Patterns of Circular And Radial Mantle Muscle Activity in Respiration and Jetting of the Squid Loligo Opalescens

The basic mechanism of jet locomotion in squids has been explained in terms of an antagonistic arrangement of circular and radial muscles in the animal’s mantle. Contraction of powerful circular muscles reduces the diameter of the mantle, and the water expelled from the mantle cavity forms the jet. The refilling phase of the jet cycle is brought about by the contraction of radial muscles. These muscles extend across the thickness of the mantle wall and insert onto robust layers of collagen fibres (the tunics) that line the inner and outer surfaces of the mantle. The action of the radial muscles in mantle refilling is facilitated by a hydrostatic mechanism in which the muscle tissue itself is the working fluid. That is, contraction of the radial muscles causes the mantle wall to become thinner, and since the volume of the muscle tissue remains constant, the circumference of the mantle must increase. Thus, the contention of the radial muscles extends the circular muscles and provides the driving force for drawing water into the mantle cavity, in preparation for another jet. The anatomical basis for this mechanism was first suggested by Young (1938), and later described in detail by Ward & Wainwright (1972), Ward (1972), and Packard & Trueman (1974). In addition to this muscular antagonism, Ward & Wainwright (1972) and Bone, Pulsford & Chubb (1981) found several connective tissue fibre systems placed within the mantle musculature. These fibres should be capable of storing elastic energy, and hence they may provide an ‘elastic recoil’ system that could complement or even replace the action of muscles under some circumstances.

Although the anatomy of this system is well known, we have no detailed understanding of how or when the various components of this elaborate skeletal system are used. For example, Ward (1972) suggested that the basic jet cycle consisted of a circular muscle contraction followed immediately by the contraction of radial muscles to bring about refilling. On the other hand, Packard & Trueman (1974) have observed that squid hyper-inflate their mantle just prior to a powerful ‘escape jet’, and they suggest that the radial muscles are activated immediately before the circular muscles contract. Mantle refilling is presumably brought about by tissue elasticity or by a second period of radial muscle activity following the contraction of the circular muscles. Finally, ultrastructural (Bone et al. 1981) and biochemical (Mommsen et al. 1981) analyses of squid mantle indicate that there are two distinct types of muscle fibres: (1) glycolytic, ‘burst’ fibres which are presumably involved in escape-jets, and (2) mitochondria-rich oxidative fibres which are presumably used for steady swimming and respiratory movements. Since the radial musculature is composed almost exclusively of the glycolytic fibres (Bone et al. 1981), it seems likely that the radial musculature is not used at all in slow swimming or respiratory movements, and this implies that mantle refilling is brought about primarily by elastic recoil.

Two previous attempts to record the electrical activity of squid mantle muscles during jetting (Wilson, 1960; Ward, 1972) have been only partially successful. In both cases, large muscle potentials associated with the contraction of circular muscles were seen, but it was not possible to discern the timing or even the existence of radial muscle activity in the jet cycle. In this study we present clear records of the electrical activity of squid mantle muscles that allow us to determine unambiguously the role of the radial muscles in jet locomotion and in respiratory movements. We find that radial muscle activity is indeed involved with hyper-inflation just prior to an escape jet. In addition, the reduction in or total lack of electrical activity associated with radial muscles during refilling indicates a major contribution from elastic energy stored in the connective tissue fibre-lattice.

Experimental animals, Loligo opalescens, were obtained by netting from Bamfield Inlet on the West Coast of Vancouver Island and were maintained in large circular tanks with running sea water at the Bamfield Marine Station. Although the animals were in spawning condition and would therefore not be expected to live very long it was possible to keep them alive in these tanks for several weeks. Only heal animals that were capable of vigorous swimming were used in these experiments.

Our primary objective was to record electromyographic (EMG) activity from the swimming muscles of the mantle and to determine the phase of the jet cycle in which the radial muscles were active. As shown in Fig. 1A and B, the circular and radial muscles in squid mantle are arranged in alternating bands, with the spacing between radial muscle bands being about 0·1 mm. Because of this close spacing, it was not possible to make separate extra-cellular recordings from the two muscle groups. Rather the EMG of both muscle types was recorded by a single pair of extracellular electrodes. Information on the direction of mantle movement made it possible to distinguish circular from radial muscle activity. This information was provided by recording changes in mantle diameter with a video monitoring system. The use of this video system required that the animal be held stationary relative to the video camera so that a single portion of the mantle could be ‘measured’. Thus, the animal was loosely restrained in a dissecting pan by placing dissecting needles across (not through) the head just anterior to the mantle aperture and across the mantle just anterior to the fins.

EMGs were recorded by impaling the mantle in the mid-ventral region with 32 gauge enamelled copper wire (diameter = 0·20 mm), which had been stripped for the last 2–3 mm. These electrodes were placed approximately 1 cm apart, and electrical activity was recorded differentially using a Grass P-15 preamplifier and displayed on a storage oscilloscope or recorded on tape, as described below.

Mantle cavity pressure was recorded using a Narco Telecare RP-1500i pressure transducer. The transducer was linked to the mantle cavity through an approximately 50 cm length of sea water filled polyethylene tubing (PE-160) that was ‘sewn’ through the mantle wall. ‘Pop’ tests of the transducer-catheter assembly typically gave resonant frequencies in the range of 30 Hz and damping factors of the order of 0·26. Since the fundamental frequency of the pressure pulse is of the order of 1 Hz and since Fourier analysis of the pressure waveform indicates that there are only very minor contributions to the total waveform from components above the fifth harmonic, we conclude that the pressure transducer is capable of accurately representing the true waveform.

Changes in mantle diameter were monitored with a Video Dimension Analyser (Model 303, Instruments for Physiology and Medicine, San Diego, California), which is capable of giving a d.c. voltage proportional to the separation of two contrast boundaries on any horizontal line of a video image. The squid was held against a black background and orientated so that the transverse axis of the animal was parallel to the horizontal sweep of the video image. Thus, providing the mantle remains circular in cross-section, the output from the video dimension analyser provides a direct indication of the diameter of the mantle. The framing rate of a video camera is 60 Hz, and the output from the analyser is smoothed with a 15 Hz filter. A test of the frequency response of the video dimension analyser using a transfer function analyser (Model SM272-DP, S. E. Labs, Ltd, Feltham, England) indicated that the output filter caused a 2% drop in amplitude and a 16 degree phase shift at 1 Hz. The waveform of the change in diameter with time is fairly sinusoidal, with a fundamental frequency of the order of 1 Hz, and therefore there should be no major distortion of the amilitude of the diameter waveform; however there may be a small but significant phase fifth of the diameter waveform relative to the pressure wave form.

All records of pressure and diameter were taken about one third of the way be from the anterior margin of the mantle. With the exception of experiments in which two sets of EMG electrodes were placed at extreme ends of the mantle to test the timing of muscle activity in escape jets, the single set of EMG electrodes was always placed as close as possible to the site where pressure and diameter data were being collected. If the video ‘window’ crossed either the pressure catheter or the electrode wires the video monitoring system had difficulty in following the outline of the animal. We therefore reduced the video window to its minimum height (approx. 2 mm) and placed it immediately adjacent to the catheter and electrode wires. In all cases pressure, diameter and EMG data were collected from within 5 mm of each other, on an animal with a total mantle length of about 10 cm. Thus, the records can reasonably be expected to represent the activity of a single region of the mantle.

Output from the three signal sources was recorded on a Hewlett Packard instrumentation tape recorder (model 3964A) and displayed on a Hewlett Packard oscillographic recorder (model 7402A). In some cases a 60 Hz notch filter (Model AP-Sl, A. P. Circuit Corp., 865 West End Ave., N.Y.) was used on the EMG channel to reduce background noise and reveal fine detail of the small radial muscle electrical activities.

Figs 1C, D, 2, 3 and 4 show several traces of electrical activity in mantle muscle associated with escape jetting. Although there is some variation in the details of the pattern of electrical activity for different animals and at different times for the same animal, the basic pattern which we feel typifies jetting is shown in Fig. 1C. This trace, taken directly from the screen of a storage oscilloscope, was triggered at a burst of large muscle potentials which undoubtedly correspond to the activation of the circular muscles. These circular muscle potentials are spaced with a period of about 1·5 s, and in between each burst of circular muscle potentials is a burst of much smaller potentials that are probably associated with the activation of the radial muscles. Note that the radial muscle activity seems to come just before the circular muscle activity, indicating that the radials are probably involved in hyper-inflation of the mantle, as suggested by Packard & Trueman (1974). This suggests that we may need to divide the jet cycle into three discrete phases, rather than just jetting and refilling. The three phases would be: (i) hyper-inflation, powered by the radial muscles; (ii) the jet, powered by circular muscles; and (iii) refilling, powered mainly by tissue elasticity. However, direct correlation of electrical activity with specific muscle groups requires additional information.

Fig. ID shows more complete data for a train of 11 jets, with a single respiratory movement coming between jet 10 and 11. In this figure the top trace gives mantle pressure, the middle trace provides a measure of mantle diameter, and the bottom trace is the electrical activity of the muscles (EMG). Although the zero pressure line and the zero volts line on the pressure and EMG traces are obvious, the mantle diameter trace provides only an indication of change in diameter and does not indicate the absolute diameter of the animal.

That the animal was healthy and vigorous at the time of the experiment is indicated the fact that it was able to achieve a large mantle pressure of around 200 cm of water (20kPa). In all cases the large muscle potentials on the EMG trace coincide with a sharp rise in mantle pressure and a large, rapid decrease in diameter. These changes must be due to the contraction of the circular muscles. In most cases the smaller muscle potentials come just before the large circular muscle potentials, a pattern shown particularly well in the last six jets in this series. These small muscle potentials undoubtedly represent radial muscle activity because they occur when the diameter of the mantle is increasing, and it is not possible for the diameter of the mantle to increase when the circular muscles are active. In most cases the radial activity appears to be associated with an obvious hyper-inflation of the mantle just prior to the jet. This can be seen clearly on the diameter trace in many of the jets, where a small upwardly directed peak (increasing radius) is seen just prior to the abrupt decrease in diameter (the jet) brought about by the contraction of the circular muscles.

Fig. 2 shows an oscilloscope trace similar to that in Fig. 1C, but instead of having a single electrode record, data presented are from two sets of electrodes placed at extreme ends of the mantle. In this case, the separation would be of the order of 10 cm. Pumphrey & Young (1938) first suggested that the graded conduction velocities of the different diameter giant axons would insure a near simultaneous contraction of the mantle. Our EMG records show that the activation of the circular muscles during jetting is almost simultaneous along the entire length of the mantle. However, the sweep rate for our records was quite slow. It is not possible to determine how closely the two sets of muscle potentials coincide, but we estimate that they are matched to better than about 10 ms. It would be nice to be able to demonstrate that this matching was due to the graded conduction velocities of the giant fibres, as Pumphrey & Young (1938) suggested. However, the conduction velocity of squid giant axons ranges from 5–20 m/s (Pumphrey & Young, 1938), and therefore the longest delay for the transmission of impulses to the most distant part of the mantle will be only approximately 10 ms. Clearly, we must repeat our experiment with better time resolution before we can truly demonstrate the matching of conduction velocity to the activation of distant muscles.

Although the general pattern described above in Fig. 1 seems to be characteristic escape jetting in most squid, some differences are seen. In addition, some low amplitude muscle potentials associated with radial muscle activity may have been lost in the recording noise. We have therefore provided expanded traces of the same three parameters for two additional trains of jets from two other animals (Figs 3, 4). In both figures, the EMG data in the expanded traces have been processed with a 60 Hz notch filter, as described in the Methods section. The reader should note that the filter alters the relative size of the circular and radial muscle potentials, making them seem more equal in amplitude. In unfiltered traces the circular muscle potentials are about an order of magnitude larger than the radial muscle potentials.

The expanded trace in Fig. 3B shows four jets taken, as indicated, from the series of jets in Fig. 3A. In this example there is a respiratory movement between each jet. The muscle activity pattern consists of a discrete burst of radial muscle potentials just prior to the activation of the circular muscles (hyper-inflation), as well as a discrete burst of radial muscle potentials immediately following the contraction of the circular muscles (refilling). The radial activity in the refilling phase of the cycle presumably complements the action of the mantle elasticity in this process. The combined action of radial muscles and tissue elasticity undoubtedly explains the very rapid increase in diameter seen following the jets in this trace. Indeed, the rapid increase in diameter seen in the third jet of this series is as fast or faster than the decrease in diameter due to the contraction of the massive circular musculature.

Close analysis of these expanded traces indicates that there is virtually no delay (less than about 10 ms) between the start of the circular muscle potentials and the start of the rise in mantle pressure. There is, however, a significant delay of about 40—80 ms between the circular muscle potentials and the decrease in mantle diameter, a delay which is probably due largely to the phase shift introduced to the diameter trace by the video dimension analyser (see Methods section). On the other hand, there is a much larger delay between the start of radial muscle potentials and obvious increases in mantle diameter (of the order of 100–200ms). Although we cannot establish the times precisely, it seems likely that there are longer time delays for the activation of the radial muscles than for the circular muscles.

Fig. 4 shows two expanded segments from a third series of escape jet movements. The expanded segments (Fig. 4B, C) reveal more clearly what we believe to be the normal pattern of radial muscle activity in the refilling and hyper-inflation phases of the jet cycle. Note in particular the final three jets in Fig. 4B, which do not have low amplitude respiratory movements intervening between the powerful jets. The burst of radial activity associated with hyper-inflation is readily apparent, but in addition the filtered trace reveals a low level of radial activity in the refilling phase. Even though the EMG trace is somewhat distorted by a movement artifact at the end of the jet and the start of the refilling phase, it is clear that there is usually a low level of radial activity in refilling. However, in some cases the activity in this region of the trace is difficult to distinguish from baseline activity. The diameter trace clearly reflects the reduction in, or lack of, radial muscle activity in the refilling phase of the jet cycle. The increase in diameter at the start of refilling is much slower in these traces than in Fig. 3B, where strong radial activity worked together with tissue elasticity to bring about very rapid refilling. The refilling phase following the first jet in Fig. 4C clearly illustrates the effect of radial activity on the rate of mantle expansion. There is an abrupt increase in the rate of expansion which coincides with a large burst of radial muscle potentials. Presumably, prior to this burst of radial activity the mantle was being expanded by tissue elasticity alone. Finally, inspection of the respiratory movements in Fig. 4C reveals a very interesting feature of the muscular activity in respiration. Although the mantle diameter decreases and the pressure increases during these respiratory movements, there does not seem to be any electrical activity associated with the activation of circular muscles. This leads us to expect an elaborate interplay of muscle contraction and tissue elasticity in respiration, just as we have seen in jetting.

Fig. 5 shows three sets of representative traces of pressure, diameter and muscle activity obtained from three different animals during respiration. Packard & Trueman (1974) showed that respiratory movements are primarily restricted to the anterior half of the mantle, where the gills of the animal are located. Respiratory movements in the posterior portion of the mantle are smaller, of different frequency, and thus may be out of phase with movements in the anterior mantle. To avoid problems in interpretation due to possible phase differences in respiratory movements we collected pressure, diameter and EMG data from the same small region of the anterior mantle. The reader should note that the amplitude of the pressure pulse during respiration is very much smaller than in jetting, being only about 2–5 cm of water (0·2–0·5 kPa) as compared to more than 200 cm of water in jetting. In some instances the pressure amplitude was so small that it was difficult to distinguish the ‘jet’ from the refilling phase. However, in all cases the mantle diameter trace clearly indicates the direction of mantle movement. Thus, it is possible to distinguish unequivocally the activity of circular muscles from radial muscles. All of the EMG traces have been processed with the 60 Hz notch filter to reveal the details of the low amplitude muscle potentials.

The EMG associated with the mantle muscles in each of the three sets of traces appears quite similar to that observed in Fig. 5C. There are obvious bursts of muscle potentials separated by regions with little or no electrical activity. However, inspection of the mantle pressure and diameter data reveal two patterns of muscular activity associated with respiratory movements. In pattern I (Fig. 5A, B) electrical activity must be associated with the contraction of radial muscles because the electrical activity occurs when mantle diameter is increasing. In pattern II (Fig. 5C) the EMG is associated with the contraction of circular muscles because the electrical activity occurs when mantle diameter is decreasing. Interestingly, both patterns apparently stabilize a single set of muscles, with tissue elasticity presumably antagonizing these muscles. We do not know which of these patterns represents the ‘normal’ pattern for the animal, but the compressed diameter traces located to the left side of Fig. 5 provide some possible insights.

These compressed diameter traces show a series of respiratory movements before and after a group of escape jets, and the expanded traces shown in Fig. 5A, B and C were taken from the respiratory movements following these escape jets. The compressed diameter traces make it possible to determine which part of the total range of diameter change (as indicated by the escape jets) was used for respiratory movements. For example, in Fig. 5A and B (pattern I) the outer limit of respiratory movements is very close to the maximum diameter achieved during hyper-inflations associated with escape jets. The inner limit of these respiratory movements, however, comes nowhere near the minimum diameter achieved during the escape jets. Thus, in pattern I animals respire at a diameter close to the maximum diameter achieved in escape jets, and it is possible that whatever limits mantle expansion to this level might also be capable of storing elastic energy to antagonize the radial muscles. In Fig. 5 pattern II) the inner limit of respiratory movements is quite close to the minimum diameter achieved in the escape jets. Thus, in pattern II it seems likely that the circular muscles are being antagonized by the same tissue elasticity which is responsible for limiting the contraction of circular muscles and powering the refilling phase of escape jets.

The data presented up to this point were obtained from animals that had been restrained in a very unnatural position, and it is difficult to determine which pattern, if any, represents the ‘normal’ one. However, the diameter trace in Fig. 5D was obtained from an animal that was tethered by gluing its dorsal mantle surface to a submerged Plexiglass rod. Since animals could be maintained for at least 2 days in this posture and since the animals appeared to ‘swim’ normally, we think that trace 5D provides a closer approximation to the normal condition. Although pressure and EMG traces are not available, the diameter data suggest that the ‘normal’ animal uses pattern II. Confirmation of this conclusion will require pressure and EMG data from untethered animals. Pattern I probably reflects the tethering method used. Animals pinned out in a pan are certainly disturbed and are most likely intent on escape. They appear to have shifted their respiration range towards the outer limits imposed by the mantle connective tissue fibre-lattice so that they are ready to give a maximal escape jet with a minimum of delay.

As expected from our understanding of squid mantle morphology, radial muscle activity is indeed involved in the jet cycle. We have shown that it is possible to record and distinguish the electrical activity of circular and radial muscles from one another and also to assign a functional role to these muscle groups. Although previous workers (Wilson, 1960; Ward, 1972) were unable to record electrical activity from the radial muscles, they both assumed that the radial muscle potentials were present but too small to be seen above recording noise. The most likely reason for our success in recording radial muscle activity is improved technology, with a great deal of credit going to the 60 Hz notch filter used to reduce baseline noise.

Several characteristic patterns of muscular activity in squid locomotion emerge from this study. In escape jets there appear to be three distinguishable phases in the full jet cycle. PHASE (i): HYPER-INFLATION. The jet cycle begins with the mantle at some intermediate, resting diameter, and it is initiated with a period of hyper-inflation, powered by strong contractions of the radial muscles. Hyper-inflation serves to maximally fill the mantle cavity, allowing a more powerful jet to follow. PHASE (ii): THE JET. Immediately after hyper-inflation the mantle wall contracts, expelling the sea water in the mantle cavity and creating the jet. This phase is powered by the contraction of the circular muscles. PHASE (iii): REFILLING. Following the jet the mantle re-expands to its resting diameter. This expansion is powered largely by the elastic recoil of the connective tissue fibre-lattice. There also may be a relatively small contribution from the radial muscles which will increase the rate of refilling (see Figs 3B, 4C). Thus, in what we believe is the ‘basic’ pattern of muscular activity in escape jetting, the major role of the radial muscles is primarily the hyper-inflation phase.

Although there appears to be a basic pattern, a certain amount of variation observed, and it would be unwise to conclude that this is the only pattern exhibited by free-swimming animals. The animals used in this study were severely restrained, and it is certainly possible that the normal jetting pattern is quite different from the one described here. For example, in some instances we observed a pattern in which there was a burst of radial muscle activity both before and after the circular muscle contractions (see Fig. 3B). However, this pattern was seen usually when respiratory movements came between individual escape jets. In an uninterrupted series of escape jets (see Figs 1D, 4B) the ‘basic’ pattern emerged. Since escape jets in free-swimming animals are unlikely to be separated by respiratory movements, we tentatively conclude that this basic pattern of muscle activity is quite similar to the normal pattern in free-swimming animals. Clearly, confirmation of this conclusion will be possible only when we complete measurements on untethered animals. One other fact which must be kept in mind is that cephalopods have very highly developed, complex nervous systems (Wells, 1978). It is possible, indeed likely, that squid are capable of several different patterns of muscle activity in escape jetting, and that they use these different patterns in a variety of different activities which require some form of maximal jet.

The other major motor behaviour which we have studied is respiration, and two basic muscle activity patterns appear in our records. In respiration pattern I, radial muscle activity is antagonized by tissue elasticity, and there does not appear to be any involvement of the circular muscles. In respiration pattern II circular muscle activity is also antagonized by tissue elasticity, and there does not appear to be any involve ment of the radial muscles. In addition, it is possible that there is a third pattern of respiration, employing reciprocally contracting circular and radial muscles, but this pattern appeared very infrequently in our records. Again, we do not know which of these patterns represents the normal pattern of a free-swimming animal, and further we do not know how respiratory movements relate to slow swimming movements. We suspect that respiratory movements in a tethered animal are similar to slow swimming movements in an untethered animal. Thus, the two patterns we saw for respiration may both represent normal patterns used by free-swimming animals for different activities. For example, it seems likely that under most circumstances the animal uses pattern II, since this allows the use of the non-fatiguing, aerobic circular muscle fibres. The oscillations in mantle diameter are around some mid-position so that the circular muscle contractions can be antagonized by the elastic system which is involved in the refilling phase of the escape jet cycle. The animal does not expand its mantle out near the maximum diameter achieved during hyper-inflation. However, in conditions where the animal anticipates having to make maximal escape jets, for example when the squid is stalking prey or when it is being stalked, it may switch to pattern I. The greatest diameter achieved in this pattern is very close to the maximum diameter attained during hyper-inflation, just prior to a powerful jet. Thus, the animal maintains its mantle cavity at nearly full volume and is poised for a rapid jet, but the animal must now use its radial muscles for respiration. Since virtually all of the radial muscle fibres appear to be of the glycolytic, ‘burst’ type (Bone et al. 1981), this pattern of respiration/slow swimming will probably be less efficient. Again confirmation of these tentative conclusions will require the completion of studies on untethered animals. Finally, let us consider how these patterns of muscle activity relate to the connective sue fibre-lattice, which appears to play such an important role in all modes of squid locomotion. The anatomical studies of Ward & Wainwright (1972) and Bone et al. (1981) indicate that there are three distinct fibre systems within the mantle musculature. The general organization of these intermuscular fibre systems have been reviewed recently by Gosline & Shadwick (1982). Two of the fibre systems (inter-muscular fibre systems 1 and 2; IMF-1 and IMF-2) are placed to antagonize the circular muscles and power the refilling phase of the escape jet cycle and respiration pattern II. The fibres in these two systems appear to be collagenous in nature and are seen to run across the mantle thickness, inserting on the inner and outer collagen tunics that line the mantle. These collagen fibres are orientated so that they are stretched when the circular muscles contract, and thus energy from the contraction of the circular muscles can be stored in these fibres. The third fibre system (IMF-3) runs parallel to the circular muscles, and contraction of the circular muscles causes these fibres to buckle. They are therefore not able to store energy from the contraction of the circular muscles. However, the circular muscles are extended by the contraction of the radial muscles, and thus the fibres in IMF-3 should be stretched and therefore able to store energy from the contraction of the radial muscles. During maximal escape jets both the elastic system antagonizing the circular muscles and the elastic system antagonizing the radial muscles are used, presumably to maximize the power output during escape manoeuvres. During respiration/slow swimming the animal can select either of two patterns, each of which employs a different elastic system. Clearly, the connective tissue fibre-lattices of squid mantle play a very important role in the mechanics of squid locomotion, and a great deal remains to be known about these systems. For this reason we are currently investigating the mechanics and function of the intermuscular fibre systems in much greater detail.

This research was supported by grants from the Natural Sciences and Engineering Research Council to JMG (67-6934) and JDS (67-7160).