Patterns of blood flow during the postprandial response in ball pythons, Python regius
Department of Biology II, University of Munich (LMU), Großhaderner Strasse 2, D-82152 Planegg-Martinsried, Germany
* Author for correspondence (e-mail: starck{at}uni-muenchen.de)
Accepted 31 December 2004
![]() |
Summary |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: postprandial response, ball python, Python regius, Doppler-ultrasonography, organ size change
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The hydraulic hypothesis predicts that the size increase of the small
intestine and the liver is correlated with increasing blood flow volume into
these organs within 24 h after feeding. Testing for such a relationship has
only recently become possible by using noninvasive ultrasonographic imaging,
which allows repeated and synchronized measurements of organ size and blood
flow volume in unrestrained snakes (Starck
and Burann, 1998; Starck et
al., 2001
). In the present work, we measured changes of blood flow
volume in the major vessels supplying the small intestine and the liver in
response to feeding and fasting during several postprandial periods.
Synchronized with those measurements of blood flow volume we measured organ
size changes and oxygen consumption to characterize SDA. To study changes of
tissue configuration and cell morphometry, tissue samples were taken from a
second group of snakes exposed to the same conditions.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Doppler-Ultrasonography
A Technos MP (Esaote Biomedica, Genua, Italy) was used for ultrasound
imaging and Doppler-ultrasonography. The Technos MP was equipped with a
broad-band linear array scanner head (LA424, 814 MHz), which was used
for all ultrasonography functions in this study, i.e. imaging, continuous wave
Doppler (CW, for detection of blood flow in vessels) and pulsed wave Doppler
(PW, for measurement of blood flow velocity). The LA424 scanner head is
designed for high-resolution ultrasonography at close range (skin diagnosis
and surface vessels) and has a spatial resolution of 0.4 mm laterally and 0.15
mm axially. The scanner head was operated with a penetration depth of 21 mm
and the focus level adjusted flexibly. When using the PW-Doppler, a pulse
repetition frequency of 4.0 kHz was used, allowing for the detection of low
flow velocities. The PW-Doppler function provides a spectrum of velocities,
which can be integrated over time to render time-averaged velocity
(TAV, in m s1). If the cross-sectional area of the
vessel is known, blood flow volume can be calculated as
TAVxcross section of vesselx60 (ml
min1). PW-Doppler measurements were taken from: (1) the A.
mesenterica, (2) the liver portal vein (Vena portae
hepaticae) and (3) the liver vein (Vena hepatica). See below
for anatomical and physiological implications of the measurement
positions.
Dissections
Two preserved Python regius specimens (snoutvent length,
SVL, 105 cm and 90 cm) were obtained from the Zoologische
Staatssammlung, Munich, for macroscopic dissections of the vascular system.
The two specimens were of unknown origin and had no inventory number; however,
they were in perfect anatomical condition and had neither external nor
internal injuries, nor pathologies.
Respirometry
Oxygen consumption was measured using an open flow system (FOX Field Oxygen
Analyzer, Sable Systems, Las Vegas, NV, USA). Oxygen consumption of six snakes
(mean body mass=147±34 g) was measured once per day starting 3 days
before feeding and continued until 11 days after feeding. Thereafter,
measurements were made every third day until 2 days before the next feeding,
when daily measurements resumed. Measurements were taken at 30°C in the
dark for 90120 min; to minimize circadian effects each individual snake
was measured at the same time of day, i.e. between 10:00 and 13:00 h. The
volume of the metabolic chamber was 1200 ml. The air stream (35 ml
min1) was dried (using silica gel blue, Roth GmbH, Germany)
before entering the metabolic chamber. The air stream vented from the
metabolic chamber to the O2-analyzer was redried before entering
the oxygen analyzer. We calculated mass-specific rate of oxygen consumption
o2 (ml
g1 h1), corrected for standard temperature
and pressure, by taking the lowest 10 min interval that did not change for
more than 0.01% O2-concentration. Metabolic data were analyzed
using Data Can software (Sable Systems Inc.). Each day, rate of oxygen
consumption was measured at the same time to avoid circadian pattern in
metabolic rate affecting the results. A respiratory quotient of 0.8 was
assumed for the strictly carnivorous animals.
Histology
The eight snakes from the Aarhus laboratory were killed by an overdose of
pentabarbiturate (Nembutal). Immediately thereafter, they were dissected
macroscopically and the small intestine and the liver preserved in 5%
paraformaldehyde in 0.1 mol l1 phosphate buffer at pH 7.4
and 4°C for at least 48 h. For histology, tissue samples of the small
intestine and liver were washed in buffer, dehydrated through a graded series
of ethanol to 96% ethanol and embedded in hydroxyethyl methacrylate
(Historesin Leica Microsystems, Nußloch, Germany). Embedded material was
sectioned into short series of 50 sections per sample (section thickness was 2
µm), mounted on slides, and stained with Methylene-Blue Thionine.
Histological sections were studied using an Axioplan research microscope
(Zeiss) equipped with a digital camera (Nikon Coolpix 990) and connected to
the image-analysis and morphometry system. SigmaScanPro (v. 4.0, Jandel
Scientific, SPSS Inc., Chicago, IL, USA) was used for imaging and
morphometry.
Ultrasound morphometry and statistics
Ultrasound morphology and landmarks for morphometry of thickness of mucosa
and cross-section of the liver have been described in previous studies (Starck
and Beese, 2001,
2002
; Starck et al.,
2001
,
2004
;
Starck, 2005
). Briefly, a
minimum of five ultrasound images per session were taken from small intestine
and liver, respectively. All ultrasound images were saved in tif-format during
ultrasound sessions. Ultrasound images were analyzed using the morphometry
program SigmaScan v. 5.0 (Jandel Scientific). The cross section of the liver
and the thickness of the mucosa of the small intestine were measured from
ultrasonographs. One measurement was taken from the liver cross sections;
multiple measurements were taken from the small intestinal mucosa.
Doppler-ultrasound images and measured PW-spectra were saved in the Technos MP
internal format and later analyzed using the morphometry options of the Esaote
Technos MP. Backup files in the original format and tif-format-converted files
of all images were saved on DVD for documentation. For all repeated
measurements, daily means from multiple measurements were entered into
statistics. None of the variables differed from normal distribution. Values
given are means ± standard deviation (S.D.) of N
individuals. Data were analyzed using repeated-measures analysis of covariance
(RM-ANCOVA). Feeding was the inter-subject factor, day after feeding was the
within-subject factor, and body mass was the covariate.
Measurements from histological slides
From histological sections, the surface magnification of the mucosal
epithelium can be measured as a functional parameter of absorptive capacity of
the tissue. Epithelial surface magnification was measured as the epithelial
surface over a baseline defined by the inner circular muscle layer.
Measurements were taken by tracing the epithelial surface with a cursor on a
digitizing tablet and calculation of its total length divided by the length of
the baseline, expressed as a dimensionless ratio. The amount of lipid droplets
incorporated into enterocytes or hepatocytes was measured as the area occupied
by lipid droplets per area mucosal epithelium. This measure does not render
absolute data of the amount of lipid in enterocytes, but allowed us to
calculate the relative amount of lipid in enterocytes which, for the purpose
of this study, was sufficient. Similarly, the lipid loading of hepatocytes was
measured as the area occupied by lipid droplets per number of hepatocytes.
Five measurements were taken per section and five sections studied per
individual snake. Tissue measurements were averaged by section and within each
animal to avoid pseudoreplication of data. SPSS v. 11.0 was used for all
statistical analyses.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Vascular ultrasonography must accommodate the variable vascular anatomy and the scanner head position needs to be adjusted to the changing positions of the vessels. Reproducible images can be obtained by placing the scanner head on the left side of the large ventral scales in the position of the gall bladder (i.e. about 1/3 SVL). The ultrasonographic image captures the posterior part of the gall bladder and the aorta in the dorsal region of the image, just below the vertebral column, as morphological landmarks. The aorta can readily be recognized by its anatomical position in the CW/PW-Doppler image (Fig. 1C) and by its pulsed pattern of arterial blood flow in caudal direction. The mesenteric arteries emerge as serial arteries from the aorta and lead to the small intestine (Fig. 1AC). For Doppler-measurements, we consistently used the first artery that branches off from the aorta directly caudal to the gall bladder. In this artery, the blood flow has the same direction and arterial pattern as in the aorta (Fig. 1C). The Vena cava posterior is anatomically next to the aorta, but direction of flow and pattern of pulse are clearly different.
The Vena subintestinalis collects blood from the small intestine and drains it to the liver portal vein (Vena portae hepaticae). The portal vein enters the liver at its caudal tip. At this position, it carries all blood that drains from the small intestine to the liver. The liver of a snake is cigar-shaped, with an oval cross section. The portal vein extends along the entire length of the liver, where it runs along the medial (inner) side of the liver (Fig. 1D), opposite to the liver vein. Along its course, smaller vessels branch off from the portal vein and drain into the sinusoids of the liver. The liver vein (Fig. 1E) runs along the lateral (outer) side of the liver opposite to the portal vein. It collects blood from the liver and drains it to the Vena postcava before it enters the much reduced Sinus venosus of the heart. Vascular ultrasonography of the liver portal vein and the liver vein requires ultrasound cross-sections through the caudal end of the liver. The liver can easily be found in right ventro-lateral position with its caudal end just anterior to the gall bladder.
From the vascular anatomy, i.e. the serial organization of vessels and the shunts between arteries, it is evident that measurements of blood flow volume in just one mesenteric artery cannot provide quantitative data of the complete blood supply to the small intestine. Such quantitative measurements of the entire blood flow volume are, however, possible in the liver portal vein, where all blood from the small intestine enters the liver (see below).
Respirometry
The mass-specific rate of oxygen consumption of fasting snakes was
0.02±0.01 ml O2 g1 h1
(average of N=6 animals, each animal was measured during four
successive feeding cycles). Within 24 h after feeding the oxygen consumption
increased to 0.074±0.006 ml O2 g1
h1 and reached peak oxygen consumption 48 h after feeding at
an average of 0.08±0.021 ml O2 g1
h1 (Fig. 2).
Thereafter, oxygen consumption declined rapidly, returning to fasting values
around 10 days after feeding (0.021±0.004 ml O2
g1 h1). Feeding had a highly significant
effect on mass-specific metabolic rate (RM-ANOVA; repeat was the inner-subject
factor and feeding was between-subject factor:
F1,12=26.28, P<0.001).
|
Organ size changes
Within 24 h after feeding, the thickness of the intestinal mucosa
increased, as observed in ultrasonographic grey level images. In fasting
snakes, the thickness of the mucosa was on average 3.73±0.27 mm (2 days
before feeding; average of N=6 animals, each animal was measured
during four successive feeding cycles). Within 3 days after feeding, the
thickness of the intestinal mucosa reached peak size of 5.98±4.3
mm(Fig. 3). 34 days
after feeding the thickness of the mucosa began to decline, reaching fasting
values at about 2 weeks after feeding (3.78±0.28 mm). The described
organ size changes were highly significant (RM-ANCOVA with body mass as
covariate; repeat was the inner-subject factor and feeding was between-subject
factor: F1,12=24.56, P<0.001; body mass was
not significant as a covariate).
|
The cross sectional diameter of the liver showed a similar pattern of increase and decrease. During the fasting periods the cross sectional diameter of the liver was 10.3±1.1 mm (2 days before feeding; average of N=6 animals, each animal was measured during four successive feeding cycles). 2 days after feeding the cross sectional diameter of the liver reached a peak value (12.9±1.6 mm). 2 weeks after feeding liver size had returned to prefeeding values (10.5±1.2 mm). Feeding was a highly significant effect (RM-ANCOVA with body mass as covariate, repeat was the inner-subject factor and feeding was between-subject factor; main effect feeding: F1,12=1.91, P=0.05; body mass was a significant covariate: F1,1=4.63, P=0.035). The factorial size increase of the liver was much smaller (on average 25%) than in the small intestine (on average 300%) and the variances were much higher. Also, the response curve was less steep and the liver took longer returning to fasting values (Fig. 4).
|
Patterns of blood flow
In all vessels studied, the patterns of blood flow changed in response to
feeding. In the mesenteric artery, blood flow velocity and consequently blood
flow volume increased after feeding, remaining elevated for 23 days and
then slowly returning to fasting values. The measurements of blood flow volume
in the first mesenteric artery had relatively large variances. The absolute
values ranged between 0.4 ml min1 and 1.2 ml
min1 (Fig.
5A), thus values were relatively small. However, the serial
arrangements of mesenteric arteries and the shunts between them indicate that
one does not capture the entire blood flow volume if measurements are only
taken in one artery. During fasting, blood flow volume in the first mesenteric
artery was 0.69±0.11 ml min1. 1 day after feeding the
blood flow volume had increased to an average of 1.0±0.12 ml
min1 and returned to pre-feeding values within 6 days after
feeding (N=6 snakes, each animal being measured during four feeding
cycles). Feeding as main effect and body mass as a covariate were both
significant in a repeated-measures analysis (RM-ANCOVA, between-subject
factor: F1,12=4.772, P0.001; covariate body
mass: F1,1=13.7, P
0.001).
|
When measured at the caudal end of the liver, the liver portal vein
transports all blood that drains from the intestines to the liver. The vessel
is easy to find by Doppler-ultrasonography and the signal to noise ratio is
very good (Fig. 5B). During
fasting, blood flow volume was on average 5.28±1.06 ml
min1. Within 24 h after feeding, blood flow volume reached a
peak at an average of 14.01±4.56 ml min1. Within 1
week after feeding, the value declined again to fasting values
(5.87±1.36 ml min1). Feeding as main effect and body
mass as the covariate were both highly significant on blood flow volume in the
liver portal vein (RM-ANCOVA, repeat was the inner-subject factor and feeding
was between-subject factor; between-subject factor:
F1,12=12.56, P0.001; covariate body mass:
F1,1=36.2, P
0.001). Blood flow in the liver
portal vein does not show a pulse; flow is continuous.
Not unexpectedly, the liver vein showed a very similar pattern to that
described for the liver portal vein (Fig.
5C). However, the absolute values of measurement were smaller than
in the portal vein because at the measurement position only about 50% of the
blood flowing from the liver to the heart was captured. The blood flow volume
in fasting snakes was on average 5.5±2.19 ml min1.
Within 24 h after feeding, it increased to 9.29±1.81 ml
min1 (N=6 snakes, each animal being measured four
times). Within 2 weeks after feeding, it returned to fasting values; however,
our measurements showed relatively high variances that made it difficult to
determine precisely when values had returned to fasting values. Effects of the
between-subject factor `feeding' and the covariate `body mass' on blood flow
volume in the liver vein were both highly significant (RM-ANCOVA repeat was
the inner-subject factor and feeding was between-subject factor;
between-subject factor: F1,12=5.72, P0.001;
covariate body mass: F1,1=107.7; P
0.001). A
continuous blood flow as typical for veins was observed.
The liver portal vein showed the best Doppler-signal of blood flow volume.
This is based on its easy accessibility to Doppler-ultrasound imaging and to
the fact that it carries all blood passing from the small intestine to the
liver. Therefore in the following analyses, blood flow volume in the liver
portal vein was used for correlations to the observed changes of organ size of
small intestine and liver. To account for possible body size effects, all
values were transformed into relative values with the fasting value (i.e. 2
days before feeding) being 100%. A nonlinear regression of mucosal thickness
as dependent variable of blood flow volume resulted in a highly significant
correlation (Fig. 6; ANOVA,
F=68.29, P0.001, r2=0.54). The
correlation explained over 50% of overall variation in mucosal thickness. A
nonlinear regression of liver size on blood flow volume also showed a tight
relationship between both variables and explained more than 50% of the overall
variation in liver size (Fig.
7; ANOVA, F=61.24, P
0.001,
r2=0.51). Interestingly, the relationship between blood
flow volume and organ size followed an exponential function for small
intestine and liver indicating limitations, i.e. increase in blood flow volume
resulted in only a limited range of values in an increase of mucosal thickness
or liver size.
|
|
Histological changes and morphometry
When fasting, the mucosal epithelium was pseudostratified and the
enterocytes contained no lipid droplets, as described previously for other
snake species (Starck and Beese,
2001,
2002
). Within 24 h after
feeding the enterocytes were loaded with lipid droplets and the epithelium
changed into a single layered configuration
(Fig. 8A,B). Also, the brush
border of enterocytes was much more prominent after feeding and the connective
tissue core of the villi showed enlarged capillaries and lymphatic vessels. No
differences were found in the tissues of the snakes that had been digesting
their prey for either 24 or 48 h, respectively.
|
Associated with lipid loading of enterocytes the surface magnification increased significantly from 13.82±3.0 to 62.9±2.6, i.e. by a factor 4.5 (t-test, T=19.62, P<0.001). In digesting snakes, lipid droplets occupied on average 35% of the epithelial area (Fig. 8B). When surface magnification was regressed as on area of lipid droplets, a highly significant linear correlation was detected, i.e. surface magnification increased significantly with increasing lipid loading of enterocytes. The relationship explained approximately 60% (regression: slope=0.09, intercept=8.45, r2=0.62; ANOVA, F=33.98, P<0.001) of the overall variation in the small intestine's surface magnification.
Similar changes were observed in the liver. The hepatocytes of fasting snakes contained no lipid droplets (Fig. 8C). Within 24 (48) h after feeding the hepatocytes were filled with lipid droplets. The area per section occupied by lipid droplets increased from a fasting value of 2.8±1% to 17±1.3% after feeding (Fig. 8D). It was not possible to calculate a correlation of liver cross-sectional diameter on area of lipid droplets because these measurements were not taken from the same individuals. However, the results are so clear that a relationship between loading of hepatocytes with lipid droplets and liver size can safely be postulated; i.e. liver size increases as hepatocytes become loaded with lipid droplets after feeding.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In ectotherm sauropsids, postprandial upregulation of metabolic rate is
affected by a variety of factors, e.g. temperature, meal size, composition of
diet and feeding frequency (Secor and
Diamond, 1997; Wang et al.,
2003
; Andrade et al.,
2005
; McCue et al.,
2005
). A 4-factorial increase of rate of oxygen consumption as
reported here for snakes feeding a prey of 25% of their body mass, and kept at
temperatures of 2530°C, is well within the range of previously
reported values (Overgaard et al.,
1999
,
2002
;
Starck et al., 2004
;
Andrade et al., 2005
). Thus,
the ball pythons in this experiment performed a regular and undisturbed
postprandial upregulation of metabolic rate. Also, the organ size changes
observed here showed the same pattern and same values of up- and
downregulation as reported earlier for another python species (Starck and
Beese, 2001
,
2002
;
Starck et al., 2004
). The
histological samples necessarily had to be taken from a different group of
individual snakes. The configuration changes observed in the tissue and the
loading of cells with lipid droplets followed the pattern described earlier
for Burmese pythons (Starck and Beese,
2001
), garter snakes (Starck
and Beese, 2002
) and other ectotherm sauropsids (Starck,
2003
,
2005
). Thus, it is safe to say
that the ball pythons in the experiment performed a postprandial response that
did not differ from that seen in other snake species.
Changes in blood flow volume were clearly correlated with the postprandial upregulation of organ size in ball pythons. The correlations explain about 50% of the variation in organ size changes as a correlated response to increased blood flow volume. Thus, the nonlinear regressions in Figs 6 and 7 support the hypothesis that the postprandial increase of mucosal thickness and liver size is partially related to increasing blood flow volume to these organs. Although the correlations support the hydraulic pump hypothesis, the relationship is not linear and not straight. Obviously, a change in blood flow volume results in increasing mucosal surface only in a restricted range, reaching an upper limit at about 150%200% blood flow volume of fasting value. Above this value, a further increase in blood flow volume does not result in continued increase of mucosal thickness. Identical observations have been made for the liver. A second factor contributing to organ size increase is loading of cells with lipid droplets. Enterocytes and hepatocytes are loaded with lipid droplets. Organ size/cell size is clearly correlated with the amount of lipid droplets in the cells, explaining approximately another 50% of overall variation in organ size.
In summary, the data presented here support the hydraulic pump hypothesis,
i.e. by pumping more blood into the vessels of the small intestine the villi
elongate and provide a larger absorptive surface to the digesta. In addition,
loading of the enterocytes with lipid droplets also contributes considerably
to an increase of the mucosal surface magnification. Of course, data on blood
flow volume and increasing enterocytes size were necessarily obtained from
different animals and therefore cannot be combined in one analysis. However,
both data sets are clear enough to safely conclude that increasing blood flow
the organ and loading of cells with lipid droplets both contribute to the
organ size increase of small intestine and liver after feeding. The exact
partitioning of the contribution of both processes ought to be determined.
Also, increased flow of lymphatic fluid into the villi might contribute to the
overall enlargement of the small intestine. The results of this study match
the previously published observation that cell proliferation is not involved
in the upregulation of small intestine size or liver (Starck and Beese,
2001,
2002
). Recently published
analyses showed that the increase of small intestine size is an energetically
cheap process (Overgaard et al.,
2002
; Secor, 2003
;
Starck et al., 2004
).
Downregulation of organ size is obviously associated with declining blood flow
volume and transport of lipid droplets to the adipose tissue depots. On that
basis, one might even speculate that the up- and downregulation of organ size
are just byproducts of digestion, with blood being pumped to the gut to
enhance transport of absorbed nutrients from the gut to the liver, while
loading of enterocytes with lipid droplets occurs during lipid absorption. For
the liver, the processes are supposedly very similar and may also be
understood as byproducts of general activation of the digestive system. The
delayed response in size change of the liver as compared to the small
intestine supports this interpretation.
![]() |
Acknowledgments |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Andrade, D. V., Cruz-Neto, A. P., Abe, A. S. and Wang, T. (2005). Specific dynamic action in ectothermic vertebrates: a general review on the determinants of post-prandial metabolic response in fishes, amphibians, and reptiles. In Physiological and Ecological Adaptations to Feeding in Vertebrates (ed. J. M. Starck and T. Wang), pp. 305-324. Enfield, NH, USA: Science Publishers.
Bedford, G. S. and Christian, K. A. (2001). Metabolic response to feeding and fasting in the water python (Liasis fiscus). Austral. J. Zool. 49,379 -387.[CrossRef]
Hafferl, A. (1933). Das Arteriensystem. In Handbuch der vergleichenden Anatomie der Wirbeltiere, Vol. 6 (ed. L. Bolk, E. Göppert, E. Kallius and W. Lubosch), pp. 563-684. Berlin: Urban and Schwarzenberg.
McCue, M. D., Bennett, A. F. and Hicks, J. W. (2005). The effect of meal composition on specific dynamic action in Burmese pythons (Python molurus). Physiol. Biochem. Zool. In press.
Overgaard, J., Andersen, J. B. and Wang, T. (2002). The effects of fasting duration on the metabolic response to feeding in Python molurus: an evaluation of the energetic costs associated with gastrointestinal growth and upregulation. Physiol. Biochem. Zool. 75,360 -368.[CrossRef][Medline]
Overgaard, J., Busk, M., Hicks, J. W., Jensen, F. B. and Wang, T. (1999). Respiratory consequences of feeding in the snake Python molurus. Comp. Biochem. Physiol. 124A,359 -365.
Secor, S. M. (2003). Gastric function and its
contribution to the postprandial metabolic response of the Burmese pythons
Python molurus. J. Exp. Biol.
206,1621
-1630.
Secor, S. M. and Diamond, J. (1995). Adaptive responses to feeding in Burmese pythons: pay before pumping. J. Exp. Biol. 198,1313 -1325.[Medline]
Secor, S. M. and Diamond, J. (1997). Determinants of postfeeding metabolic response in Burmese python, Python molurus. Physiol. Zool. 70,202 -212.[Medline]
Secor, S. M. and Diamond, J. (1998). A vertebrate model of extreme physiological regulation. Nature 395,659 -662.[CrossRef][Medline]
Secor, S. M. and Diamond, J. (2000). Evolution of regulatory response to feeding in snakes. Physiol. Biochem. Zool. 73,123 -141.[CrossRef][Medline]
Secor, S. M., Stein, E. D. and Diamond, J. (1994). Rapid upregulation of snake intestine in response to feeding: a new model of intestinal adaptation. Am. J. Physiol. 266,G695 -G705.[Medline]
Starck, J. M. (1999). Structural flexibility of the gastro-intestinal tract of vertebrates. Implications for evolutionary morphology. Zool. Anz. 238,87 -101.
Starck, J. M. (2003). Shaping up: how vertebrates adjusts their morphology to changing environmental conditions. Animal Biol. 53,245 -257.[CrossRef]
Starck, J. M. (2005). Structural flexibility of the digestive system of tetrapods Patterns and processes at the cellular and tissue level. In Physiological and Ecological Adaptations to Feeding in Vertebrates (ed. J. M. Starck and T. Wang), pp. 175-200. Enfield, NH, USA: Science Publishers.
Starck, J. M. and Beese, K. (2001). Structural
flexibility of the intestine of Burmese python in response to feeding.
J. Exp. Biol. 204,325
-335.
Starck, J. M. and Beese, K. (2002). Structural
flexibility of the small intestine and liver of garter snakes in response to
feeding and fasting. J. Exp. Biol.
205,1377
-1388.
Starck, J. M. and Burann, A. K. (1998). Noninvasive imaging of the intestinal tract of snakes: A comparison of normal anatomy, radiography, magnetic resonance imaging, and ultrasonography. Zoology 101,210 -223.
Starck, J. M., Dietz, M. and Piersma, T. (2001). Ultrasound scanning. In Body Composition Analysis: A Handbook of Non-Destructive Methods, Chapter 7 (ed. J. R. Speakman), pp. 188-210. Cambridge: Cambridge University Press.
Starck, J. M., Moser, P., Werner, R. and Linke, P. (2004). Pythons metabolize prey to fuel the response to feeding. Proc. R. Soc. Lond. B 271,903 -908.[CrossRef][Medline]
Thompson, G. G. and Withers, P. C. (1999). Effect of sloughing and digestion on metabolic rate in the Australian carpet python, Morelia spilota impricata. Austral. J. Zool. 47,605 -610.
Wang, T., Andersen, J. B. and Hicks, J. W. (2005). Effects of digestion on the respiratory and cardiovascular physiology of amphibians and reptiles. In Physiological and Ecological Adaptations to Feeding in Vertebrates (ed. J. M. Starck and T. Wang), pp.279 -303. Enfield, NH, USA: Science Publishers.
Wang, T., Zaar, M., Arvedsen, S., Vedel-Smith, C. and Overgaard, J. (2003). Effects of temperature on the metabolic response to feeding in Python molurus. Comp. Biochem. Physiol. 133A,519 -527.
Related articles in JEB: