used for calculation of mass specific power was 0.05 gms (± 0.0073
gms). The mean jaw length was 2.29 cm (± 0.072 cm), and the mean retroarticular
process length was 1.6 mm (± 0.045 mm). Anatomical values are reported with their
standard errors.
16
Discussion
Prey Capture Kinematics
All feedings recorded took place with the frogs elongating their tongues inertially,
in which the dropping of the jaw propels the flap-like tongue in a targeted direction,
generally that in which the animal is facing (Nishikawa, 2000; Mallett et al., 2001;
Lappin et al., 2006; Deban and Lappin, 2011). When attempts were made to elicit
feedings at 5°C, the frogs captured the crickets through jaw prehension, which is
otherwise associated with the acquisition of larger, more unwieldy prey such as
earthworms (Anderson, 1993). In this case, tongue use was not observed. The thermal
sensitivity of the duration of tongue projection suggests that the tongue’s projection is
driven by muscle contraction as well as elasticity.
The effect of prey distance upon duration of mouth closing (p = 0.0020, Table 2)
is curious, and it suggests a behavioral coupling between the animal’s closing its mouth
and “resetting” itself at rest after withdrawing from its feeding lunge. Because final gape
is not significantly affected by prey distance, this effect would not be driven by a greater
gape at a longer lunge distance. Maximum tongue reach was not significantly affected by
prey distance, suggesting that it also does not necessarily influence the duration of mouth
closing.
The overall thermal sensitivity of mouth closing (duration 1/Q10 = 2.08,
17
p = < 0.0001; mean velocity Q10 = 2.17, p = < 0.0001) relative to ballistic mouth opening
(duration 1/Q10 = 1.16, p = 0.0082; mean velocity Q10 = 1.20, p = 0.0011) is interesting
for its comparative value (Figures 2, 4, and 6; Table 2). Inertial elongation is generally
associated with elastic recoil in the m. depressor mandibulae, which applies a torque
about the jaw joint to drop the jaw and inertially elongate the tongue (Lappin et al., 2006;
Deban and Lappin, 2011). Elastic recoil-based ballistic opening is believed to be
associated with a reduced sensitivity to temperature based upon the properties of the
components necessary for elastic energy storage, such as connective biomaterials like
collagen that accompany muscle tissue (Alexander 1966; Denny and Miller 2006; Rigby
et al., 1959; Deban and Lappin, 2011). The expectation is that greater reliance on elastic
mechanisms reduces thermal sensitivity; conversely, the more the movement relies on
muscular drive, the greater its thermal sensitivity will be (Bennett, 1985; Barnes and
Ingalls, 1991).
Whereas depressor activity is associated with elastic recoil, levator activity, which
closes the mouth, is not. Based upon the inverse Q10 value (1.16) found, the duration of
ballistic mouth opening appears to have more of a thermally insensitive (and ostensibly
elastic) drive than does the duration of mouth closing (1/Q10 = 2.08) (Figure 6). Since the
gape width at the end of ballistic opening showed low thermal response (Q10 = 1.04), the
thermally insensitive duration of ballistic opening represents the animal widening its gape
at nearly the same rate and degree, regardless of temperature. Were the movement more
muscle-dependent, temperature effects on shortening velocity would have changed the
18
duration significantly (Bennett 1984, 1985). The Q10 of mean ballistic opening velocity
(1.20) supports this relationship.
Prey Capture Dynamics
The relatively low thermal sensitivity for both mean (Q10 = 1.20, p = 0.0011) and
instantaneous velocity (Q10 = 1.19, p = 0.0016) of ballistic mouth opening suggests
elastic mechanisms driving the rapid, initial jaw depression in Rana pipiens, in a manner
similar to that seen in Bufo terrestris (Lappin et al., 2006; Deban and Lappin, 2011).
Given the dramatic, thermally-induced changes in velocity observed in purely muscular
movement (Bennett, 1984; Barnes and Ingalls, 1991), as well as the more modest (or
absent) changes seen with elastic recoil systems (Anderson and Deban 2010, Deban and
Lappin, 2011, Deban and Richardson 2011), we can compare performance in two gape
movements and assess which is more elastically or muscularly powered. Based on this
pattern, the mean velocities of ballistic and transport mouth opening (Q10 = 1.20 and
1.23, respectively) show greater elastic involvement than does the mean velocity of
mouth closing (Q10 = 2.17), which is more muscularly driven.
The mean velocities of tongue projection (Q10 = 1.66) and tongue retraction
(Q10 = 1.65) show intermediate thermal sensitivity. The results suggest a muscular
involvement in projection and an elastic involvement in retraction, resulting in
movements that are neither completely elastic nor completely muscular.
Electromyographic data from the retraction musculature would be useful in determining
the motor control of the tongue during the feeding cycle, and particularly during
19
projection, when no muscular input from the tongue was expected, according to the
inertial elongation model of tongue projection (Nishikawa 2000). These findings suggest
a possible difference in feeding motor control between Rana pipiens and Bufo terrestris
(Deban and Lappin, 2011).
Lunge Kinematics and Dynamics
The dynamics and kinematics of the lunge are statistically significant when
considered relative to initial prey distance, which is to be expected. Thermal sensitivity
(mean velocity Q10 = 1.64, duration 1/Q10 = 1.31) (Table 2), however, suggests that the
lunge itself is not entirely muscularly driven (Bennett, 1984). Jumping, a related
explosive behavior, has been shown to rely strongly on an elastic element arranged in
series with the musculature of the lower leg (Roberts and Marsh, 2003). In turn, jumping
is temperature-sensitive over a subset of the overall range of biologically relevant
temperatures (Hirano and Rome, 1984). In this case, however, the animal is not actually
producing a jump so much as it is extending its body quickly toward its prey. The need
for jumping power would then likely be damped by decreased muscular loading of the
elastic element (i.e., the Achilles’ tendon). Were the lunge completely reliant on muscle
and without a contributing elastic element, the Q10 and inverse Q10 values corresponding
to the kinematics and dynamics of the behavior (0.85 for distance, 1.31 for duration, and
1.64 for mean velocity) would probably be considerably higher (Bennett, 1984). These
results suggest modulated recruitment of elastic recoil based upon the movement
required, although electromyographic data is necessary for confirmation.
20
Anatomy and convergence with Bufo
Rana is considered an inertial elongator convergent with Bufo for its tongue
protraction (Gans, 1961; Gans and Gorniak, 1982; Nishikawa, 2000; Lappin et al., 2006),
with numerous non-inertial elongators more closely related to each genus than Rana and
Bufo are to each other, a conclusion borne out by continually improving phylogenies
(Nishikawa, 2000; Hoegg et al., 2004; Van der Meijden, 2006). Both Rana pipiens and
Bufo terrestris manage similar performance through the same behavior, but with
considerably different dynamic properties (Lappin et al., 2006; Deban and Lappin, 2011).
The peak depressor mass-specific power output found in Bufo when tested across the
temperature range of 10° to 38°C was determined to span 449.6 to 4,348.5 W kg-1 of
depressor muscle mass (Deban and Lappin, 2011). Over all the feedings conducted for
this study with Rana, the peak depressor mass-specific power output was appreciably
lower; in fact, Rana’s values (115.1 - 1,783.2 W kg-1) were less than half that produced
by Bufo. The quality of the m. depressor mandibulae may itself provide the answer that
explains the difference in dynamics. First, Bufo appears to have more muscle to use than
does Rana. The mean mass of the paired mm. depressores mandibulae in sampled Bufo
was found to be 0.13 g (Deban and Lappin, 2011), while Rana’s mean paired depressor
mass (0.045 g) was less than half of Bufo’s, with animals of comparable size (mean Rana
SVL = 8.42 cm (± 0.15 cm), Bufo SVL range = 3.9 - 8.2 cm (Deban and Lappin, 2011)).
Additionally, the ratio of depressor mass to jaw/tongue mass was considerably greater for
Bufo (17 - 22% (19±1%)) (Deban and Lappin, 2011) than for Rana (2.85 - 6.0%
(4.77±0.68%)).
21
Both Rana and Bufo demonstrate some power amplification from elastic
elements, as their mass-specific power output surpasses 373 W kg-1, or that of muscle
alone (Lutz and Rome, 1984; Altringham and Block, 1997; Aerts, 1998; Roberts and
Marsh, 2003). Rana, however, achieves successful performance with appreciably less
muscle and power. The difference in the depressors may also explain the kinematic
differences between the protrusion of the tongue in specimens from these genera.
Although this study does not include quantitative, comparative results relating both Rana
and Bufo, based upon appearance in the feeding films, Rana projects its tongue in a
higher arc than that seen in Bufo.
The higher performance values observed in Bufo do not necessarily impart an
additional advantage in the pursuit of the same prey Rana would consume. Higher
performance values do not appear to limit feeding efficacy, although more powerful
feeding might not become a constraint if typical prey items tend to move quickly. The
result may be a neuromechanical “arms race” of sorts in which the capacity to feed more
powerfully could broaden the range of possible prey that are themselves more powerful
in their capacities to escape, although answering that question would involve more
research into the effects of temperature on the prey species on which the inertial
elongators feed.
Feeding ecology is informative in this regard. Rana is more aquatic, with an
occasional need to lunge over water to capture aerial prey (Gans, 1961), while Bufo is
more terrestrial and feeds more cryptically on prey that may not jump or fly. In any case,
Rana appears to feed with a convergent, yet less powerful system than Bufo employs,
22
with apparently little functional constraint imposed by the difference, except for a higher
temperature sensitivity (Q10 of 1.06 and 1.20 for the mean velocity of ballistic mouth
opening in Bufo terrestris and Rana pipiens, respectively (Deban and Lappin, 2011)).
23
Conclusion
Rana pipiens possesses an inertial elongation system for tongue prehension of
prey. The mechanism’s performance rates demonstrate graded thermal sensitivity
depending upon whether a given movement is more elastically or muscularly. More
muscularly dependent movements, such as mouth closing (duration 1/Q10 = 2.08, mean
velocity Q10 = 2.17) have a greater thermal sensitivity than do movements thought to be
more reliant on elastic recoil, such as ballistic mouth opening (duration 1/Q10 = 1.16,
mean velocity Q10 = 1.20), suggesting that Rana relies upon some elastic recoil to power
its jaw depression (and therefore feeding) mechanism. Elastic mechanisms in feeding are
an emerging pattern in ectotherms, and the mechanism in Rana is convergent with that
found in toads, chameleons, and salamanders (Lappin et al., 2006; Anderson and Deban,
2010; Deban and Lappin, 2011; Deban and Richardson, 2011). Insensitivity to
temperature in feeding mechanisms allows ectotherms to live in thermally variable
environments without consequences for prey capture (Anderson and Deban, 2010).
Rana pipiens may use the elastic elements in its legs for some resilience against
thermal effects when it lunges at its prey. Q10 (for mean velocity, 1.64) and 1/Q10 (for
duration, 1.31) values for lunge-related kinematic and dynamic values are not
commensurate with an entirely muscularly driven lunge. A partially elastic lunge may
offer an advantage over the thermal dependencies of the prey species Rana lunges to
24
capture. The intermediate thermal sensitivity may be a side effect of the engagement of
the legs to propel the lunge in a manner similar to jumping (Roberts and Marsh, 2003).
Rana and Bufo have converged on the inertial elongation mechanism for tongue
projection, and measurements of mass-specific power output of the m. depressor
mandibulae suggest that the differences that distinguish inertial elongation between the
taxa accompany differences in the relative mass of the depressor muscle. Bufo
demonstrates more than twice the mass-specific power output Rana does, and Bufo
appears to have a greater depressor muscle mass-specific mass overall (Deban and
Lappin, 2011). The difference in mass-specific power output suggests that the
outstanding power values Bufo produces are not mandatory for inertial elongation
performance, but may explain differences in overall kinematics between the tongue
movement in Rana and Bufo. Why these convergent feeding mechanisms are so divergent
in dynamics is still open to question, but Rana pipiens appears to have less reliance on
elastic recoil than does Bufo terrestris, as evidenced by its lower peak mass-specific
depressor power output and its higher thermal dependence throughout the gape cycle.
25
References
Abramoff, M.D., Magalhaes, P. J., and Ram, S. J. (2004). Image processing with
ImageJ. Biophot. Intl. 11(7), 36-42.
Aerts, P. (1998). Vertical jumping in Galago senegalensis: the quest for an obligate
mechanical power amplifier. Proc. R. Soc. B 353, 1607-1620
Alexander, R. McN. (1966). Rubber-like properties of the inner hinge-ligament of
Pectinidae. J. Exp. Biol. 44, 119-130.
Altringham, J. D., and Block, B. A. (1997). Why do tuna maintain elevated slow
muscle temperatures? Power output of muscle isolated from endothermic and
ectothermic fish. J. Exp. Biol. 200, 2617-2627.
Anderson, C. W. (1993). The modulation of feeding behavior in response to prey type
in the frog Rana pipiens. J. Exp. Biol. 179, 1-12.
Anderson, C. V. and Deban, S. M. (2010). Ballistic tongue projection in chameleons
maintains high performance at low temperature. Proc. Natl. Acad. Sci. 107,
5495-5499.
Barnes, W. S., and Ingalls, C. P. (1991). Differential effects of temperature on
contractile behavior in isolated frog skeletal muscle. Comp. Biochem. Physiol. A
100, 575-580.
Bennett, A. F. (1984). Thermal dependence of muscle function. Am. J. Physiol. Reg.
Integ. Comp. Physiol. 247, R217-229.
Bennett, A. F. and John-Alder, H. B. (1984). The effect of body temperature on the
locomotory energetics of lizards. J. Comp. Physiol. B 155, 21-27.
Bennett, A. F. (1985). Temperature and muscle. J. Exp. Biol. 115, 333-344.
Bennett, A. F. (2004). Thermoregulation in African chameleons. In Animals and
Environments: Proceedings of the Third International Conference of Comparative
26
Physiology and Biochemistry, International Congress Series, Vol 1275 (ed. S.
Morris, A. Vosloo). Amsterdam: Elsevier. pp 234-241.
Burrows, M. (2011). Jumping mechanisms and performance of snow fleas (Mecoptera,
Boreidae). J. Exp. Biol. 214, 2362-2374.
Deban, S. M. and Lappin, A. K. (2011). Thermal effects on the dynamics and motor
control of ballistic prey capture in toads: maintaining high performance at low
temperature. J. Exp. Biol. 214, 1333-1346.
Deban, S. M. and Richardson, J. C. (2011). Cold-blooded snipers: thermal
independence of ballistic tongue projection in the salamander Hydromantes
platycephalus. J. Exp. Zool. A doi:10.1002/jez.708.
de Groot, J. H. and van Leeuwen, J. L. (2004) Evidence for an elastic projection
mechanism in the chameleon tongue. Proc. R. Soc. B 271, 761-770.
Denny, M. and Miller, L. (2006). Jet propulsion in the cold: mechanics of swimming in
the Antarctic scallop Adamussium colbecki. J. Exp. Biol. 209, 4503-4514.
Dreisig, H. (1984). Control of body temperature in shuttling ectotherms. J. Therm. Biol.
9, 229-233.
Else, P. L. and Bennett, A. F. (1987). The thermal dependence of locomotor
performance and muscle contractile function in the salamander Ambystoma
tigrinum nebulosum. 128, 219-233.
Gans, C. (1961). A bullfrog and its prey. Nat. Hist. 70 (2), 26.
Gans, C. and Gorniak, G. C. (1982). How does the toad flip its tongue? Test of two
hypotheses. Science 216, 1335-1337.
Hill, A. V. (1951). The influence of temperature on the tension developed in an isometric
twitch. Proc. R. Soc. B 138, 349-354.
Hirano, M. and Rome, L. C. (1984). Jumping performance of frogs (Rana pipiens) as a
function of muscle temperature. J. Exp. Biol. 108, 429-439.
Hoegg, S., Vences, M., Brinkmann, H., and Meyer, A. (2004). Phylogeny and
comparative substitution rates of frogs inferred from sequences of three nuclear
genes. Mol. Biol. Evol. 21, 1188-1200.
27
Huey, R. B., and Hertz, P. E. (1984). Is a jack-of-all-temperatures a master of none?
Evolution 38, 441-444.
John-Alder, H. B., Barnhart, M. C., and Bennett, A. F. (1989). Thermal sensitivity of
swimming performance and muscle contraction in northern and southern
populations of tree frogs (Hyla crucifer). J. Exp. Biol. 142, 357-372.
Korsmeyer, K. E., Lai, N. C., Shadwick, R. E., and Graham, J. B. (1997). Heart rate
and stroke volume contribution to cardiac output in swimming yellowfin tuna:
response to exercise and temperature. J. Exp. Biol. 200, 1975-1986.
Lappin, A. K., Monroy, J. A., Pilarski, J. Q., Zepnewski, E. D., Pierotti, D. J., and
Nishikawa, K. C. (2006). Storage and recovery of elastic potential energy
powers ballistic prey capture in toads. J. Exp. Biol. 209, 2535-2553.
Lutz, G. J. and Rome, L. C. (1994) Built for jumping: the design of the frog muscular
system. Science 263, 370-372.
Mallett, E. S., Yamaguchi, G. T., Birch, J. M., and Nishikawa, K. C. (2001). Feeding
motor patterns in anurans: insights from biomechanical modeling. Amer. Zool. 41,
1364-1374.
Nishikawa, K. C. and Gans, C. (1996). Mechanisms of tongue protraction and narial
closure in the marine toad Bufo marinus. J. Exp. Biol. 199, 2511-2529.
Nishikawa, K. C. (2000). Feeding in frogs. In Feeding: Form, Function, and Evolution
in Tetrapod Vertebrates (ed. Kurt Schwenk). San Diego: Academic Press.
Putnam, R. W. and Bennett, A. F. (1982). Thermal dependence of isometric contractile
properties of lizard muscle. J. Comp. Physiol. B 147, 11-20
R Development Core Team. (2008). R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria.
http://www.r-project.org.
Rao, K. R. and Bullock, T. H. (1954). Q10 as a function of size and habitat
temperature in poikilotherms. Amer. Naturalist 88, 33-44.
Renaud, J. M. and Stevens, E. D. (1984) The extent of short-term and long-term
compensation to temperature shown by frog and toad sartorius muscle. J. Exp.
Biol. 108, 57-75.
28
Rigby, B. J., Hirai, N., Spikes, J. D., and Eyring, H. (1959). The mechanical
properties of rat tail tendon. J. Gen. Physiol. 43, 265-283.
Roberts, T. J. and Marsh, R. L. (2003). Probing the limits to muscle-powered
accelerations: lessons from jumping bullfrogs. J. Exp. Biol. 206, 2567-2580.
Roberts, T. J. and Azizi, E. (2011). Flexible mechanisms: the diverse roles of biological
springs in vertebrate movement. J. Exp. Biol. 214, 353-361.
Rome, L. C. and Kushmerick, M. J. (1983). Energetics of isometric contractions as a
function of muscle temperature. Am. J. Physiol. Cell Physiol. 244 (1),
C100-C109.
Rome, L. C. and Sosnicki, A. A. (1990). The influence of temperature on mechanics of
red muscle in carp. J. Physiol. 427, 151-169.
Sutton, G. P. and Burrows, M. (2008). The mechanics of elevation control in locust
jumping. Comp. Physiol. A 194, 557-563.
Van der Meijden, A. (2006). Molecular phylogeny and biogeography of ranoid frogs.
Ph.D. dissertation, University of Konstanz.
Wilson, R. S., James, R. S., and Johnston, I. A. (2000). Thermal acclimation of
locomotor performance in tadpoles and adults of the aquatic frog Xenopus laevis.
J. Comp. Physiol. B 170, 117-124.
29
Appendix: Figures and Tables
Figure 1. Anatomical features and digitized points. The m. depressor
mandibulae is drawn in approximate anatomical position. Digitizing
points are drawn where they were found for analysis. The tips of the
upper and lower jaws, the jaw joint, the anatomical tongue tip, and the
nearest edge of the cricket were digitized. This figure represents an
advanced stage in the feeding cycle, and the nearest edge of the cricket
was used for measurement of initial prey distance on only the first frame
of the image sequence. The distance between the nearest edge of the
cricket and the tip of the upper jaw is the initial prey distance.
30
Avg Lunge Vel (m/sec)
0.3
0.25
0.2
0.15
0.1
0.05
JMP
with 0symbols:
Y by
X
Bivariate
Fit of Fit
Dur
Ballistic
Opening (msec) By Temperature
Page 3 of 10
-0.05 Fit of Max Vel Ballistic Open (m/sec) By Temperature
Bivariate
10
Max Vel(msec)
Ballistic Open (m/sec)
Dur Ballistic Opening
0.8
15
20
25
Temperature
0.7
JMP
with symbols:
Y by
X
Bivariate
Fit of Fit
Dur
Ballistic
Opening (msec) By Tempera
Transformed Fit Log
Bivariate
Fit of Dur Ballistic Opening (msec) By Temperature
Bivariate Fit of Max Vel Ballistic Open (m/sec) By Tempe
0.6
30
0.5
Max (cm)
Dep MS Power (W/kg)
Max
Ballistic
Ballistic
EndAcc
Gape
(cm)Open (m/sec^2) Max Vel Ballistic Open (m/sec)
Max Tongue Reach
Avg Vel Ballistic Open
(m/sec)
0.8
25
0.4
0.3
20
0.2
15 10
15
20
25
0.7
0.6
0.5
0.4
JMP
with symbols:
Y by
Bivariate
Fit of Fit
Avg
VelX Ballistic Open (m/sec) By Temperature
Temperature
Bivariate
Fit of Max Accel Ballistic Open (m/sec/sec) By Temper
0.3 symbols:
JMP
with
Y by
X Mass-Spec Power (W/kg) By Tem
Bivariate
Fit of Fit
Max
Dep
10 Transformed Fit Log
200
Bivariate
Fit of Ballistic End Gape (cm) By Temperature
Bivariate Fit of Avg Vel Ballistic Open (m/sec) By Temperature
0.2
Avg Vel Ballistic Open (m/sec)
1.8
10
15
20
25
Temperature
1.6
Transformed Fit Log
1.4
1.6 10
15
1.5
1.4
20
25
Temperature
150
Transformed Fit Log
100
Bivariate Fit of Max Vel Ballistic Open (m/sec) By Temperature
Bivariate
Fit of Avg Vel Ballistic Open (m/sec) By Temper
Max Acc Ballistic Open (m/sec^2)
1.3
1.8
1.2
1
50
1.6
1.1
0.8
1.4
10
15
20
25
1
0.6
Temperature
1.2
JMP
with
Y by
Page 4 of 10
0.9
0.4 symbols:
Bivariate
Fit of Fit
Avg
VelX Ballistic Open (m/sec) By Temperature
1 Transformed Fit Log
Bivariate
Fit
of
Max
Accel
Ballistic
Open
(m/sec/sec)
By
Temperature
0.2
0.8
Bivariate
Fit of Max Dep Mass-Spec Power (W/kg) By Temperatu
0.8 10
15
20
25
15
20
25
2000
200 10
1.2
Temperature
Transformed Fit Log
0.4
1500
150
Temperature
0.6
Transformed Fit Log
Bivariate Fit of Max Accel Ballistic Open (m/sec/sec) Bivariate
By Temperature
Fit of Max Tongue Reach (cm) By Temperature
0.2
4 10
15
100
3.5
3
10
15
20
25
Temperature
Fit of Max Accel Ballistic Open (m/sec/sec) By T
2.5
20
10
1.5
15
20
25
Temperature
Transformed Fit Log
1
of ballistic mouth opening
kinematic and dynamic results at
Bivariate Fit of Ballistic End Gape (cm) By Temperature
Bivariate
Fit
of
Max
Dep
Mass-Spec
Power
(W/kg)
By
Temperature
0.5
experimental temperatures. Q10 trends are represented by red logarithmic fit lines
10
15
20
25
2000
applied to each response variable. Scatterplot columns are aligned
along common
Temperature
temperature gridlines. Different markers represent different
individuals,
and all
Transformed
Fit Log
1500 analyzed feeding events are represented by single
points
in Dur
eachTongue
plot. Proj (msec) By Temperature
Bivariate
Fit of
Max Dep MS Power (W/kg)
Transformed
Fit Log
Figure 2. Scatterplots
25
Transformed Fit Log
500
Bivariate
50
20
Temperature
1000
1000
500
0
10
15
20
25
31
Temperature
Transformed Fit Log
Bivariate Fit of Ballistic End Gape (cm) By Temperature
JMP
with symbols:
Y by
X
Bivariate
Fit of Fit
Max
Tongue
Reach (cm) By Temperature
Page
Bivariate Fit of Dur Tongue Proj (msec) By Temperature
100
90
Dur Tongue Proj (msec)
Dur Tongue Proj (msec)
80
JMP
with symbols:
Y by
X
Bivariate
Fit of Fit
Max
Tongue
Reach (cm) By Temperature
70
Page 6 of 10
Bivariate Fit of Dur Tongue Proj (msec) By Temperature
60
100
50
90
80
40
70
30
60 symbols: Fit Y by X
JMP
with
20
Bivariate
Fit of Max Dep Mass-Spec Power (W/kg) By Temperature
50
Bivariate
Fit of Ballistic End Gape (cm) By Temperature
10
40 10
1.6
15
20
Ballistic
End Gape (cm)
Avg Proj
Vel (m/sec)
Tongue Proj
(m/sec)
Avg Vel Tongue
30
1.5
20
1.4
25
Temperature
Transformed Fit Log
Bivariate
Fit of Avg Vel Tongue Proj (m/sec) By Temperature
10
1.3 10
15
20
25
Temperature
1.2
2
1.1 Transformed Fit Log
Bivariate Fit of Avg Vel Tongue Proj (m/sec) By Temperature
1
1.5
0.92
0.81
1.5
10
15
20
0.5
1
25
Temperature
Transformed Fit Log
Max Tongue Reach (cm)
Bivariate
Fit of Max Tongue Reach (cm) By Temperature
0
10
0.5
4
3.5
0
20
25
Temperature
Transformed Fit Log
10
3
Bivariate
2.5
15
15
20
25
Fit of Final
gape distance (m) By Temperature
Temperature
Transformed Fit Log
Bivariate
Fit of Final gape distance (m) By Temperature
2
1.5
1
0.5
10
15
20
25
Temperature
Figure 3. Scatterplots
of tongue
projection kinematic and dynamic
Transformed
Fit Log
results atBivariate
experimental
temperatures.
TheProj
plots(msec)
are formatted
as
Fit of
Dur Tongue
By Temperature
described in the legend of figure 2.
32
Page
Final gape dista
0.025
0.02
0.015
10
15
20
25
Temperature
JMP
with symbols:
Y by
Transformed Fit Log Page 7 of 10
Bivariate
Fit of Fit
Avg
VelX Tongue Proj (m/sec) By Temperature
Bivariate Fit of Dur Transport Opening (msec) By Tempera
Bivariate Fit of Final gape distance (m) By Temperature
Final gape distance (m)
0.03
0.025
0.02
Dur Transport Opening (msec)
200
175
150
125
100
JMP
with symbols:
Y by
X
Bivariate
Fit of Fit
Dur
Transport
Opening (msec) By Temperature
0.5 10
15
Page 8 of 10
By Temperature
JMP
with symbols:
Y by
X 50
Bivariate
Fit of Fit
Dur
Transport
Opening (msec) By Temperature
20
25
10
15
20
25
Page 8 of 10
Bivariate Fit of Avg Vel Transport Open (m/sec)
By Temperature
Temperature
Temperature
0.4
Transformed Fit Log
Avg Vel Transp Open (m/sec)
Dur Transport Opening (msec)
Avg Vel Transp Open (m/sec)
Bivariate
Fit of Avg Vel Transport Open
0.015
75
(m/sec)
Transformed Fit Log
0.5
Bivariate Fit of Dur Transport Opening (msec)
By Temperature
Bivariate
Fit of Avg Vel Transport Open (m/sec) By Tempe
0.3
200
0.2
175
1500.1
125
10
15
0.4
0.3
0.2
20
25
Temperature
100
0.1
Transformed Fit Log
10
15
JMP
with symbols:
Y by
Bivariate
Fit of Fit
Avg
VelX Tongue Retr (m/sec) By Temperatu
20
25
75
Bivariate
Fit of Dur Tongue Retr (msec) By Temperature
Bivariate
Fit of Avg Vel Tongue Retr (m/sec)
By Temperature
Temperature
20
Transformed Fit Log
25
180
160
Bivariate Fit of Avg Vel Tongue Retr (m/sec) By Temperature
Temperature
Dur Tongue Retr (msec)
15
0.4
Transformed Fit Log
Avg Vel Tongue Retr (m/sec)
Avg Vel Tongue Retr (m/sec)
50
0.5
10
0.5
140
Bivariate
Fit of Avg Vel Transport Open (m/sec)120
By Temperature
0.3
0.2
0.1
10
15
0.4
0.3
0.2
20
100
80
60
40
25
10
15
Temperature
20
25
Temperature
0.1
Transformed
Fit Log
Figure
4. Scatterplots
of prey transport and tongue
retraction
Transformed
Fit Log kinematic and
10
15
20
25
Bivariate
Fit of results
Dur Tongue
Retr (msec)
By Bivariate
Temperature
Fit
Durare
Mouth
Closing
dynamic
at experimental
temperatures.
Allof
plots
formatted
as (msec) By Temperature
Temperature
described in the legend of figure 2.
Transformed Fit Log
250
33
Dur Mouth Closing (msec)
Bivariate Fit of Dur Tongue Retr (msec) By Temperature
200
150
100
50
10
15
20
Temperature
25
JMP with symbols: Fit Y by X
Pa
Bivariate Fit of Max Lunge Dist (cm) By Temperature
14
12
Max Lunge Dist (cm)
Max Lunge Dist (cm)
JMP with symbols: Fit Y by X
10
Bivariate
Pa
Fit of Max Lunge Dist (cm) By Temperature
14
8
12
6
10
4
8
2
10
6
15
20
25
Temperature
Transformed Fit Log
4
Bivariate Fit of Max Lunge Vel (m/sec) By Temperature
2
1.3 10
1.2
15
20
1.1
Transformed Fit Log
1
Bivariate
Fit of Max Lunge
0.9
Avg Lunge Vel (m/sec)
Max Lunge Vel (m/sec)
Max Lunge Vel (m/sec)
25
Temperature
Vel (m/sec) By Temperature
0.8
1.3
0.7
1.2
0.6
1.1
0.51
0.4
JMP
with
Y by
X
0.9symbols:
Bivariate
Fit of Fit
Max
Lunge
0.3
0.8
Bivariate
0.2
0.7
0.4 10
0.6
Vel (m/sec) By Temperature
Fit of Avg Lunge Vel (m/sec) By Temperature
15
20
Temperature
0.35
0.5
0.4 Transformed Fit Log
0.3
0.3
0.25
Bivariate
Fit of Avg Lunge Vel
0.2
0.2
10
15
20
0.15
Temperature
0.1
25
(m/sec) By Temperature
25
Transformed Fit Log
0.05
Bivariate Fit of Avg Lunge Vel (m/sec) By Temperature
0
-0.05
10
15
20
25
Temperature
Dur Ballistic Opening (msec)
Figure 5. Scatterplots
of lunge kinematic
and dynamic results at
Transformed
Fit Log
experimental
temperatures.
plots
are formatted
as described
in Temperature
the
Bivariate
Fit ofAll
Dur
Ballistic
Opening
(msec) By
legend of figure302.
25
20
15
34
10
10
15
20
Temperature
25
Page
10°C
25°C
10°C
0
150
10
180
20
210
30
240
60
270
90
300
120
330
25°C
Figure 6. Montage comparing feedings at 10°C and 25°C. Both image sequences
feature the same animal. Projection is shown in 10 msec timesteps, and retraction
is shown in 30 msec timesteps. The line drawn after the frames at 30 msec
distinguishes the ballistic initial feeding stages from the remainder of the gape
cycle. For brevity, the sequences shown begin with the frame immediately before
the start of ballistic mouth opening. The timesteps shown may have missed peak
or landmark events (e.g., end of tongue projection or ballistic opening). For the
sequences shown, the animal finished ballistic mouth opening after approximately
14 msec at 10°C and 10 msec at 25°C. The frog reached peak tongue projection
after approximately 27 msec at 10°C and 26 msec at 25°C. The animal completed
the full feeding sequence in approximately 282 msec at 10°C and 149 msec at
25°C.
35
Table 1. Minimum and maximum values for variables measured in the Rana pipiens feeding experiments.
Minimum
Maximum
Feedings
Body Temperature (°C)
10.10
26.70
46 (9, 10, 9, 9, 9)
Duration of ballistic mouth opening (s)
0.01
0.02
46 (9, 10, 9, 9, 9)
Gape at end of ballistic mouth opening (m)
0.01
0.02
46 (9, 10, 9, 9, 9)
Duration of tongue projection (s)
0.02
0.09
46 (9, 10, 9, 9, 9)
Maximum tongue reach (m)
0.01
0.04
46 (9, 10, 9, 9, 9)
Duration of transport mouth opening (s)
0.06
0.19
46 (9, 10, 9, 9, 9)
Final gape at end of transport mouth opening (m)
0.01
0.03
46 (9, 10, 9, 9, 9)
Duration of tongue retraction (s)
0.06
0.17
46 (9, 10, 9, 9, 9)
Duration of mouth closing (s)
0.05
0.26
46 (9, 10, 9, 9, 9)
Mean velocity of ballistic opening (m/s)
Maximum instantaneous velocity of ballistic opening
(m/s)
Maximum instantaneous acceleration of ballistic
opening (m/s/s)
Maximum depressor mass-specific power of ballistic
opening (W/kg)
Mean velocity of tongue projection (m/s)
Mean velocity of tongue retraction (m/s)
Mean velocity of transport opening (m/s)
Mean velocity of mouth closing (m/s)
0.37
1.58
46 (9, 10, 9, 9, 9)
0.25
0.80
46 (9, 10, 9, 9, 9)
31.30
99.30
46 (9, 10, 9, 9, 9)
115.09
1783.23
46 (9, 10, 9, 9, 9)
0.14
0.06
0.10
0.06
2.10
0.54
0.54
0.46
46 (9, 10, 9, 9, 9)
46 (9, 10, 9, 9, 9)
46 (9, 10, 9, 9, 9)
46 (9, 10, 9, 9, 9)
Lunge distance (m)
0.02
0.09
41 (8, 10, 8, 8, 7)
Lunge duration (s)
0.08
0.35
41 (8, 10, 8, 8, 7)
Average lunge velocity (m/s)
0.01
0.36
40 (8, 10, 8, 7, 7)
The number of involved feedings is shown for each variable, in total and relative to each animal.
36
37
<0.0001
12.17
27.91
16.51
14.89
2.90
Mean velocity of tongue projection (m/s)
Mean velocity of tongue retraction (m/s)
Mean velocity of transport opening (m/s)
Mean velocity of mouth closing (m/s)
40.99
56.86
42.50
72.98
13.68
13.42
11.50
12.32
F-ratio
7.76
2.90
22.05
36.53
21.24
1.20
34.83
104.63
<0.0001
<0.0001
<0.0001
<0.0001
0.0007
0.0007
0.0016
0.0011
P-value
0.0082
0.0963
<0.0001
<0.0001
<0.0001
0.2794
<0.0001
<0.0001
Temperature Effects
1.42
6.62
0.89
9.94
5.62
5.31
4.01
0.01
F-ratio
0.43
1.62
4.16
0.03
3.02
1.58
2.58
10.96
0.2403
0.0140
0.3522
0.0031
0.0228
0.0266
0.0522
0.9058
6, 45
6, 45
6, 45
6, 45
6, 45
6, 45
6, 45
6, 45
DF
P-value
0.5138 6, 45
0.2102 6, 45
0.0482 6, 45
0.8708 10, 45
0.9010 6, 45
0.2160 6, 45
0.1165 6, 45
0.0020 6, 45
Intial Prey Distance
0.022
0.022
0.009
0.034
0.020
0.012
0.007
0.008
Temperature
Slope
-0.006
0.002
0.009
-0.015
-0.013
0.002
-0.012
-0.032
1.66
1.65
1.23
2.17
1.57
1.33
1.19
1.20
Q10
0.86
1.04
1.24
0.70
0.74
1.04
0.75
0.48
0.60
0.61
0.81
0.46
0.64
0.75
0.84
0.83
1/Q10
1.16
0.96
0.81
1.43
1.35
0.96
1.33
2.08
Lunge distance (m)
2.81
0.0409
8.11
0.0074
98.50
<0.0001 6, 40
-0.007
0.85
1.18
Lunge duration (s)
1.25
0.3099
11.09
0.0021
25.23
<0.0001 6, 40
-0.012
0.76
1.31
Average lunge velocity (m/s)
3.88
0.0109
7.11
0.0118
11.09
0.0021 6, 39
0.021
1.64
0.61
P-values are presented in bold type for those values that were statistically significant against a Bonferroni-corrected alpha of 0.0029 (0.05 / 17). Inverse Q10 values are
highlighted for significant non-rate values, while Q10 values are highlighted for significant rate values. Q10 and inverse Q10 values associated with significance for effect of
temperature are shown in bold type.
<0.0001
<0.0001
<0.0001
0.0341
0.0002
<0.0001
16.29
7.29
<0.0001
18.03
P-value
0.0057
<0.0001
<0.0001
<0.0001
0.1301
<0.0001
0.1022
0.5867
Mean velocity of ballistic opening (m/s)
Maximum instantaneous velocity of ballistic opening
(m/s)
Maximum instantaneous acceleration of ballistic
opening (m/s/s)
Maximum depressor mass-specific power of ballistic
opening (W/kg)
Duration of ballistic mouth opening (s)
Gape at end of ballistic mouth opening (m)
Maximum tongue reach (m)
Duration of tongue projection (s)
Duration of transport mouth opening (s)
Final gape at end of transport mouth opening (m)
Duration of tongue retraction (s)
Duration of mouth closing (s)
F-ratio
4.28
35.47
25.76
13.15
1.90
9.60
2.08
0.72
Table 2. ANCOVA results for kinematic and dynamic variables examined.
Individual Effects