Inherent Activity in the Sea-Anemone, Metridium Senile (L.)

Similar direct responses to stimuli are to be found in the feeding reactions of Anemonia sulcata (Pantin & Pantin, 1943), and in both the feeding and the protective reactions of Metridium senile (Hall & Pantin, 1937). Parker (1919) noted the apparent absence of spontaneous and of rhythmic activity in M. marginatum. Direct observation would at first suggest this to be true of M. senile. But careful and prolonged observation by various methods reveals that the animal is in a state of continual and varied activity, which is, however, too slow to be easily perceived directly by the eye (Batham & Pantin, 1950). The investigations which we will now describe show that a large part of the behaviour of these animals is based upon activity which is not a direct response to external stimuli. It is inherent in the apparently unstimulated animal.

The character of the activity varies from time to time, the animal passing through different ‘phases’ of activity, which may endure for a considerable period. Thus the activity of an animal in a phase of elongation may differ greatly from one in a phase of general contraction. Hence there are two sets of phenomena to be investigated: the continual activity itself, and the variation of this activity with the phase of the animal.

The observed changes in shape which constitute the activity of the animal are brought about by muscular action. In a system such as we have here, the relation of muscular action to change of form is not as simple as it is in the body of an animal with a hard skeleton. The mechanics of muscular action in Metridium are discussed by us in a previous paper (Batham & Pantin, 1950). We will merely point out here that the main body of the anemone, the column, consists of a plastically extensible cylinder whose walls carry a sheet of circular muscle and a system of longitudinal parietal muscles (see Fig. 1 of the paper quoted). The ends of the cylinder are closed by the oral disk and the attached foot. Between these run longitudinally the numerous mesenteries, the majority of which bear longitudinal retractor muscles. These various muscular sheets act upon the fluid contained in the gastral cavity; contraction of one layer thus tending to enforce distension of the others. The pressure developed in the gastral cavity is extremely low (1-6 mm. of water), and normally muscular action produces the maximum spatial displacement with the minimal performance of external work.

Three complementary methods of observation have been used : (i) prolonged direct observation; (ii) cinematography, particularly at slow speeds, about one exposure every sec. (‘time-lapse’ cinematography); (iii) smoked drum records. Each method has special advantages. Direct observation leaves the animal less disturbed than with any other method. It is, however, difficult to obtain quantitative data in this way and continual observation over periods of hours or even days is exacting to the judgement of the observer. We have found that the value of such observation is greatly increased by viewing the anemone against a grid of squares behind it, and by making freehand sketches of the animal every half minute or minute over long periods. Cinematography not only gives permanent records, but when a film is viewed the apparent activity can be accelerated so that imperceptibly slow movements can be easily interpreted. When the apparent speed of events is increased in this way about 6o-fold, Metridium seems to be a very active animal ; and it is remarkable that the speeded movements of the animal as seen on the screen have a decidedly purposive character.

Smoked drum records are invaluable for obtaining quantitative measurement of the time relations and extent of movement. For the production of such records it is necessary to remember that the body executes contractions of great size, but that the tensions exerted by the muscles are normally extremely low. As we have said, the pressure in the coelenteron is only to be measured in millimetres of water. Consequently recording levers must be both free to move and so lightly balanced that they cause minimal deformity of the weak body. Happily, the movements are also slow, and they are therefore easily recorded by long isotonic levers writing lightly on rather thinly smoked paper. In all the figures the records read from left to right, and unless otherwise stated contractions are registered downwards.

With an animal capable of such great change of shape as Metridium, drum records must be continuously accompanied by direct observation. Only thus can a recorded movement be surely attributed to a particular system of muscles and not to the indirect consequence of some other deformation.

The attachment of a recording lever to an animal even of the simplicity of an actinian at once raises the question of how far the activity is ‘normal ‘and how far this treatment is itself a source of stimulation. This kind of difficulty is also present in cinematography of the animals, and even to some extent with direct observation.

For in both these cases light is needed, and the animals show some degree of sensitivity to light. The difficulty is naturally acute when it is necessary to determine the behaviour during darkness. Fortunately, comparison of observations by the different methods, and particularly comparison with extensive direct observations both in aquaria and in the field made over the course of three years, leads to the conclusion that provided care is taken to have fine recording levers on the one hand, or to use constant lights of comparatively low intensity on the other, there is no obvious influence of the method of recording on the essential character of the behaviour.

Throughout this paper it must not be forgotten that we are dealing with a whole animal. We are not employing an isolated ‘physiological preparation’, the activity of which can be easily compared with that of controls ‘at rest’. As in other ‘whole’ animals, the activity may vary even under apparently constant conditions, and a response to a stimulus may include a large and variable component supplied by the animal of itself. Often, therefore, a causal relationship between activity and environmental conditions is difficult to establish or to disprove except after prolonged observation of many instances such as we have made.

Most of the experiments were conducted at the Marine Biological Laboratory, Plymouth, on Metridium collected locally, particularly from Millbay Dock. During experiments they were kept in large vessels (2-10 1.) of clean, aerated sea water. Experiments were performed in an ordinary laboratory at room temperature and also in a rock-hewn cellar virtually free of vibration and under constant conditions of light and very small temperature fluctuation.

We shall show elsewhere (Batham & Pantin, 1951) that the body of Metridium may be divided into three more or less separate systems of muscles. Of these, the muscular system of the foot is rarely active; that of the oral disk commonly shows some activity; while the muscular system of the column is always active. In the present paper we shall particularly consider this last system.

Fig. 1 shows the simultaneous activity of various parts of an unstimulated anemone, recorded isotonically. The lowest tracing (C) records the activity of the mesenteries by means of a thread attached to the lip of the mouth. The two tracings A and B above this record the vertical activity of the parietal musculature by threads from two opposite sides of the body wall. In A, B and C a contraction is registered downwards. Above this is recorded the activity at the marginal sphincter by a lateral thread (Y), the movement of which was transmitted to the recording point by a right-angled lever.

The uppermost tracing (X) records contraction of the circular muscle at a point about half way up the column. In this last case the tracing was recorded from the movement of a pivoted horizontal arm of straw resting lightly against the column of the anemone. This method was employed here because it was found that threads inserted into the column occasionally cause a curious inhibition of the circular muscle at that level so that a ring-like ‘shelf’ of weaker tone bulges out all round the column. This ‘shelf’ seems to act as a barrier to the conduction of peristaltic waves up or down the column. The records begin with a retractor contraction following a brief stimulus, indicated on the time scale.

It will be seen at once from Fig. 1 that all parts of the column system are in continual activity. It will also be seen from the time scale that the activity is extremely slow. This applies to all the muscle systems including those of the mesenteries; though the latter can also respond with much greater rapidity to strong stimulation which initiates the retraction response. It will be shown in a later paper that there is some evidence that the mesenteric retractors, and perhaps also the sphincter, can contract at two different rates; though whether this is due to the possession of two distinct kinds of muscle fibre in the muscles is unknown.

The pattern of activity varies in different animals, but that recorded in Fig. 1 is quite commonly to be seen whether or not the animal has levers attached to it. The major events are more or less periodic contractions of the parietal body wall. Each of these is then followed by an elongation of the column. This is not a mere passive relaxation of the parietals but an active extension due to contraction of the circular muscle layer (Batham & Pantin, 1950). The parietal contraction, which takes about i min. to reach its maximum, is followed by a slow contraction of the circular muscle immediately below the marginal sphincter. From the sphincter, a wave of contraction of the circular muscle now spreads down peristaltically, narrowing the column and causing it again to elongate. The successive elements of this sequence of contractions vary greatly in their intensity and their duration in different cases ; and the activity may be complicated by the occurrence of more than one parietal or circular contraction in succession. Indeed, there is generally a background of irregular, though less intense, activity.

A characteristic pattern of events is shown in Fig. 2, recorded on a faster drum than that used for the record shown in Fig. 1. Contractions are registered downwards. The record shows the activity of a specimen displaying slight parietal contractions and vigorous activity of its circular musculature. It will be seen that each parietal contraction (a, b, c, d and e) is followed about 0·5-1 min. later by a constriction of the sphincter region. This initiates a peristaltic wave which, in passing down the column, records itself at X between 1 and 2 min. later. However, such a sequence is by no means invariable. For instance, at O a spontaneous contraction of the circular muscle arose locally, without any preceding contraction of the parietal or sphincter.

These features of naturally occurring activity in the unstimulated animal are also to be seen when a response is elicited by electrical stimulation. We have shown (Batham & Pantin, 1950) that a battery of stimuli at low frequency (1 in 3 to 1 in 10 sec.) causes a contraction of the parietal musculature, which is followed by one or more waves of peristaltic circular muscle contractions over a period of several minutes.

One of the most striking features of the column activity is its frequent semi-rhythmic character. Fig. 1 shows the fairly regular occurrence of major parietal contractions at intervals round about 10 min. Approximate rhythms of about this frequency are often to be seen. Their low frequency is remarkable.

But the rhythm is sometimes entirely absent. Fig. 3 shows essentially a-rhythmic activity of three parts of the parietal musculature recorded from points round the upper edge of the body wall. Though a-rhythmic, the activity illustrated in Fig. 3 shares one important character in common with the rhythmic activity seen in Fig. 1 : the activity of each part shows evident co-ordination with the rest. Such coordination is by no means complete. The parietal contractions recorded from the two points of the body wall in Fig. 1 do not all correspond. Rather more are to be seen originating on the side recorded by tracing A than on that recorded by tracing B. Lack of co-ordination is almost always evident in the smaller contractions, particularly of the circular muscle layer. At times, large unco-ordinated contractions very obviously occur. The asymmetric activity of the parietal musculature recorded in Fig. 4 caused the animal to bend right over to one side.

Co-ordination is greater between different parts of the parietal system than it is between the parietals and the mesenteric retractors. The middle region of the column is apt to show the greatest departure from co-ordination (Fig. 5). The co-ordination between different parts of the parietal body wall may be surprisingly complete.

In experiments of this kind it is important to guard by direct observation against the possibility that co-ordination is simply of mechanical origin due to the drag of one part of the body upon another. The most direct proof that this is not the case is found in the co-ordinated activity of strips of body wall partly severed from connexion with the rest of the animal. These experiments will be discussed in a later paper. We may note, however, in Fig. 5 that the co-ordination of different points of the parietal wall is greater than that between the parietal wall and the mid-point of the column on the same radius; or between the wall and the mouth region which lies between them. Further, comparison of the activity of the two parietal regions of Fig. i shows that by no means all the contractions of one region are accompanied by a response of the other.

One feature of co-ordination is of especial interest: its extreme slowness. We have seen that contractile activity initiated in the parietals may take minutes before it reaches the circular muscle of the column as a peristaltic wave. But even the passage of the contractile wave round the parietal system is slow. The onset of contractions does not begin simultaneously at different parts of the body wall. In some parts they may not begin till more than half a minute after the ‘leading’ part has begun to contract. This may result in a temporary asymmetry of the body, and is another illustration of the independence between co-ordination and mechanical factors.

The following table shows the succession of the natural contractions of two parts of the body wall (Table 1, A and C), of the mesenteries at the mouth region (B), and of the sphincter (D). They were recorded in an anemone in a similar condition to that of Fig. 1, and with similarly arranged recording threads. The part which ‘led ‘the contraction is recorded with the value o, whilst the other figures show the number of seconds’ delay before each part began to contract.

These delays are far longer than can be accounted for by conduction of excitation in the nerve-net in the manner described by Pantin (1935) in Calliactis. Using the technique employed on that occasion, preliminary experiments were made by us on the conduction time of the through-conduction system of the nerve-net in Metridium. They were made by recording the latent period of contraction of the powerful retractors on opposite sides of the animal following stimulation by condenser shocks at a frequency of 1 per second applied to the base of the column. Values for the conduction time across animals some 5 cm. in diameter varied between 50 and 80 msec. Such values are larger than for the through-conduction system of Calliactis, but they are trivial in comparison with delays of the order of 15 sec. or more.

There is other evidence that these long delays are not simply due to slow progressive conduction across the nerve-net. If four or more threads are attached in order round the edge of the column, the order of the recorded responses of each sector of the column does not follow the order of their position. The part diametrically opposite the leader may often begin its response before the intermediate regions. It seems that a large part of these very long delays is of local origin and is not due to slow progressive conduction across the whole animal.

Table 1 shows that different parts of the anemone may act as ‘leader’ on different occasions. The observations illustrated here are rather exceptional in that ‘leadership ‘more usually remains with one particular sector of the animal for long periods without change. It should be said in passing that direct observation shows that the ‘leading’ section of the column is as often located between two parts from which records are taken as not. There is no reason to associate initiation of activity with the trauma of an inserted recording thread.

Sudden illumination or mechanical shocks may cause a response in an anemone. It is therefore necessary to consider whether the observed normal activity is really the simple consequence of random external stimulation. This does not seem to be so.

In the first place, animals placed side by side in the same aquarium show no correlation in the occurrence of their movements, except after evident common stimulation such as a large change in light intensity or the addition of food solution. In the second, removal so far as possible of external stimuli does not appear to affect activity. Animals were placed under comparatively constant environmental conditions in large bowls of clean sea water in a cool cellar. The bowls stood on shock-absorbing material on a firm base. The temperature of the water varied by about 0·5° C. over 24 hr., and readings during a day often differed by less than 0·1° C. Animals were left in darkness or in constant artificial light. The activity continued under these constant conditions, and indeed was not diminished in comparison with animals in aquaria on laboratory benches subject to variations of intensity of daylight or the ordinary mechanical vibrations of the room.

The activity was equally evident with direct observation or with cinematography of intact animals to which no threads were attached. The one form of stimulus which could not be avoided for very prolonged periods was the mechanical agitation associated with aeration of sea water. Fig. 6, however, shows that absence of aeration produces no evident effect upon the character of the activity registered from the parietal body wall.

All these facts indicate that the activity is inherent in the animal and is not the result of successive responses to random external stimuli. This conclusion is fully borne out by the frequently rhythmic character of the activity (Fig. 1), for random stimuli will not recur at more or less regular intervals.

When we term the activity ‘inherent’ we do not imply that it is necessarily due to self-exciting elements, as in a self-exciting nerve or muscle cell or the rhythmically active ganglia which cause the regular excitation of the bell in Scyphomedusae (Bozler, 1926). The activity of Metridium causes changes of shape, and this raises the possibility that movement of the animal may itself excite subsequent movement after the manner of a chain reflex. It might be suggested that the rise in coelenteric pressure set up by a local contraction might automatically excite a stretch receptor or muscle fibre elsewhere in the animal. Parker (1919) has shown in Condylactis that sudden distension by fluid of an isolated tentacle may cause it to contract, and this is also true of the isolated tentacles of Anemonia sulcata. However, the pressures required to produce such contractions are considerable, whilst we have seen in the previous paper (Batham & Pantin, 1950) that active contractions are normally accompanied by extremely small changes of pressure (generally less than 6 mm. of water). Moreover, we found in the experiments recorded in the same paper that great distension of the coelenteron with fluid caused no general contraction. It sometimes caused a specific reflex contraction of the radial muscles of the mesenteries, but even this only occurred at very great distensions and, for the anemone, comparatively large pressures (40 mm. water). For a stretch to act as a stimulus in anemone tissue it must be very large—far larger than that which normally occurs during inherent activity.

It is difficult therefore to ascribe normal activity to a simple sequence of a response which engenders a stretch stimulus which in turn engenders a fresh response, after the manner of a chain reflex. Such an hypothesis would also have to contend with the difficulty of accounting for the fact that major contractions may occur regularly at long intervals of the order of 10 min. and that the intervening period between these contractions is spent in activity of a more irregular kind.

So far as our present observations go, they would be consistent with the supposition that activity arises spontaneously within the tissues. But while activity is evidently no simple chain reflex we cannot as yet certainly conclude that it arises entirely independently of stimuli or changes of excitability generated by the animal’s own movements. It is for this reason that we have preferred to term the activity ‘inherent’. We intend to imply by this simply that the activity is an observed property of the animals which does not arise directly from external stimuli.

From our experiments we may conclude that Metridium shows continual muscular activity in the column. This is so slow as scarcely to be perceptible to the eye as movement, and varies considerably in character and extent in different individuals and at different times. It is this activity which maintains the average state of tone balancing the coelenteric pressure (Batham & Pantin, 1950). The activity may involve a co-ordinated contraction sequence of the reciprocally acting circular and parietal muscles. Different parts of the parietal system may show co-ordinated contraction to a varying extent. The activity appears to be inherent and independent of external stimuli, and of the more obvious internal ones; such as stimulation by stretching of one part through contraction of another. The activity is often more or less rhythmic, giving fairly regular contractions at a frequency of the order of one in 10 min. Such very slow rhythms call to mind the rhythmic outbursts of organs composed of plain muscle such as the mammalian uterus and the proboscis of Arenicola (Wells, 1950). We have no evidence yet of the mechanism of the ‘time keeper’ in such slow activities.

Continual activity is to be found at times in all the muscular systems of Metridium, and not only in the column. Cinematograph records show that the tentacles of an open unstimulated quiescent anemone may exhibit either almost complete immobility or a continual active twitching. This contrast may be observed in two unstimulated anemones side by side in the same aquarium.

While Metridium by great modification of its shape shows activity very strikingly, even an apparently inert anemone like Calliactis parasitica is, in fact, in a state of continual activity. Prolonged records of the movement of the column of Calliactis show its state of continual though slight movement.

It will be shown later that inherent activity plays an essential part in the behaviour of these Actinozoa. It is no irrelevant by-product. In this we see an evident analogy with activity in higher animals possessed of a central nervous system. Thé activity cycles of Polychaete worms investigated by Wells (1950) provide an example.

The one striking contrast between this Actinian inherent activity and that of higher animals possessing a central nervous system is its extremely low frequency. Notwithstanding the similar part which inherent activity may play in both grades of nervous system, we must be careful not to suppose at present that identity of effect implies identity of mechanism ; the systems differ too greatly in the degree of their structure, and in their time scale. The possibility that to achieve the same behavioural requirement, different kinds of process are utilized in the two grades of nervous system cannot as yet be dismissed (Pantin, 1950).

1. The sea-anemone Metridium senile shows continual muscular activity. The activity is so slow that it is rarely appreciated by the eye as movement. Methods of observing and analysing such activity are discussed.

2. The activity of the column of the anemone has been analysed. It consists of a sequence of reciprocal contractions of the parietal muscles and the circular muscle coat. A sequence of activity commonly begins with a contraction of the parietals, followed by contraction of the marginal sphincter, which in turn initiates a peristaltic wave. The whole sequence lasts several minutes. The size and duration of its components may vary greatly. Activity may show a more or less regular rhythm with a period of the order of 10 min. between each major contraction. It may, however, show no trace of rhythm.

3. The activity of different parts of the body wall may show striking co-ordination. A contraction of one part of the parietal musculature is usually followed by contraction of the others. In other cases there may be no trace of co-ordination. The parietal muscles of one side may contract without contraction of those opposite, so that the animal bends over.

4. Co-ordination takes place through one part of the body wall acting as ‘leader’. The other parts of the body wall follow this contraction with long delays (up to 30 sec. or more). The delay is far greater than the through-conduction time in the nerve-net (50-80 msec. in Metridium). There is evidence that it is of local origin. One sector usually maintains leadership for long periods; but from time to time the site of leadership changes.

5. Evidence is given that the activity continues unaltered in the absence of external stimulation. It is inherent. The evidence does not suggest that it is maintained through self-stimulation by preceding contractions after the manner of a chain reflex.

6. The activity varies greatly in character and extent in different animals and in the same animal at different times. This remains true even under apparently constant environmental conditions.

Much of this work was done at the Marine Biological Laboratory, Plymouth, to the Director and staff of which we are most grateful for the many facilities given us. Part of the work was done during the tenure by one of us (E.J.B.) of a Shirtcliffe Fellowship of the University of New Zealand. We wish to thank the Department of Scientific and Industrial Research for a grant for the development of a special research which enabled this work to be concluded.