The association between cardiac accelerations and movements in fetal sheep

Observations in 5 fetal sheep at 120-134 days gestation, using either real time ultrasound or direct emg recordings to detect fetal movements, showed an association between such movements and cardiac accelerations (greater than 2 s duration and 4 beats/min amplitude). Neuromuscular blockade with gallamine abolished the fetal movements but not the accelerations.     

Fetal acidosis and hypotension during repeated umbilical cord occlusions are associated with enhanced chemoreflex responses in near-term fetal sheep.

This study examined the hypothesis that repeated episodes of brief but severe hypoxia would not attenuate the chemoreflex-mediated rapid initial fall in fetal heart rate (FHR) and, further, that greater hypoxic stress, as shown by hypotension and metabolic acidosis, would be associated with an enhanced chemoreflex response. Chronically instrumented, near-term fetal sheep received 1 min total umbilical cord occlusion either every 5 min for 4 h (1:5 group; n = 8) or every 2.5 min (1:2.5 group; n = 8) until mean arterial blood pressure fell to <20 mmHg on two successive occlusions. Umbilical cord occlusion caused variable decelerations, with sustained hypertension in the 1:5 group and little change in acid-base status (pH 7.34 +/- 0.03 after 4 h). In contrast, the 1:2.5 group showed progressive hypotension and metabolic acidemia (pH 6.92 +/- 0.04 after the last occlusion). The 1:2.5 group showed a significant increase in the rate of initial fall in FHR during the occlusion series, which was greater than the 1:5 group in the last 30 min of the occlusion series (9.4 +/- 1.4 vs. 3.5 +/- 0.3 beats.min(-1).s(-1); P < 0.01), with a greater fall in FHR (71.9 +/- 6.5 vs. 47.0 +/- 8.7 beats/min; P < 0.05). In summary, this study demonstrated that repetitive laborlike cord occlusions, which led to severe fetal compromise, were associated with an increase in the slope and magnitude of the initial FHR deceleration. These findings support the concept of the chemoreflex as a central, robust component of fetal adaptation to severe hypoxia.     

Human maternal and fetal response to graded exercise

Six healthy active women in the third trimester of pregnancy participated in a graded exercise protocol to levels of exertion perceived to be equivalent to that of their usual exercise regimen. Fetal heart rate response (FHR) was documented by ultrasound transducer and confirmed (n = 1) by ultrasonic visualization. Resting maternal O2 consumption was 277 +/- 50 (SD) ml/min and rose to 1,132 +/- 202 ml/min at a mean final exercise intensity of 79 +/- 9 W after 12.8 +/- 1.7 min on a cycle ergometer. There was no significant change in maternal serum insulin, growth hormone, glucose, or pH values. Maternal leukocyte count, hemoglobin, and venous lactate levels rose significantly during the exercise (P less than 0.05). FHR prior to exercise was 142 +/- 4 beats/min and decreased to 84 +/- 34 beats/min during exercise. The decrease in FHR was documented within 1 min of initiating exercise in all cases. During exercise, fetal movements were not accompanied by FHR accelerations. Within 1 min following the cessation of exercise, FHR rose to 143 +/- 8 beats/min and fetal movements were accompanied by FHR accelerations. Since the recovery of FHR occurred immediately after cessation of maternal exercise, this level of maternal exercise does not appear to be harmful to the fetus.     

Determination of heart rate and heart rate variability in the equine fetus by fetomaternal electrocardiography

Heart rate is an important parameter of fetal well-being. We have analyzed fetal heart rate (HR) and heart rate variability (HRV) by fetomaternal electrocardiography (ECG) in the horse (Equus caballus) from midpregnancy to foaling. It was the aim of the study to detect changes in the regulation of fetal cardiac activity over time and to establish normal values in undisturbed pregnancies. A total of 22 mares were available for the study. Fetomaternal electrocardiography was a reliable technique to detect cardiac signals in fetuses between Day 173 of gestation and foaling. Fetal HR decreased from 115 ± 4 beats/min (Days 170 to 240 of gestation) to 83 ± 3 beats/min (Day 320) to 79 ± 1 beats/min (1 d before foaling; P < 0.001). Mean beat to beat (RR) interval and standard deviation of the RR interval (SDRR) increased (P < 0.001). Gestational age thus affects RR interval and HR in the equine fetus. From Days 270 to 340 of gestation, SDRR increased from 11.4 ± 1.3 msec on Day 270 to 27.8 ± 3.6 msec on Day 340 (P < 0.05), and the root mean square of successive RR differences (RMSSD) tended to increase (P = 0.07), indicating maturation of the fetal autonomous nervous system. For the last 10 d before foaling, fetal HR and HRV remained constant and did not allow predicting the onset of parturition in the horse. Only during the last 30 min before the foal was born, in 4 of 5 fetuses, HR decreased and RR interval increased. Accelerations and decelerations in HR were detectable at all times, but neither their number nor duration changed over time.     

Respiratory and non-respiratory thoracic movements in the fetal pig.

The aim of this study was to characterize the pre-natal activity of the respiratory muscles in a non-ruminant, the pig. Tracheal pressure was recorded from 11 unanaesthetized fetal pigs in utero during late gestation in 9 sows. Two types of inspiratory effort occurred episodically in each fetus during each recording period. Episodes of breathing movements lasted 8.6 +/- 1.1 min and their overall incidence was 41.1 +/- 3.4% of recording time. The mean amplitude of the respiratory efforts was 7.6 +/- 1.1 mmHg and the mean inspiratory time was 0.8 +/- 0.1 s. Episodes of non-respiratory inspiratory efforts, considered to be fetal hiccups, lasted 5.6 +/- 0.8 min. Individual hiccups had a duration of 150-200 msec, a mean amplitude of 41.4 +/- 2.4 mmHg and a mean frequency, during episodes, of 21.9 +/- 2.0 min-1. In contrast to fetal breathing movements, hiccups appeared to be stereotyped events. It is concluded that, in common with other ruminant and non-ruminant species, two types of inspiratory effort occur in the fetal pig. Fetal hiccups, which have been observed in non-ruminant species, may be analogous to deep inspiratory efforts in the sheep fetus. The stimulus for, and function of, fetal hiccupping remain to be determined.     

Umbilical artery flow velocity waveforms and placental vascular resistance during maternal placental outflow obstruction in sheep

This study was designed to test the hypothesis that the pulsatility index (PI) of the umbilical artery flow velocity waveform varies as a function of placental vascular resistance. Placental vascular resistance was raised by a one-minute occlusion of the maternal inferior vena cava. Occlusion of the maternal inferior vena cava resulted in a decrease in fetal heart rate from 183 +/- 7.8 beats/min to 142 +/- 8.6 beats/min at the end of occlusion (P less than 0.05). Placental vascular resistance increased from 0.113 +/- 0.021 mmHg.ml-1.min during control to 0.151 +/- 0.033 mmHg.ml-1.min (P less than 0.05) during occlusion. The pulsatility index increased from 1.05 +/- 0.05 to 1.85 +/- 0.4 (P less than 0.05) during occlusion. After parasympathetic blockade with atropine fetal heart rate did not change during occlusion. Placental vascular resistance increased from 0.091 +/- 0.014 before to 0.121 +/- 0.021 mmHg.ml-1.min during occlusion (P less than 0.05). The pulsatility index increased from 0.98 +/- 0.1 before to 1.12 +/- 0.12 during occlusion (P less than 0.05). These results support the hypothesis that, in the fetal sheep, placental vascular resistance is one of the determinants of the pulsatility index of the umbilical artery.     

Skeletal muscle beta-receptors and isoproterenol-stimulated vasodilation in canine heart failure

To investigate whether heart failure alters beta-adrenergic receptors on skeletal muscle and its associated vasculature, the density of beta-adrenergic receptors, isoproterenol-stimulated adenylate cyclase activity, and coupling of the guanine nucleotide-binding regulatory protein were compared in 18 control dogs and 16 dogs with heart failure induced by 5-8 wk of ventricular pacing at 260 beats/min. Hindlimb vascular responses to isoproterenol were compared in eight controls and eight of the dogs with heart failure. In dogs with heart failure, the density of beta-receptors on skeletal muscle was reduced in both gastrocnemius (control: 50 +/- 5; heart failure: 33 +/- 8 fmol/mg of protein) and semitendinosus muscle (control: 43 +/- 9; heart failure: 27 +/- 9 fmol/mg of protein, both P less than 0.05). Receptor coupling to the ternary complex, as determined by isoproterenol competition curves with and without guanosine 5'-triphosphate (GTP), was unchanged. Isoproterenol-stimulated adenylate cyclase activity was significantly decreased in semitendinosus muscle (control: 52.4 +/- 4.6; heart failure: 36.5 +/- 9.5 pmol.mg-1.min-1; P less than 0.05) and tended to be decreased in gastrocnemius muscle (control: 40.1 +/- 8.5; heart failure: 33.5 +/- 4.5 pmol.mg-1.min-1; P = NS). Isoproterenol-induced hindlimb vasodilation was not significantly different in controls and in dogs with heart failure. These findings suggest that heart failure causes downregulation of skeletal muscle beta-adrenergic receptors, probably due to receptor exposure to elevated catecholamine levels, but does not reduce beta-receptor-mediated vasodilation in muscle.     

Effect of hypoxia on arterial baroreflex control of heart rate and muscle sympathetic nerve activity in humans.

We tested the hypothesis that acute hypoxia would alter the sensitivity of arterial baroreflex control of both heart rate and sympathetic vasoconstrictor outflow. In 16 healthy, nonsmoking, normotensive subjects (8 women, 8 men, age 20-33 yr), we assessed baroreflex control of heart rate and muscle sympathetic nerve activity by using the modified Oxford technique during both normoxia and hypoxia (12% O(2)). Compared with normoxia, hypoxia reduced arterial O(2) saturation levels from 96.8 +/- 0.3 to 80.7 +/- 1.4% (P < 0.001), increased heart rate from 59.8 +/- 2.4 to 79.4 +/- 2.9 beats/min (P < 0.001), increased mean arterial pressure from 96.7 +/- 2.5 to 105.0 +/- 3.3 mmHg (P = 0.002), and increased sympathetic activity 126 +/- 58% (P < 0.05). The sensitivity for baroreflex control of both heart rate and sympathetic activity was not altered by hypoxia (heart rate: -1.02 +/- 0.09 vs. -1.02 +/- 0.11 beats. min(-1). mmHg(-1); nerve activity: -5.6 +/- 0.9 vs. -6.2 +/- 0.9 integrated activity. beat(-1). mmHg(-1); both P > 0.05). Acute exposure to hypoxia reset baroreflex control of both heart rate and sympathetic activity to higher pressures without changes in baroreflex sensitivity.     

Leg vasoconstriction during dynamic exercise with reduced cardiac output.

We evaluated whether a reduction in cardiac output during dynamic exercise results in vasoconstriction of active skeletal muscle vasculature. Nine subjects performed four 8-min bouts of cycling exercise at 71 +/- 12 to 145 +/- 13 W (40-84% maximal oxygen uptake). Exercise was repeated after cardioselective (beta 1) adrenergic blockade (0.2 mg/kg metoprolol iv). Leg blood flow and cardiac output were determined with bolus injections of indocyanine green. Femoral arterial and venous pressures were monitored for measurement of heart rate, mean arterial pressure, and calculation of systemic and leg vascular conductance. Leg norepinephrine spillover was used as an index of regional sympathetic activity. During control, the highest heart rate and cardiac output were 171 +/- 3 beats/min and 18.9 +/- 0.9 l/min, respectively. beta 1-Blockade reduced these values to 147 +/- 6 beats/min and 15.3 +/- 0.9 l/min, respectively (P < 0.001). Mean arterial pressure was lower than control during light exercise with beta 1-blockade but did not differ from control with greater exercise intensities. At the highest work rate in the control condition, leg blood flow and vascular conductance were 5.4 +/- 0.3 l/min and 5.2 +/- 0.3 cl.min-1.mmHg-1, respectively, and were reduced during beta 1-blockade to 4.8 +/- 0.4 l/min (P < 0.01) and 4.6 +/- 0.4 cl.min-1.mmHg-1 (P < 0.05). During the same exercise condition leg norepinephrine spillover increased from a control value of 2.64 +/- 1.16 to 5.62 +/- 2.13 nM/min with beta 1-blockade (P < 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)     

Bradykinin-induced chemoreflexes from skeletal muscle: implications for the exercise reflex

We examined the cardiovascular response to bradykinin stimulation of skeletal muscle afferents and the effect of prostaglandins on this response. Intra-arterial injection of 1 microgram bradykinin into the gracilis muscle of cats reflexly increased mean arterial pressure by 16 +/- 2 mmHg, left ventricular end-diastolic pressure by 1.6 +/- 0.6 mmHg, maximal dP/dt by 785 +/- 136 mmHg/s, heart rate by 11 +/- 2 beats/min, and mean aortic flow by 22 +/- 3 ml/min. The hemodynamic responses were abolished following denervation of the gracilis muscle. The increases in mean arterial pressure and maximal dP/dt were reduced by 68 and 45%, respectively, following inhibition of prostaglandin synthesis with indomethacin (2-8 mg/kg iv). Treatment with prostaglandin E2 (PGE2, 15-25 micrograms ia) restored the initial increase in mean arterial pressure, but not dP/dt, caused by bradykinin stimulation. Injection of PGE2 (15-30 micrograms ia) into the gracilis, without prior treatment with indomethacin, augmented the bradykinin-induced increases in mean arterial pressure and dP/dt. We conclude that small doses of bradykinin injected into skeletal muscle are capable of reflexly activating the cardiovascular system and that prostaglandins are necessary for the full manifestation of the corresponding hemodynamic response. The pattern of hemodynamic adjustment following bradykinin injection into skeletal muscle is very similar to that induced by static exercise. Therefore, it is possible that intense exercise provides a stimulus for this bradykinin-induced reflex in vivo.     

Effect of acutely-induced lactic acidemia on fetal breathing movements, heart rate, blood pressure, ACTH and cortisol in sheep.

Experiments were conducted in 8 chronically-catheterized fetal sheep at 125-135 days gestation in order to determine the effect of exogenously administered lactic acid to the fetus on fetal heart rate, blood pressure, breathing movements (FBM), electrocortical activity (ECOG), plasma immunoreactive (IR-ACTH) and cortisol concentrations. When fetal arterial pH decreased from 7.37 +/- 0.01 during the control period to 7.20 +/- 0.01, there was an initial bradycardia followed by tachycardia but no change in blood pressure. The amplitude of FBM increased 2-fold initially in association with an increase in PCO2 from 47.9 +/- 2.1 mmHg to 58.8 +/- 3.6 mmHg at 5 min into the lactate infusion. There was no change in the incidence of FBM or low-voltage ECOG and there was no change in the plasma concentrations of IR-ACTH and cortisol with the infusion of lactate. We conclude that the major effects of acutely elevating circulatory lactate concentrations in fetal sheep are to increase the amplitude of FBM and to cause an initial bradycardia followed by a tachycardia.     

Plasma adenosine and cardiovascular responses to dipyridamole in fetal sheep.

The effects of dipyridamole infusion on fetal arterial plasma adenosine level, [ADO], and the systemic cardiovascular system were studied in 10 fetal sheep at 130-135 days gestational age. Dipyridamole (0.25 mg/kg) was infused into the fetuses intravenously during normoxia and hypoxia. Plasma [ADO] was measured using high-performance liquid chromatography, (HPLC), and fetal heart rate and arterial blood pressure were monitored throughout the study. These studies were performed in the absence and presence of theophylline, an adenosine receptor antagonist. During normoxia (PO2, 23.8 +/- 2.0 Torr), dipyridamole infusion increased fetal plasma [ADO] from 0.82 +/- 0.10 microM to 1.41 +/- 0.16 microM within 1 min (P < 0.01) and fetal heart rate from 157 +/- 6 bpm to 174 +/- 7 bpm (P < 0.01), but did not change mean blood pressure. Fetal plasma [ADO] and fetal heart rate returned to basal levels quickly. Treatment with theophylline did not alter the elevation of plasma [ADO] after dipyridamole infusion, but abolished responses of fetal heart rate to dipyridamole infusion. After 15 min of hypoxia with an average arterial PO2 of 15.4 +/- 1.1 Torr, fetal plasma [ADO] increased to 1.15 +/- 0.14 microM (P < 0.01). Dipyridamole infusion then further raised fetal plasma [ADO] to 1.67 +/- 0.27 microM (P < 0.01). The duration of the increase of fetal plasma [ADO] after dipyridamole infusion was no longer in hypoxia than in normoxia, however there was no significant change in the pattern of transient fetal bradycardia and persistent hypertension.(ABSTRACT TRUNCATED AT 250 WORDS)     

Repetitive reduction of uterine blood flow and its influence on fetal transcutaneous PO2 and cardiovascular variables

The influence of repeated asphyxia on fetal transcutaneous PO2, relative local skin perfusion, heart rate, blood gases and pH was investigated in 15 experiments on 8 acutely instrumented sheep fetuses in utero between 125 and 145 days gestation (term is 147 days). Uterine blood flow was intermittently arrested (11 times within 33 min) by intra-vascular maternal aortic occlusion, exposing the fetuses to repeated episodes of asphyxia of 30 (n = 3), 60 (n = 9) and 90 (n = 3) s duration. The fetal transcutaneous PO2 fell as the duration of asphyxia (2 alpha less than 0.01), heart rate deceleration area (2 alpha less than 0.01) and acidaemia (2 alpha less than 0.01) increased. With decreasing skin perfusion, which was dependent on the duration of asphyxia (2 alpha less than 0.001) and acidaemia (2 alpha less than 0.001), a discrepancy developed between transcutaneous and arterial PO2. The increase (delta) in transcutaneous-arterial PO2 difference was related linearly to the duration of asphyxia (2 alpha less than 0.01), the mean haemoglobin oxygen saturation (2 alpha less than 0.001), acidaemia (2 alpha less than 0.001) and relative local skin flow (2 alpha less than 0.05). It was highest after severe episodes of asphyxia (90 s), when O2 saturation, skin blood flow and arterial blood pH values were low. Fetal heart rate deceleration area was only correlated with the cutaneous-arterial PO2 difference when the mean fetal haemoglobin oxygen saturation was below 35%. Thus, a discrimination of heart rate decelerations that are significant for the fetus seems to be possible, when associated with low transcutaneous PO2 values. We conclude that in the sheep fetus transcutaneous PO2 measurements during repeated asphyxial episodes yield information on fetal oxygenation and on the skin vasomotor response.     

Influence of static magnetic fields on pain perception and sympathetic nerve activity in humans.

Static and pulsed magnetic fields have been reported to have a variety of physiological effects. However, the effect of static magnetic fields on pain perception and sympathetic function is equivocal. To address this question, we measured pain perception during reproducible noxious stimuli during acute exposure to static magnets. Pain perception, muscle sympathetic nerve activity, mean arterial pressure, heart rate, and forearm blood velocity were measured during rest, isometric handgrip, postexercise muscle ischemia, and cold pressor test during magnet and placebo exposure in 15 subjects (25 +/- 1 yr; 8 men and 7 women) following 1 h of exposure. During magnet exposure, subjects were placed on a mattress with 95 evenly spaced 0.06-T magnets imbedded in it. During placebo exposure, subjects were placed on an identical mattress without magnets. The order of the two exposure conditions was randomized. At rest, no significant differences were noted in muscle sympathetic nerve activity (8 +/- 1 and 7 +/- 1 bursts/min for magnet and placebo, respectively), mean arterial pressure (91 +/- 3 and 93 +/- 3 mmHg), heart rate (63 +/- 2 and 62 +/- 2 beats/min), and forearm blood velocity (3.0 +/- 0.3 and 2.6 +/- 0.3 cm/s). Magnets did not alter pain perception during the three stimuli. During all interventions, no significant differences between exposure conditions were found in muscle sympathetic nerve activity and hemodynamic measurements. These results indicate that acute exposure to static magnetic fields does not alter pain perception, sympathetic function, and hemodynamics at rest or during noxious stimuli.     

Steady-state maternal and fetal plasma concentrations of glyceryl trinitrate (GTN) in the preterm sheep

The administration of glyceryl trinitrate (GTN; nitroglycerin) is increasing during preterm pregnancies, yet its disposition and, importantly, the extent of fetal exposure remain to be elucidated. When used as a tocolytic (pharmacological agent that stops uterine contractions), it is administered transdermally (24-48 h). Here, we quantified the maternal and fetal steady-state plasma concentrations of maternal intravenous GTN in preterm sheep and continuously monitored maternal and fetal vascular parameters to observe possible dose-dependent vascular effects. Preterm (120 days gestation) pregnant sheep (n = 6) were instrumented with maternal femoral arterial (MA) and venous (MV) and fetal femoral arterial (FA) and umbilical venous (UV) polyethylene blood-sampling catheters. During maternal GTN infusion (3.0 micro g.kg-1.min-1, 60-min duration) the steady-state GTN concentrations ([GTN]) were as follows: MA, 98.6 +/- 9.0 nM; UV, 17.4 +/- 7.6 nM; and FA, <5 nM. There were no changes in maternal and fetal mean arterial pressure and heart rate or in uterine activity. Overall, the steady-state [GTN] was established by 5 min, and the UV/MA ratio of [GTN] was 0.18. The FA [GTN] (<5 nM) indicates that the fetus cleared essentially all GTN in the UV, and the maternal and fetal heart rate and mean arterial pressure appear to be independent of maternal GTN infusion.     

Sleep and wakefulness in the green iguanid lizard (Iguana iguana)

The reptile Iguana iguana exhibits four states of vigilance: active wakefulness (AW), quiet wakefulness (QW), quiet sleep (QS) and active sleep (AS). Cerebral activity decreases in amplitude and frequency when passing from wakefulness to QS. Both parameters show a slight increase during AS. Heart rate is at a maximum during AW (43.8+/-7.9 beats/min), decreases to a minimum in QS (25.3+/-3.2 beats/min) and increases in AS (36.1+/-5.7 beats/min). Tonical and phasical muscular activity is present in wakefulness, decreases or disappears in QS and reappears in AS. Single or conjugate ocular movements are observed during wakefulness, then disappear in QS and abruptly reappear in AS. Although these reptiles are polyphasic, their sleep shows a tendency to concentrate between 20:00 and 8:00 h. Quiet sleep occupies the greater percentage of the total sleep time. Active sleep episodes are of very short duration, showing an average of 21.5+/-4.9 (mean+/-SD). Compensatory increment of sleep following its total deprivation was significant only for QS. Reaction to stimuli decreased significantly when passing from wakefulness to sleep. It is suggested that the lizard I. iguana displays two sleep phases behaviorally and somatovegetatively similar to slow wave sleep and paradoxical sleep in birds and mammals.     

Increased cardiac sympathetic nerve activity in heart failure is not due to desensitization of the arterial baroreflex

Increased sympathetic drive to the heart worsens prognosis in heart failure, but the level of cardiac sympathetic nerve activity (CSNA) has been assessed only by indirect methods, which do not permit testing of whether its control by arterial baroreceptors is defective. To do this, CSNA was measured directly in 16 female sheep, 8 of which had been ventricularly paced at 200-220 beats/min for 4-6 wk, until their ejection fraction fell to between 35 and 40%. Recording electrodes were surgically implanted in the cardiac sympathetic nerves, and after 3 days' recovery the responses to intravenous phenylephrine and nitroprusside infusions were measured in conscious sheep. Electrophysiological recordings showed that resting CSNA (bursts/100 heartbeats) was significantly elevated in heart-failure sheep (89 +/- 3) compared with normal animals (46 +/- 6; P < 0.001). This increased CSNA was not accompanied by any increase in the low-frequency power of heart-rate variability. The baroreceptor-heart rate reflex was significantly depressed in heart failure (maximum gain -3.29 +/- 0.56 vs. -5.34 +/- 0.66 beats.min(-1).mmHg(-1) in normal animals), confirming published findings. In contrast, the baroreflex control of CSNA was undiminished (maximum gain in heart failure -6.33 +/- 1.06 vs. -6.03 +/- 0.95%max/mmHg in normal sheep). Direct recordings in a sheep model of heart failure thus show that resting CSNA is strikingly increased, but this is not due to defective control by arterial baroreceptors.     

Effects of moderate hypoxia on fetal electrocortical activity, eye movements, and breathing activity in sheep

Isocapnic hypoxaemia (delta PaO2 = -8.0 +/- 0.5 mmHg; delta CaO2 = -2.86 +/- 0.20 ml/dl) was produced in fetal sheep by having the ewe breathe for one hour a gas mixture (v/v) of 10.5% O2 and 1.5% CO2 in N2. Mean fetal heart rate, blood pressure, and incidence of low voltage electrocortical activity were not affected. However, the incidence of rapid-eye-movements and breathing activity was reduced by about 40%. Breathing movements during hypoxaemia had a mean inspiratory time, breath interval, and tracheal pressure amplitude which did not differ significantly from those during control experiments in which the ewe breathed air from the plastic bag. These observations suggest that hypoxia decreases the incidence of breathing movements but does not affect the amplitude or pattern of breathing activity and that it may reduce the incidence of eye movements and breathing activity through a common mechanism.