Of the large number of published studies on the ion relations of invertebrate rhythmic muscles, the great majority concern preparations from Crustacea and molluscs—largely because the animals of these groups are provided with vigorous and experimentally convenient hearts. Data on members of other phyla are relatively few and, from the comparative point of view, their multiplication is evidently desirable. Recently, one of us published a description of the so-called “isolated extrovert” of the lugworm, Arenicola marina L., a preparation consisting of the proboscis and part of the oesophagus (Wells, 1937). Suspended in sea water, the isolated extrovert maintains for many hours a very regular and characteristic activity pattern, consisting of a series of outbursts of vigorous rhythmical contraction separated by periods of relative rest. That this is the normal activity pattern was shown in the paper already mentioned. Because of its vigour, its regularity, its long survival after excision and the ease with which that operation is performed, it is an excellent object for experiments on the action of chemical factors. The writers have undertaken a study of ion actions on this preparation. The results obtained by varying the Mg concentration form the subject of the present communication. Experiments with K and Ca have been begun and it is hoped to complete them shortly.

The dissection of the isolated extrovert was carried out as already described (Wells, 1937, p. 122). Care was always taken to leave the circum-oesophageal nerve ring and supra-oesophageal ganglia out of the preparation, as their presence may introduce irregularities into its behaviour. It is of great importance, when dissecting out an extrovert, to avoid stretching the tissue, especially in the region where oesophagus joins the proboscis, near the insertion of the retractor muscle into the gut wall. The movements were recorded with very light isotonic levers.

The preparations were exposed to Mg concentrations which changed either abruptly (“constant exposure” experiments) or gradually (“drift” experiments) from one value to another, using methods described elsewhere (Wells & Ledingham, 1940).

All solutions applied to the extroverts were made of A.R. reagents, with water from an ordinary block tin condensing tube. We have not attempted to work with any preparations for much longer than 24 hr., and in experiments of this duration further precautions against traces of impurities do not seem to be necessary. Solutions were made up in distilled water to which Na bicarbonate N/400 had been added, and which had been vigorously aerated overnight. The resulting salines were adequately buffered at pH 8·0–8·2, i.e. near the normal pH of sea water.

It is already known that sea water keeps the extrovert in a vigorous and apparently normal condition for a long time (Wells, 1937), and we started with the assumption that a balanced solution closely resembling sea water in composition would be a normal medium. Reliable analyses of Arenicola blood or body fluid are not available. The artificial sea water was made by mixing a number of solutions each approximately isotonic with sea water (McClendon, 1916; Pantin, 1926), so that the proportions of the various ions could readily be varied when desired, without greatly altering the osmotic pressure of the resulting mixture. The proportions finally adopted were as follows :

This mixture supported the rhythm as well as natural sea water, and change from one to the other produced no obvious modification in activity. In making up solutions with “abnormal” Mg concentrations (it should be remembered that the “normality” of the above saline is assumed, simply because the mixture works) the proportion of MgCl2 taken was varied as desired. Since all mixtures were made up to 100 c.c. with NaCl, variation in Mg entailed an opposite variation in Na. In other words, the following factors were held constant throughout the work : K, Ca, sulphate, pH, osmotic pressure. To ensure constancy of the latter factor, increase in Mg was compensated by decrease in Na, and vice versa. The inclusion of sulphate as the sodium, instead of the more usual magnesium, salt obviously facilitates matters when low Mg concentrations are wanted.

To simplify the statement of Mg concentrations, the “normal” saline, whose composition is given in detail above, and which was used as the starting point in our experiments, is referred to in the following pages as “Mg 1”, and other solutions are characterized simply by their Mg content, which is given in multiples of that in the “normal” mixture. Thus “Mg 2”, for example, is a mixture containing twice as much Mg as artificial sea water, and so on. “Mg 1” contains 0.058 M MgCl2.

The essential facts can be summarized in two statements: (1) Mg tends to inhibit the preparation ; (2) the preparation has a considerable power of accommodating itself to a new Mg concentration, so the effects of a sudden change in the amount of Mg in the bathing fluid are greatest at first, then gradually become less. These points are brought out in Figs. 1, 2.

Fig. 1.

Accommodation to a high Mg concentration. First extract begins in Mg 1, and shows a short exposure (lomin.) to Mg 2, and the beginning of a long exposure (4 hr.), also to Mg 2. Second extract shows return to Mg 1 after the long exposure. Third extract shows behaviour of the preparation, still in Mg 1, 2 hr. later. In all records: read from left to right; upstroke of lever means contraction of preparation ; time signal marks minutes ; Mg concentrations given as multiples of that in sea water.

Fig. 1.

Accommodation to a high Mg concentration. First extract begins in Mg 1, and shows a short exposure (lomin.) to Mg 2, and the beginning of a long exposure (4 hr.), also to Mg 2. Second extract shows return to Mg 1 after the long exposure. Third extract shows behaviour of the preparation, still in Mg 1, 2 hr. later. In all records: read from left to right; upstroke of lever means contraction of preparation ; time signal marks minutes ; Mg concentrations given as multiples of that in sea water.

Fig. 2.

Accommodation to a low Mg concentration. First extract begins in Mg 1, and shows a short exposure (10 min.) to Mg 0·5. Second extract shows the beginning of a long exposure (4 hr.), also to Mg 0·5. Third extract shows return to Mg 1 after the long exposure. Fourth extract shows behaviour of preparation, still in Mg 1, 3 hr. later; also a second short exposure to Mg 5,

Fig. 2.

Accommodation to a low Mg concentration. First extract begins in Mg 1, and shows a short exposure (10 min.) to Mg 0·5. Second extract shows the beginning of a long exposure (4 hr.), also to Mg 0·5. Third extract shows return to Mg 1 after the long exposure. Fourth extract shows behaviour of preparation, still in Mg 1, 3 hr. later; also a second short exposure to Mg 5,

Fig. 1 shows that on changing from Mg 1 to Mg 2, the extrovert is at first completely relaxed. In four hours’ exposure to the high concentration, a considerable degree of recovery has occurred (accommodation to high Mg). On returning to Mg 1, the effect depends on the duration of the exposure to Mg 2. After a short exposure, during which the extrovert had no time to adjust itself to the new conditions, the original behaviour pattern is resumed smoothly and directly on returning to Mg 1. But after long exposure, when accommodation has occurred, return to Mg 1 causes hyperactivity, i.e. the typical reaction to low Mg. This gradually passes off (accommodation to Mg 1).

Fig. 2 shows the corresponding experiment with a low Mg concentration. On changing to Mg 0-5, there is an increase in activity, which passes off to a large extent if the extrovert is given time to accommodate itself to the new conditions. On returning to Mg 1, the result is a temporary inhibition—i.e. the typical reaction to high Mg—if, and only if, the exposure to Mg 0-5 was long enough for accommodation to occur.

Accommodation, then, can be detected in two ways. As an extrovert adjusts itself to a new solution its behaviour improves, but at the same time the old solution ceases to be appropriate. The immediate effect of applying any mixture depends on the previous history of the preparation; thus Mg. 1 causes excitement in Fig. 1 and inhibition in Fig. 2.

These accommodation phenomena are sometimes exceedingly striking. With small changes—e.g. Mg 1 to Mg 1·5 or to Mg 0·75—the difference between the immediate response and the final equilibrium condition is not great, but with such changes as Mg 1 to Mg 3, or Mg 1 to Mg 0, the activity may be completely suspended, or very greatly increased, for many hours before accommodation results in comparatively normal behaviour.

About the physiological nature of the accommodation process we know little. The problem is discussed below. It is however clear that accommodation does not depend on special events, such as permeability changes, associated with contraction. In very high Mg concentrations, spontaneous activity never appears, but the accommodation process can nevertheless be shown to take place. Fig. 3 illustrates an experiment in which the extrovert was alternated between Mg 1 and a solution containing no NaCl, the necessary amounts of KC1, CaCl2 and Na2SO4 being made up to 100 with MgClg. This gave a Mg value of 6-2. It will be seen that return to normal causes great temporary excitement after long exposure to high Mg, but not after short. This proves that the accommodation process has taken place, although in the presence of so much Mg it could not lead to resumption of mechanical activity. Incidentally, it is interesting to note that Mg inhibition is promptly and completely reversible, even with such high concentrations. After recording the second half of Fig. 3, we put the preparations—a group of four, of which the record of one is shown—back into Mg 6-2 and left them in it, at room temperature, for 15 hr. After this time they were completely limp and toneless, and we thought them dead. They were however mounted for recording, and, on returning them to Mg 1, there was immediate recovery of tone followed by great hyperactivity slowly settling down into the typical pattern, very much as in the second half of Fig. 3.

Fig. 3.

Accommodation to a concentration of Mg, so high that activity is impossible. First extract begins in Mg 1; and shows a short exposure (10 min.) and the beginning of a long one (4 hr.) to Mg 6·2. Second extract shows return to Mg 1 after the long exposure. Compare Fig. 1.

Fig. 3.

Accommodation to a concentration of Mg, so high that activity is impossible. First extract begins in Mg 1; and shows a short exposure (10 min.) and the beginning of a long one (4 hr.) to Mg 6·2. Second extract shows return to Mg 1 after the long exposure. Compare Fig. 1.

The effects of different Mg concentrations will now be described in fuller detail.

Over the range from Mg o to Mg 1, the chief effects of variation in Mg concentration are as follows. Normally, the extrovert exhibits a regular series of activity outbursts, each consisting of a tone wave upon which a number of vigorous rhythmic strokes are superposed. If the Mg is high, the period of rhythmic strokes tends to be confined to the top of the wave. If the Mg is low, the active period spreads over the whole wave and even into the intervals between the waves. The general tone level is depressed by high, and raised by low, Mg. A big downward change, e.g. from Mg i to Mg o, produces violent, tetanus-like contraction, with minute strokes superposed on it, but even in the total absence of Mg, the contraction slowly passes off and, in a few hours, the typical intermittent behaviour pattern can be seen again (Fig. 4).

Fig. 4.

Two extroverts exposed, at signal, to Mg 0. First extract shows the change and the first 2 hr. of the exposure. Second extract shows behaviour 6 hr. later, still in Mg 0.

Fig. 4.

Two extroverts exposed, at signal, to Mg 0. First extract shows the change and the first 2 hr. of the exposure. Second extract shows behaviour 6 hr. later, still in Mg 0.

More intense Mg inhibition can be seen in extroverts equilibrated with high Mg concentrations (e.g. Mg 2 upwards), or, at lower concentrations, as temporary effects following such sharp upward changes as Mg o to Mg 1. In these conditions, the interval between successive outbursts lengthens, and the amplitude of the contractions may diminish.

It seems probable, from the general nature of these effects, that the action of Mg is primarily on the excitor mechanism. We do not seem however to be dealing with a simple action on one site, for the type of effect produced by raising the Mg varies with the Mg concentration. Confining our attention to extroverts in equilibrium with their bathing media, we find that, over the lower part of the concentration range, increasing Mg restricts the extent of each activity period ; that at higher Mg concentrations, increase lengthens the interval between outbursts ; and that at still higher concentrations, the amplitude of the strokes falls off.

Above a certain limiting concentration, spontaneous activity ceases altogether. The exact determination of this limit is however difficult. Our first experiments on this problem (made with the “constant exposure” technique) established the following points: (1) there is an upper limit, apparently at about Mg 3, above which the normal outburst pattern cannot appear; (2) there are considerable differences in Mg tolerance between different preparations ; (3) under normal conditions, preparations often show a certain amount of “background” activity between the rhythmic outbursts, this background activity is less sensitive to high Mg than the outburst pattern and may be present up to about Mg 5 ; (4) temporary inhibition following upward changes may be very prolonged, especially as the upper limit is approached.

These points were well brought out in an experiment on ten extroverts, which were exposed, first for 4 hr. to Mg 1 and then for 19 hr. to higher Mg concentrations. During the latter period, they behaved as follows.

Three preparations exposed to Mg 2-5 were first inhibited for periods of 2-4 hr., and then comparatively normal outbursts gradually returned. However, the final frequency remained about half what it was in Mg 1, the intervals between outbursts now occupying about 10-15 min.

Four preparations exposed to Mg 3 were inhibited for 7-8 hr., then outbursts reappeared. In two preparations, the outbursts, although weaker than in Mg 1 and occurring at intervals of 20 min. or more, were still fairly vigorous. In the others, the outbursts were exceedingly feeble and occurred at very long intervals.

Three preparations exposed to Mg 3-5 showed complete inhibition for several hours, after which “background” activity gradually returned. No distinct outbursts appeared during the whole of the 19 hr. exposure. One preparation showed a single brief tone wave near the end of this period, which was perhaps an abnormal outburst.

The latter observation raises the question, if the exposure to Mg 3-5 had been longer, would the extroverts have become active again? To attempt to answer this question by further prolonging the experiment would be hardly justifiable. As described above, it already involves continuous recording of the excised preparations for 23 hr. It is not possible to use extroverts for much longer than this, owing to the intervention of weakness, irregularity and other results of the long survival period. If the question is to be answered, it must be in some other way.

We therefore carried out a number of “drift” experiments, in which the Mg concentration changed very gradually, using methods described in detail elsewhere (Wells & Ledingham, 1940). In some cases, the Mg started at the sea water concentration and was made to rise slowly. In others, the extroverts were first accommodated to Mg concentrations well above the limit and then exposed to conditions in which the Mg gradually fell towards the sea water value. The object of these experiments was to avoid temporary effects of quick change. If the rate of drift is slow enough, the tissue will be able to keep itself accommodated all the time. Failure to drift slowly enough will be revealed by a difference between the limits shown by upward and by downward drifts ; for an upward drift, by inhibiting the cells, will yield a low limit, while a downward drift will have the opposite effect.

Fig. 5 gives the Mg time curves of our various drifts. Each curve represents a single experiment made on two or (more usually) three extroverts simultaneously. Upward drifts A, B, C were too rapid; the extroverts were inhibited, more or less sharply, at concentrations varying from Mg 1-5 to 2-5. Upward drifts D, E were satisfactory; at Mg 3, the preparations were still giving quite vigorous outbursts, although at long intervals, but above this value the falling away was marked. At the ends of D and E outbursts still occurred at very long intervals, but they had degenerated into brief tone waves without superposed strokes, and were usually so small as to be nearly lost in the “background” activity of the preparation. The downward drifts (a-d) were apparently a little too fast; the extroverts became active, rather abruptly, in the range Mg 2·9–3·6 In this case, as already stated, a slight raising of the apparent limit is to be expected, since a downward change has an exciting effect.

Fig. 5.

Graphs of the drift experiments, to show how the Mg concentration varied with time.

Fig. 5.

Graphs of the drift experiments, to show how the Mg concentration varied with time.

Taken together, the results of the drifts and those of the constant exposures described above point to the existence of an upper Mg limit, situated, for most preparations, a little above Mg 3. The amount of variation between preparations is however considerable, and it would be unprofitable to attempt to localize the limit more exactly.

Most workers on Mg actions make no mention of accommodation. The only case known to us, besides the present one, of accommodation to Mg in an isolated organ, is that of the mammalian intestine described by Jendrassik (1924). He states: “In Lôsungen mit normalem K-und Ca-Gehalt bewirkt eine Steigerung der Mg-Konzentration eine voriibergehende bzw. bleibende Lahmung, die Abnahme vor-übergehende Reizung.” Unhappily, he gives no figures of these Mg effects and no details, e.g., of the duration of the temporary inhibition or excitement. He also describes temporary effects of changes in the K and Ca concentrations, and these (which he illustrated) are in general very much briefer than those encountered by ourselves with Mg. He was, however, working with a mammalian organ at 37°C. On the whole his results with Mg seem to resemble ours, indicating that the property of accommodation to this ion is not a peculiarity of Arenicola tissue. It may even be very widespread. Most workers on isolated rhythmic muscles, at least with invertebrate preparations, have contented themselves with short exposures to the various salt mixtures tried, and in many cases the inhibition generally produced by Mg excess might be followed by some degree of accommodation if the exposure were longer. Perhaps tissue-accommodation plays a part in the spontaneous recovery of whole animals from the anaesthesia produced by injection of Mg salts—a phenomenon noted by many authors (e.g. Meltzer & Auer, 1905, 1906; Wiki, 1934).

Turning now to the problem of the mechanism of accommodation, the fact that a sudden change produces (1) immediate response, followed by (2) slow adjustment, suggests at once that the first is a direct action of the new chemical environment on the surface of the cell, while the second is due to the penetration of some substance across the cell membrane. This is the viewpoint adopted by Jendrassik (1924). The substance which actually crosses the membrane need not necessarily be Mg. One can assume, on the one hand, that excitability depends in some way on the relative amounts of Mg within and without the cell, in which case Mg penetration will cause accommodation, or, on the other hand, that increased external Mg lowers excitability by entering into reversible combination with some organic constituent of the cell surface, whose uncombined fraction is then partly restored by slow outward diffusion from reserves within the cell. Both hypotheses have an essential point in common—they ascribe the slowness of accommodation to the difficulty of penetrating the cell membrane.

While there is at present no critical experimental evidence for or against such hypotheses, it should be emphasized that the experimental results can be otherwise explained. In a recent theoretical discussion of the properties of the steady state compared with those of equilibrium, Burton (1939) has shown that what he calls “overshoot” will occur in steady state systems under certain conditions. It is only necessary to introduce a slight modification of his theoretical system, to arrive at a simple and attractive explanation of our results, along quite different lines from the “surface action and penetration” mechanisms described above.

Burton discusses the properties of “the simplest imaginable steady state system”, i.e.
where AB is an irreversible chemical reaction of velocity constant k ; 5 is a source, of constant concentration, from which A is continually replenished by a process similar to diffusion (diffusion constant k0); Z is a sink to which B is removed (diffusion constant kc). If, as a result of some external agency, velocity constant k is suddenly increased to a new value, the concentration of B will rise; if k0 <k2 it will overshoot and then settle down to a new steady level. Burton illustrates the properties of this system by means of a simple model, in which the concentrations of 5, A, B, Z are represented by the levels of water in vessels (that of B being written on a smoked drum by means of a float and lever), and the diffusion and reaction constants by cocks of variable resistance. He also discusses more complicated steady state systems, and the conditions under which they will overshoot.

Suppose now that, instead of diffusing away to Z, B behaves as a relaxation oscillator, and accumulates until a critical concentration is reached, when it is suddenly and completely removed. The result will be a rhythmic system in which the frequency of the rhythm behaves as the concentration of B does in Burton’s case. The frequency will rise and fall with k, and overshoot after a sudden change in the value of that constant.

To make this clear, we have constructed a water model, similar to Burton’s as regards 5 and A, but with a relaxation oscillator for B (Fig. 6). Velocity constant k is represented by either of two capillaries of different diameters, so its value can be changed by turning the two-way tap. To emphasize the analogy with a rhythmic muscle, B may be called the “excitor mechanism”, and vessel V, overflow O and tambour T may together be called the “contractile mechanism”, since their function is to inscribe a contraction of appropriate form on the drum for every discharge of B. The apparatus gives records closely paralleling those of the extrovert (Fig. 7). If the system is running steadily with a fast capillary at k, the level of A will be fairly low. If a slow capillary is now substituted (Fig. 7, upper line), the flow rate into B falls at once, but it then shows a gradual “accommodation” as the level of A rises to a new steady position. On returning to the original capillary after the new position is reached, there is great acceleration followed once more by “accommodation” as A falls. It is of course possible to imitate the action on the extrovert of Mg concentrations so high that mechanical activity is impossible, by simply closing the tap for a time (Fig. 7, lower line).

Fig. 6.

Water model to show accommodation in a rhythmic steady state system.

Fig. 6.

Water model to show accommodation in a rhythmic steady state system.

Fig. 7.

Records made with the apparatus of Fig. 6. Tap k changed at X. Upper line: fast flow rate, then change to slow, then change back. Lower line: imitation (by temporarily closing tap k) of the experiment of Fig. 3.

Fig. 7.

Records made with the apparatus of Fig. 6. Tap k changed at X. Upper line: fast flow rate, then change to slow, then change back. Lower line: imitation (by temporarily closing tap k) of the experiment of Fig. 3.

If then we suppose that the extrovert is a steady state system comparable to the model, our accommodation phenomena can be explained without assuming the gradual penetration of any substance across the cell membrane. It is only necessary to postulate that a new Mg concentration alters the value of k as from the moment of its application. The sequence of “overshoot” and “accommodation” will follow as a matter of course, from the resultant changes in concentration of substance A.

  1. The reactions of the isolated extrovert of Arenicola marina to variations in the external Mg concentration are described and discussed.

  2. Artificial sea water supports a vigorous rhythm for many hours and was therefore arbitrarily taken as the “normal” saline. In all mixtures, the following were held constant: pH, K, Ca, sulphate, osmotic pressure. Increase in Mg was osmotically compensated by decrease in Na, and vice versa.

  3. High Mg concentrations depress, and low ones raise, the spontaneous activity level of the preparation.

  4. The preparation can accommodate itself, to a large extent, to a new Mg concentration. The effects of abruptly changing to a new mixture are greatest just after the change, and gradually become less as accommodation occurs.

  5. After accommodation to a new Mg concentration, the old is no longer appropriate. Return to artificial sea water evokes Mg-deficiency reactions after accommodation to high Mg, and Mg-excess reactions after accommodation to low Mg. In either case, the preparation slowly accommodates itself back again to normal.

  6. The accommodation process occurs in mixtures whose Mg concentration is so high that spontaneous activity cannot reappear. This is shown by the fact that Mg-deficiency reactions are evoked by changing back to normal after long exposure to such mixtures, and proves that accommodation does not depend on special events (such as permeability changes) associated with functional activity.

  7. If time enough for accommodation is allowed, or if the change of Mg concentration is made very slowly (“drift” experiments), it is found that fairly normal activity can occur over a wide range—from Mg-free mixtures to mixtures containing about three times the amount of Mg in sea water. At the upper end of this range, the preparations are, however, markedly depressed.

Burton
,
A. C.
(
1939
).
J. cell. comp. Physiol
.
14
,
327
.
Jendrassik
,
L.
(
1924
).
Biochem. Z
.
148
,
116
.
Mcclendon
,
J. F.
(
1916
).
J. biol. Chem
.
28
,
135
.
Meltzer
,
S. J.
&
Auer
,
J.
(
1905
).
Amer. J. Physiol
.
14
,
366
.
Meltzer
,
S. J.
&
Auer
,
J.
(
1906
).
Amer. J. Physiol
.
15
,
387
.
Pantin
,
C. F. A.
(
1926
).
J. exp. Biol
.
3
,
275
.
Wells
,
G. P.
(
1937
).
J. exp. Biol
.
14
,
117
.
Wells
,
G. P.
&
Ledingham
,
I. C.
(
1940
).
J. exp. Biol
.
17
,
337
.
Wiki
,
B.
(
1934
).
Bev. méd. Suisse Rom
.
54
,
973
.