Development of Contractile and Energetic Capacity in Anuran Hindlimb Muscle during Metamorphosis

Article (PDF Available)inPhysiological and Biochemical Zoology 76(4):533-43 · July 2003with29 Reads
DOI: 10.1086/376422 · Source: PubMed
Anuran larvae undergo water-to-land transition during late metamorphosis. We investigated the development of the iliofibularis muscle in bullfrog tadpoles (Rana catesbeiana) between Gosner's stage 37 and stage 46 (the last stage). The tadpoles began staying in shallow water at least as early as stage 37, kicking from stage 39, active hindlimb swimming from stage 41, and emerging onto shore from stage 42. For control tadpoles kept in water throughout metamorphosis, muscle mass and length increased two- to threefold between stages 37 and 46, with rapid increases at stage 40. Large, steady increases were found in femur mass, tetanic tension, contraction rate, and power between stages 37 and 46. Concentrations of ATP and creatine phosphate and rates of the phosphagen depletion and the activity of creatine kinase increased significantly, mainly after stage 43. Shortening velocity, tetanic rise time, and half-relaxation time varied little. Energy charge (the amount of metabolically available energy stored in the adenine nucleotide pool) remained unchanged until stage 43 but decreased at stage 46. Compared with the control, experimental tadpoles that were allowed access to both water and land exhibited 1.2- to 1.8-fold greater increases in femur mass, tetanic tension, power, phosphagen depletion rates, and creatine kinase activities at late metamorphic stages but no significant differences for other parameters measured. In sum, most hindlimb development proceeds on the basis of the increasingly active use of limbs for locomotion in water. The further increases in tension, mechanical power, and "chemical power" on emergence would be advantageous for terrestrial antigravity performance.

Full-text (PDF)

Available from: Inho Choi, Jul 30, 2014
Development of Contractile and Energetic Capacity in Anuran
Hindlimb Muscle during Metamorphosis
Jin Cheol Park
Han Suk Kim
Masamichi Yamashita
Inho Choi
Department of Life Science, College of Liberal Arts and
Science, Yonsei University, 234 Maeji-Ri, Heungup-Myon,
Wonju, Gangwon-Do 222-710, Republic of Korea;
of Space and Astronautical Science, Sagamihara 229-8510,
Accepted 3/17/03
Anuran larvae undergo water-to-land transition during late
metamorphosis. We investigated the development of the iliofi-
bularis muscle in bullfrog tadpoles (Rana catesbeiana) between
Gosner’s stage 37 and stage 46 (the last stage). The tadpoles
began staying in shallow water at least as early as stage 37,
kicking from stage 39, active hindlimb swimming from stage
41, and emerging onto shore from stage 42. For control tadpoles
kept in water throughout metamorphosis, muscle mass and
length increased two- to threefold between stages 37 and 46,
with rapid increases at stage 40. Large, steady increases were
found in femur mass, tetanic tension, contraction rate, and
power between stages 37 and 46. Concentrations of ATP and
creatine phosphate and rates of the phosphagen depletion and
the activity of creatine kinase increased significantly, mainly
after stage 43. Shortening velocity, tetanic rise time, and half-
relaxation time varied little. Energy charge (the amount of met-
abolically available energy stored in the adenine nucleotide
pool) remained unchanged until stage 43 but decreased at stage
46. Compared with the control, experimental tadpoles that were
allowed access to both water and land exhibited 1.2- to 1.8-
fold greater increases in femur mass, tetanic tension, power,
phosphagen depletion rates, and creatine kinase activities at
late metamorphic stages but no significant differences for other
parameters measured. In sum, most hindlimb development
proceeds on the basis of the increasingly active use of limbs
for locomotion in water. The further increases in tension, me-
* Corresponding author; e-mail:
Physiological and Biochemical Zoology 76(4):533–543. 2003. 2003 by The
University of Chicago. All rights reserved. 1522-2152/2003/7604-2115$15.00
chanical power, and “chemical power” on emergence would be
advantageous for terrestrial antigravity performance.
From the perspective of gravitational biology, amphibian met-
amorphosis poses interesting questions about limb muscle de-
velopment during the water-to-land transition. Amphibians
were the first vertebrate to emerge on land. The life history of
most anurans repeats this transition during late metamorpho-
sis. During the transformation, anuran tadpoles undergo a
complete restructuring of the body plan to fit terrestrial life
(Burggren and Just 1992). One of the most prominent sets of
changes is that of the structures needed for terrestrial loco-
motion. According to the staging method of Gosner (1960),
tadpole hindlimbs emerge at stage 26, toes begin to differentiate
at stage 37, forelimbs emerge at stage 42, and the tail is resorbed
between stages 40 and 46 (the last stage). Thyroid hormone
(e.g., T
) begins to rise from stage 42 and peaks at stage 43–
44, which represents the climax of metamorphosis (Burggren
and Just 1992). During the transition, tadpoles shift from using
the tail fin for propulsion to using the hindlimbs for propulsive
thrusts in water or on land. However, the transformation of
the locomotory structures is not enough for successful emer-
gence onto land. Because of the loss of aquatic buoyancy at
emergence, the limbs must also adjust to bearing their own
weight for postural balance and jumping. How does the limb
muscle develop as tadpoles emerge for the first time? In ad-
dition, does weight bearing render an additional effect on the
contractile and energetic development of the limb muscles?
The metamorphic transition of anuran muscles (i.e., from
relatively unloaded to steady loading) provides an animal model
that has some similarities to overload training or reloading after
space flight. Previous studies have shown that muscle mass,
contractility, excitability, and accumulation of electrolytes (e.g.,
) increase markedly in vertebrate skeletal muscle dur-
ing the early growth period (Hazlewood and Nichols 1969;
Swynghedauw 1986). Moreover, muscle function often develops
quickly, in parallel with locomotory or behavioral maturity dur-
ing growth. For instance, tension, shortening velocity, power
production, and oxidative and glycolytic capacities of skeletal
muscles rapidly reach the adult level during development de-
pending on the functional precocity of the species (Swynghe-
dauw 1986; Choi et al. 1993; Choi and Ricklefs 1997). Precocity,
534 J. C. Park, H. S. Kim, M. Yamashita, and I. Choi
one of the developmental patterns opposite to altriciality, em-
phasizes differentiation of cells at the cost of growth, which
results in early functional maturity.
In addition to the developmental process, overload or reload
of muscles (e.g., by exercise or after space flight) produces
compensatory adjustments in muscle structure and function
(Fitts et al. 2001). Overloaded muscles become hypertrophic,
with increases in their fiber diameter and mass, density of mi-
tochondria, sarcoplasmic protein content, and interstitial space
(Freeman and Luff 1982; Pottle and Gosselin 2000). Muscle
tension, shortening velocity, and power of limb muscles also
change according to the type and intensity of loading, muscle
size, and myofiber type (Tsika et al. 1987; Pette 2001). Muscle
force, for example, increases with overloading, but tension
(normalized by cross-sectional area) decreases because of in-
creased fiber size (Roy et al. 1982).
Like muscle contractility, energetic capacity is also highly
responsive to work load. Muscles trained by endurance exercise
increase their numbers of mitochondria and, concomitantly,
the capacity to produce adenosine triphosphate (ATP) by ox-
idative phosphorylation (Gollnick 1986). ATP is the immediate
energy source for myofibrillar contraction and is hydrolyzed to
adenosine diphosphate (ADP) and inorganic phosphate by my-
osin ATPase. ATP is resynthesized by creatine phosphate (CrP),
the energy storage molecule that delivers high-energy phosphate
to ADP via creatine kinase (CK; Gollnick 1986). Because anuran
hindlimbs usually exhibit burst locomotion as in swimming or
jumping, CK activity would determine the rate of energy trans-
fer from CrP to ATP for such rapid locomotion. Thus, it is
likely that, in addition to the aforementioned developmental
changes, muscle overloading during metamorphosis would also
increase the muscles’ CK activity and capacity to liberate phos-
phagen (ATP, CrP).
Little is known about the development of muscle function
in anuran larvae. During the transitional stages, tadpoles use
both the tail and hindlimbs, but as the tail degenerates at the
climactic stages, they use limbs (mainly hindlimbs) for loco-
motion, either kicking/swimming in water or walking/hopping
on land (Stehouwer 1992). Thus, the hindlimb muscles of the
mature tadpoles experience two types of loading: swimming
thrust in water and weight bearing on shore. The loading to
the muscles increases further when the tadpoles jump (Marsh
and John-Alder 1994). In addition, tadpoles experience sub-
stantial dehydration after emergence onto land, which causes
significant decreases in body mass, plasma volume, oxygen con-
sumption, and muscle tension (Hillman 1982; Gatten et al.
1992). Muscle performance would be complicated by the effect
of dehydration whenever the tadpoles emerge.
In this article, we explored the development of contractile
and energetic capacities of a hindlimb muscle in tadpoles of
the bullfrog, Rana catesbeiana. We hypothesized that the hind-
limb muscles of tadpoles would exhibit intrinsic development
that would match the normal locomotory demand even in the
absence of emergence. We further hypothesized that in re-
sponse to weight loading, the tadpoles would show even faster
Material and Methods
This study was approved by the Yonsei University Animal Care
and Use Committee. We collected bullfrog tadpoles (Rana ca-
tesbeiana) in June–September 2000–2002 from a local reservoir
in southwestern Korea. We collected tadpoles in stages before
37 because this stage would be the earliest at which it is possible
to obtain a sizable muscle mass from the hindlimb. Tadpoles
were kept in a pool in an environmental chamber (24–25C)
with a 13L : 11D light : dark cycle and were fed ad lib. with
Tetra fish feeders (Tetra, Melle, Germany).
Behavior and Locomotion
To find the specific stage at which tadpoles begin to emerge,
we randomly selected 20 tadpoles at each stage from 37 to 46.
The tadpoles were placed in an aquarium (length # width #
) containing a slope of gravelheight p 1m# 0.5 m # 0.5 m
to mimic the natural shore (Stehouwer 1992). The tadpoles
were able to move from deep water to shallow water to land
at will. A remote-controlled Sharp Slimcam video camera was
set at an angle of about 45 above the aquarium in a position
where it could record all three niches simultaneously. The 20
tadpoles introduced were given 4 h for acclimation and were
recorded for5sattheendofthe4-hacclimation period. We
repeated the video recording for each stage and counted the
number of tadpoles in each of the three zones. From the be-
havioral observation of individuals, we also determined the first
stages at which the hindlimb activities (e.g., kicking, swimming)
occurred in water (Stehouwer 1992).
Control versus Experimental Subjects
We chose stages 37, 40, 43, and 46 for the muscle contraction
and energetics experiments. Tadpoles in the pool (at stages
before 37) were randomly divided into a control group and an
experimental group and were reared in two separate aquaria
( ). The control tadpoles were grown only1m# 0.5 m # 0.5 m
in water so that they never had a chance to experience their
own weight or a terrestrial environment. Small Styrofoam
pieces were provided in the aquarium so that the mature tad-
poles could access them with their forelimbs for air breathing.
The experimental tadpoles were grown in the same aquarium
used for the behavior and locomotion experiment. Both aquaria
were aerated continually and covered with an acrylic plate that
kept the relative humidity of the air at 97.0%–99.9% at all
times (Cole-Parmer 37550-10 thermohygrometer). Body tem-
perature was measured from the abdominal cavity of three stage
Muscle Development in Anuran Larvae 535
Figure 1. Preference of tadpoles for the three niches (deep water, shal-
low water, or shore) through the late metamorphic stages.
46 tadpoles reared in the land/water aquarium using a Cole-
Parmer 91100-20 thermometer and a copper-constantan ther-
mocouple (0.025-mm diameter, California Fine Wire, Grover
City, Calif.). Body temperature was within 0.3–0.5C of am-
bient air and water temperature.
Muscle Contraction
Preparation. At each stage, a tadpole was pithed and its iliofi-
bularis muscle was dissected from the right hindlimb. The mus-
cle was immediately immersed in a petri dish containing cooled
oxygenated Ringer’s solution (115 mM NaCl, 2.5 mM KCl, 1.8
mM CaCl
, 0.85 mM NaH
glucose [pH 7.2]). The tendons at each end of the muscle were
tied tightly with silk threads. The length of each tendon was
kept as short as possible in order to minimize its stretch at
contraction. The muscle was then placed horizontally in a mus-
cle bath (80 mL). One end of the muscle was connected to a
Harvard 60-2995 force transducer (natural frequency, 60 Hz;
maximum capacity, 50 g) and the other end to a vertical beam
of a T-lever attached to a Harvard 52-9511 rotary transducer.
During isometric contraction, the muscle was held immobile
by a stop on the horizontal side of the T-lever. For isotonic
shortening, the stop was quickly removed by a 24-W solenoid
actuator (Choi et al. 1998). The muscle bath temperature was
maintained at C by a refrigerated circulator (Kookje10.0⬚Ⳳ1
Scien 33-WBF-15) and was continuously monitored using a
Cole-Parmer 91100-20 digital thermocouple thermometer.
Muscles were stimulated with a pair of bright platinum elec-
trodes connected to a Grass S48 stimulator that supplied a 1.0-
ms square wave pulse or pulse train. All electrical signals from
the transducers were digitized by a Biopac MP100 A/D con-
verter and stored on computer disk.
Experiments. We determined the optimum muscle length (ML)
and supramaximal voltage needed to produce maximum twitch
force for each preparation as in Choi et al. (1998). A stimulus
frequency of 200 Hz was used to attain fully fused tetanic force
). A rest period of at least 20 min was allowed between
contraction trials. In after-loaded isotonic contractions, the
muscle was stimulated at the same frequency. When the muscle
exerted a maximal isometric contraction, the stop on the hor-
izontal arm of the T-lever was removed quickly by a solenoid
actuator (24 W). The muscle then shortened against the im-
posed load, ranging from 10% to 80% F
. Shortening velocity
(V) was determined from a constant length change over a time
period of 10–15 ms after shortening started. From the data of
and of isotonic series, a force-velocity relationship was es-
tablished for that muscle. We adopted a curve-fit equation de-
veloped by Marsh and Bennett (1986). We then obtained a
power-velocity relationship with every pair of force and velocity
measurements. From the two relationships, we determined a
particular force (F) and a velocity (V) at maximum power,
which are known to be the effective contractile capacities ex-
erted by animals in peak performance (Lutz and Rome 1994).
After each experiment was completed, we measured the op-
timum ML in place using a micrometer under a light micro-
scope. Muscle mass (M
) was measured with a chemical balance
after the tendons were removed. Cross-sectional area (CSA) of
the muscle was approximated by the equation CSA p
, assuming the specific gravity of the .tissue 1.0
Tetanic force (F or F
) was converted to tension (T or T
dividing the value by the CSA. Shortening velocity V was nor-
malized with ML and power with M
. From the stored tension
curves, tetanic rise time (TRT), half-relaxation time (HRT),
and rate of tetanic tension production ( ) were obtained.dT/dt
Because the time points of rising and falling of each tension
curve were unclear, we defined TRT as the time length between
0.1 and 0.9 T
in the rising phase of the curve and HRT as the
time length between 0.9 and 0.5 T
in the falling phase. The
was calculated from 0.8 T
divided by TRT.dT/dt
Energetic Study
Tissue Preparation. To determine the energetic capacity of the
iliofibularis muscle, we measured the phosphagen (ATP, CrP)
depletion rate of the muscle using general principles adopted
from Cain et al. (1962) and Spriet (1989). The iliofibularis
muscles were dissected from both hindlimbs. The left limb
muscle was frozen directly (unstimulated) in liquid nitrogen.
The right limb muscle was frozen after a 3-s electrical stimu-
lation using the following sequence of stimulation and quick-
freezing methods. The muscle was prepared as in the isometric
contraction study, but in this case, the muscle was placed in
536 J. C. Park, H. S. Kim, M. Yamashita, and I. Choi
Table 1: Summary of morphological variables in control and experimental groups of bullfrog tadpoles at
four stages
Groups and Variables
37 40 43 46
Body mass (g) 24.87 1.18 (14) 26.95 1.28 (14) 26.43 .66 (14) 22.18 1.16 (14)
Muscle mass (mg) 12.4 1.0 (14) 26.7 2.3 (14) 29.4 3.8 (14) 39.4 1.8 (14)
Muscle length (mm) 8.03 .44 (7) 13.44 .57 (7) 14.71 1.21 (7) 17.54 .48 (7)
Femur mass (mg) 6.7 1.2 (11) 10.6 1.2 (12) 18.3 3.0 (12) 28.8 3.9 (11)
Body mass (g) 24.10 .86 (14) 25.54 .96 (14) 27.19 .58 (14) 20.31 .96 (14)
Muscle mass (mg) 12.7 1.1 (14) 18.0 1.9 (14) 28.6 1.2 (14) 40.1 2.3 (14)
Muscle length (mm) 8.16 .27 (7) 10.87 .65 (7) 15.13 .64 (7) 16.86 .71 (7)
Femur mass (mg) 9.0 2.8 (10) 11.2 1.2 (14) 24.1 2.0 (12) 39.2 5.0 (11)
Note. Values are SE. Sample size is given in parentheses. All variables had significant differences within experimental groupmeans 1
(one-way ANOVA with stage entered as a main effect). There were significant differences between groups for femur mass (ANCOVA with
the stage entered as a covariate).
an aluminum cell instead of the muscle bath. The cell tem-
perature was kept at C by a water jacket connected10⬚Ⳳ0.5
to a water circulator. One end of this muscle was tied to a silk
thread and connected to the Harvard force transducer. The
other end was tied to a silver chain and hung to a short pro-
jection of a robust beam fixed adjacent to the aluminum cell.
The top and bottom of the cell were open so that the muscle
could drop down through the cell when released. Two electrodes
were introduced inside from the top. A solenoid actuator was
set near the projection and was controlled by the electric signal
of a Grass S48 stimulator. After a 3-s muscle stimulation, the
solenoid was activated, pulling the silver chain from the pro-
jection, allowing the muscle to rapidly drop down. Right below
the cell, a liquid nitrogen bath was placed so that the fallen
tissue was frozen less than 0.1 s after the end of stimulation.
The depletion rate of each phosphagen was determined from
the difference in the phosphagen concentration between un-
stimulated and stimulated muscles divided by stimulation time
(3 s).
Assay. Assays for adenine nucleotides and CrP were adopted
from Spriet (1989) and Lin et al. (1998). Each frozen tissue
was ground into powder in liquid nitrogen and 0.8 N perchloric
acid. The powder was allowed to warm up slowly to 0C. The
homogenate was centrifuged for 15 min at 600 g (4C). The
supernatant was neutralized with 6 N KOH, and the precipi-
tated potassium perchlorate was removed by centrifugation for
15 min at 15,500 g (4C). The supernatant was filtered through
a cellulose filter (0.45-mm pore size), and the filtrate was assayed
for [ATP], [ADP], [AMP], [Cr], and [CrP] by high-perform-
ance liquid chromatography (HPLC).
We used a reverse-phase ion-pairing HPLC with gradient
elution to assay adenine nucleotides. The system (Younglin In-
strument, Seoul, Korea) consisted of an injector/mixer, two
solvent delivery systems (MP930), and a dual absorbance de-
tector (M730D). We used an RP-18 column (Perkin Elmer) for
the stationary phase. The column had a particle size of 5 mm
and inner dimensions of mm. The mobile phase
220 mm # 4.6
consisted of two buffers: buffer A (100 mM KH
tetrabutylammonium hydrogen sulfate [pH 6.0]) and buffer B
(20% methanol, 20% acetonitrile, and 60% buffer A [by vol-
ume]). Gradient elution was conducted in the following steps.
Pure buffer A was delivered for 5 min, and then buffer B was
linearly increased to 35% over a 10-min period. The elution
was followed by a 10-min isocratic elution with 35% buffer B,
then linear increments of buffer B to 55% over a 7-min period,
and again to 100% over the next 5 min. Finally, 100% buffer
B was delivered for 10 min to complete the elution. A period
of 12 min was given for column equilibration between runs.
The flow rate was fixed at 1 mL min
throughout the elution
cycle. A sample of 20 mL was introduced into the column for
each determination. The elution was monitored with a UV
detector at 254 nm. For Cr and CrP assays, ion-pair HPLC was
performed on the same reverse-phase column with isocratic
elution. Buffer A was used for the mobile phase. The flow rate
was fixed at 1 mL min
throughout the elution cycle. The
elution was monitored with a UV detector at 218 nm. After
completing the assays, we calculated the concentration of each
chemical in the tissue from a relationship between known con-
centrations of standard solutions for that chemical and areas
under HPLC curves obtained from corresponding standard so-
lutions. We normalized each sample’s concentration with the
tissue mass.
Muscle Development in Anuran Larvae 537
Figure 2. Typical tension-velocity (A) and power-velocity (B) curves for tadpoles of the control group at stages 40 and 46. The curve-fit equation
used for the tension-velocity relationships was ), where B and C have dimensions of velocity and AV p B(1 T/T )/(A T/T ) C(1 T/T
oo o
is dimensionless (Marsh and Bennett 1986). For the tadpole at stage 40, muscle mg, muscle mm, cross-sectionalmass p 21.8 length p 11.9
, , , and . For the tadpole atarea p 1.83 T p 83.13 V p 6.51 P p 69.77 A p 0.112 B p 0.437 C p 2.614
stage 46, muscle mg, muscle mm, cross-sectional mm
,mass p 30.7 length p 14.4 area p 2.13 T p 169.49 V p 7.45
, , , and . The T and V at P
(the point at which peak locomotory performance wouldP p 158.76 A p 0.118 B p 0.548 C p 2.820
be exerted) were 37.08 kN m
and 1.89 ML s
, respectively, for stage 40 and 75.0 kN m
and 2.12 ML s
, respectively, for stage 46.
CK Activity
General procedures for CK activity measurement followed those
of Krause and Wegener (1996). The iliofibularis muscles of both
limbs in each tadpole were dissected and ground into powder
in liquid nitrogen. The powder was allowed to warm up slowly
to 0C and then was sonicated on ice in 10 volumes of a
medium containing 50 mM triethanolamine-HCl, 2 mM EDTA,
30 mM b -mercaptoethanol (pH 7.5). We used a Sigma DG147-
K assay kit and determined CK activity at nm follow-l p 340
ing the kit procedure. The assay temperature was kept at 25C.
Each homogenate was diluted to 1 : 150 because of extremely
high CK activity in the muscle. We recorded absorbance at 0,
1, 2, and 3 min versus water as reference. We determined mean
absorbance change per min (DAmin
) for the 3-min exam-
ination and calculated CK activity by multiplying DAmin
the dilution factor.
Data Analyses
Data are presented as SE unless otherwise noted.mean 1
The developmental change in contractile and energetic capacity
within each group was examined by one-way ANOVA with
metamorphic stage entered as the main effect. Scheffe’s mul-
tiple-comparison test was used to examine the significance of
differences among stages. Intergroup difference was examined
by ANCOVA with the stage as a covariate. All the statistical
procedures were conducted with SPSS/PC.
Behavior and Locomotion
Tadpoles preferred to stay in deep water yet began to venture
into shallow water at least as early as stage 37 (Fig. 1). The
proportion of tadpoles moving from deep water to shallow
water (where their dorsal parts emerged from the water surface)
rose slightly at stages 41–42. The tadpoles increased their pref-
erence for the shore from stage 42 onward. The tadpoles began
kicking from stage 39 and active hindlimb swimming from stage
41 onward.
Morphological data for control and experimental groups are
summarized in Table 1. In both groups, body mass (M
creased slowly with stage but decreased at the last stage of
metamorphosis when the tail fin and many organs are remod-
eled dramatically and the tadpoles cease feeding. Although there
may have been some dehydration in the experimental tadpoles,
it was not enough to have a measurable effect on mass. M
ML, and femur mass (M
) increased 3.1-, 2.1-, and 4.3-fold,
respectively, between stages 37 and 46, with fairly rapid in-
creases in M
and ML at stage 40. In particular, the increment
in M
was more rapid for the experimental group than for the
control after stage 40 (ANCOVA, , ),F p 5.03 P p 0.027
1, 92
which results in a 1.3-fold difference at stages 43–46 between
538 J. C. Park, H. S. Kim, M. Yamashita, and I. Choi
Figure 3. Changes in maximum tetanic tension (A) and maximum
power (B) for control and experimental groups between stages 37 and
46. Both tension and power increased significantly over the meta-
morphic stages within each group (one-way ANOVA with the stage
given as a main effect) and increased more rapidly in the experimental
group than in the control after stage 40 (ANCOVA with the stage as
a covariate). Symbols represent means of seven individuals, and vertical
bars represent 1 SE.
the two groups. Intergroup differences were not statistically
significant for M
, M
, and ML.
Muscle Contractility
Figure 2 presents typical curves demonstrating tension-velocity
and power-velocity relationships for stages 40 and 46 in the
control group. Our data were fit well by the hyperbolic-linear
equation in Marsh and Bennett (1986; see Fig. 2 legend for
equation). Figure 3 illustrates maximum tetanic tension and
maximum power as a function of metamorphic stage. The max-
imum tension of the control group increased steadily from 33.9
to 158.7 kN m
between stages 37 and 46 (Fig. 3A; one-way
ANOVA, , ). The maximum tension ofF p 140.9 P
! 0.001
3, 27
the experimental group increased more rapidly after stage 40
than did the control (ANCOVA, , ),F p 12.10 P p 0.001
1, 55
which resulted in 1.2- to 1.3-fold greater tension at stages 43
and 46.
The pattern of change in maximum power across the stages
was similar to that in maximum tension for both groups (Fig.
3B). Maximum power of the control group increased steadily
from 27.6 to 133.1 W kg
between stages 37 and 46 (F p
3, 27
, ). The experimental group gained power ca-119.4 P ! 0.001
pacity more rapidly after stage 40 (ANCOVA, ,F p 11.08
1, 55
), resulting in 1.3- and 1.4-fold differences at stagesP p 0.001
43 and 46, respectively.
Rate properties of the iliofibularis muscle are summarized
in Table 2. There were no significant alterations in shortening
velocity (V), TRT, and HRT for the control groups over the
metamorphic stages ( ). Rate of tetanic tension pro-P
1 0.05
duction ( ) increased 10.0-fold between stages 37 and 46dT/dt
( , ). For all the rate variables (V,TRT,F p 34.33 P
! 0.001
3, 27
HRT, and ), there were no statistically significant differ-dT/dt
ences between the groups (ANCOVA, ).P
1 0.05
Biochemical Alterations
The energy depletion capacity of the left (unstimulated) iliofi-
bularis muscle is summarized in Table 3. In the control group,
the concentration of ATP increased slowly from 3.39 to 4.44
mmol g
between stages 37 and 43 and rapidly to 12.34 mmol
at stage 46 ( , ). CrP concentration alsoF p 191.4 P ! 0.001
3, 27
showed a similar pattern of increments, with the levels of 20.51–
30.37 m mol g
at stages 37–43 and 62.05 mmol g
at stage 46
( , ). Energy charge, representing the frac-F p 151.1 P
! 0.001
3, 27
tion of total adenylate pool “charged” by ATP or its equivalent
(Segel 1975), remained at 0.88 during stages 37–43 but de-
creased to 0.85 at stage 46 ( , ). The levelsF p 40.1 P
! 0.001
3, 27
of ATP and CrP of the experimental group were slightly lower
than the corresponding variables of the control, but the overall
pattern of change of each phosphagen did not differ from the
control (ANCOVA, , for ATP;F p 0.29 P p 0.59 F p
1, 55 1, 55
, for CrP). Energy charge for the experimental0.03 P p 0.87
group decreased from approximately 0.87 to 0.88 at stages 40
43 to 0.85 at stage 46, and this pattern of variation was similar
to that of the control ( ; Table 3).P
1 0.05
Rates of ATP and CrP depletion are illustrated as a function
of stage in Figure 4. For the control, the ATP depletion rate
remained at 0.53–0.65 mmol g
between stages 37 and 43
and increased rapidly to 1.27 mmol g
at stage 46
( , ). The CrP depletion rate also showedF p 19.11 P
! 0.001
3, 27
a similar pattern during the four-stage period, with values of
3.70–3.55 mmol g
at stages 37–43 and 6.29 mmol g
at stage 46 ( , ). The ATP depletion rateF p 40.72 P ! 0.001
3, 27
of the experimental group increased slightly faster than that of
the control after stage 40 (ANCOVA, , ),F p 5.87 P p 0.019
1, 55
resulting in 1.44- and 1.29-fold differences at stages 43 and 46,
respectively. The rate of CrP depletion also showed a more
rapid increase in the experimental group than in the control
after stage 40 (ANCOVA, , ), with 1.37-F p 11.02 P p 0.002
1, 55
and 1.55-fold differences at stages 43 and 46, respectively.
Last, the change in muscle CK activity is shown in Figure 5.
CK activity in the control changed significantly from 158.93 U
Muscle Development in Anuran Larvae 539
Table 2: Summary of muscle contractile variables for control and experimental groups of bullfrog tadpoles at four
Groups and Variables
37 40 43 46
Velocity at P
(muscle length s
) 1.99 .09 2.01 .07 2.00 .05 1.88 .05
Tetanic rise time (ms) 142.1 8.6 143.2 10.3 158.4 17.4 162.9 22.7
Half-relaxation time (ms) 70.5 5.5 105.0 17.2 87.7 7.0 83.0 13.3
(kN m
dT/dt .129 .02 .568 .06 .749 .14 1.297 .06
Velocity at P
(muscle length s
) 1.63 .28 1.90 .06 1.95 .06 2.13 .05
Tetanic rise time (ms) 138.2 2.2 121.8 5.2 146.5 6.1 174.8 8.2
Half-relaxation time (ms) 61.0 2.0 109.4 22.7 82.1 15.3 77.7 7.2
(kN m
dT/dt .160 .02 .301 .06 .923 .07 1.543 .16
Note. Values are SE. Sample size was for all the variables, except for half-relaxation time at stage 43 in the experimental groupmeans 1 n p 7
().n p 6
The only variable with a group showing significant differences among the four stages (one-way ANOVA with stage entered as a main effect);
between groups, there were no significant differences for each stage.
at stage 37 to 1,857 U g
at stage 46 ( ,F p 64.73 P !
3, 27
), with a rapid increase between stages 43 and 46. For the0.001
experimental group, the CK activity elevated more rapidly than
the control after stage 40, rendering 1.76- and 1.31-fold dif-
ferences at stages 43 and 46, respectively (ANCOVA, F p
1, 55
,).5.36 P p 0.025
Development without Emergence
Our data demonstrate that the bullfrog tadpoles began active
use of hindlimbs for swimming at stage 41 and weight-bearing
emergence on the shore at stage 42 (Fig. 1). We therefore ex-
pected that intrinsic muscle development might be triggered at
these stages and that rapid increases in growth and functional
development would occur even in the absence of actual emer-
gence. In the control group, a rapid increase was seen in M
and ML at stage 40 (Table 1), where we observed unexpectedly
massive muscles while bone growth was relatively stable (Table
1). There were steady increases in muscle tension (4.7-fold),
rate of tension production (10-fold), and power (4.7-fold),
while little change was seen in shortening velocity and time
variables (TRT, HRT) between stages 37 and 46 (Fig. 3; Table
2). The steady increases in tetanic tension and power may reflect
the gradual switch of locomotion from the tail to hindlimbs
(step-on exercise in shallow water, kicking, swimming) during
the transformation (Huey 1980; Stehouwer 1992). These muscle
tension and power outcomes also imply that contractile and
associated proteins (i.e., myosin, actin, ATPase, etc.) accumulate
steadily across the fibers during these stages. Thus, the number
of sarcomeres and crossbridges per CSA probably increase as
well. Because shortening velocity and time variables were rel-
atively constant, the number of sarcomeres per fiber length
probably remained the same during the late metamorphic
stages. Moreover, the anuran iliofibularis muscle is known to
be composed of five fiber types (three twitch types and two
tonic types; Gans and Gueldre 1992). But because the rate
variables varied little over the late stages, the relative proportion
of these fibers probably did not change much.
The rapid increase was seen most clearly in the measures of
biochemical capacity, such as phosphagen concentration, phos-
phagen depletion rate, and CK activity (Table 3; Figs. 4, 5).
Interestingly, the most rapid increase occurred at the end of
metamorphosis (stage 46) rather than at stages 41–42. We spec-
ulate that the demand for energy utilization in the limb muscles
rises substantially at the time of complete tail loss, at which
time hindlimb thrust becomes the only form of motor activity
(Huey 1980). The small but significant decrease in energy
charge at stage 46 might also reflect the upregulated energy
utilization of the limbs as they become the sole means of lo-
comotion. The increase in the phosphagen concentrations and
CK activities during transformation (Table 3) may indicate a
corresponding increase in mitochondrial content per M
myofibers differentiated to more red fibers during the period.
Development with Free Emergence
Many species of animals move from water to land (or vice
versa) at some time during their life history. Most anurans
undergo such a transition at the end of metamorphosis. Rep-
tiles, birds, and mammals emerge onto land at hatching or
birth when young depart from the amniotic environment (Wal-
ker and Luff 1995). After transition, the animal must suddenly
bear its own full weight, which presents some challenges in
postural maintenance, feeding, predator avoidance, and other
ecological interactions (Wassersug and Sperry 1977). Natural
540 J. C. Park, H. S. Kim, M. Yamashita, and I. Choi
Table 3: Biochemical changes in the left (unstimulated) iliofibularis muscle in control and
experimental bullfrog tadpoles at four stages
Groups and Variables
37 40 43 46
[ATP] (mmol g
3.39 .03 3.99 .05 4.44 .02 12.34 .61
Energy charge
.879 .002 .878 .001 .879 .003 .850 .008
[CrP] (mmol g
20.51 .51 23.80 .54 30.37 .19 62.05 3.00
[ATP] (mmol g
3.11 .04 3.53 .04 4.56 .05 11.74 .48
Energy charge .856 .005 .865 .005 .876 .002 .850 .003
[CrP] (mmol g
22.07 .71 23.77 .77 27.35 .90 61.94 2.16
Note. Values are SE. Sample size was for all groups. Energymeans 1 n p 7 charge p ([ATP]
(Segel 1975).1/2[ADP])/([ATP] [ADP] [AMP])
Significant differences within a group over the four stages (one-way ANOVA with stage given as a main effect).
selection would favor individuals that possess an intrinsic com-
pensatory solution to overcome this transitional problem.
In this study, structural and functional alterations in the
experimental tadpoles relative to the control represent the effect
of emergence on limb growth and development after stage 42,
even though the extent of loading was not monitored quan-
titatively in the two groups. Bone mass and several functional
variables (muscle tension, power, ATP and CrP depletion rates)
exhibited such an effect during the metamorphic climax. How-
ever, the experimental group failed to show a significant effect
of emergence on M
, M
, and ML. The experimental tadpoles
also showed no significant difference from the control in con-
tractile rate variables (shortening velocity, TRT, HRT, rate of
tension production) and phosphagen concentrations. The ob-
servation of similar M
and M
between the two groups may
indicate that dehydration in the experimental animals, partic-
ularly at the last stages, was not marked in our aquaria because
the relative humidity was nearly 100%. Evaporative cooling in
the experimental tadpoles staying on the shore also seemed
insignificant because body temperature was almost the same as
that of the ambient air or water temperature. Under natural
conditions, tadpoles emerging on the shore may experience a
significant reduction in M
and muscle tension because of de-
hydration (Hillman 1982; Gatten et al. 1992). Convective cool-
ing may also lower body temperature of tadpoles. This dec-
rement in temperature would cause further reduction in muscle
and locomotory performance.
The effect of weight loading on muscle adjustments during
metamorphosis may be conceptually similar to that of over-
loading by heavy exercise or reloading after space flight. Ex-
ercise-induced overloading causes muscle hypertrophy with in-
creases in M
and diameter (Freeman and Luff 1982; Pottle
and Gosselin 2000). This was not the case in our study. It has
also been found that reloading or overloading causes serious
muscle damage (e.g., sarcomere disruption and interstitial
edema), which initially brings about force and power decreases
(Fitts et al. 2001). In this study, however, experimental tadpoles
after the first emergence trials showed more rapid elevation of
muscle tension (T
) and power (P
) above that of the controls.
The elevation of these contractile properties may be due to the
fact that emergence was preceded by other forms of limb ex-
ercise (stepping, kicking, swimming). These activities allow
them to adjust to subsequent weight loading and seem im-
portant in maintaining myofibrillar integrity in the developing
muscle. Because M
, ML, and CSA did not differ between the
two groups within each stage, the 1.2- to 1.4-fold greater tension
and power after stage 40 suggests that the experimental group
had more compact packing of myofilaments in the limb muscle.
The rapid increments in M
and rates of energy utilization (ATP,
CrP) after stage 40 may also be the result of the preceding
adjustments in concert with greater muscle contractility.
In our study, rate and time variables (shortening velocity,
tension production rate, time variables) remained similar to
those of the control even after emergence. This intergroup sim-
ilarity implies that the number of sarcomeres per fiber length
and/or myofiber type was not altered by weight bearing. Our
recent study on the soleus muscle in rats demonstrates that
shortening velocity increases and response times (TRT, HRT)
decrease for 1 d of reloading following unloading and then
gradually approach the level of control (a normal loading
group) during 2-wk reloading (I. Choi, unpublished data). In
the case of exercise overloading, shortening velocity usually
decreases in comparison with its control counterpart (Tsika et
al. 1987), but in the other case (e.g., the fast-twitch extensor
digitorum longus), shortening velocity increases, with more
sarcomeres developing in series along myofibers (Freeman and
Luff 1982). These data together indicate that the rate responses
of a muscle vary depending on loading history, loading con-
dition, and the composition of myofiber types.
Maximum tetanic tension of the froglets seemed quite close
Muscle Development in Anuran Larvae 541
Figure 4. Changes in ATP depletion rate (A) and CrP depletion rate
(B) for control and experimental groups between stages 37 and 46.
Statistics are as in Figure 3. Symbols and vertical bars represent
SE ( ).means 1 n p 7
Figure 5. Changes in CK activity for control and experimental groups
between stages 37 and 46. Statistics are as in Figure 3. Symbols and
vertical bars represent SE ( ).means 1 n p 7
to the levels found in adult frogs and toads. While the tension
ranged from 45 to 62 kN m
at stages 37–40, it increased to
186 kN m
at stage 46 in our experimental tadpoles. Shin et
al. (2000) measured a tension of 181 kN m
in the bullfrog
at 1 d posttransformation. These values are similar to the ten-
sion of 190–210 kN m
measured in the semimembranous
muscle and sartorius muscle of adult Rana pipiens and Bufo
americanus at 10–15C (Renaud and Stevens 1984; Lutz and
Rome 1994).
It is interesting to note that in terms of tension production
capacity, bullfrog muscle development seems to parallel pre-
cocial avian muscle development. With regard to the initiation
of pedal locomotion, metamorphosis of anuran larvae and
hatching of avian embryos may be considered as analogous
physiological incidents. The precocial avian muscle generates
maximum tension of 55 kN m
at 3 d prehatching, 70 kN
at 1 d prehatching, and 149 kN m
at 1 d posthatching
(Reiser et al. 1982 [chicken]; Choi and Bakken 1991 [bob-
white]). In contrast, the limb muscle of an altricial bird (red-
winged blackbird) generates maximum tension of 53 kN m
at 1 d posthatching and 217 kN m
at 8 d posthatching (Choi
and Bakken 1991). With the rapid maturity of the muscle func-
tion, pedal locomotion can take place immediately after trans-
formation or hatching in froglets or precocial chicks, whereas
such functional maturity takes several days to weeks in altricial
chicks before fledging (Ricklefs 1983; Choi et al. 1993).
Emergence on the land during the late larval stages influ-
enced biochemical development as well, evidenced by the ATP
and CrP depletion rates and CK activity, which increased more
rapidly after stage 40 in the experimental animals compared
with the controls (Figs. 4, 5). However, the phosphagen con-
centrations increased in a similar pattern between the two
groups. The results may suggest that myofiber types and mi-
tochondrial content per M
were similar between the two
groups within each stage. Furthermore, it would be the enzyme
catalytic capacity (e.g., CK activity) that potentially increased
the energy depletion rates more rapidly in the experimental
group after stage 40 (Figs. 4, 5). In association with muscle
contractile capacity, the rate of energy depletion (“chemical
power”) appears to correlate with mechanical power rather than
shortening velocity during the transitional period (Figs. 3B,4;
Table 2). Because locomotory capacity in anurans is constrained
by the mechanical power of the limb muscles (Lutz and Rome
1994), it will be worth examining energy depletion rates and
muscle mechanics (e.g., power) simultaneously within the same
tissue to test such a correlation.
In conclusion, most hindlimb development proceeded in ac-
cordance with the increasingly active use of limbs for loco-
motion in water. Other than rapid increments in M
and ML
at stage 40, tension, contraction rate, and power increased
steadily, while shortening velocity and contraction and relax-
ation times changed little between stages 37 and 46. Phosphagen
concentrations, phosphagen depletion rates, and CK activities
542 J. C. Park, H. S. Kim, M. Yamashita, and I. Choi
increased substantially at the last stage of transformation. In
response to weight loading, M
, tension, power, energy deple-
tion rates, and CK activities increased more rapidly after the
stage of the first emergence. Intergroup differences were not
evident in M
, M
, ML, shortening velocity, contraction rate
and times, phosphagen concentrations, and energy charge. Be-
cause weight-bearing activities are novel and crucial steps to-
ward terrestrial life for newly emerging tadpoles (Wassersug
and Sperry 1977), the current study may provide an invaluable
clue to understanding how early animals made the transition
from a weightless water environment to the gravitational chal-
lenges of a land environment.
We thank Drs. R. Wassersug, K. Park, and U. J. Jung and two
anonymous reviewers for critical comments and encourage-
ment on this manuscript; K. S. Lee and H. S. Lee for main-
tenance of tadpoles; and Jennifer Macke for editing the text.
This work was supported by Yonsei University Research Fund
of 2001 to I.C.
Literature Cited
Burggren W.W. and J.J. Just. 1992. Developmental changes in
physiological systems. Pp. 467–530 in M.E. Feder and W.W.
Burggren, eds. Environmental Physiology of the Amphibians.
University of Chicago Press, Chicago.
Cain D.F., A.A. Infante, and R.E. Davies. 1962. Chemistry of
muscle contraction: adenosine triphosphate and phospho-
creatine as energy supplier for single contractions of working
muscle. Nature 196:214–217.
Choi I. and G.S. Bakken. 1991. Locomotion and muscle func-
tion during postnatal development of the northern bobwhite
(Colinus virginianus): effect of body temperature. Physiol
Zool 64:653–672.
Choi I., Y. Cho, Y.K. Oh, N.-P. Jung, and H.-C. Shin. 1998.
Behavior and muscle performance in heterothermic bats.
Physiol Zool 71:257–266.
Choi I. and R.E. Ricklefs. 1997. Changes in protein and elec-
trolyte concentrations in the pectoral and leg muscles during
avian development. Auk 114:688–694.
Choi I., R.E. Ricklefs, and R.E. Shea. 1993. Skeletal muscle
growth, enzyme activities, and the development of ther-
mogenesis: a comparison between altricial and precocial
birds. Physiol Zool 66:455–473.
Fitts R.H., D.R Riley, and J.J. Widrick. 2001. Functional and
structural adaptations of skeletal muscle to microgravity. J
Exp Biol 204:3201–3208.
Freeman P.L. and A.R. Luff. 1982. Contractile properties of
hindlimb muscles in rat during surgical overload. Am J Phys-
iol 242:C259–C264.
Gans C. and G.D. Gueldre. 1992. Striated muscle: physiology
and functional morphology. Pp. 277–313 in M.E. Feder and
W.W. Burggren, eds. Environmental Physiology of the Am-
phibians. University of Chicago Press, Chicago.
Gatten R.E., Jr., K. Miller, and R.J. Full. 1992. Energetics at rest
and during locomotion. Pp. 314–377 in M.E. Feder and W.W.
Burggren, eds. Environmental Physiology of the Amphibians.
University of Chicago Press, Chicago.
Gollnick P.D. 1986. Metabolic regulation in skeletal muscle:
influence of endurance training as exerted by mitochondrial
protein concentration. Acta Physiol Scand 128:53–66.
Gosner K.L. 1960. A simplified table for staging anuran em-
bryos and larvae with notes on identification. Herpetologica
Hazlewood C.F. and B.L. Nichols. 1969. Changes in muscle
sodium, potassium, chloride, water and voltage during mat-
uration in the rat: an experimental and theoretical study.
Johns Hopkins Med J 125:119–133.
Hillman S.S. 1982. The effects of in vivo and in vitro hyper-
osmolality on skeletal muscle performance in the amphibians
Rana pipiens and Scaphiopus couchii. Comp Biochem Physiol
Huey R.B. 1980. Sprint velocity of tadpoles (Bufo boreas)
through metamorphosis. Copeia 1980:537–540.
Krause U. and G. Wegener. 1996. Exercise and recovery in frog
muscle: metabolism of PCr, adenine nucleotides, and related
compounds. Am J Physiol 270:R811–R820.
Lin A.T.-L., K.-K. Chen, C.-H. Yang, and L.S. Chang. 1998.
Effects of outlet obstruction and its reversal on mitochondrial
enzyme activity in rabbit urinary bladders. J Urol 160:2258
Lutz G.J. and L.C. Rome. 1994. Built for jumping: the design
of the frog muscular system. Science 263:370–372.
Marsh R.L. and A.F. Bennett. 1986. Thermal dependence of
contractile properties of skeletal muscle from the lizard Sce-
loporus occidentalis with comments on methods for fitting
and comparing force-velocity curves. J Exp Biol 126:63–77.
Marsh R.L. and H.B. John-Alder. 1994. Jumping performance
of hylid frogs measured with high-speed cine film. J Exp Biol
Pette D. 2001. Plasticity in skeletal, cardiac, and smooth muscle:
historical perspectives: plasticity of mammalian skeletal mus-
cle. J Appl Physiol 90:1119–1124.
Pottle D. and L.E. Gosselin. 2000. Impact of mechanical load
on functional recovery after muscle reloading. Med Sci
Sports Exercise 32:2012–2017.
Reiser P.J., T. Bradford, and J.A. Rall. 1982. Isometric contractile
properties and velocity of shortening during avian myoge-
nesis. Am J Physiol 243:C177–C183.
Renaud J.M. and E.D. Stevens. 1984. The extent of short-term
Muscle Development in Anuran Larvae 543
and long-term compensation to temperature shown by frog
and toad sartorius muscle. J Exp Biol 108:57–75.
Ricklefs R.E. 1983. Avian postnatal development. Pp. 1–83 in
D.S. Farner, J.R. King, and K.C. Parkers, eds. Avian Biology.
Vol. 7. Academic Press, New York.
Roy R.R., I.D. Meadows, K.M. Baldwin, and V.R. Edgerton.
1982. Functional significance of compensatory overloaded
rat fast muscle. J Appl Physiol 52:473–478.
Segel I.H. 1975. Biochemical Calculations: How to Solve Math-
ematical Problems in General Biochemistry. Wiley, New York.
Shin J.S., J.C. Park, M. Yamashita, and I. Choi. 2000. Anuran
metamorphosis: a model for gravitational study on motor
development. Korean J Biol Sci 4:223–229.
Spriet L.L. 1989. ATP utilization and provision in fast-twitch
skeletal muscle during tetanic contractions. Am J Physiol 257:
Stehouwer D.J. 1992. Development of anuran locomotion: eth-
ological and neurophysiological considerations. J Neurobiol
Swynghedauw B. 1986. Developmental and functional adap-
tation of contractile proteins in cardiac and skeletal muscles.
Physiol Rev 66:710–771.
Tsika R.W., R.E. Herrick, and K.M. Baldwin. 1987. Interaction
of compensatory overload and hindlimb suspension on my-
osin isoform expression. J Appl Physiol 62:2180–2186.
Walker D.W. and A.R. Luff. 1995. Functional development of
fetal limb muscles: a review of the roles of activity, nerves
and hormones. Reprod Fertil Dev 7:391–398.
Wassersug R.J. and D.G. Sperry. 1977. The relationship of lo-
comotion to differential predation on Pseudacris triseriata
(Anura: Hylidae). Ecology 58:830–839.
    • "In support of this, mortality during the metamorphic period was higher at the low temperature (Orizaola and Laurila 2009b), likely reflecting the strong dependence on high temperatures in R. lessonae (Sinsch 1984). Amphibians lose body mass during metamorphosis mainly because of tissue dehydration (Hensley 1993; Park et al. 2003 ). Mass loss during this process could be important because juvenile size is correlated with survival, as well as starvation and desiccation resistances (Tracy et al. 1993; Beck and Congdon 1999; Semlitsch et al. 1999 ). "
    [Show abstract] [Hide abstract] ABSTRACT: Understanding the factors that affect the process of metamorphosis in species with complex life cycles, and in particular their variation within and among populations, has been rarely explored until recently. We examined the effects of temperature environment on several metamorphic characteristics in three populations of the pool frog (Rana lessonae Camerano, 1882) by rearing individuals at two temperature environments (20 and 25 degrees C). Higher temperature shortened the metamorphic period and reduced the absolute mass loss, although there was no difference between the temperatures in the percentage of mass lost. No differences among the populations were detected, but there was significant intrapopulation variation both in the mean and in the plasticity for the duration of metamorphosis. These results indicate that several aspects of metamorphosis are plastic in amphibians, these traits may have considerable intrapopulation variation, and that temperature is a strong factor affecting the process of metamorphosis.
    Full-text · Article · Jul 2009
    • "There could be several explanations for this: (1) this species is fully aquatic, and so the transition costs between terrestrial and aquatic locomotion are not present. Hind limb development, which has been shown during the larval stage to aid swimming (Park et al., 2003), may confer a greater advantage to metamorphosing Xenopus because of the large, well-developed webbed feet found in this species. It has been shown that drag forces are considerably increased during metamorphic climax due to forelimb emergence (Dudley, King & Wassersug, 1991), but the advantage of the thrust generated by the developing hind limbs could offset this cost. "
    [Show abstract] [Hide abstract] ABSTRACT: In organisms with complex life cycles, such as amphibians, selection is thought to have minimized the duration of metamorphosis, because this is the stage at which predation risk is presumed to be highest. Consequently, metamorphic duration is often assumed to show little if any environmentally induced plasticity, because the elevation in the extrinsic mortality risk associated with prolonging metamorphosis is presumed to have selected for a duration as short as is compatible with normal development. We examined the extent to which metamorphic duration in the anuran amphibian Xenopus laevis was sensitive to environmental temperature. Metamorphic duration was influenced by body size, but independent of this effect, it was strongly influenced by environmental temperature: the duration at 18 °C was more than double that at 24 and 30 °C. We also compared the vulnerability of larval, metamorphosing and post metamorphic Xenopus to predators by measuring their burst swimming speeds. Burst swim speed increased through development and while we found no evidence that it was reduced during metamorphosis, it did increase sharply on completion of metamorphosis. We therefore found no evidence of a substantial increase in vulnerability to predators during metamorphosis compared with larval stages, and hence the slowing of metamorphosis in response to temperature may not be as costly as has been assumed.
    Full-text · Article · Sep 2007