The Effects of Ions and Drugs on Neuromuscular Preparations of Sea Anemones: I. On Preparations of the Column of Calliactis and Metridium

Following Pantin’s (1935) demonstration of facilitation in the quick closing response of the sea anemone, Calliactis parasitica, the effects of ions and drugs on the facilitated response in this animal, and in Metridium senile, were studied in detail (Ross & Pantin, 1940; Ross, 1945, 1952). There was evidence that some ions had specific effects on facilitation at the neuromuscular junctions but, as the work was done on whole animals, some of the effects observed might have arisen at other places. Moreover, many drugs had no effect; it was possible that they failed to reach the neuromuscular junctions.

Techniques have now been developed for the study of isolated preparations of anemones (Batham & Pantin, 1954; Ross, 1957a), which offer more favourable material for investigating the effects of such treatments at definite sites of action and enable substances to reach the neuromuscular junctions more easily. In these preparations effects on the muscles which give quick responses can be distinguished from effects on muscles involved in slow movements. The earlier work gave almost no information about the action of ions and drugs on slow activities and responses, which have many interesting features of their own. For these reasons, it was desirable to extend the study of ion and drug effects to isolated preparations of Calliactis and Metridium.

A preliminary report has been published on the effects of tryptamine and 5-hydroxytryptamine on both the quick and slow types of preparations (Ross, 1957 b). Of the present papers, the first deals with preparations giving only slow responses to stimuli and showing only slow spontaneous activities, viz. preparations taken from the columns of these animals. The second will deal with work on preparations which give both quick and slow responses to stimuli, and in which inherent activity is weak or non-existent, viz. the marginal sphincters of Calliactis and Metridium (Ross, 1960).

Preparations were made by cutting rings of Calliactis and Metridium after several hours anaesthesia in a mixture of equal parts sea water and MgCl₂ (0.4 M) (Batham & Pantin, 1954; Ross, 1957a). After about 2 days’ recovery in sea water the rings were set up in a muscle bath for recording on a slowly rotating kymograph. Using a Perspex electrode-hook (Ross, 1957a), responses to stimuli could be recorded as well as the normal activity of the preparation. Stimuli came from a condenser discharged at appropriate frequencies (intervals between shocks ranging from 1 to 10 sec.) and reached the preparation through platinum tips on the electrode-hook.

For some experiments ring preparations with a strip of column attached were used to differentiate between the effects of treatments applied directly on or near the contractile tissue, and effects on the conducting system and on receptors at some distance from the muscle. This was done in a shallow bath with a paraffin wax inset across which ran a slightly raised barrier. There was a difference in level of about 5 mm. in the wax on either side of the barrier. On the low side, the strip of column was pinned down and immersed in fluid. On the upper side beyond the barrier, the preparation was fixed where the loop joined the strip. The other end of the loop, attached to the writing lever of the kymograph, was suspended in air and irrigated continuously by fluid from an overhead supply. This fluid ran away on the far side of the wax inset without meeting or mixing with the fluid immersing the strip of column, which could also be changed independently. A short piece of column at the barrier was unavoidably out of contact with either fluid and had to be kept moist by flooding occasionally with sea water. Although this arrangement was not ideal, it enabled the contractile and non-contractile parts of the strip preparations to be tested independently as desired.

Isolated ring preparations of the column of Metridium or Calliactis show intermittent slow contractions and relaxations. Batham & Pantin (1954) described this activity in Metridium, and also the responses to electrical stimuli in preparations with longitudinal strips of column attached. Needler & Ross (1958) showed that column preparations from Calliactis have similar properties. While this work has been in progress, Ewer (1960) has been investigating the activities of preparations from the column of Calliactis and has shown up differences in activity and responses at different levels of the column and in preparations of different ages after the operations.

Compared with most smooth muscle preparations from isolated organs, these isolated rings of the column show a good deal of variety in their activities. The following is a general description of the mechanical record, examples of which are seen in the figures accompanying this paper. A preparation attached to a lever under load soon takes up a stable basic length. Periodically, and quite suddenly, a contraction begins which, under light load, may close the ring completely. Equally suddenly, without the maintenance of tension, relaxation sets in and brings the preparation back to the basic length. The rate of relaxation decreases as the basic length is approached, and except for the occasional occurrence of small contractions this basic tonus is maintained for some time before the next powerful contraction begins. Over long periods, in the absence of stimulation, the record shows an irregular rhythm of big movements every 10-20 min., each one lasting 2-3 min., alternating with much longer periods of relaxation interrupted by a highly variable number of small movements.

Electrical stimuli, delivered when the preparation is extended, evoke responses which reproduce many features of the spontaneous movements. Ten to twenty stimuli separated by intervals of between 1 and 10 sec. will usually elicit a maximal contraction. Fewer stimuli produce submaximal contractions and as a rule three or four stimuli are required to cause any shortening at all.

An outstanding feature of these responses to stimuli, as Batham & Pantin (1954) pointed out, is their long latency, usually not less than 30 sec. and often as long as 2 min. or more. In these column preparations the latency of the response varies with frequency and number of stimuli, as does the slow response of Calliactis (Ross, 1957a). Table 1 gives latent periods of responses to five stimuli at four frequencies in eight different preparations. At 1 stimulus per sec. the mean latencies of responses to 2, 3, 4 and 5 stimuli were respectively: 32, 28, 27 and 21 sec.

The latency of the response of a given preparation to standard groups of stimuli is consistent over periods of time up to many hours. However, when preparations are studied over several days (this is possible as good preparations may survive for more than 10 days) changes in excitability and latency occur. Ewer (1960) has shown that the latent period usually gets shorter with the passage of time. He attributes this to a decline of an inhibitory condition which is particularly evident in freshly prepared rings from the pedal edge or lower part of the column. Rings including the pedal edge have not been used in this work and most of the experiments have been done on rings from the submarginal and mid-column levels, where changes in excitability are less conspicuous.

These preparations are also sensitive to stretch, to reduction of load and to general tactile stimulation ; all of these cause vigorous contractions. The latency of responses to strong tactile stimuli or reduced load is of the same order of magnitude as that of responses to electrical stimuli. The contractions elicited by stretching the preparations, however, have much shorter latent periods, as the following figures show. In five preparations, each tested five times, the mean latent periods of responses were: to ten electrical stimuli (1.1 sec. between stimuli)-36 sec.; to strong tactile stimulation—30 sec.; to unloading—30 sec.; to stretch—10 sec. This difference may be important in determining the relation between responses to stimuli and normal inherent activity.

Given preparations which are so variable in their inherent activity and their responses, the effects of treatments can only be determined for each preparation after a long period of preliminary recording and testing of responses. Data on the following are considered to be significant in estimating the effects of any treatment applied: (1) ‘direct’ contractions (see below); (2) the tonus, or basic length, of the preparation when fully relaxed between contractions; (3) the frequency and magnitude of the contractions in the inherent activity; (4) the size, and (5) the latency of the response to a given number of electrical stimuli at a given frequency.

The effects of altering the ionic environment of the column preparations were studied in two ways: (1) adding measured amounts of isotonic solutions of KC1, CaCl₂ and MgCl₂ to natural sea water; (2) treating preparations with artificial sea waters lacking various constituent ions using the data compiled by Pantin (1946). The results are summarized in Tables 2 and 3, which deal with single and multiple ion effects, respectively.

In these tables, and in Tables 4-6 dealing with drug effects, an attempt has been made to express the results quantitatively, even though the standards for such an expression are very inexact and arbitrary. The ‘o’, ‘+ ‘and ‘— ‘ratings used in the tables are based on the following scheme. A treatment was presumed to have no effect if no change of any kind was observed in the course of four preliminary tests. If any effects, or possible effects, were observed in the preliminary tests, four more tests were carried out, or more if it were necessary to decide whether an effect occurred or not. Most of the results were based, therefore, on eight tests. Treatments were usually given when the preparation was relaxing after one of its normal spontaneous contractions, when another movement was not to be expected for perhaps at least another 10 min. A contraction was presumed to be a ‘direct contraction’ in response to the treatment if it occurred within 5 min. of the introduction of the solution or the drug. If such contractions were observed in only 2-3 of the tests, it was rated ‘+ ‘; if in 4-5 of the tests, it was rated ‘+ + ’, and if in 6 or more of the tests, it was rated ‘ + + + ’. Similarly, a rise of tonus amounting to 25-50 % of the full contraction was rated ‘+ ’, a rise of 50-75 % was rated ‘+ + ’ and a rise of 75 % or more was rated ‘ + + + ’. The same arbitrary gradings, 25-50 %> 50-75 %> and 75 % ^and over were used to provide ‘+ ’, ‘+ + ’ and ‘+ + + ’ratings for comparing spontaneous activity and responses to stimuli before and during treatments. Where negative effects were evident, the same grades were used to provide ‘— ’, ‘–’ and ‘–––’ratings for depressant treatment.

Exposing preparations to excess K+ affects both tonus and activity. The change in tonus takes the form of an apparent direct contraction, or series of contractions, which usually occur within 1-2 min. With such an early response it is not difficult to distinguish between direct contractions and the spontaneous contractions which can occur at any time, although they are usually at least 10 min. apart.

Fig.1 a shows the effect of increasing K+ · 8 on a ring of the column of Metridium. The direct contraction is always followed by a maintained rise in tonus and an increased frequency of beat at the new level. All these effects are more pronounced the greater the excess of K+. Effects on Calliactis are similar. This type of activity continues for at least an hour (longer exposures were not tested) without showing any depressant effects (cf. Mayer, 1906). Recovery from excess K+ usually takes the form seen in Fig. 1 a, viz. steady relaxation without spontaneous activity.

The size and latency of the responses to stimuli after exposure to excess K⁺ cannot be compared with those of the controls. This is because there is so much more movement and the preparations are almost fully contracted at the new level of tonus. When K+ ions are only slightly increased ( · 2) and the direct effects are small, no significant changes in the size and latency of the responses are apparent.

The effects of K+-free sea water are difficult to assess. Immediate effects are slight, but after about $12$ hr. activity begins to decline or becomes erratic. Effects on tone are not consistent and the diminished or erratic activity may be associated with a moderate rise in tonus, as in Fig. 1,b, or with a relaxed condition. Even after prolonged exposure to K+-free sea water preparations still respond to stimuli. On return to natural sea water normal activity begins almost at once (Fig. 1 b). Potassium is, therefore, not essential for the responsiveness of the column musculature, though in its absence normal activity cannot continue indefinitely.

The effects of excess Ca²⁺ are problematical. In at least half my experiments, increasing Ca²⁺ up to · 8 had no detectable effects. Possible direct contractions are almost non-existent, but in some cases increased activity and incomplete relaxation produced a gentle rise in tone (Fig. 2,a). In some cases, also, responses to stimuli are enhanced (Fig. 2 b). Although the actions shown in the illustrated examples might rank as major effects, the large proportion of negative results gives excess Ca^a+ the overall low rating ascribed to it in Table i, since inconsistency of action may be regarded as a sign of low activity.

In the absence of Ca²⁺ the activity and the responsiveness of the column preparations soon disappear. Fig. 2 c shows an example of this effect. During this treatment the preparations relax steadily; they do not remain inactive while holding the original basic length. Responses to stimuli in Ca²+-free sea water are at first much smaller and have a longer latency while, after about 10 min., the preparations very rapidly as in a partial tetanus. Fig. 3b shows this phenomenon with contractions occurring at a frequency of about 1 per 1-2 min. Recovery takes the form of a rapid return to the original basic length and renewal of the original slower level of activity. As with excess K+, it was impossible to carry out any tests on responses to stimuli in Mg²⁺ free sea water because of the greatly increased tonus and intense activity of the preparations.

Replacing the sodium chloride of sea water by sucrose has a prompt depressant effect on the activity of the preparation but leaves the responses to electrical stimulation virtually unimpaired (Fig. 4a) at least for exposures up to 30 min. If the sodium chloride is replaced by choline chloride, in order to maintain Cl ions at their normal concentration, the result is different. An early direct contraction followed by enhanced activity at a raised level of tonus is the usual effect of the treatment (Fig. 4b). Large doses of choline alone (4X 10^-3) also cause direct contractions but do not raise tonus or effect activity to any extent. The enhanced activity when NaCl is replaced by choline chloride shows that absence of Na alone does not seriously impair the working of the column musculature; it would seem that the depression seen in Fig. 4a is due to Cl^- lack and not to the absence of Na+.

Column preparations of Metridium and Calliactis soon shorten when exposed to sea water of pH 5 or 4, but after these initial contractions, activity continues at the shorter length without any marked increase in frequency. Recovery was generally slow. Alkaline sea water at pH 10 had similar effects.

Artificial sea waters lacking sulphate and bicarbonate ions were also tested but no significant effects on tonus or activity were observed.

In review, the pattern of single ion effects is as follows. With K⁺ and Ca²⁺, excess and deficiency do not have simple opposing effects. Excess K+ is highly excitatory but K⁺-free sea water is only moderately inhibitory in its effects. Excess Ca²⁺ is only slightly excitatory whereas Ca²+-free sea water is completely inhibitory. Only Mg²⁺ shows simple opposing effects with excess and deficiency, the former being highly inhibitory and the latter causing a spectacular hyperactivity. The ionic requirements for normal working seem to be that K⁺ should not exceed, that Ca²⁺ should not fall below and that Mg²⁺ should neither exceed nor fall below, their concentrations in sea water. Obviously Mg²⁺ is particularly important in providing the ionic conditions for normal excitation and neuromuscular activity.

The results of tests on the effects of excess or lack of two or more cations simultaneously are set out in Table 3 and illustrated in Fig. 5. The effects of such treatments were never more profound than those produced by excess or lack of particular single ions. The special significance of Mg²⁺ is indicated by the fact that the effects of Mg²⁺-excess and Mg²⁺-lack cannot be offset by simultaneously raising or lowering Ca²⁺ or K+ concentrations. The hyperactivity associated with Mg²⁺-free sea water is only abolished when both Ca²⁺ and K+ are removed, i.e. in NaCl only. The effect of NaCl only, i.e. without Ca²⁺, K⁺, Mg²⁺, SO^3-, HCO₃^-, Br⁻, is interesting. After an initial direct response, tone is usually maintained at a higher level with activity continuing for about 15-30 min. and then gradually coming to a standstill (Fig. 5d). Recovery afterwards is rapid compared to that following excess Mg²⁺ or Ca²⁺-free sea water. The NaCl effect as a whole is slower and less specific than the single ion effects described above.

It is noteworthy that Ca²⁺-free sea water only stops all activity when Mg ions are present; sea water lacking both Ca and Mg acts exactly like Mg²⁺-free sea water alone (Fig. 5,a). Partial restoration of Mg²⁺ in such circumstances produces the inhibition associated with Ca²+-free conditions (Fig. 5 b). This suggests that the Ca²⁺-free inhibition is due to the great excess of Mg²⁺ over Ca²⁺, and perhaps only another form of the inhibitory effect of Mg²⁺ excess.

It appears that K+, Ca²⁺ and Mg²⁺ have specific functional roles in maintaining the balanced conditions for the sequence of contractions and relaxations that make up the normal activity of these muscles. These roles might be summarized as follows : K⁺ is excitatory but not essential for contraction ; Ca²⁺ is not excitatory or inhibitory but is essential for contraction in the presence of Mg²⁺; Mg²⁺ is essential for relaxation in the presence of K⁺ and Ca²⁺.

The results with ions also show the limits of performance of the column musculature. Direct effects, as with Mg²⁺-free sea water and excess K⁺, seldom appear in less than 2 min. and then shortening is slow and usually stepwise (Figs. 1a, 3b), producing a state of high tonus reached over several minutes. In the most enhanced activity observed, contractions occur about once every 2 min. Responses to stimuli are less sensitive to the action of ions than might have been expected. Inhibitory treatments depress and delay responses, but their size and latency are not affected much by excitatory treatments like excess K⁺ and reduced Mg²⁺, though perhaps excess Ca^a+ can double the size and speed of the response (Fig. 2b). It is clear that in the column the excitable and contractile system is basically slow and cannot be made to function much more quickly than it does under normal conditions.

The drugs which were tested on the column preparations were mostly amines of general biological occurrence and widespread activity. They will be treated in four groups:

• (1) acetylcholine and drugs acting at cholinergic junctions;

• (2) sympathomimetic amines and associated substances;

• (3) indolealkylamines and their associates;

• (4) histamine, and some miscellaneous substances.

A standard pattern of test was employed in each case. The drug was introduced into the bath in concentrations of 1 × 10^-5 and 1 × 10^-4 as trial doses. Separate experiments were usually carried out to study the effect of a treatment on the inherent activity without stimulation and on the responses to electric shocks. If a substance were totally ineffective in four tests at these concentrations, no further tests were carried out. If, however, there were signs of even slight activity, further tests were conducted at these and other concentrations. Some substances giving negative results in the first tests were also given additional trials because of their importance and activity on nerve and muscle in other animals.

The results of the drug experiments are presented in Tables 4, 5 and 6 and they are expressed quantitatively in the same way as the ion effects above. A glance at these tables shows the following general points: (1) The great majority of drugs and substances employed in these tests had no detectable effects on the column preparations. (2) No drugs acted at dilutions comparable to the minimum effective doses in most other systems; the minimum effective dose in this case was 1 × 10^-8 and most of the effects were observed at concentrations of 1 × 10^-5 or 1 — 10^-4. (3) The responses to stimuli were very little influenced by any of the drugs used in the tests. (4) No substances were found which blocked, or more than slightly diminished, the inherent activity of the preparations. A detailed description of the results with the different classes of drugs follows.

The experiments showed clearly that the movements and responses of the column preparations, like the quick responses of the whole animal (Ross, 1945), are unaffected by concentrations of acetylcholine (chloride—Roche) up to 1 × 10^-4. Introducing eserine (physostigmina or physostigmin sulphas—B.D.H.), at the same time or before acetylcholine, did not alter this negative result. Eserine itself, in very high concentrations (1 × 10^-3), caused a slow rise in tonus but did not alter the frequency or size of the slow movements and had no effect on the responses to stimulation. Atropine (sulphate—B.D.H.) and nicotine (acid phosphate—B.D.H.) at 1 × 10^-4 were ineffective, as were the specific neuromuscular and ganglionic blocking agents, curare (d-tubocurarine chloride—Savory and Moore) and hexaméthonium (tartrate—May and Baker). One other ester or choline was tested, carbachol (carbamyl choline chloride—B.D.H.) at 1 × 10^-4 but this, like acetylcholine, had no effect. Tétraméthylammonium (iodide—B.D.H.), known to be present in extracts of Actinia equina (Ackermann, Holtz & Reinwein, 1923) though not identified in extracts of Calliactis and Metridium (Mathias, Ross & Schachter, 1960), was also ineffective.

Thus, there is no suggestion that this class of substance is important in slow neuromuscular activity in Calliactis or Metridium. Clearly, the sea anemones, and perhaps the whole Phylum Coelenterata, do not employ acetylcholine as a neuromuscular transmitter or neurohumour (Bacq, 1947).

Unlike the earlier tests with adrenaline on the quick response of whole Calliactis and Metridium (Ross, 1945), these experiments on the column preparations showed that adrenaline (chloride—Parke Davies; acid tartrate—B.D.H.) has an action on the column preparations. It causes direct contractions and a state of maintained tonus in which the musculature is shortened almost to its full extent. These effects appear in less extreme form at concentrations as low as 1 × 10⁶ (Fig. 6,a) ; at 1 × 10^-4 they are striking (Fig. 6 c). At higher concentrations, the small contractions in tonus follow each other more frequently than the full contractions in the preliminary record. At concentrations too low to cause direct effects (1 × 10^-6 to 1x 10^-8), the size of the response and its latent period did not differ appreciably from the controls..

Table 5 summarizes the results for this class of substances and shows that the adrenaline effect is unique. Noradrenaline (acid tartrate—Bayer) has no significant effect. Tyramine (hydrochloride—B.D.H.), previously shown to have profound effects on the size of the quick response (Ross, 1945) and now known to have similar effects on excised sphincters (Ross, 1960), is only slightly active on column preparations. Ephedrine (hydrochloride—B.D.H.), nor-sympatol, 1-p-sympatol, 3-hydroxytyramine (gifts of Dr H. Blaschko) and amphetamine (benzedrine), which might have been expected to display adrenaline-like effects, had no action on the tonus or activity of the preparations or their responses to stimulation. The only substances of this group which did affect the preparations in an adrenaline-like way were isoamylamine (B.D.H.), which was moderately active, and phenylethylamine (hydrochloride—Roche) which was slightly so (Fig. 6 d).

Although cocaine (hydrochloride—B.D.H.) enhances the quick facilitated response of the whole animal (Ross, 1945), it has no effects on the tonus or activity of the column preparations or on the response to stimuli. Moreover, it does not enhance the response to adrenaline or make the preparations more sensitive to it. Thus the potentiating action of cocaine in relation to adrenaline and the sympathetic system in the vertebrates is completely lacking here.

The effects of several adrenergic blocking agents were investigated. They included ergotoxine (ethanesulphonate—Burroughs Wellcome). 933 F (piperi-dinomethylbenzodioxane), dibenamine (gift of Dr H. Blaschko) and regitin (methanesulphonate—CIBA). With the exception of 933 F, these substances are only slightly soluble in sea water and the doses could not be exactly determined; they were probably less than 1 × 10^-5. 933 F at 1 × 10^-4 and saturated solutions of the others, however, had no depressant or inhibitory effects on the responses to stimulation, on the activity and tonus of the preparations or on the response of the preparation to adrenaline at 1 × 10^-5.

There is little to add here to the earlier brief paper on the action of tryptamine and 5-hydroxytryptamine (5-HT) on the column preparations of Calliactis and Metridium. Tryptamine (hydrochloride—Roche and B.D.H.) at concentrations of 1 × 10^-4 causes direct contractions and maintains a state of high tonus in the preparations without having much effect on activity (Fig. 7 a). At lower concentrations the contractions are bigger and at higher concentrations, when the muscle is much shorter due to the direct effects, the latent period of the response to stimulation is reduced. 5 HT (serotonin creatinine sulphate—Roche), on the other hand, has little effect on these preparations.

Other derivatives of tryptamine and substances with related effects have been tested on the column preparations (Table 6). 5-Methoxytryptamine (Roche) caused direct contractions and a rise in tonus and was almost as effective as tryptamine at the same concentration (Fig. 7 b). 5-Benzyloxytryptamine (Roche), at an undetermined concentration (only slightly soluble in water) also caused a rise in tonus. It is clear that these preparations are generally sensitive to certain types of indolealkylamine molecules, but it would require a more extensive survey than could be made here to indicate what features the effective substances of this group have in common.

Substances which affect serotonin activity elsewhere have also been tested. Lysergic acid diethylamide (L.S.D.), which antagonizes the response to 5 HT in a number of muscle preparations, and reserpine which releases 5 HT from brain and other tissues (Page, 1958), had no detectable effects on these preparations. Thus, as with adrenaline, no clear parallel exists between the sensitivity of the anemone column musculature to tryptamine and some of its derivatives and the effects on nerve and muscle in other animals. Welsh’s (1957) suggestion that 5-HT is a neurohumour in a number of invertebrates cannot be applied to anemones, although Calliactis contains large amounts in the tissues lining the coelenteron (Mathias, Ross & Schachter, 1957).

The remaining substances whose effects are summarized in Table 6 are a miscellaneous group chosen for their effects on other preparations or because they are known to occur in a wide range of animals, especially amongst marine invertebrates. They include histamine (acid phosphate—Burroughs Wellcome), strychnine (sulphate—B.D.H.), adenosine triphosphate (Light), y-amino-butyric acid (Roche), and guanidine (sulphate—B.D.H.). With the exception of ATP, which had moderate contractile and tonic effects, all these substances were totally ineffective at concentrations of 1 × 10^-4. The failure of these substances emphasizes once again that the actininian neuromuscular system has few pharmacological links with nerve and muscle in other animals.

Discussion of the results will be confined to the three topics: comparative aspects of the ion and drug effects; site of action of effective substances; the bearing of the results on the problem of rhythmical activity. The mechanism of slow responses to stimulation will be discussed together with slow responses of the sphincters in the next paper (Ross, 1960).

The effects of ions on contractile rhythms in different animals show few consistent patterns. Prosser (1950) has summarized the data from experiments on a good many vertebrates and invertebrates without finding any common features distinguishing, for instance, neurogenic from myogenic hearts, or marine from non-marine species. Comparing the anemone results with Prosser’s table and with data on other organs, for example, Cucumaria cloaca (Wells, 1942), Arenicola extrovert (Wells & Ledingham, 1942), shows that some of the effects obtained on the column preparations are not unique. The lively action of Mg²⁺-free sea water on tone and rhythm is found also in Ostraea, Aplysia and Arenicola. The effects of excess and lack of Ca²⁺ are more unusual. Only in the frog heart does Ca²⁺-lack depress the rhythm as it does in the column preparations.

The only close parallel with the anemone results is in another coelenterate, the medusa Cassiopea, studied by Mayer (1906). He described the effects of various artificial sea waters on the rate of pulsation of Cassiopea before and after the removal of the marginal sense organs which initiate the rhythm in the normal medusa. The disk deprived of sense organs gave results that correspond very closely with the effects of similar sea waters on the column preparations. With the whole medusa there were differences in the effects of sea water lacking both Ca²⁺ and Mg²⁺ and sea water lacking both K⁺ and Mg²⁺ (Table 7). Perhaps there is a suggestion here that column preparations correspond with the medusa without its marginal sense organs, i.e. without nervous pacemaker, and that the ion effects are due to actions on the contractile system itself.

Information on the site of action of effective ion and drug treatments was obtained by the technique described above (p. 733) for distinguishing effects on the contractile system from effects on more remote sensory and nervous elements. When preparations consisting of a loop of column attached to a longitudinal strip of column were treated differentially with excess K+, Mg²⁺-free sea water, adrenaline or tryptamine, effects were obtained only when these substances were brought into contact with the contractile loop. No effects occurred when the strip alone was treated. This shows that the effects arise either in the muscle itself or in that part of the nerve net in immediate contact with the muscle.

The drug effects show no obvious parallels with other animals. Although the preparations are sensitive to adrenaline and tryptamine, substances which influence the activity of these two drugs on other muscles have no corresponding effects on the column preparations. Therefore, the adrenaline effect cannot be assumed to have a functional significance linking it with the action of adrenaline on sympathetic effectors. Yet because the few substances that are effective tend to be sympathomimetic in their effects on vertebrates (cf. Ross, 1945) some affinity between the two is indicated. In this connexion, Östlund’s (1954) survey of catechol amines in lower animals is important. In extracts of Metridium prepared by a specific catechol adsorption method he detected a substance on paper chromatograms of which the spray reactions and pharmacological properties were those of a catechol amine. It differed in R_f value from adrenaline, noradrenaline, and dopamine, so he called it ‘catechol-4’.

It is not possible from the results with ions and drugs to throw much light on the problem of the origin and maintenance of the slow periodic activity seen in isolated preparations. Simple enhancing and inhibitory effects on the rhythm were not found. Increased frequency of rhythm, particularly in Mg²⁺-free sea water and with adrenaline (1x 10^-4), was always associated with higher tone, and cessation of activity with relaxation. These observations are consistent either with a myogenic or neurogenic origin of the rhythm. However, there were two situations, K+-free and NaCl-free sea waters, where depression and eventual cessation of the rhythm occurred without the abolition of responses to electrical stimulation. Thus the nerve net was excitable and could conduct, the muscle could contract, and yet there was no periodic activity. This is more in keeping with an external nervous pacemaker than with a rhythm built into the contractile mechanism itself. Ewer’s (1960) approach to this problem, based on carefully designed stimulation experiments on column preparations, has provided more reliable information on the factors controlling the rhythm. Any further work with ions and drugs must take into account his demonstration of the interplay of excitation and inhibition in the rhythmical activities and the responses of these animals.

Some of this work was done at the Marine Biological Association’s Laboratory at Plymouth and at the Stazione Zoologica, Naples. My thanks are due to the Directors of these institutions and their staffs for their assistance and kindness on many occasions. The visit to Naples was made possible by a grant of £200 from the Browne Fund of the Royal Society which I gratefully acknowledge. I also thank Dr B. Heath-Brown of Messrs Roche Products, Ltd., Welwyn Garden City, for the gift of several derivatives of tryptamine, Dr Hugo Bein, of CIBA, Basel, for a sample of regitin, and Dr H. Blaschko, of the Pharmacology Department, Oxford, for samples of a number of sympathomimetic drugs and adrenergic blocking agents not commercially available, and the last named also for his good advice and interest on a number of occasions. Mr L. Sutton provided invaluable technical assistance during the later phases of the work. Prof. D. W. Ewer and Dr H. O. Schild read the manuscript and made many helpful criticisms and suggestions for which I am most grateful.