!"#$%$#&'()*%")+,-.'%$#&'()/$&0.*1'-%)-2)3%1#*(*#"'*()4,0%1'%&
!01,-#5678)9*:$6);<)='(>6
4-0#($8)?,@6'-.-&'(*.)A--.-&@B)C-.<)DEB)F-<)D)5F-G<)H)I$(<B)JKKL7B)MM<)JNOPHJNLD
?0Q.'6,$")Q@8)R,$)S%'G$#6'1@)-2)+,'(*&-)?#$66
41*Q.$)S/T8)http://www.jstor.org/stable/30163900
!(($66$"8)JDUJJUOVVK)OO8OE
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=ucpress.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to
Physiological Zoology.
http://www.jstor.org
1325
of
andCholinergic
Regulation
Adrenergic
Intracardiac
Shunting
W.Hicks
James
Departmentof Ecology and EvolutionaryBiology, School of Biological Sciences,
Universityof California,Irvine, Irvine, California92717
Accepted 1/25/94
Abstract
Reciprocal variations in the heart rate (HR), pulmonary vascular resistance
(Rpud)and pulmonary blood flow (Q,,,) are associated with intermittent lung
breathing in reptiles. During ventilation Rpu~decreases and the HR and the Qpu1
increase. In contrast, Rputincreases and the HR and Qput decrease during apnea.
Besides these changes, intermittent ventilation is associated with changes in the
distribution of blood flow between the pulmonary and systemic circulations. A
right-to-left (R-L) intracardiac shunt predominates during apnea, while during
ventilation a left-to-right intracardiac (L-R) shunt predominates. These changes
in the HR, Qpu, Rpul,and intracardiac shunting may be under adrenergic and
cholinergic control. Recent experiments in the turtle Pseudemys scripta support
this hypothesis. In this species, cholinergic stimulation resulting from electrical
stimulation of vagal efferent nerves or infusion of acetylcholine resulted in a
bradycardia, an increased Rput,a reduced Qpuj,and the development of a net R-L
intracardiac shunt. The net R-Lshunt flow was 6 mL/min/kg and represented approximately 40% of the systemic blood flow. The changes in HR, Rpul,and Qpul
were eliminated by atropine. The R-Lshunt was also eliminated by atropine. In
contrast, adrenergic stimulation, resulting from electrical stimulation of vagal afferent nerves or infusion of epinephrine resulted in a tachycardia, a decrease in
Rp,,, an increased Q,,, and the development of a net L-Rshunt. The net L-Rshunt
flow was 28 mL/min/kg, representing 58% of the Q,,u. Preliminary evidence suggested that the changes in the HR, Qput,and Rpulduring vagal aferent stimulation were reduced by propranolol. The results of this study support the hypothesis
that the size and direction of the intracardiac shunt is determined by the relative
resistances offered by the pulmonary and systemic circulations.
Introduction
Intracardiacshuntingnormallyoccurs in chelonians (turtles) and squamates
(lizards and snakes). The complex cardiacanatomyof these animalsallows
1994.C 1994byTheUniversity
of Chicago.
Physiological
Zoology
67(6):1325-1346.
Allrightsreserved.
0031-935X/94/6706-9369$02.00
1326 J.W.Hicks
the systemic venous blood to bypassthe lungs and pulmonaryvenous blood
to bypass the systemic circulation.The magnitude of cardiacshunting may
be affected by the physiological state of the animal. In particular,cardiac
shunting may varyas a function of ventilatorystate. Manyreptiles are intermittent lung breathers. Systemic venous blood bypassing the lungs may
predominateduring apnea, while recirculationof pulmonaryvenous blood
throughthe lungs mayoccur duringthe briefventilatoryperiods. The ability
to regulate intracardiacshunting may provide several adaptivefunctions for
intermittentlybreathingreptiles (table 1; Burggren 1987).
The mechanismscontrollingthe directionand size of intracardiacshunting
are not well understood. Factorsthat affectcardiacfunction (heartrate [HR]
and contractility)and the vascularresistances of variousvascularbeds can
potentially regulate intracardiacshunting. These factorsmay include adrenergic and cholinergic mechanisms as well as nonadrenergicnoncholinergic
(NANC)systems.The purpose of this articleis to review the cardiacanatomy
of noncrocodilian reptiles and to describe the possible role of cholinergic
and adrenergicregulationin cardiacshunting. Experimentalevidence supportingcholinergic and adrenergiccontrol of intracardiacshuntingin turtles
will be presented.
Cardiac Anatomy
The hearts of noncrocodilian reptiles are made up of two separate atrial
chambersand a single ventricle. The ventricle is subdivided into three anatomicallyinterconnected chambersor cava (fig. 1). A distinctivefeatureof
the ventricularanatomy is the presence of a septum-like structurecalled
the muscularridge or Muskelleiste.The muscularridge originatesfrom the
ventralventricularwall, runningfromapex to base, and divides the ventricle
into a smallercavumpulmonale (CP) and largercavumdorsale (VanMierop
and Kutsche 1981). The dorsolateralborder of the muscularridge is free,
allowing potential communicationbetween the CPand cavumdorsale (Van
Mierop and Kutsche 1981). A second incomplete vertical septum, which
originates from the dorsal aspects of the muscularridge to the dorsal wall
of the cavum dorsale, furthersubdivides the cavum dorsale into the cavum
arteriosum(CA) and the cavumvenosum (CV).
In all reptiles, three greatvessels arise from the ventricle:the pulmonary
artery,the right aortic arch (RAo), and the left aortic arch (LAo). The pulmonary arteryemerges to the left of the two aortic arches and originates
from the CP. The RAo and LAo arise from the CV. In addition, the RAo
divides into subclavianand carotid arteries, and a third branch of the RAo
Smits,
1985
Burggren,
1979;
White
1979;
White
Shelton
1989
1982 and
1987
and
1984
1985
Evans
reptiles
and
in
Reference Burggren
Burggren
Burggren
White
Wood
Ackerman
ventilation
Direction R-L
R-LR-LR-L
L-R
L-R
Shunt
intermittent
during
due
shunting
intracardiac
of
1
lung
lunga
lungsa the
the in
the
lungs
stores. into
the
into
shunt
(1987).
into
oxygen
filtration
elimination
energya flux
Burggren
lung
physiological
CO2
mismatching
significance
from
CO2
Significance
plasma
cardiac
V/Q
to
Saves "Metering"
Reduces
Reduces
Facilitates
Minimizes
Summarized
Functional
Functional
TABIEI
Ventilation:
a
Apnea:
1328 J.W.Hicks
RAt
RAo
LAt
LAo
PA
CV
CA
MR
AVvalves
CP
Fig. 1. Schematic representation of the heart of noncrocodilian reptiles.
PA, pulmonary artery; RAt, right atrium; LAt, left atrium. Drawing is
modified from original line drawing by F. N. White (unpublished).
unites with the LAomediocaudally to form the dorsal aorta (Van Mierop
and Kutsche 1985).
Intracardiac Shunts
These anatomicalfeatures of the reptilian heart result in the potential for
intracardiacshunting. Cardiacshunts are defined as right-to-left (R-L) or
left-to-right(L-R).An R-Lshunt representssystemicvenous blood bypassing
the pulmonarycirculationand reentering the systemic circulation.In contrast,recirculationof pulmonaryvenous blood into the pulmonaryarterial
circulation is called an L-Rshunt. If the systemic blood flow (Qsys)and
pulmonaryblood flow (Qpui)are measured, then intracardiacshunting can
be represented as the difference between these two flow rates (Qpu - Qsys).
This difference is referredto as the net shunt flow.
In reptiles, the net shunt flow does not provide a complete descriptionof
intracardiacshunting. The anatomicalarrangementof the ventricularcava
(see fig. 1) allows for bidirectionalshunting, in which an L-Rand R-Lshunt
both occurduringthe cardiaccycle (Heislerand Glass 1985;Ishimatsu,Hicks,
and Heisler 1988; Hicks and Comeau 1994). In hearts with bidirectional
shunting, the net shunt flow may underestimatethe actual shunt flows. A
is equal to
simple model illustratesthis point (fig. 2). In this model, the QpuI
inReptiles 1329
Cardiac
ShuntRegulation
Qpul = 100 ml/min
Qsys = 50 mI/min
Systemic
Pulmonary
-80
L-R
ml/min
OR-L
Right Atrium
QRAt= 50 ml/min
-
30
ml/min
LeftAtrium
= 100ml/min
QLAt
Fig. 2. Model of bidirectional intracardiac shunting in reptiles. Open arrows represent oxygenated blood; filled arrows represent deoxygenated
blood. See text for detailed description.
the systemicvenous return(Qt) plus the actualL-Rshuntflow (-R) minus
the actual R-Lshunt flow (Q-L ). Accordingly,the Qsys is the sum of the
pulmonaryvenous return(Q~t) and the QRL minus the Q-R.In this example,
the net L-Rshunt flow is 50 mL/min. Howeverthe L-R is 80 mL/min, which
This contributionof the L-R to the pulmonary
represents 80%of the QpuI.
circulationis called the shunt fraction (QL-R/Qpu). In this model, the Qsys
resultsfroma QR-L
of 30 mL/min and the R-Lshuntfraction(QR-L/ys)
is 60%.
Underthe flow conditionsprovidedin this example,the R-Lshuntflow cannot
be detected from measurementsof net shunt flow. Bidirectionalshunting
can be quantifiedby measurementsof blood oxygen content from arterial
and cardiacsites (Ishimatsuet al. 1988) or by the simultaneousinjection of
radioactivelylabeled microspheres into the left and right atrium (Heisler,
Neumann,and Maloiy1983). The measurementof net shunts, either L-Ror
R-L,requirecarefulinterpretationif bidirectionalshunting is present.
Intracardiac Shunting Patterns and VentilatoryState
Intermittentventilatorypatterns are a characteristicfeature of many chelonians and squamates.Such patternsconsist of brief periods of ventilation
1330 J.W.Hicks
interspersedamong apneasof variableduration(Shelton,Jones, and Milsom
1986). The ventilatorystate (ventilation or apnea) is often associated with
changes in the size of the R-Land L-Rintracardiacshunt. However, the
precise flow ratiosand absolute magnitudesof these intracardiacshunts are
virtuallyunknown. Regardless, for a few species of chelonians and squamates, the cardiovascularevents that occur during intermittentventilation
have been described.
Apnea
During apnea, associated with quiet breathing or diving, there is a bradycardiacoupled with an increase in the pulmonaryvascularresistance (Rpu1),
which leads to a reduction in the Qp~l.For example, in the freely diving
turtle Pseudemys scripta, the HR decreases by 80%and the Rpu1
increases
(Shelton and
by 150%.These changes result in an 80%reduction in the Qpul
Blood
flow
measurements
that
net
R-Lshunt
a
suggest
Burggren 1976).
develops during apnea (Whiteand Ross 1966; Shelton and Burggren1976).
The net R-Lshunt flow is estimated to be 5-10 mL/min/kg. In the turtle,
analysis of measurements of the partialpressure of oxygen (Po2) of blood
from arterial and cardiac sites confirms an R-Lintracardiacshunt during
apnea (Burggrenand Shelton 1979; White, Hicks, and Ishimatsu 1989).
The actual shunt flows during apnea have been estimated in a varanid
lizardand in turtles (Heisler and Glass 1985;White et al. 1989). In the lizard
Varanusexanthematicus, at a body temperatureof 300C,the R-Lshunt flow
is 11 mL/min/kg, or approximately31%of the Qsys(Heisler and Glass 1985).
In the turtle Chrysemyspicta at 300C, the R-Lshunt is 18 mL/min/kg, or
approximately61%of the Qyy,(Heisler and Glass 1985). In this turtle, the
R-Lshunt flow decreases to 8 mL/min/kg at a body temperatureof 150C.
The R-Lshunt flow at this temperatureis 60%of the Qsys. In these reptiles,
the fractionalR-Lshunt did not change duringventilation (Heisler and Glass
1985). In contrast,a microsphere study of the turtle P. scripta shows that
the fractionalR-Lshuntis greaterduringapneathanduringventilation(White
et al. 1989).
Several studies have shown that an L-Rshunt also occurs during apnea.
In the lizard V exanthematicus and in the turtle C picta a fractional L-R
shunt of 11%-20%occurs during apnea (Heisler and Glass 1985). Measurements of blood Po2 from the pulmonaryarteryand the left and right atrium
confirms an L-Rintracardiacshunt during apnea (White et al. 1989). The
mechanism responsible for this L-Rshunt has been discussed in detail
(Heisler and Glass 1985; Hicks and Comeau 1994).
Cardiac
Shunt
inReptiles1331
Regulation
Ventilation
During ventilatoryperiods, the cardiovascularchanges are the reciprocals
of those occurring during apnea. For example, in the turtle P. scripta the
HR doubles and the Rpuldecreases by more than 50%,which leads to a
threefold increase in the QpuI
(Shelton and Burggren 1976). Blood flow
measurementsfrom selected arteriesestimate thata net L-Rshunt develops
duringventilation (White and Ross 1966; Shelton and Burggren1976). The
net L-Rshunt is 30-50 mL/min/kg. Recent studies of the snakeAcrochordus
is 10-fold higher than Qssduringventilation (Lilgranulatus show that Qu0,
lywhite and Donald 1989). This may be the largest net L-Rshunt measured
for an intermittentlybreathingreptile. In the turtle P. scripta analysisof the
blood Po2 from the left atrium,right atrium,and pulmonaryarteryconfirms
an L-Rintracardiacshunt during ventilation (White et al. 1989).
The magnitudeof the L-Rshunt flow duringventilationhas been estimated
in the lizard V exanthematicus and the turtle C.picta at 30C. In the lizard,
the QLR is approximately3 mL/min/kg and is 11%of the Qpu.In the turtle,
the QLRshunt is 5.5 mL/min/kg, which represents 18%of the Qpul(Heisler
and Glass 1985).
An R-Lintracardiacshunt also occurs during ventilation. In the lizard
V exanthematicus and the turtle C. picta, the fractional R-Lshunt is
10%-25%during ventilation (Heisler and Glass 1985). In the turtle P.
scripta, measurements of blood Po2from cardiacand arterialsites indicate
a 20%-30%fractional R-Lshunt during ventilation (White et al. 1989).
The mechanism responsible for this R-Lintracardiacshunt has been discussed in detail (Heisler and Glass 1985; Hicks and Comeau 1994).
Adrenergic and CholinergicRegulation of the CardiovascularSystem
The size and directionof intracardiacshuntingwill be determinedby factors
that control cardiacfunction (HR and myocardialcontractility)and the vascular resistance of the pulmonaryand systemic circulations.These factors
certainly include adrenergic and cholinergic mechanisms (Berger and
Burnstock 1979; Nilsson 1983). In addition, NANCsystems may play an
importantrole in controlling cardiovascularfunction in reptiles (Lillywhite
and Donald 1989; Conlon, Hicks, and Smith 1990).
Cholinergic Regulation
The vagus may be the primary regulator of the HR during ventilation and
apnea (Burggren 1975). A great deal of evidence shows that both the atria
1332 J.W.Hicks
and the ventricle are innervatedby the vagus nerve (see Berger and Burnstock [1979]for review). In the turtle P. scripta the bradycardiathat occurs
duringapnea is eliminatedby an intravenousinfusion of atropine(Burggren
1975). In contrast,intravenousadministrationof propranololhas no effect
on the bradycardiathatoccurs duringapnea (Burggren1975). In P. scripta,
electrical stimulationof the vagus nerve results in a bradycardiathat is abolished by atropine (Comeau and Hicks 1994).
The pulmonaryvasculatureof chelonians and squamatesexhibits a cholinergic vasoconstrictorinnervation.Electricalstimulationof the vagusnerve
results in an increase in the Rpj,in turtles, lizards, and snakes (Luckhardt
and Carlson1921; Berger 1972, 1973;Burggren1977; Milsom, Langille,and
Jones 1977;Smithand MacIntyre1979;Donald, O'Shea,and Lillywhite1990;
Hicks and Comeau 1994; Comeau and Hicks 1994). This increase is abolished by atropine. The development of an R-Lshunt may be under cholinergic control (White 1976). In the turtle P. scripta electrical stimulation of
the vagus nerves results in a reduction in the Po2 of both the LAoand RAo
(Burggren 1978; Hicks and Comeau 1994). This reduction in the arterial
Po2 occurs even though the pulmonaryvenous and systemic venous Po2
remain unchanged. The systemic arterialPo2 is also decreased by the intravenous infusion of acetylcholine (ACh) (Hicks and Malvin1992). The effects
of cholinergic stimulation on the R-L or the net R-Lshunt have not been
determined.
Adrenergic Regulation
In turtles cardiac sympathetic nerves leave the spinal cord at the level of
the tenth spinal nerve, run forwardthroughthree gangliaof the sympathetic
chain, and extend towardthe heart (Gaskell and Gadow 1884). Sympathetic
observationsconfirmthatfibers
fibersinnervatethe atriumand ultrastructural
containing vesicles typical of adrenergic nerves occur in the ventricular
myocardium of turtles (Yamauchiand Chiba 1973). Stimulationof these
fibersresult in an accelerationof the HR.This effect is blocked by bretylium
(an adrenergic neurone-blockingagent) (see Berger and Burnstock[1979]
for review). Increases in the HR occur after intravenousadministrationof
epinephrine and norepinephrine (Bergerand Burnstock1979; Comeauand
Hicks 1994).
Histochemical studies show adrenergic nerves in the pulmonary vasculature of the lacertilian Trachysaurus rugosus (McLeanand Burnstock
1967; Furness and Moore 1970) and the file snake, A. granulatus (Lillywhite and Donald 1989). In the rat snake Elaphe obsoleta, adrenergic
innervation is present on the pulmonary artery, the smaller pulmonary
Cardiac
Shunt
inReptiles1333
Regulation
arteries and veins, and the main pulmonary vein (Donald et al. 1990).
The functional role of adrenergic nerves in the pulmonary vasculature is
contradictory. Luckhardtand Carlson (1921) reported a differential effect
of epinephrine on the pulmonary vasculature. In the turtles Chrysemys
and Malacoclemys small doses of epinephrine produce a vasodilation,
whereas larger doses produce a vasoconstriction. In contrast, the adrenergic agents norepinephrine, isoprenaline, or phenylephrine have no
effect on the pulmonary arterial tone in the tortoise (Berger 1972). In
the turtle C. scripta electrical stimulation of the cervical sympathetic
nerves produces no change in the pulmonaryperfusion pressure (Milsom
et al. 1977). In addition, infusion of epinephrine, in a dose range from
1 to 100 ytg, does not affect the pulmonary arterypressure (Milsom et al.
1977). In contrast, a marked pulmonary vasodilation results from the
injection of epinephrine into the pulmonary arteryof the rat snake (Donald et al. 1990). Administration of propranolol eliminates this response.
In the rat snake, electrical stimulation of the vagus nerve produces a
biphasic response in Rpu. During the stimulation there is a pulmonary
vasoconstriction, followed by a poststimulatory pulmonary vasodilation
(Donald et al. 1990). The administration of bretylium or propranolol
eliminates this poststimulatory vasodilation. This later observation suggests the Qpuis regulated by the reciprocal interaction of adrenergic and
cholinergic nerves (Donald et al. 1990).
Recent evidence from the turtle P. scripta shows that electrical stimulation of vagal afferent fibers results in an increase in HR, a reduction
in Rpul,and an increase in systemic vascular resistance (Rsys).These
changes result in an increase in
Qpul
and a decrease in
Qsys
(Comeau
and Hicks 1994). These cardiovascular changes are reduced following administration of bretylium. Finally, an intravenous infusion of
epinephrine (0.1 gtg/kg) produces cardiovascular changes similar
to those measured during vagal afferent stimulation (Comeau and
Hicks 1994).
Adrenergic control of intracardiacshunting is not well studied. Analysis of blood Po2 from the systemic arteries and from cardiac sites indicates that adrenergic stimulation may abolish the R-Lshunt. An intravenous infusion of epinephrine
eliminates the systemic venous
admixture in the aortic arches of the turtle P. scripta (Hicks and Malvin
1992). A similar result is obtained during electrical stimulation of vagal
afferent nerves in this same species (Hicks and Comeau 1994). Intravenous infusion of epinephrine may produce a large net L-R shunt in
turtles (Comeau and Hicks 1994).
1334 J.W.Hicks
andMethods
Material
In our laboratorywe have focused on the role of the vagus nerve on control
of centralvascularblood flow in the turtle Pseudemysscripta. Recently,we
have addressed the hypothesis that the size and direction of intracardiac
shunting results from the reciprocalinterplayof the adrenergicand cholinergic mechanisms (Comeau 1992; Hicks and Malvin 1992; Comeau and
Hicks 1994; Hicks and Comeau 1994). The purpose of this study was to
quantifythe effects of vagal nerve stimulation on the size of the net R-Lor
net L-Rshunt flow. Studieswere conducted in five animals (X= 1.65 + 0.34
kg, mean + SD), anesthetized (Nembutal,30 mg/kg), in the supine position
and ventilated (tidal volume = 25 mL/kg and breathingfrequency = 7-10
beat/min). The following cardiovascularvariables were measured or calculated:HR, Qpu,Qsys,pulmonaryarterialpressure (Ppa),pulmonaryvenous
pressure (Ppv),centralvenous pressure,arterialpressure (Psys),Rpuly, net
L-Rshunt flow, and net R-Lshunt flow. The Qpuiand Qyys
were determined
with ultrasonictransit-timeflow probes (2R;Transonic).The Qpuiwas measured by placing a 2R flow probe on the left and right pulmonaryarteries.
The Qs~ywascalculated by measuring the LAoflow (2R flow probe) and
multiplying by a factor of 2.8. This flow factor was previously determined
in this preparation(Comeau 1992) and was in good agreement with a previous study (Shelton and Burggren 1976). Centralvascularpressureswere
measuredin four sites. A small section (3-5 mm) of the common pulmonary
arterywas exposed and a 20-gauge i.v. catheterwas gently inserted downstream for the measurement of Ppa.The Ppvwas measured by inserting a
catheter into the left atriumby a method previously described (Heisler et
al. 1983). The centralvenous pressurewas measuredby insertinga catheter
(PE 50) into the right jugularvein and advancing2-3 cm towardthe heart.
Finally,the Psyswas measuredby insertinga catheterinto the rightcommon
carotid artery.The pressure catheterswere connected to four straingauge
pressure transducers (P23XL; Spectramed, Oxnard, Calif.). The HR was
measured by inserting needle electrodes into the right and left foreleg and
the left hindlimb and connecting to a cardiotachometer(Type 9857; Sensormedics, YorbaLinda,Calif.). The R~,1and XR,were calculated from the
Poiseuille equation. The net intracardiacshunt flow was calculated by the
differencebetween Qpuiand Qsys.Analog outputsfrom the flowmeters,pressure transducers,and cardiotachometerwere connected to a dataacquisition
system (Series 500; Keithley Metrabyte).All cardiovascularvariableswere
sampled at 1 Hz and stored onto disk for lateroff-line analysis.Experiments
were conducted at room temperature,210-23C.
Cardiac
Shunt
inReptiles1335
Regulation
In this preparation,the right and left cervicalvagus nerves were isolated
and bilateral sectioned (Comeau 1992). Bipolar, silver, stimulating electrodes were placed on either the efferent or afferentend of the sectioned
vagus nerve. Stimulationwas provided by an A310 Accupulser pulse generatorcoupled with an A360 D/R constant currentstimulus isolator (World
PrecisionInstruments,New Haven,Conn.). All experimentswere conducted
at room temperature(210-230C). The protocol for these experiments consisted of control periods followed by periods of electrical stimulation of
either the vagal afferentsor vagal efferents. Stimulationlevels were at 1040 pA;2-5 V, 200 ms duration, 1-3 Hz, and lasted up to 1 min. In addition,
intravenousinfusion of epinephrine, ACh,propranolol,or atropinewas periodically conducted.
Results
Vagal Eferent Stimulation and AChInfusion
The cardiovascularvariablesduring control conditions are shown in table
2. Electricalstimulationof the rightVEFresulted in a large increase in Rpu,
a reduction in HR and a reduction in Qp~,(fig. 3). Similarchanges in Qpu,,
were observed following a bolus injection of ACh (200 nmol/
HR,and Rpu~
kg; fig. 3). These cardiovascularchanges were completely blocked after
intravenousadministrationof 6 gmol/kg atropine (fig. 3).
Magnitude of R-LIntracardiac Shunting
A net R-Lshunt developed during vagal efferent stimulation (fig. 4). The
size of the net shunt was a function of the ratio Rpou/RPy
(fig. 5). As the Rpui/
ratio
the
R-L
net
shunt
increased.
Individualmeasurements
increased,
Rys
of net R-Lshunt ranged up to 25 mL/min/kg. The average net R-Lshunt
was 6 + 5 mL/min/kg (X + SD, n = 5). The net R-Lrepresented 40%+ 7%
of the Qsys. A net R-Lshunt did not develop afteradministrationof atropine.
Vagal Afferent Stimulation and Epinephrine Infusion
Electrical stimulation of vagal afferents resulted in an increase in HR, a
reduction in
and an increase in Qp, (fig. 6). An intravenousinfusion of
Rl,,
epinephrine (4-5 nmol/kg/min) resulted in similarchanges in these cardiovascularvariables(fig. 6). Preliminaryevidence showed that these changes
were abrogatedby administrationof propranolol(10 ltmol/kg; fig. 6).
HR(beat/
min)
43
4
C
220
of
.75 .20
(mmHg/
mL/min/kg)
Rsys
temperature
19
body
a
at
3
Pys(mmHg)
scripta
sys
24
6
(mL/
min/kg)
Pseudemys
.40 .13
turtle
the
in
(mmHg/
mL/min/kg)
Rpul
artery.
15 3
conditions
(mmHg)
Ppul
pulmonary
common
in
control
30
6
(mL/
Qpul
min/kg)
during
pressure
blood
(kg)
variablesMass
.34
1.65
represents
2
5)
=
Cardiovascular
(n
TABLE
Value
mean
Pp,,
Mean
SD
Note.
Cardiac
Shunt
inReptiles1337
Regulation
Stimulation
VagalEfferent
VEF+ Atropine
AChInjection
Qpul
E 4o[
0
I
HR1HR
RVEF---
1RVEEF
ACh-_
I
RpulI
.E 1o
I
E 4o
1-.4---4I
0
20
40
60
Time (s)
80
100 1200
2
50
Time (s)
75
100
0
10
20
30
40
50
0
Time (s)
Fig. 3. Changes in Qpu,,HR, and Rp,tin a single turtle, Pseudemys
scripta, during electrical stimulation of the right vagal efferent nerve
(RVEF),after AChinjection and during RVEFafter administration of
atropine. In this series of experiments the level of nerve stimulation was
20 ptA,2.5 V,200 ms duration, 1.5 Hz. The injection ofACh was intravenous at a dose of 200 nmol/kg and the dose of atropine was 6pmol/kg.
Theanimal was 1. 73 kg.
Magnitude of L-RIntracardiac Shunting
A net L-Rshunt developed during vagal afferentstimulation (fig. 4). The
size of the net L-Rshunt flow was a function of the ratioRpu~/Rsys
(fig. 5). As
the Rpufl/ysratio decreased, the net L-Rshunt flow increased. Individual
measurements of net L-Rshunt ranged up to 55 mL/min/kg. The overall
magnitude of the net L-Rshunt was 28 + 7 mL/min/kg (X + SD, n = 5).
This represented 58%+ 8%of the Qpu.The size of the net L-Rshunt was
reduced after administrationof propranolol (10 pmol/kg) (L-Rshunt = 15
+ 10 mL/min/kg; X+ SD, n = 3).
Qpu,versus Q,,s
Figure 7 shows a summaryof all flow measurements made in this study.
The increase in the total cardiacoutput resulted primarilyfrom an increase
in the OpuI.
1338 J.W.Hicks
7
6
5
I
f
3
E2
1VEF
0
(3V,2Hz,200ms,50/A)
0
20
40
60
Time(sec)
80
100
120
50
40
30
S*O.
E 20
VAF
110
0
200ms, 60/A)
(3.5V,5H;z,
0
50
Time(sec)
100
150
Fig. 4. Upper panel, the development of a net R-Lshunt during electrical
stimulation of vagal efferent nerves in the turtle Pseudemys scripta. Electrical stimulation was 3 V, 2 Hz, 200 ms, 50 gA. Bottom panel, the development of a net L-Rshunt during electrical stimulation of vagal aferent
nerves in the turtle. Electrical stimulation was 3.5 V, 5 Hz, 200 ms, 60
pA. The animal was 1.8 kg.
Discussion
These experimentssupportthe hypothesisthata net R-Lshuntis of cholinergic
originand thata net L-Rshuntis of adrenergicorigin.In addition,these results
shuntsresult
supportthe hypothesisthatthe size and directionof intracardiac
from the reciprocalinterplayof cholinergicand adrenergicmechanisms.
Regulation of lntracardiac Shunt
The results of this study strongly support the hypotheses that the direction
and size of intracardiacshunts are dictatedby the relativeresistancesoffered
inReptiles1339
Cardiac
Shunt
Regulation
60
o
o
4~0 200
C
CtO
0
0
o
20
E
0
o
00
oo o
Rpu
0.0
0.5
oo
o
o oo
o o
ooooo
o
o
0
1.0
1.5
2.0
Rpul/Rsys
Fig. 5. The effects of the ratio Rp,/Rsys, on the net L-Rshunt (upper panel)
and net R-Lshunt (bottom panel).
by the pulmonaryand systemic circulations (White and Ross 1966; White
1976; Burggren 1985). The precise mechanism for the effects of vascular
resistanceon intracardiacshunting is controversial.Currently,there are two
hypotheses addressing the mechanism of intracardiacshunt: "pressure
shunting"and "washoutshunting."The pressureshunting hypothesis states
that low-resistance connections exist between the ventricularcava during
systole that permit nearly unimpeded blood flow through the ventricle.
Consequently,when the Rpulincreases relativeto the Rys, a portion of blood
1340 J.W.Hicks
VagalAfferentStimulation
Qpul
EPIInfusion
VAF+ /3-blockade
I
60
30
R I
I
555FHR
EoRpul
I
I
500
45o
0.7
0.8
0.4Rpul
E 0.2
,I
50
25
0.1,,,_
I
75
Time(s)
100
125
150 0
50
100
150
0.2
.0.3
200
_,0,.__1o.,
Time(s)
250 0
25
50
75
100
125
150
Time(s)
Fig. 6 Changes in Qp,,,,HR, and Rputin a single turtle, Pseudemys
scripta, during electrical stimulation of the left vagal afferent nerve
(LVAF),following epinephrine (EPI) infusion and during (LVAF)after
administration ofpropranolol. In this series of experiments the level of
nerve stimulation was 20 gA, 3 V, 100 ms duration, 5 Hz. The infusion of
EPI was intravenous at a dose of 4.5 nmol/kg/min and the dose ofpropranolol was 10 gtmol/kg. The animal was 1.6 kg.
within the CP (see fig. 1) will be ejected around the muscular ridge into
decreases relative to Rys,a
the systemic circulation. Conversely,when Rp~1
will
in
be
of
the
CV
and
blood
CA
ejected around the muscular
portion
ridge into the pulmonarycirculation.Evidence supportingthe pressure hypothesis has been reported (Shelton and Burggren 1976). In contrast,the
washout hypothesis states that during systole, the muscular ridge forms a
functional separationof the CP from the CVand CA,and thus blood cannot
flow from the CP into the systemic circulation during systole. In addition,
the functional separationresulting from the muscularridge prevents blood
within the CVand CAfrom flowing into the pulmonarycirculation during
systole. Therefore, the size of the R-Lshunt results from the end-diastolic
volume of blood in the CV that is "washed" into the systemic circulation
during systole. Conversely, the size of the L-Rshunt is determined by the
end-systolic volume of blood in the CV that is washed into the CP during
the subsequent diastole. Variationsin the size of the intracardiacshunt result
from factors affecting diastolic filling and/or systolic ejection from the CV
and CA(Heisler and Glass 1985). Such factorswould include the resistance
offeredby the pulmonaryand systemic circulations(Heisler and Glass 1985).
Cardiac
Shunt
inReptiles1341
Regulation
80
o
o
o
o
20 -
0
o
cs, oo
.oEo o
S4
o
oV
o
oo o o a o
o
oz
o
o
o
o
o
o
o
o
1,.1p
o o
40
20o
0%0
0
0
o
0 0o
oo
~e
4 000
%00
OhD 4k4bo
oB
o
o
ooo
00
0
0 6p
008
0m/mnkgl
0
20
40
60
80
Q~sys(milmin kg- )
Fig. 7. The relationship between total Qputand total Q,ysin the turtle
Pseudemys scripta during electrical stimulation of the vagus nerve. Individual data points are from five animals and represent both afferent and
eferent nerve stimulation.
Evidence supporting the washout hypothesis has been recently reported
(Hicks and Malvin1992; Hicks and Comeau 1994).
Regardlessof the precise mechanism for intracardiacshunting, it is clear
from this study thatfactorsthat affectthe relativevascularresistances in the
pulmonaryand systemic circulationswill influence the size and direction
of the net intracardiacshunt. Therefore, it is not surprisingthat the administrationof atropine, which blocks cholinergic vasoconstrictionof the pulmonary circulation, eliminated the net R-Lshunt that developed during
vagal efferent stimulation.
The effects of adrenergicblockade are more difficultto interpret.The net
L-Rshunt results primarilyfrom the large increase in Qpul(fig. 7). It is
possible that P-adrenergicblockade prevented the development of a pulmonaryvasodilationand the subsequent increase in Qpul.Thus the net L-R
shunt was reduced. Alternatively,P-adrenergicblockade may have reduced
the effects of adrenergicstimulationon myocardialcontractility.This study
could not differentiatebetween these factors.In addition, this study did not
1342 J.W.Hicks
address the effects of a-adrenergic blockade. The a-adrenergic receptors
play an importantrole in the regulationof the Rys(Nilsson 1983). It is likely
thata-adrenergicblockade will also reduce the size of the net L-Rshunt by
preventing an increase in Rysduring adrenergic stimulation.
Finally, this study did not address the role of NANCmechanisms in controllingintracardiacshunt.Variouspeptides mayplay a role in cardiovascular
function in reptiles. In the snake Acrochordusgranulatus there is the presence of vasoactive intestinal peptide immunoreactivityin the pulmonary
vasculature(Donald and Lillywhite 1989). In the turtle Pseudemys scripta
results in a systemic vasodiintravenousadministrationof Thr6-bradykinin
lation and a increase in the LAoblood flow. Clearlyany NANCfactor that
effects the Rpul/Rsys ratio will influence the direction and size of the net
intracardiacshunt. The NANCcontrol of cardiovascularfunction in reptiles
remains a topic for research.
Intracardiac Shunts and OtherPhysiological States
The number of studies that have simultaneouslymeasuredthe Q~,iand Qsys
in chronicallyinstrumentedanimals is limited. Therefore,the direction and
size of intracardiacshunting during other physiological states, such as activity, changes in body temperature,feeding, and digestion, is virtuallyunratio
known. Undoubtedly,these physiological states may alter the Rpul/Rys
and therefore affectthe size and direction of the net shunt. Recently,West,
Butler,and Bevan (1992), measuredleft pulmonaryarteryblood flow (Qpul)
and left aorticarchblood flow (Qo) duringrest and swimming in the green
sea turtle, Chelonia mydas.This study revealed thatduringexercise the HR
increased, the Rpu1
decreased, and the Qp~iincreased. The total increase in
that is,
was
accounted for primarilyby an increase in QLpul;
cardiacoutput
,pul increasedmore than Qo duringexercise. Qualitatively,the relationship
between Qypu1 and yQo during exercise in C. mydas is very similar to the
relationship between Qp~,and Qsysin P. scripta measured in this study
(fig.7).
Functional Significance ofL-RIntracardiac Shunting
Several hypotheses have addressed the adaptive functions of intracardiac
shunting (table 1). Many of these hypotheses have emphasized the R-L
shunt that develops during apnea. Of the currentset of hypotheses, none
appear to satisfactorilyexplain the functional advantageof a large net L-R
shunt. In terms of oxygen transportefficiency, a large L-Rshunt associated
with an increase in QI would minimize physiological shunting due to the
Cardiac
inReptiles1343
Shunt
Regulation
ventilation-perfusionratio (f7/Q) mismatching in the lung (Wood 1984;
Westet al. 1992). This would improvethe level of oxygenationof pulmonary
venous blood. The V7/Qheterogeneity within the reptilian lung has been
described only for the alligatorAlligator mississipiensis(Powell and Gray
1989) and the teju lizard Tupinambisnigropunctatus (Hlastalaet al. 1985).
The effects of increasing Qpulor increasing the size of the L-Rshunt on
V/Q heterogeneity have not been determined. The effects of an increase in
Qui or L-Rshunt on the expired P02 to pulmonaryvenous Po2 gradienthas
not been reported.Recent evidence suggests thatthe increase in 6QpU
during
adrenergicstimulationdoes not significantlyimprovethe Po2of pulmonary
venous blood (Hicks and Malvin1992;Hicksand Comeau 1993). In contrast,
an increase in the level of the L-Rshunt may adversely affect oxygen exchange. The mixing of pulmonaryvenous and systemic venous blood would
increase the pulmonary arterialP02. This increase would reduce the Po2
gradient from lung gas to blood and offset any benefit that would result
from the increase in Qpul.
In terms of CO2exchange, the L-Rshunts may be beneficial (Ackerman
and White 1979;White 1985). The mixing of pulmonaryvenous blood with
systemic venous blood increases the oxygen saturationof the pulmonary
arterialblood. This mixing diminishes the capacity of blood to carryCO2
(the Haldane effect). These changes promote CO2elimination from the
lung (White 1985). Theoretically, the amount of CO2eliminated depends
on the minute ventilation, the Qpu,the size of the L-Rshunt, the slope of
the CO2dissociation curve, and the magnitudeof the Haldaneeffect (White
1985). The effects of an L-Rshunt on CO2elimination have not been experimentallyverified.
An additional consequence of an L-Rshunt, not previously appreciated,
may be an improvementin systemic oxygen transport.The systemic oxygen
transportis the product of mean arterialoxygen content (CaO2) and Qsys.
Duringperiods of increased oxygen demand, the systemic oxygen transport
can be improved by increasing CaO2,sys, or both. In turtles, the capacity
to increase Qsys may be constrained (West et al. 1992; fig. 7). This probably
resultsfromthe adrenergicvasodilationof the pulmonarycirculation,making
the pulmonarycircuit the favoredroute for blood flow. Recent studies have
shown that an adrenergicallyinduced L-Rshunt eliminates the R-Lshunt
and subsequentlyraises the systemic arterialoxygen saturationto levels that
are equal to pulmonaryvenous values (Hicks and Malvin 1992; Hicks and
Comeau 1994). Thus, total systemic oxygen transportis improved. These
observationssuggest thatthe eliminationof the R-Lshuntand the subsequent
increase in the CaO2may be importantduring activityor when there is an
increased demand for oxygen.
1344 J.W.Hicks
This article has shown thatthe direction and size of intracardiacshunting
mayresultfromthe reciprocalinterplayof adrenergicandcholinergicfactors.
Specifically,these factors may alter the relative resistances offered by the
pulmonaryand systemic circulations.The development of an R-Lshunt is
of vagalorigin and is undercholinergic control. In contrast,the development
of an L-Rshunt is underadrenergiccontrol.The manipulationof intracardiac
shunting, either by nerve stimulation or the action of specific agonist and
antagonist,can be used in both acute and chronicanimalpreparations.Future
studies will use these techniques to experimentallytest the hypotheses that
address the functional role of intracardiacshunting.
Acknowledgments
The authoracknowledges Dr. Atsushi Ishimatsu (NagasakiUniversity)and
Dr. NorbertHeisler (Max-Planck-Institute)for manyhours of insightful discussions on intracardiacshunting during a recent visit to Gittingen. These
studies have been supported by National Science Foundationgrant DCB900458.
Literature
Cited
F. N. WHITE.1979. Cyclic carbon dioxide exchange in the turtle,
Pseudemys scripta. Physiol. Zool. 52:378-389.
P.J. 1972.Thevagalandsympathetic
innervation
of the isolatedpulmonary
BERGER,
ACKERMAN, R., and
3:113-124.
arteryof a lizardanda tortoise.Comp.Gen. Pharmacol.
of the visceralandvascularsmoothmuscleof
. 1973.Autonomicinnervation
4:1-10.
the lizardlung.Comp.Gen.Pharmacol.
P. J., and G. BURNSTOCK.
1979. Autonomic nervous system. Pages 1-47 in
and P. ULINKSI,eds. Biology of the Reptilia. Vol. 10.
C. GANS,G. NORTHCUTT,
Neurology. Academic Press, New York.
BURGGREN, W. W. 1975. A quantitative analysis of ventilation tachycardia and its control
in two chelonians, Pseudemys scripta and Testudo graeca. J. Exp. Biol. 63:367380.
BERGER,
..
..
of thechelonianreptile:morphology,
circulation
1977.Thepulmonary
pharmacology and hemodynamics.J. Comp. Physiol. 116B:303-324.
1978. Influence of intermittentbreathingon ventriculardepolarizationpatterns in chelonian reptiles.J. Physiol. 378:349-364.
. 1982. Pulmonaryplasmafiltrationin the turtle:a wet vertebratelung?Science
215:77-78.
-.
1985. Hemodynamicsand regulationof centralcardiovascularshunts in repeds. Cardiovascular
and W. W. BURGGREN,
tiles. Pages 121-142 in K. JOHANSEN
shunts: phylogenetic, ontogenetic and clinical aspects. Munksgaard, Copenhagen.
Cardiac
Shunt
inReptiles1345
Regulation
. 1987. Formand function in reptiliancirculation.Am. Zool. 27:5-20.
W. W., and G. SHELTON.1979. Gas exchange and transportduring interBURGGREN,
mittent breathingin chelonian reptiles. J. Exp. Biol. 82:75-92.
BURGGREN,W.W.,A. SMITS,and B. EVANS.1989.Arterial02 homeostasis duringdiving
in the turtle Chelodina longicollis. Physiol. Zool. 62:668-686.
S. G. 1992. Vagal regulation of central vascularblood flow in the turtle,
COMEAU,
Pseudemysscripta. M.S.thesis. CreightonUniversity.
COMEAU,S. G., and J. W. HICKS. 1994. Regulationof central vascularblood flow in
the turtle.Am.J. Physiol. 267:R569-R578.
1990. Isolation and biological activity
CONLON,J. M.,J. W. HICKS, and D. D. SMITH.
of a novel kinin ([Thr6]bradykinin)from the turtle, Pseudemys scripta. Endocrinology 126:985-991.
1989. Vasoactiveintestinal polypeptide-immuDONALD,J. A., and H. B. LILLYWHITE.
noreactive nerves in the pulmonaryvasculatureof the aquatic file snake Acrochordusgranulatus. Cell Tissue Res. 255:585-588.
DONALD, J. A., J. E. O'SHEA, and H. B. LILLYWHITE. 1990. Neural regulation of the
pulmonaryvasculaturein a semi-arborealsnake, Elapheobsoleta.J. Comp. Physiol.
159B:677-685.
innervation
of the cardiovascular
FURNESS,J. B.,andJ. MOORE. 1970.Theadrenergic
system of the lizard, Trachysaurusrugosus.Z. Zellforsch.Mikrosk.Anat.108:150176.
W. H., and H. GADOW.1884. On the anatomyof the cardiacnerves in certain
GASKELL,
cold-blooded vertebrates.J. Physiol. Lond.5:362-372.
1985. Mechanismsand regulation of central vascular
HEISLER,
N., and M. L. GLASS.
shunts in reptiles. Pages 334-347 in K.JOHANSENand W. W. BURGGREN,
eds. Cardiovascularshunts: phylogenetic, ontogenetic and clinical aspects. Munksgaard,
Copenhagen.
HEISLER,N., P. NEUMANN,and G. M. O. MALOIY.1983. The mechanism of intracardiac
shunting in the lizard, Varanus exanthematicus.J. Exp. Biol. 105:15-31.
HICKS,J. W., and S. C. COMEAU.1994. Vagal regulation of intracardiacshunting in
turtles.J. Exp. Biol. 186:109-126.
of intracardiac
HICKS, J. W.,and G. M. MALVIN.1992.Mechanism
shuntingin the
turtle Pseudemysscripta.Am.J. Physiol. 262:R986-R992.
1985.The matching
HLASTALA,M. P.,T. A. STANDAERT, D. J.PIERSON, and D. L.LUCHTEL.
of ventilation and perfusion in the lung of the tegu lizard, Tupinambis nigropunctatus. Respir.Physiol. 60:277-294.
ISHIMATSU,
A.,J. W. HICKS,and N. HEISLER.1988. Analysisof intracardiacshunting in
the lizard Varanus niloticus: a new model based on blood oxygen levels and
microsphere distributions.Respir.Physiol. 71:83-100.
H. B., andJ. A. DONALD. 1989.Pulmonary
blood flowregulationin an
LILLYWHITE,
aquaticsnake. Science 245:293-295.
A. B., and A. J. CARLSON.1921. Studies on
LUCKHARDT,
the visceral sensory nervous
system. VIII. On the presence of vasomotorfibres in the vagus nerve to the pulmonaryvessels of the amphibianand reptilianlung. Am. J. Physiol. 56:72-112.
MCLEAN,J. R., and G. BURNSTOCK.1967. Innervation of the lungs of the sleepy lizard
(Trachysaurusrugosus)-1.Fluorescenthistochemistryof catecholamines.Comp.
Biochem. Physiol. 22:809-813.
1346 J.W.Hicks
W. K., B. L. LANGILLE,
and D. R. JONES.1977. Vagal control of pulmonary
MILSOM,
vascularresistance in the turtle, Chrysemysscripta. Can.J. Zool. 55:359-367.
S. 1983. Autonomicnerve function in the vertebrates.Springer,New York.
NILssoN,
POWELL,F. L., and A. T. GRAY.1989. Ventilation-perfusion relationships in alligators.
Respir.Physiol. 78:83-94.
1976. Cardiovasculardynamicsof the Chelonia
SHELTON,G., and W. W. BURGGREN.
and
ventilation.
J. Exp. Biol 64:323-343.
lung
during apnea
W.
D.
R.
K.
and
1986. Controlof breathingin ectothermic
MILSOM.
SHELTON,G.,
JONES,
and J. G. WIDDICOMBE,
eds. Handvertebrates. Pages 857-909 in N. S. CHERNIACK
book of physiology. Sec. 3, The respiratorysystem. Vol. 2, Controlof breathing.
AmericanPhysiologicalSociety, Bethesda, Md.
1979. Autonomic innervation of the visceral and
SMITH,D. G., and D. H. MACINTYRE.
vascularsmooth muscle of a snake lung (Ophidae: Colubridae).Comp. Biochem.
Physiol. 62C:187-191.
1981. Comparativeanatomyof the ventricular
VANMIEROP,L.H. S., and L.M.KUTSCHE.
A.
A. OPPENHEIMER-DEKKER,
C.
and A.J. MOULERT,
G.
35-46
in
WENINK,
septum. Pages
eds. The ventricular septum of the heart. Leiden University Press, Boston.
. 1985. Some aspects of comparative anatomy of the heart. Pages 38-56 in K.
JOHANSENand W. BURGGREN,eds. Cardiovascular shunts: phylogenetic, ontogenetic
and clinical aspects. Munksgaard,Copenhagen.
R. M. BEVAN.1992. Pulmonaryblood flow at rest and
WEST, N. , P. J. BUTLER,and
during swimming in the green turtle, Chelonia mydas. Physiol. Zool. 65:287-310.
WHITE,F. N. 1976. Circulation. Pages 275-334 in C. GANS, ed. Biology of the Reptilia.
Vol. 5. Academic Press, New York.
. 1985. Role of intracardiacshunts in pulmonarygas exchange in chelonian
eds. Cardiovascular
and W. BURGGREN,
reptiles. Pages 296-309 in K. JOHANSEN
shunts: phylogenetic, ontogenetic and clinical aspects. Munksgaard, Copenhagen.
WHITE, F. N., J. W. HICKS,and A. ISHIMATSU.1989. Respiratory states and intracardiac
shunts in turtles.Am.J. Physiol. 256:R240-R247.
WHITE, F. N., and G. Ross. 1966. Circulatory changes during experimental diving in
the turtle.Am.J. Physiol. 211:15-18.
WOOD,S. C. 1984. Cardiovascular shunts and oxygen transport in lower vertebrates.
Am. J. Physiol. 247:R3-R14.
YAMAUCHI,A., and T.
CHIBA.
1973. Adrenergicand cholinergic innervationof the
turtle heart ventricle. Z. Zellforsch. Mikrosk. Anat. 143:485-493.