|
| |
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
First published online June 13, 2008
Journal of Experimental Biology 211, 2058-2065 (2008)
Published by The Company of Biologists 2008
doi: 10.1242/jeb.018044
Why go bipedal? Locomotion and morphology in Australian agamid lizards
1 Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2
3EJ, UK
2 Zoology, School of Animal Biology, University of Western Australia, Western
Australia, Australia
3 Centre for Ecosystem Management, Edith Cowan University, Western Australia,
Australia
4 School of Sport science, Exercise and Health, University of Western Australia,
Western Australia, Australia
* Author for correspondence (e-mail: cc498{at}cam.ac.uk)
Accepted 17 April 2008
 |
Summary |
|---|
 TOP Summary INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Key words: bipedal, sprint speed, acceleration, endurance, locomotion, agamids, dragon lizards
|
INTRODUCTION |
|---|
|
TOPSummaryINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
). Both the
thecodont–dinosaur group and modern-day lizards are capable of a bipedal
walking gait, as are birds and some primates (including humans). Hopping on
two legs has risen independently in placental mammals, marsupials and birds
(Snyder, 1962). Bipedalism in
these groups may therefore convey some advantage over quadrupedal locomotion,
or may be a consequence of the front limbs being used for a different purpose.
In birds, the forelimbs have been modified for flight, requiring bipedal
terrestrial movement, whereas hopping in mammals may provide an energetic
advantage (Bennett, 1985). In
some primates, the forelimb may be used for tool making, food handling and
carrying articles when standing or walking
(Howell, 1944). For lizards,
the front limbs are not adapted for a function other than quadrupedal
locomotion, thus the advantages of bipedalism for these taxa are not
clear.
In extant lizards the bauplan with potential for bipedalism seems to have
arisen only once (Fig. 1).
Based on the literature (e.g. Irschick and
Jayne, 1999; Aerts et al.,
2003), personal communications (J. B. Losos, Harvard University)
and personal observations (C.J.C.), bipedal locomotion has not been observed
in Sphenodontida, Gekkonidae, Scincidae, Cordylidae or Xantusiidae, with
quadrupedal locomotion almost certainly being ancestral for these groups.
Bipedal locomotion first appears in some Teiioidea and in some Lacertidae
(Irschick and Jayne, 1999;
Aerts et al., 2003). It is
retained in some members of the linage leading to Iguanidae and Acrodonta,
although it is reduced or lost in some groups within these families (e.g.
Chamaeleonidae, Phrynosomatidae). The lineage leading to Anguimorpha, however,
seems to have lost the ability to run bipedally (e.g. Shinisaurus,
Lanthanotus) only to reappear within the Varanidae. To our knowledge, the
re-emergence of facultative bipedalism within varanids is limited to three
species within the gouldii group of the Indo-Australian clade of
Varanus (V. gouldii, V. panoptes and V. giganteus;
C.J.C., personal observation). Therefore, bipedalism seems to have emerged to
be lost and regained in subsequent groups, which suggests it is an adaptive
trait. So what are the advantages (or disadvantages) of bipedal locomotion. In
other words, why do lizards go bipedal?
|
;
Snyder, 1952;
Snyder, 1954;
Snyder, 1962). He argued that
bipedal lizards differed from their quadrupedal counterparts in morphology,
speed and economy of movement. Of the morphological features cited, long
hindlimbs, short forelimbs, a narrow pelvis and a long tail were thought to
aid bipedalism mostly through increased stride length. If stride rate remains
constant, then increased stride length leads to an increase in a second trait
associated with bipedalism, increased speed. Differences in speed between
bipedal and quadrupedal lizards were noted, but analysis was difficult since
Snyder (Snyder, 1962) was
forced to compare lizards that ran only quadrupedally with those that ran only
(or seemingly so) bipedally. Moreover, it was often difficult to differentiate
between morphological adaptations that were for speed rather than bipedalism
per se.
Snyder also suggested that bipedalism would increase the economy of
movement as no internal work was done to move the front limbs
(Snyder, 1949;
Snyder, 1952;
Snyder, 1954;
Snyder, 1962), although he did
not test this idea. This issue was examined by Fedak and Seeherman
(Fedak and Seeherman, 1979) who
showed similar costs of transport for ponies and ostriches (of similar size),
and further by Roberts et al. (Roberts et
al., 1998) who found similar economy for similar-sized dogs and
turkeys. Thus bipedalism does not seem to convey any economic advantage in
species studied so far. However, these studies were often forced to compare
animals that differed greatly, both anatomically and phylogenetically.
Recently, Aerts et al. proposed a more radical explanation for the cause of
bipedalism (Aerts et al.,
2003). Manoeuvrability, they suggested, benefits from a caudal
shift of the body centre of mass (body-COM). They modelled a small lacertid
lizard (Acanthodactylus erythrurus), to show that acceleration
combined with this caudal shift in the body-COM would cause the front legs of
the lizard to lose traction with the ground, resulting in bipedal locomotion.
Thus, Aerts et al. (Aerts et al.,
2003) proposed that bipedalism may just be a consequence of
acceleration and a caudal shift of the body-COM. For their theory to be
supported, there would be a maximum threshold of acceleration for quadrupedal
movement beyond which the lizard would have no choice but to run bipedally,
with the front limbs losing contact with the ground. Aerts et al. predicted
that this acceleration threshold would be influenced by the relationship
between the horizontal body-COM and the vertical body-COM
(Aerts et al., 2003).
We used high-speed cinematography to analyse quadrupedal and bipedal
locomotion in Australian agamid lizards, in the light of the hypothesis of
Aerts et al. (Aerts et al.,
2003). The comparative method was used to test these ideas and
explore this evolutionary puzzle in a phylogenetic context. The fundamental
question that we addressed was whether bipedalism is dependent on phylogeny,
morphology, energetic cost of locomotion, speed and/or acceleration.
|
MATERIALS AND METHODS |
|---|
|
TOPSummaryINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Various morphological dimensions were measured for each lizard. These were,
snout-to-vent length (SVL), tail length (TAIL), head–neck length (HN),
thorax–abdomen length (TA), upper forelimb length (UFL), lower forelimb
length (LFL), forefoot length (FFOOT), upper hindlimb length (UHL), lower
hindlimb length (LHL), hindfoot (HFOOT) and body mass. These linear
measurements were made to ±0.05 mm using digital callipers; body mass
was measured using electronic scales to ±0.05 g. Species means were
used in all analyses. For morphological dimensions, the effects of size were
removed using a variation of Somers' size free analysis
(Sommers, 1986) developed by
Thompson and Withers (Thompson and
Withers, 2005) for Ctenophorus. %Bipedal data were
corrected for size dependence by computing residuals from logSVL.
The horizontal body-COM was measured for each species on a subset of the lizards (either live or dead). Each lizard was placed in a `natural' position with its hindlimbs under the hips, and the forelimbs under the shoulder. Lizards were placed on a ruler that was suspended by string, in perfect balance. The lizard was moved forward or backwards until the balance of the ruler was restored. The body-COM was recorded as the anterior distance from the body-COM to the vent of each lizard.
We were unable to train lizards to run at various constant speeds on a
treadmill to calculate the net cost of transport and efficiency of locomotion.
Instead we used endurance as a surrogate for locomotor efficiency. Endurance
was determined in a separate trial on the motorised treadmill at 280 mm
s–1 in a controlled-temperature room at 35°C. Time to
exhaustion for each lizard was recorded with a stopwatch. Exhaustion was
indicated by a lack of righting response, once turned on its back. Since mass
is thought to have a significant effect on endurance
(Garland, 1994), residuals from
a log–log plot of mass and endurance were regressed against
%bipedal.
Speed and acceleration of both bipedal and quadrupedal strides was measured
for four species; C. caudicinctus (N=5, 66 strides), C.
femoralis (N=4, 16 strides), C. nuchalis (N=4,
25 strides) and R. adelaidensis (N=4, 12 strides). These
species were chosen for speed and acceleration studies because they were
representative of the overall %bipedal range. Speed and acceleration were
determined by digitizing a mark on the lizard over the pelvis and a mark on
the treadmill, at 200framess–1, using Peak Motus software
(V2000; Peak Performance Technologies, Inc.). The position of the pelvis was
calculated relative to the treadmill to remove the effect of speed of the
treadmill. The resulting three-dimensional coordinates were smoothed using a
mean square error algorithm in Matlab (ver. 7.1, Mathworks, Inc., Natick, MA,
USA) since this approach was least biased and error prone, and outperformed
other available methods (Walker,
1998).
Since bipedalism may be the result of sustained speed or acceleration (as opposed to maximal speed and acceleration that may occur for very short periods of time), both maximal and average values for speed and acceleration were analysed. The highest recorded speeds and accelerations for each stride were used for maximum speeds and accelerations. Average speed was calculated as the mean value for speed scores over the entire stride. The derivative of the regression for speed over time was used to calculate average acceleration.
The log-likelihood statistic was used to calculate the presence and strength of a threshold between bipedal and quadrupedal for speed and acceleration. This statistic describes the probability that the observed values of the dependent variable (speed or acceleration) may be predicted from the observed values of the independent variables (bipedal or quadrupedal). This statistic ranges from 0 to minus infinity, where for this test values closer to zero indicate a tighter threshold. This statistic was calculated using the probit analysis in Minitab (ver. 15, Minitab, Inc., State College, PA, USA).
The acceleration thresholds for bipedalism were calculated for each species from acceleration data. The 50th percentile from the probit analysis was used as an estimate of the acceleration threshold. All strides from each individual in a species were grouped together for analysis.
Lizard acceleration thresholds were compared to those predicted by the
model of Aerts et al. (Aerts et al.,
2003), which relates bipedalism to both the position of the
horizontal body-COM and vertical body-COM by the equation
 |
) were
calculated from the midpoint of strides that showed obvious bipedal
behaviour.
Statistical analysis
All statistical analyses were carried out using StatistiXL (ver. 1.5,
Statistixl.com,
Perth, WA, Australia), unless otherwise stated. All data were tested for the
presence of a normal distribution using Lilliefors test implemented in Matlab.
If this condition was met, then parametric tests were performed; otherwise
their non-parametric equivalent was used. The relationship between
morphological traits and %bipedal was tested using Pearson's correlations. To
test the effect of bipedalism on speed and acceleration, one-tailed
t-tests were used. In each case, quadrupedal strides were assumed to
have a lower speed and acceleration than bipedal strides. Strides from all
individuals from each species were grouped for analysis.
|
|
), and
k* calculation is based on
(Blomberg et al., 2003) for a
Brownian motion model (d=1).
The phylogenetic relationships among Australian agamids has been largely
resolved (Melville et al.,
2001; Hugall et al.,
2008), but no single tree is available with branch lengths for all
species included in this study. A distance matrix was calculated from a
maximum likelihood tree using 1748 bp of mitochondrial gene sequences (ND1,
ND2 and COI) for the species published in GenBank from Melville et al.
(Melville et al., 2001).
Moloch horridus was used as an outgroup based on Hugall et al.
(Hugall et al., 2008).
Sequences were aligned using ClustalX
(Thompson et al., 1997). The
phylogenetic tree was constructed using the maximum likelihood algorithm by
the computer program PAUP* [Sinauer Associates, Inc., Sunderland,
MA, USA, ver. 4.0b2a (Swofford,
2000)]. The appropriate model of molecular evolution for the
maximum likelihood analysis was evaluated by the likelihood ratio test
implemented by the computer program MrModeltest 3.7
(Posada and Crandall,
1998).
|
RESULTS |
|---|
|
TOPSummaryINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
;
Hugall et al., 2008). The
Ctenophorus clade was retained by this tree, as was the nesting of
Rankina within this clade. The close association between
Typanocryptis and Pogona was also apparent. Therefore this
tree is useful in phylogenetic analysis.
%Bipedal
All species measured were capable of bipedal running, but differed in their
proportion of bipedal strides. Variation in bipedalism among species spanned
the entire range in almost a continuum
(Fig. 3). Some lizards would
readily run bipedally, e.g. up to 85% of the strides for L. gilberti,
whereas others rarely ran bipedally, e.g. C. rubens. The
k* value for %bipedal was 0.789 with a P value of
0.532 suggesting that bipedalism is not strongly associated with phylogeny
within this group of agamids.
Morphology
Morphological data for the 16 species of Australian agamid
(Table 1) showed a significant
relationship to %bipedal. %Bipedal was significantly correlated with log
snout–vent length (r2=0.44, P=0.005,
N=16; Fig. 4A), with
the larger lizard species running bipedally more frequently. When
phylogenetically independent data were examined, contrasts for bipedalism were
still significantly correlated with contrasts for logSVL
(r2=0.46, P=0.004, N=15).
|
|
Few of the relative body dimensions (after the effect of size was removed using Somers' size-free analysis) were significantly correlated with %bipedal (Table 2). Size-free features such as a relatively long tail or short forelimbs were not correlated with size-corrected %bipedal in these lizards. Only size-free UHL, size-free HN length and size-free TA had a significant correlation with %bipedal (Table 2). Both HN and TA were negatively related to %bipedal, whereas UHL was positively related to %bipedal (Fig. 4B). However, these were not significant after Sidàk–Bonferonni correction for multiple tests was applied.
|
The horizontal body-COM position was significantly and positively correlated with size (log body-COM vs logSVL, r2=0.47, P=0.004, N=15). After the effect of size was removed by computing residuals, differences in body-COM were significantly related to %bipedal (r2=0.38, P=0.015, N=15; Fig. 4C). Lizards for which body-COM was relatively closer to the hip were bipedal more frequently than lizards with the body-COM located more anteriorly. This result was still significant when analysed in a phylogenetic context, using independent contrasts for size-corrected body-COM and size-corrected %bipedal (r2=0.40, P=0.012, N=14).
Endurance
Size-corrected endurance was negatively correlated with %bipedal
(r2=0.49, P=0.008, N=13;
Fig. 4D). This was also true
for phylogenetically corrected data (r2=0.40,
P=0.012, N=12). Lizards that were more bipedal generally had
a lower endurance, whereas those running predominately quadrupedally could do
so for longer. Endurance was also positively related to body-COM. Lizards with
relatively higher endurance tended to show an anterior shift in the body-COM
(r2=0.53, P=0.007, N=12).
Speed and acceleration
Average speed of bipedal strides was not significantly higher than average
speed for quadrupedal strides for any of the four species tested
(Table 3). However, for average
acceleration there was a significant difference between quadrupedal strides
and bipedal strides for three of the four species tested
(Table 3). Average acceleration
for bipedal strides was significantly higher than for quadrupedal strides,
except for C. femoralis, which had no significant difference between
average acceleration of bipedal and quadrupedal strides. This suggests that
acceleration rather than speed is important for bipedalism.
|
To determine the presence of a threshold between quadrupedal and bipedal strides logistic regression was used. Here results of the log-likelihood statistic close to zero suggest a more distinct threshold. For each of the four species tested, average acceleration had a more distinct threshold between quadrupedal and bipedal strides, than did average speed (Table 3).
Maximal speed of bipedal strides was not significantly different from the maximal speed of quadrupedal strides for three of the four species tested (Table 4). The exception was R. adeladiensis, which had a higher speed for bipedal strides. Similarly, maximal acceleration of bipedal strides was not significantly different from maximal acceleration of quadrupedal strides for three of the four species examined (Table 4). The exception was C. nuchalis, which had significantly higher acceleration for bipedal strides compared with quadrupedal strides.
|
Acceleration thresholds differed for each species (Table 5, Fig. 5). For both maximum and average accelerations, C. femoralis showed the lowest acceleration threshold (i.e. ran bipedally at lower accelerations), whereas R. adelaidensis showed the highest acceleration threshold (i.e. higher accelerations for bipedal strides). Acceleration thresholds were not significantly related to %bipedal, nor any morphological trait.
|
|
The acceleration threshold predicted from the model of Aerts et al.
(Aerts et al., 2003) also
differed for each species (Table
5). The order for threshold values was the same as those
calculated from acceleration scores above, with C. femoralis and
C. caudicinctus showing lower thresholds and C. nuchalis and
R. adelaidensis showing higher thresholds. For R.
adelaidensis the model closely agreed with the maximum acceleration
threshold, however, for the three remaining species, the thresholds predicted
from the model of Aerts et al. (Aerts et
al., 2003) were higher when compared to the maximum acceleration
thresholds calculated from acceleration data
(Table 5).
|
DISCUSSION |
|---|
|
TOPSummaryINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
). The latter were thought to be advantageous and hence
bipedalism was considered an adaptation. We have shown that some of these
morphological features are related to bipedalism, but most are related to a
caudal shift of the body-COM. Snyder suggested that long hindlimbs, short
forelimbs, long tails and a narrow pelvis were adaptations to bipedalism
(Snyder, 1962). Of these, only
upper hind limb length was related to %bipedal for Australian dragons. Body
size was also strongly related to %bipedal, complicating the issue, since both
size and long hindlimbs are also widely associated with sprint speed
(Garland and Losos, 1994), thus
as Snyder (Snyder, 1962) noted
it is often difficult to separate whether these features have developed as an
adaptation to bipedalism or speed.
Snyder also suggested that running bipedally may provide a significant
economic advantage over running quadrupedally since no work needs to be done
to move the front limbs (Snyder,
1962). Endurance data from the current study do not support this
notion. Lizards that ran bipedally did not run for as long, which suggests,
(1) lizards that run bipedally are not doing so to conserve energy, and (2)
lizards that run bipedally may incur an energetic cost. The latter point
suggests that bipedalism may provide some other benefit(s) to balance this
cost. Whether endurance is a useful proxy for estimating the energetic cost of
transport is still unclear. However, previous studies have shown that while
endurance capacity generally increases with mass
(Garland, 1994), the energetic
cost of transport usually decreases with mass
(John-Alder et al., 1986).
Thus, what little evidence exists suggests that low costs of transport are
associated with higher endurance capacities.
Since %bipedal and body-COM are related, it is possible that body-COM may influence endurance. For instance, a quadruped may benefit from an anterior shift of the body-COM since both the fore and hindlimb can contribute to forward propulsion, and may reduce the role of fatigue for the hind limbs. A posterior shift in the body-COM may result in the hindlimbs contributing most of the propulsive force, and thus fatiguing earlier.
Snyder also proposed that bipedalism would allow an increase in speed,
since the hindlimbs were able to move through a larger stride without being
interrupted by the forelimbs (Snyder,
1962), however the current data do not support consistently higher
speeds for bipedal strides.
The conclusion of Aerts et al. (Aerts et
al., 2003), that lizards run bipedal because forward acceleration
creates a lift on the front part of the body, receives stronger support. We
show that acceleration was a better indicator than speed of posture during the
stride, at least for average acceleration. The cause of the disparity between
maximal and average acceleration data is unknown. Perhaps this suggests that
bipedalism is dependent on longer periods of constant acceleration, or it may
reflect the higher error rate inherent in calculating maximum acceleration
(Walker, 1998).
The models developed by Aerts et al.
(Aerts et al., 2003) to predict
acceleration thresholds matched the data order for our Australian agamids,
i.e. it predicted species with low thresholds, from those with high
thresholds. However, the absolute values for the threshold predicted by Aerts
et al. (Aerts et al., 2003)
were considerable higher. This shows that while the point of application of
the hindlimb, and the horizontal and vertical positions of the body-COM may be
important to bipedalism, there may be other factors which allow lizards to run
bipedally at lower accelerations. One possibility is that lizards are able to
modify these factors during a stride. Certainly the point of application of
the hindlimb changes relative to the hip throughout the stride, but
incorporating this into the model will only produce a predicted threshold
higher than the ones presented here.
Another possibility is that lizards actively manipulate the body-COM which
may cause bipedalism at lower than predicted accelerations. Aerts et al.
(Aerts et al., 2003) noted that
active tail lifting during the acceleratory phase will affect trunk rotation
positively, causing the front limbs to lose contact with the ground earlier
than predicted by the model. Furthermore, the forelimb position can change
from one stride to the next. Some species actively tuck their fore limbs along
the side of the body during some bipedal strides (compare C.
caudocinctus Movie 1 in supplementary material, where arms are held away
from body, with L. gilberti Movie 2 in supplementary material, where
arms are tucked into the body). This may have the effect of moving the
body-COM closer to the hip, allowing bipedal locomotion at lower
accelerations, extending a bipedal stretch. These observations are important
as they suggest that some lizards actively attempt to run bipedally, which
implies an advantage; however the advantage does not appear to be speed or
endurance.
In any case, the presence of an acceleration threshold seems likely, and
the conclusions of Aerts et al. (Aerts et
al., 2003), that bipedalism evolved as a consequence of
acceleration, seems probable. However, the exact position of these thresholds,
a model to predict them and any derived benefit of bipedalism requires further
investigation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TOPSummaryINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Aerts, P., Van Damme, R., D'Aout, K. and Van Hooydonck, B. (2003). Bipedalism in lizards: whole-body modelling reveals a possible spandrel. Philos. Trans. R. Soc. Lond. B Biol. Sci. 358,1525 -1533.[CrossRef][Medline]
Bennett, A. F. (1985). Energetics of locomotion. In Functional Vertebrate Morphology (ed. M. Hilderbrand, D. M. Bramble, K. F. Liem and D. B. Wake), pp.173 -184. Cambridge, MA: Harvard University Press.
Blomberg, S. P., Garland, T., Jr and Ives, A. R. (2003). Testing for phylogenetic signal in comparative data: behavioral traits are more labile. Evolution 57,717 -745.[CrossRef][Medline]
Fedak, M. A. and Seeherman, H. J. (1979). Reappraisal of energetics of locomotion shows identical cost in bipeds and quadrupeds including ostrich and horse. Nature 282,713 -716.[CrossRef][Medline]
Fry, B. G., Vidal, N., Norman, J. A., Vonk, F. J., Scheib, H., Ramjan, S. F. R., Kuruppu, S., Fung, K., Hedges, S. B. and Richardson, M. K. (2006). Early evolution of the venom system in lizards and snakes. Nature 439,584 -588.[CrossRef][Medline]
Garland, T. (1994). Phylogenetic analyses of lizard endurance capacity in relation to body size and body temperature. In Lizard Ecology: Historical and Experimental Perspectives (ed. E. R. Vitt and E. R. Pianka), pp.237 -259. Princeton: Princeton University Press.
Garland, T. and Losos, J. B. (1994). Ecological morphology of locomotor performance in squamate reptiles. In Ecological Morphology: Integrative Organismal Biology (ed. P. C. Wainwright and S. M. Reilly), pp. 240-302. Chicago: University of Chicago Press.
Garland, T., Dickerman, A. W., Janis, C. M. and Jones, J. A. (1993). Phylogenetic analysis of covariance by computer simulation. Syst. Biol. 42,265 -292.[CrossRef]
Howell, A. B. (1944). Speed in Animals: Their Specialization for Running and Leaping. Chicago: University of Chicago Press.
Hugall, A. F., Foster, R., Hutchinson, M. and Lee, M. S. Y. (2008). Phylogeny of Australasian agamid lizards based on nuclear and mitochondrial genes: implications for morphological evolution and biogeography. Biol. J. Lin. Soc. Lond. 92,343 -358.
Irschick, D. J. and Jayne, B. C. (1999). A field study of the effects of incline on the escape locomotion of a bipedal lizard, Callisaurus draconoides. Physiol. Biochem. Zool. 72,44 -56.[CrossRef][Medline]
John-Alder, H. B., Garland, T. and Bennett, A. F. (1986). Locomotory capacities, oxygen consumption, and the cost of locomotion of the shingle-back lizard (Trachydosaurus rugosus). Physiol. Zool. 59,523 -531.
Melville, J., Schulte, I. and Larson, A. (2001). A molecular phylogenetic study of ecological diversification in the Australian lizard genus Ctenophorus. J. Exp. Zool. 291,339 -353.[CrossRef][Medline]
Posada, D. and Crandall, K. A. (1998).
MODELTEST: testing the model of DNA substitution.
Bioinformatics 14,817
-818.
Roberts, T. J., Kram, R., Weyand, P. G. and Taylor, C. R. (1998). Energetics of bipedal running. I. Metabolic cost of generating force. J. Exp. Biol. 201,2745 -2751.[Abstract]
Snyder, R. C. (1949). Bipedal locomotion of the lizard Basiliscus basiliscus. Copeia 1949,129 -137.[CrossRef]
Snyder, R. C. (1952). Quadrupedal and bipedal locomotion of lizards. Copeia 1952,64 -70.[CrossRef]
Snyder, R. C. (1954). The anatomy and function of the pelvic girdle and hindlimb in lizard locomotion. Am. J. Anat. 95,1 -45.[Medline]
Snyder, R. C. (1962). Adaptations for bipedal locomotion of lizards. Am. Zool. 2, 191-203.[Medline]
Sommers, K. (1986). Multivariate allometry and removal of size with principal components analysis. Syst. Biol. 35,359 -368.
Swofford, D. (2000). PAUP*: Phylogenetic Analysis using Parsimony (* and other methods). Sunderland: Sinauer.
Thompson, G. and Withers, P. C. (2005). The relationship between size-free body shape and choice of retreat for Western Australian Ctenophorus (Agamidae) dragon lizards. Amphib. Reptil. 26,65 -72.[CrossRef]
Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F. and
Higgins, D. G. (1997). The CLUSTAL_X windows interface:
flexible strategies for multiple sequence alignment aided by quality analysis
tools. Nucleic Acids Res.
25,4876
-4882.
Townsend, T. M., Larson, A., Louis, E. and Macey, J. R. (2004). Molecular phylogenetics of squamata: the position of snakes, amphisbaenians, and dibamids, and the root of the squamate tree. Syst. Biol. 53,735 -757.[CrossRef][Medline]
Walker, J. A. (1998). Estimating velocities and accelerations of animal locomotion: a simulation experiment comparing numerical differentiation algorithms. J. Exp. Biol. 201,981 -995.[Abstract]
Related articles in JEB:
This article has been cited by other articles:
|
|
 |
|
  K. Phillips LIZARDS PULL A WHEELIE J. Exp. Biol., July 1, 2008; 211(13): i - i. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
