mise à jour du
10 décembre 2006
Appl Anim. Behav. Sci.
The Phenomenon of pandiculation in the kinetic
behaviour of the sheep fetus
AF Fraser
Surgical Research Laboratories, Department of Surgery,
Memorial University of Newfoundland, St. John's, Newfoundland (Canada)
Pandiculation: the comparative phenomenon of systematic stretching
Fraser AF
Horse stretch The experimental animal Rollin BE & Kesel ML


A long-term fluoroscopic study was performed on pregnant ewes over several lambing seasons to determine the nature of sheep fetal behaviour. This study led to the recognition of the phenomenon of fetal pandiculation, which is essentially similar to the post-natal pandiculation now seen as a comparative phenomenon of systematic stretching. The characteristic features of fetal pandiculation have now been studied in general form and as specific components in the fetuses of 22 ewes. Thirty-three events of full pandiculation were detected in 25 episodes in 222 observations. The extant definition of pandiculation was used as the frame of reference for the phenomenon as follows: "An instinctive movement, consisting in the extension of the legs, the raising and stretching of the arms and the throwing back of the head and trunk, accompanied by yawning". In addition, extension of the axial skeleton and vigour were taken as further definitive features of full pandiculation.
The phenomenon has both form and composition. In form, various salient features can be identified as typical. In the order approximating their consistency of expression in the phenomenon, they include: (I) pattern; (II) direction; (III) elongation and extension; (IV) polarity; (V) episodic occurrence. Since these features occur in the majority of cases, they can be taken as identifying the general form of major pandiculation.
No episode of fetal pandiculation incorporated all of the characteristic components. A score was made of the numbers of components occurring per episode; scores ranged from 6 to 8, to a maximum of 9. The modal average was 7. Additional features included: (a) temporal distribution distribution was even throughout the period of gestation studied (2-20 days pre-partum); (b) incidence per fetus - full pandiculation was usually encountered in each fetus on only one or two occasions; (c) partial pandiculation - this was recognised when the components were reduced in expression of vigour or in number. Partial, incomplete and minor pandiculations collectively numbered 63, and exceeded full pandiculations in incidence.
Fetal pandiculation could be related to the ends of fetal somnolent states and may be a marker of transitional phases between sleep and alert states. It can therefore indicate that the phasic nature of fetal processes is being sustained normally and may thus indicate that the status of the fetus is satisfactory in general respects. Viewed in a wider, comparative light, fetal pandiculation could be regarded as a simple signal of adequate well-being which could be identified by non-invasive methods. It is clear that the phenomenon calls for appropriate attention in further work The pursuit of research into fetal behaviour can be a profitable branch of fetology.
In the finite space of the uterus, the fetus is able to produce remarkable and various movements. The features of fetal movement call for appreciation since they probably relate to fetal well-being and perinatal competence. Contemporary studies on fetal respiratory movements have overshadowed other forms of fetal movement (Bowes et al., 1981; Harding and Poore, 1982), but all activity has significance in the course of fetal development (Liggins and Gluckman, 1984). In particular, fetal kinesis indicates forms of movement in confined space essential to normal physical function. In the course of practical situations in veterinary obstetrics, the gross kinetic movements and attitudes of the maturing fetus require to be identified so that appraisal of the fetal status can be made. Additionally, in comparative obstetrical work there is flow an acknowledgement of "the need for a simple non-invasive test of fetal wellbeing" (Harding, 1984). This may lie in fetal movement in some form (Truclinger et al., 1979). Fetal kinetics warrant more study.
The sheep has been a traditional and suitable model for both human and animal fetology, and parturient physiology from the studies of Barcroft and Barron (1937) to the present (Hecker, 1983). Some later studies on fetal kinetics were based on Doppler ultrasound before real-time ultrasound was used (Goodman and Manteli, 1980; Wittman et al., 1981). As an alternative to ultrasound, an earlier radiographic study on the principal features of fetal kinesis in the sheep reported simple and complex movements, and postural steps prior to parturition (Fraser and Terhune, 1977 a,b). At the conclusion of that report, it was stated that: "In future studies of this nature, fluoroscopy and possible cine-radiography could prove of value" (Fraser and Terhune, 1976b).
Cine-radiography was subsequently found to have technical difficulties relating to film development, but fluoroscopy proved suitable and a long-term fluoroscopic programme was embarked upon in sheep. The video-taped records from this work were accumulated over several breeding seasons before being thoroughly studied. From this work, an observation on overall sheep fetal posture at birth, including elbow flexion intrapartum, has been reported by Husa et al. (1988) which confirmed the radiographic findings of Fraser and Terhune (1977b). Review and synthesis of these fluoroscopic records led to the recognition of the phenomenon of pandiculation in the sheep fetus.
The general form of pandiculation can be given briefly as a vigorous overall extension of the body with upward and forward extension of the head, forward extension of the forelimbs and stretching of the spinal column. In some instances, there is also hind limb extension. Wide yawning is also included in some instances. This stretching behaviour in the fetus is essentially similar to the post-natal and adult pandiculation, which itself has only recently been recognised as a comparative phenomenon of systematic stretching (Fraser, 1989a). Characteristic fetal pandiculation has now been studied, in its general form and specific components, in an extensive review, involving 222 observations on 22 ewes.
Various components can be recognised in fetal pandiculations. In the order of their apparent occurrences they are as follows: (1) elbow extension; (2) elbow advance; (3) carpal and digital extension; (4) cervico-occipital extension; (5) head elevation; (6) pandiculated pose; (7) intention movements; (8 and 9) yawning and jaw movement.
Animals, materials and methods
The recordings on which this report is based were made during three sheep breeding seasons. All the ewes were of the Newfoundland breed type and were taken as available from an adjacent research flock. Each chosen animal had been previously screened to ensure that it was pregnant. In as many cases as possible, ewes were selected which carried only a single fetus. They were monitored by the fluoroscopic method on a loose schedule daily, or on alternate days, throughout the period estimated as the final 2-3 weeks of gestation. Each sampling period was of 10 min duration, during which all the fluoroscoped skeletal images on the monitor screen were also recorded on video tape by a video cassette recorder coupled to the monitor. On some occasions, repeat observations were carried out later in the same day to determine the persistence of some feature of interest, such as pre-partum orientation.
All ewes were placed, with minimal restraint, on the X-ray table in left lateral recumbency. This position was chosen to minimise any effect from rumenal weight and bulk overlying the fetus, and possibly impeding its movement, since fetal activity can be compromised by maternal forces (Harding et al., 1981). A canvas binder was used over the ewe's body to provide stability for the ewe on the table. Rope hobbles were placed on paired forelimbs and hind limbs, and secured to the side rail of the table. Blindfolds over the eyes of the ewes were found to quieten them. Ewes soon adapted to the procedure and after the initial few sessions there was little struggle during fluroscopy.
When the recordings were subsequently reviewed, very specific features, such as limb movements which have been noted by others (Natale et al., 1981; Rigatto et al., 1982), were the initial focus of attention. However, after the holistic phenomenon of pandiculation was discovered, all further reviews were aimed at singling out and interpreting this special feature. Since it was only occasionally shown by the fetus and had not been specifically sought out in the initial monitoring, locating the episodes of pandiculation required thorough searching of all the recordings. Although some of these lacked focus, many examples were discernible and complete. They were present in real time on the tapes and existed as a short phase of activity which might be analogous to the short-term cycles of activity which have been reported in the human fetus (Granat et al., 1979). The taped material did not always bridge the entire shortterm episode owing to the cut-off time of the original taping or loss of target. Viewing both fore and hindquarters at the same point in the observation was difficult, except in those cases where curving of the fetal spine brought them together.
The definition of pandiculation, as given in the Oxford English Dictionary, was used as the frame of reference for the phenomenon as follows: "An instinc tive movement, consisting in the extension of the legs, the raising and stretching of the arms and the throwing back of the head and trunk, accompanied by yawning". In addition, extension of the axial skeleton and vigour were taken as definitive features of full pandiculation. It became apparent that some episodes of this phenomenon lacked some degree of quality or some components. These instances were designated as partial pandiculation, a term which also included minor pandiculation. The phenomenon was therefore classified into full or partial categories. The definition given to the latter, for the purposes of this study, was as follows: "A bodily extension which lacks some property of major pandiculation but which retains some of its salient features such as full limb and axial extension".
General form of major pandiculations
Events of overall, holistic extension were seen as complex fetal movements of a high order of organisation. They were notable in a particular state in which the fetus exhibited maximal extension of head and neck, together with marked forelimb extension. The limb extensions were bilaterally symmetrical and sometimes bipolar, i.e. affecting both fore and hind poles of the fetal axis. The co-ordination of all these sundry extensions showed that they were related in one general kinetic feature of extension, viz. pandiculation. Episodes of this phenomenon were comparatively rare and usually isolated events, but at other times they occurred in a sequence of two or three instances in close association. Thirty-three major pandiculations were seen in a total of 22 fetuses (Table 1).
In form, various salient features can be identified as typifying the activity. Some of these are constant and, to that extent, definitive; others are variable features of general form. They are described below in the order approximating their consistency of expression in the phenomenon.
I Pattern
The progressive course of the actions in pandiculation, after being effectively initiated, exhibited a pattern. The outline of this behavioural scheme conformed to the concept of an action pattern, as defined by McFarland (1987). The several movements flowed into each other, creating one compound action of 5-10 s duration.
II. Direction
The collective effect of these fetal movements was distal extension, resulting in pandiculation having a longitudinal direction. In this, it was more often observed that the anterior parts, namely the head, neck and forelimbs, influenced the longitudinal direction in the disposition of the fetus. In a proportion of cases, both fore and hind parts participated simultaneously in extensions. It was taken as axiomatic that the phenomenon required the vigorous extension of the head, neck and forelimbs as essential features. When extension of the hind limbs occurred alone, this was not taken as an episode of pandiculation, particularly since some hind limb extensions were directed anteriorly, rather than posteriorly, in relation to the fetal body. As a result of the head and neck twisting during extension, a few pandiculations had a rotary direction. The degrees of such rotation were often 90-180', in relation to the long axis of the fetus.
III. Elongation and extension
The longitudinal direction resulted in the fetus being outstretched. It follows that these two features are related to each other, but they are subtly different in that one characterises the initial step of the phenomenon, while the other describes its final form. Linear elongation was also a consequence of rotary pandiculation which was, incidentally, more often observed in the fetus in a supine position than in a prone position.
The whole activity took the form of a vigorous action which resulted in the extension of the fetus in its totality. Vigour was a notable quality in the form of this kinetic event.
IV Polarity
Pandiculation was sometimes bipolar in form, in that both forequarters and hindquarters were involved in the extension. This appeared to take place in some instances with the foreparts extending marginally before the hind parts did. Sometimes, pandiculations appeared to involve only the anterior end of the animal and such extension was taken as one of the identifying forms of this activity. Observations on this feature were made difficult by the inability to monitor both ends of the fetus simultaneously unless the fetus presented these in one radiographic field by virtue of lateral spinal flexion.
V. Episode
It was usual for instances of full pandiculation to occur in isolation. In addition, these events were of low frequency and only occurred in 25 (- 11%) of the observations.
In five episodes, pandiculations occurred in series, with one event following almost immediately after another. In three of these cases, three pandiculations occurred, while in the two others two pandiculations occurred in close succession forming a pair of actions taking the form of one pandiculated episode.
The above features occur in the majority of cases and can therefore be taken as identifying the general form of major pandiculation. Within the general form, various components can be recognised. These are listed in Table 2a and b, and are given there and below in the order of their apparent incidence in pandiculation.
Components of fetal pandiculation
1. Elbow extension This component occurred in every instance and was considered to be one of the definitive features of the phenomenon. Typically, both elbows extended simultaneously and to an equivalent degree.
2. Elbow advance Somewhat in association with elbow extension, advance of the elbow joint was also a notable feature and occurred in each instance of full pandiculation. Again, this was taken as a definitive component. The advance of the elbow was, of course, the consequence of the shoulder joint extending, but the resultant shifting forward of the elbow was the more prominent result.
3. Carpal and digital extension In association with extension of the elbow and the shoulder (as given above in Nos. I and 2), full extension of the carpal and digital joints also occurred. This obviously resulted in the forelimbs being extended forward to their fullest extent.
4. Cervico-occipital extension Together with the above three components, this item of head and/or neck extension was also an identifying characteristic in pandiculation. Typically, in this the head was extended to its fullest possible position, chiefly as a result of maximal extension of the atlanto-occipital joint.
5. Head elevation While this component was similar to component 4 above, it was considered to be somewhat different in that the exertion of the fetus appeared to cause its head being raised as a result of elevating neck action. The impression was gained that when this occurred it tended to precede the forward extension of the head.
6. Follow-up pandiculated pose In the majority of incidences, when the pandiculation had been completed there was a follow-up to this with the fetus remaining in the extended pose passively for a variable period of time. This pandiculated pose was noted in approximately two-thirds of cases. Some of these were poses of partial pandiculation. A notable example of a pose in partial pandiculation was the characteristic birth posture, which has general tension, but elbow flexion (Fraser and Terhune, 1977b; Husa et al., 1988).
7. Intention movements In a number of instances, the phenomenon of pandiculation appeared to be preceded by initiative movements of the forelimbs (Fig. 1). These took the form of pedal probing or pedalling-like actions in which the two forelimbs alternated in performing short forward extensions and retracting movements with carpal flexion. On some occasions, the "pedalling" took the form of probing with the forelimbs, in short movements directing the feet forward with carpal extension. These actions meet the definition of intention movements provided by McFarland (1987).
8 and 9. Yawning and jaw movement Although they have not been previously reported in the literature, it was evident in this study that certain fetal jaw movements were morphological yawns. These actions took the form of slow extension of the tempero-mandibular joints to the fullest extent. The mouth was held opened in these instances for - 2 or 3 s and then closed. In some instances, such yawns occurred as a commencement to pandiculation, while in other instances they occurred as the conclusion of pandiculation. They were not a constant feature of this phenomenon, but yawning was seen clearly in several instances of full pandiculation. Yawning, per se, could be considered in the category of a partial pandiculation.
Somewhat in association with yawning, very rapid actions of tempero-mandibular articulation (TMA) were seen as a ninth component at various points in fetal activity associated with yawning and as a component of pandiculation. Rapid TMAs were, however, common at other times and it would be difficult to be certain of the actual relationship between this minor kinetic factor and the performance of pandiculation.
Component score
Most of the components, as listed in Table 2a, occurred in the majority of episodes of full pandiculation. A score was made of the numbers of components occurring per episode. These scores ranged from 6 to 8 with a maximum of 9. The modal average was 7; the typical pandiculation, therefore, had 7 of these components in its make-up. No episode of pandiculation incorporated all of the characteristic components (Table 2h).
Additional features
Temporal distribution Pandiculation episodes were noted in relation to time of subsequent birth. The distribution was even throughout the period of gestation studied; thus the occurrences of pandiculation ranged from 2 to 20 days pre-partum (Table 3).
Incidence per fetus Since fetal pandiculation is a comparatively rare event, it was usually encountered in each fetus on only one or two occasions. To some extent, the instances of multi-pandiculation in two fetuses were exceptions to this observation (Table 1).
Partial pandiculation Partial (minor) pandiculations were recognised. Some of these were instances of fetal stretching in which vigour was low and duration was brief. In many other partial pandiculations, the complement of components was similar to full pandiculation, but some of these, such as elbow extension and forward thrust of the head, were reduced in expression. Most incomplete or partial pandiculations were lacking in one of the following ways: (a) absence of extension in one forelimb, but present in the other, in association with atlantooccipital extension; (b) symmetrical and full extension of limbs, without atlanto-occipital extension, but with cervical elevation; (c) failure of followthrough in the overall action after its initiation.
Partial, incomplete and minor pandiculation, as one category, outnumbered full pandiculations in incidence. They bore a relationship to the complete form in that they were absent, or deficient, in the five fetuses which failed to show full pandiculation in any observation (Table 1)
Fetal pandiculation, with its full characteristics, having been recognised by observation and induction, must be assumed to be a qualitative item in the behaviour of the fetus. As such, the evaluation of this phenomenon is a chal-lenge from a comparative viewpoint. This fetal activity is not likely to be con-fined to the sheep and compound fetal movements at a high level of complexity,
in association with vigorous bodily extension, can be sought in other subjects as evidence of the same phenomenon. A first indication of such an activity in
P another species has already emerged in a monitored survey of kinesis and pos-ture in the pre-partum bovine fetus, using a combination of methods with Doppler ultrasound and manual palpation (Fraser, 1989b).
The phenomenon has substantial form and structural detail. The components are numerous and the majority of these are evident in each occurrence : of the action, although no single event incorporated every possible component in this study. These items collectively create a vigorous unitary movement of overall extension. In the confined space of the uterus, pandiculation must represent both an achievement and a physical need. In the post-natal circumstances of confmement husbandry of livestock, it may be no less of a need. The holistic nature of the action suggests that close physical restraint could inhibit the performance of the whole gesture and, indeed, this impression was gained in some observations. This phenomenon raises a fresh question of welfare, should the accommodation of an animal not be adequate in all dimensions to permit the occurrence of this behavioural expression.
Viewed in a wider, comparative light, fetal pandiculation could be related to fetal somnolent states which are now, themselves, well recognised (Ruckebusch, 1972; Dawes, 1973; Ruckebusch et al., 1977; Ioffe et al., 1980; Nathanielsz et al., 1980). If pandiculation is a marker of transitional phases between alert and somnolent states, it can indicate that the phasic nature of fetal processes is progressing normally. It may therefore indicate that the status of the fetus is satisfactory in general respects. Assessment of such fetal status is an objective in obstetrics (Trudinger et al., 1979; Manning et al., 1981). As such, fetal pandiculation could be regarded as a simple signal of adequate well-being which could be identified by non-invasive methods. Its detection in the human, if it occurs there also, would appear to be feasible by maternal information. Whether all of these possibilities prove of value in due course or not, it is clear that the phenomenon calls for appropriate attention in further work.
One interesting minor component of the action is fetal yawning and this has not been previously identified. Even in the adult human subject, yawning has received little research. Proctor (1964) describes the physiological minutiae of human yawning and adds the interesting comment that it is more likely to be associated with vigorous stretching after sleep than before. In addition, Proctor adds the observation that such activity is notable in the newborn baby.
Some current research speculation into aspects of fetal respiration, has been focused on so-called "gasping" (Harding, 1984). Since errors can be made in the interpretation of ultrasonic imaging of fetal actions (Farman et al., 1975) it is here suggested that such isolated gasping may be yawning, which has a gasp-like nature. The pursuit of further research into fetal pandiculation might resolve other puzzles in fetology. For such a purpose, it seems that the sheep fetus continues to be a suitable model.
Pandiculated poses were often seen to follow full and partial forms. These poses themselves were variously full or partial in pandiculated degree. Partially pandiculated poses usually differed from fully pandiculated ones in regard to limb involvement. In partially pandiculated poses, the upper or proximal joints of both pairs of limbs, namely shoulder, elbow, hip and stifle, may be flexed. Extension of the axial skeleton and pedal extremities, resulting in outward thrusting of the head and feet, however, are persistent features in most instances of pandiculated pose. The stretching phenomenon may exhibit such outreaching of the head alone. This central feature of axial extension seems appropriate in an action pattern which is a vertebrate phenomenon. Articular exercise in general, and vertebral exercise in particular, may be its mechanical value.
One partially pandiculated pose was the birth posture, which exhibited extreme extension of the atianto-occipital joint, carpal joints and digits. This was associated with very positive and fixed orientation towards the maternal pelvis in the last day of gestation. To this extent, it might be said that pandiculation primes the fetus for birth. However, the wide gestational distributions of episodes show that this is not the sole function of this action. It is likely that the function of fetal pandiculation is compound. Its associated sensory component of mild pleasure is known in the human experience. This cannot be overlooked as a possible reinforcing factor in this, part voluntary and part involuntary, action. The existence of this sensation in the animal, the fetus, can only be surmised on the basis that many of the evident components of the phenomenon are remarkable in their similarity across vertebrate species. Broad physiological benefits in the post-natal situation are recognised in human yawning and stretching.
Among its roles, fetal pandiculation has a major function as a system of orthopaedic behaviour. It operates in the determination of function in the moving parts of the musculo-skeletal system. This is performed through regional exercise. In many cases, several regions are involved and, not uncommonly, the entire body is affected in pandiculation. Articular development and maintenance are clear outcomes of such systematic function, which also enables the fetus to pose for birth.