ABSTRACT
Conduction of excitation in response to local mechanical or electrical stimulation has been studied in various hydroid species.
There are systems in the coenosarc of the stems and stolons of all species which conduct excitation at rates of from 1 to 3·5 cm./sec.
Two physiological types of conducting systems have been found.
Through-conducting systems, showing: all-or-none response, sharp threshold, reproducibility.
Local systems, showing : spread of response dependent upon stimulus strength, no sharp threshold, responses which decline with increasing distance from the stimulated point, variability.
Colonial co-ordination is better developed in those species whose colonies are structurally better developed. It is effected in most species by either (a) or (b). Both types are present together in Hydractinia echinata.
INTRODUCTION
There are many reports in the literature confirming the existence of colonial co-ordination in the Coelenterata. Studies have been made on co-ordinated polyp responses and on the spread of luminescent waves across a colony, activities presumably mediated through nerve nets. Several authors have described contraction of polyps other than the ones directly stimulated in colonial Hexacorallia (Klunzinger, 1877; Duerden, 1902; Matthai, 1918; Abe, 1939; Hyman, 1940; Horridge, 1957); and propagated waves of polyp retraction have been described for the Octocorallia (Milne-Edwards, 1835; Krukenberg, 1887; Parker, 1920 b; Hiro, 1937; Gohar, 1940; Horridge, 1956a, b, 1957; Broch & Horridge, 1957). Most of the reports merely mention the occurrence of colonial responses, however, and only Krukenberg, Parker, and especially Horridge have analysed these reactions in any detail.
The luminescent waves which traverse colonies of certain alcyonarians have been the most studied colonial responses of coelenterates, probably owing to the ease with which they can be observed and, with modern methods, accurately measured. Among the works dealing with these luminescent responses are those of Panceri (1872), Harvey (1917), Parker (1919, 1920a, b), Moore (1926), Honjo (1944), Buck (1953), Nicol (1955 a, b, 1958) and Davenport & Nicol (1956).
Among the Anthozoa, only for the alcyonarian Veretillum does there appear to be histological evidence for a colonial nervous system (Niedermeyer, 1914). Kassianow (1908) failed to find nervous connexions between polyps in Alcyonium, but the recent paper by Horridge (1956a) on this genus gives physiological evidence that such connexions exist. Hyman (1940), discussing the Madreporaria, states that, on the basis of behavioural responses, ‘the presence of a nerve net throughout the colony must be assumed’. It appears probable that the dearth of histological evidence for colonial nervous systems in Anthozoa is only a result of the scarcity of histological studies devoted to colonial forms.
The other coelenterate class with colonial representatives is the Hydrozoa. Among the Hydrozoa the siphonophores offer the most highly specialized colonies in the Animal Kingdom. There now appears to be good reason to regard Velella and its relatives (the Chondrophorae of Hyman (1940)) as single modified polyps and not colonies in the usual sense of the word (Garstang, 1946; Mackie, 1959), and these forms will not be considered here. Nervous connections between at least some of the members of a colony have been histologically detected in the remaining siphonophores (Schaeppi, 1898;Mackie, 1960 a, b), and responses of many individuals following stimulation at one point has been described by Bigelow (1891), Schaeppi (1898), Schneider (1902), and Mackie (1960a, b).
There appear to be few reports on colonial co-ordination in hydroids. Föyn (1927) described contraction of all members of Clava colonies when the stolon is mechanically stimulated, and he used such responses to delimit the bounds of one colony when several colonies grew from a common intertwined mat of stolons. Co-ordinated responses of polyps, especially the spiral zooids, have been reported by Wright (1856) and Schijfsma (1935), among others, for Hydractinia. Zoja (1891), in a commonly overlooked paper, described propagated polyp withdrawal following electrical stimulation in Corydendrium, Coryne, Eudendrium, Tabularia, and Campanularia, and he studied this reaction in some detail in Pennaria and Podocoryne. His results are discussed in the parts of this paper dealing with some of these same genera. Evidence for the presence of a neural substrate which could mediate these responses is given by the report of ganglion cells in the coenosarc of Syncoryne (Citron, 1902) and Cordylophora (Mackie, 1961). Jickeli (1883a) reported ganglion cells in Eudendrium in the ‘hydro-phyton’ (a term designating both the hydrocaulus and the hydrorhiza according to Allman (1871)). Jickeli, however, described the hydrophyton as being capable of movement and shape changes, so he may have been referring to only that portion of the hydroid immediately below the hydranth proper.
The following study was begun in an attempt to augment our limited knowledge of colonial responses in coelenterates.
MATERIALS AND METHODS
The animals used in this study were the colonial hydroids Pennaria, Syncoryne, Tubularia, Cordylophora, Podocoryne, Hydractinia aggregata and H. echinata. Species were identified with the aid of Fraser’s monographs on American hydroids (1937, 1944). All experiments were performed on colonies submerged in sea water or, in the case of Cordylophora, in dilute sea water. Mechanical stimuli were given to the colonies by pinching with forceps or by prodding with a fine glass rod. Electrical stimuli were delivered to the colonies by fine silver wires (0·2 mm. in diameter) insulated to the tip with lacquer. When stimulating colonies of Hydractinia or Podocoryne which form two-dimensional sheets, the electrodes were placed about 1 mm. apart on the stoloni-ferous mat of the colony. For the other species, the electrodes were placed one on each side of the slender stolons or stalks of the colony. The electric stimuli in the experiments on H. echinata were produced by a condenser-discharge apparatus which gave pulses with a half-amplitude duration of 0·05 msec. For all other species the electric stimuli were square pulses produced by a Grass S4 stimulator and were, unless otherwise specified, 0·5 msec, in duration.
Observations of polyp responses were made visually. Techniques used uniquely on one species will be described in the section devoted to that species.
RESULTS
I. A species with both a through-conducting and a local polyp co-ordinating system
Gastropod shells carried by hermit crabs, collected at Woods Hole, are often encrusted with colonies of Hydractinia echinata Fleming. Such colonies can contain several thousand nutritive polyps or gastrozooids, each about 1 mm. high. Scattered among the gastrozooids are numerous reproductive polyps, the gonozooids. Spiral zooids, presumably defensive polyps, are frequently found around the aperture of the gastropod shell.
(a) Mechanical stimulation
Prodding a gastrozooid with a glass rod may give no polyp responses, or a contraction of one or a few tentacles, or contraction of both the tentacles and the hydranth. This contraction often spreads to neighbouring polyps, and several hundred polyps can contract following such stimulation. Lightly pushing on the stoloniferous mat causes a similar local wave of polyp contraction. The polyp contraction often spreads in patches, with contraction in one group of polyps following contraction in another group after a brief but noticeable delay. Polyps included in a contracted area take about 30 sec. to relax. The conduction velocity of the waves of local contraction was not accurately measured, but it appeared to be about 1 cm./sec. If the area of contraction includes some spiral zooids, these become more tightly coiled and move closer to the stoloniferous base.
When a single polyp is crushed with forceps or the stolon mat is damaged, a wave of polyp contraction is initiated which affects all members of the colony. When this wave reaches a spiral zooid, the spiral zooid does not coil more tightly but suddenly uncoils, lashing out, sometimes repeatedly (Fig. 1). The conduction velocity of such excitation appears considerably faster than that associated with local responses; spiral zooids near the stimulated area and those some distance away across the shell aperture from the stimulated area lash almost simultaneously. The different conduction velocities indicate activity in different conducting systems. The polyps relax slowly following injurious stimulation and often take several minutes to re-extend. Polyps far from the injured area frequently relax before those near the point of stimulation.
(b) Electrical stimulation
Local polyp contraction similar to that seen with mild mechanical stimulation can be evoked with an electric shock. Such contraction often spreads in patches, and spiral zooids included in a responding area coil more tightly. Although this response can be produced by a single shock, there is some summation of sub-threshold stimuli and several shocks, each 20−30% below threshold, can give a local contraction. A second shock following an effective one sometimes increases the size of the locally responding area. More often, however, it either has no effect or it causes contraction of the whole colony and lashing of the spiral zooids. Increasing the stimulus intensity much above threshold usually causes such colony-wide contraction.
On many occasions a local contraction of polyps cannot be obtained; even just-threshold stimuli evoke contraction of the whole colony and lashing of the spiral zooids. This response is always seen with stimuli of much above threshold intensity. At low frequencies of stimulation the spiral zooids lash once for each of the first few shocks. At frequencies above 2/sec., however, individual spiral zooids do not follow each stimulus, and the lashing of the spiral zooids of the colony becomes asynchronous, some following every other shock, some following every third. Usually only the first few stimuli of a series are effective. In a typical case, using a 1/sec. train of stimuli, the gastrozooids contracted to the first shock and the spiral zooids lashed for each of the first four shocks. The spiral zooids then ceased responding and the remaining polyps began to relax, even though the stimuli continued.
On two occasions, following strong, repetitive shocks, relaxation was continually interrupted by spontaneous recontraction of patches of polyps. The areas spontaneously contracting became smaller and the intervals between contractions lengthened with time until the colony was again quiescent. Complete relaxation took about 15 min. Such polyp behaviour appears analogous to the spontaneous waves of luminescence seen in sea pens following repeated stimulation (Buck, 1953 ; Nicol, 1955 b, 1958 ; Davenport & Nicol, 1956).
II. A species with a through-conducting polyp co-ordinating system
Podocoryne carnea Sars resembles Hydractima echinata in appearance, and similarly has spiral zooids around the aperture of the gastropod shell upon which it grows. Only a few experiments were done on this species. The colonies used were sent to Los Angeles from Woods Hole, and, although they did survive for several weeks in the laboratory, evaluation of these experiments must be made with the reservation that the colonies may not have been in fully prime condition.
The only response to either mechanical or electrical stimuli seen in these colonies was rapidly propagated, colony-wide contraction of polyps and lashing of the spiral zooids. The contraction of the gastrozooids was sometimes incomplete, occasionally no more than a brief jerk. When the polyps did contract, relaxation was often slower near the point of stimulation. The colony-wide contraction was quite all-or-none and strength-duration curves could easily be obtained for electric stimuli. The chronaxie from such curves was between 0·5 and 1 msec.
Zoja (1891) described polyp contraction in Podocoryne carnea following electrical stimulation as being simultaneous everywhere in the colony and occurring in two or three separate contractions. In my experiments the conduction velocity of this species-was rapid but not ‘instantaneous’, and no evidence of repetitive activity following; electrical stimuli was seen.
III. Species with local polyp co-ordinating systems
(1) Cordylophora lacustris Allman
These hydroids were grown in sea water diluted to 10 % by aged water from a freshwater aquarium. Some of the original specimens were given me by Dr O. Kinne and others were personally collected from the Sacramento River near Antioch, California. They were kept at 20°−22°C. and fed several times a week with newly hatched brine shrimp. Under such conditions this species forms colonies consisting of a single stolon and, every 2−4 mm., short, upright stalks, each bearing one hydranth. The older polyps and coenosarc regress as the colony forms new polyps ; a complete colony under these conditions usually contains three to six hydranths.
(a) Mechanical stimulation
If a polyp is probed with a glass rod, the polyp bends toward the rod, or one or more tentacles retract, or all the tentacles become depressed and the polyp shortened. The latter response will be called polyp contraction. Such activities do not affect neighbouring polyps. If a polyp is pinched with forceps or torn with a glass rod, a propagated wave of polyp contraction is seen which involves all members of these small colonies. In contrast to the high intensity of mechanical stimulation to a polyp needed to cause responses of neighbours, gentle prodding of the stolon of a colony causes an immediate wave of contraction, again affecting all members of the colony.
(b) Single electric stimuli
When a portion of the stolon is stimulated with a single shock of sufficient intensity, a wave of polyp contraction is initiated which propagates with an average velocity of 2·6 cm./sec. (22°C.). This response is not all-or-none (Fig. 2); both the number of polyps responding and the degree of the contraction in any one polyp increases with increasing stimulus intensity. The range over which increasing the stimulus intensity gives a greater response is small. In over half of twenty trials of an experiment involving several different colonies, increasing the stimulus intensity 25 % above threshold gave complete contraction of all the polyps of the colony and in no instance was it necessary to double the stimulus intensity to achieve this end.
In most cases the effect of increasing the stimulus intensity is not smoothly graded. A just-threshold stimulus often causes approximately equal contraction in several polyps near the electrodes and frequently no one polyp can be made to respond individually. Slowly increasing the stimulus intensity in such a case gives no greater response until some critical level is reached, at which time more polyps respond and those of the initially responding group show greater contraction. In Cordylophora, therefore, groups of two to three polyps often form behavioural units with respect to stimuli applied to the stolon.
(c) Repetitive electric stimuli
If two sub-threshold shocks are given to the stolon within 10 sec. of each other, one or more polyps will often respond. The intensity range over which this apparent summation of sub-threshold stimuli occurs is small, usually both stimuli have to be within 10 % of the threshold intensity. The effect of a second shock following the first within 10 sec. is quite pronounced when using stimuli of intensity between threshold and that which gives maximum contraction of all the members of the colony. The second stimulus usually causes more polyps to respond and the already active polyps to contract more fully. If a second shock is given following a shock which evoked complete contraction of all members of the colony, it often delays relaxation of the polyps, although sometimes it has no apparent effect. Occasionally with such stimuli polyps near the electrodes show retarded relaxation while those farther away relax in a normal manner. It appears in such cases as though conduction of the second stimulus has been blocked somewhere in the colony.
Some of the activities of the polyps are apparently unaffected by the induced tentacle depression response. Feeding and defecation continue while a polyp is being stimulated even though they may be temporarily interrupted during the tentacle depression period. In one case, the middle of three polyps being observed captured a small crustacean. When stimuli of above threshold strength were applied to one end of the colony, all but the centre polyp showed a maximal response, the centre polyp not responding at all. This was the only observed instance where the excitation seemed to skip one polyp and continue to a more distal polyp. Usually feeding did not completely inhibit tentacle depression, but did lead to a lesser degree of contraction as compared to non-feeding polyps.
(d) Changes in sensitivity
If a colony is allowed a 1 min. rest between each stimulus or group of stimuli, the intensity required to give maximal contraction of all the members of the colony slowly increases during the experimentation and, at the end of a 2 hr. period, might be twice what it was at the beginning. Small changes in the position of the electrodes do not affect this decreased sensitivity. A change in the sensitivity to electric stimuli is not seen if 5 min. are allowed between each stimulation.
If a number of supra-threshold shocks are given to the colony at short intervals, the polyps soon cease to respond. In one experiment, for example, the hydranths contracted following each of the first five stimuli delivered to the stolon at 2 sec. intervals and then began to relax, even though the stimulation continued.
(2) Pennaria tiarella (Ayres)
Pennaria colonies were collected from floats and pilings near the Hawaii Marine Laboratory. These colonies consist of a main stem which is several centimetres long and, on alternating sides of this stem, branches bearing polyps at 2 mm. intervals. Near the base of the colony the branches are longer and may have as many as eight hydranths, while distally they become shorter and the most distal branches have but a single hydranth. A complete colony has about fifty hydranths. The hydranths are 1 mm. high, and each has a distal row of seven to thirteen long filamentous tentacles and numerous short capitate tentacles scattered over the manubrium. Only freshly collected colonies were used in this study, as even in running sea water the colonies soon degenerated in the laboratory.
(a) Mechanical stimuli
Pennaria shows much spontaneous activity. Polyps often bend from side to side and individual tentacles are frequently elevated. If a polyp is lightly touched with a fine glass rod, the hydranth bends toward the rod or one or more of the filamentous tentacles rises up over the stimulated area. More forceful prodding causes a coordinated elevation of the row of long tentacles. Strong mechanical stimulation, such as pinching with forceps, evokes contraction of the polyp body and folding of all the tentacles about the hydranth. This response will be called polyp contraction. Strong mechanical stimulation of a polyp often leads to contraction of neighbouring polyps, and frequently a wave of polyp contraction affecting all members of the colony is thus initiated. Pinching the stem or one of the branches usually causes a wave of polyp contraction which spreads to all parts of the colony.
(b) Single electric stimuli
A single shock to either the stem or a branch can cause contraction of a few to many polyps which is by no means all-or-none (Fig. 3). The hydranth responses range from slow, partial elevation of the proximal tentacles, with relaxation in 10 sec. or less, to rapid, complete closure of the tentacles with relaxation taking from 40 to 180 sec. The magnitude of the polyp response is graded with distance from the stimulating electrodes; those polyps near the electrodes contract more than those near the periphery of the responding area.
The sensitivity of the colony usually declines greatly during a series of stimulations. The threshold was often found to have more than doubled between successive determinations, even though 1 min. had been allowed between each stimulus. This decreased sensitivity was not changed by small movements of the stimulating electrodes. After a long period of stimulation, leaving the colony quiescent for periods up to 1 hr. did little to reduce the heightened threshold. Experiments with two pairs of electrodes on opposite ends of the colony indicated that the threshold change was confined to the stem and branches of that part of the colony responding to the stimulation.
The number of polyps which contract following a single near-threshold shock is quite variable. Increasing the stimulus intensity by 25 % above a shock which proved sub-threshold can cause contraction in from one to thirty polyps. Further increasing the intensity brings about greater contraction in those responding polyps which did not contract maximally, as well as causing more polyps to react. The effectiveness of increasing stimulus intensity is also quite variable. In over half the cases, increasing the stimulus intensity by 25 % more than doubled the number of responding polyps. This result was more likely, however, if few polyps had contracted at the lower intensity.
As Zoja (1891) found, the wave of polyp contraction usually spreads more easily toward the apex of the colony than toward the base. The number of responding branches is typically larger on the apical side of the electrodes. Since proximal branches bear more polyps than do distal ones, the difference in the number of polyps responding on either side of the electrodes is less marked, but usually more polyps contract distally to the electrodes than proximally.
A branch often acts as a behavioural unit; all the polyps on a responding branch generally contract to the same stimulus. Exceptions to this observation are seen, however, and occasionally only those polyps close to the main stem on a branch are affected. Infrequently the wave of polyp contraction skips one or a few polyps and these remain expanded while those on either side contract. In a similar way the wave will occasionally miss an entire branch while polyps on branches on either side respond. This skipping of individual polyps or whole branches is usually seen only in the portion of a stimulated colony where polyp contraction is slow and incomplete.
(c) Repetitive electric stimuli
Although polyp contraction can be produced by a single shock, the threshold stimulus intensity is usually lower for repetitive stimuli and the number of polyps responding can be increased by stimulating one or more times following a first shock. The decrease in threshold intensity for repetitive stimuli is greater than that seen in Cordylophora and can be as much as 30%.
The frequency usually used for experiments with repeated stimuli was one shock per 5 sec. so that polyp responses caused by each stimulus could be fully evaluated. In over half the trials, excluding those cases where no polyps responded to the first stimulus, the increment in the number of polyps contracting due to the second stimulus was greater than the number responding to the first shock, although the results of any one experiment were variable and unpredictable. Usually only the first few stimuli of a series are effective in increasing the responding area. In a typical case, for example, six polyps contracted following the first stimulus, sixteen more responded to the second, and five additional polyps contracted after the third stimulus. The fourth, fifth and sixth shocks of this series caused contraction in no additional polyps. If an interposed shock is given at a shorter than usual interval after the stimuli of such a series have ceased being effective, a further spread of excitation can often be produced. Thus it is not only the number of stimuli at a given intensity but also the interval between them which limits the spread of polyp retraction.
With repeated shocks as with single stimuli, the degree of polyp contraction is greater near the electrodes and spread of excitation proceeds more readily distally than proximally.
If the electrodes are placed on a branch rather than on the main stem, a single shock typically causes contraction of only the polyps on that branch while an additional stimulus evokes contraction of polyps on several neighbouring branches. This is further evidence that branches often act as single responding units. The junction between the stem and a branch frequently presents a barrier to the spread of excitation which, once crossing this barrier, spreads more easily in the area beyond.
(d) Conduction velocity
As Pennaria colonies are long and the conduction velocity is quite slow, reasonably accurate conduction velocity measurements can be made with a stop-watch and visual observation. By using conduction velocity measurements, Horridge’s (1957) second model of the functioning of nerve nets can be partially tested.
Horridge proposed that conduction in a nerve net may involve only a portion of the elements in the net, and that the density of these active units at any point in the net depends in part on the stimulating conditions; the more intense the stimulus the greater the density. If this is so, one should expect to find the conduction velocity a function of the stimulating conditions. Following a low-intensity stimulus, the excitation should spread along just a portion of the elements in the net, a portion determined by the field created by the electrodes and by the topology of the con ducting elements and the connexions between them. Such spread could follow either direct or meandering pathways. If these pathways are direct, the conduction velocity will be maximum ; if indirect, something less. The average conduction velocity of a number of such experiments, then, should be less than the maximum conduction velocity which would be obtained if fortuitously the fastest and most direct routes were always taken by the excitation. If, on the other hand, most of the conducting elements available are stimulated, the conduction velocity measured will be that of the first-arriving impulses, those coming along the fastest and most direct pathways of all those available. Under such conditions the measured conduction velocity should always be the maximum of which the net is capable.
To see if the conduction velocity was a function of the method of stimulation, the conduction velocity was measured in Pennaria colonies while stimulating with a single shock or by a 1 sec. burst of forty shocks. It was felt that the latter stimulus would be sufficient to excite most of the available pathways in the colony and the conduction velocity following such stimulation would be maximal. The time between the beginning of contraction of polyps on opposite ends of the microscope field was measured with a stop-watch, and the distance between these polyps was measured with an ocular micrometer. This distance was usually about 1·5 cm. Only those cases in which the wave of excitation traversed the entire microscope field were counted. Over twenty conduction-velocity measurements were made with each method of stimulation. Usually each measurement was made with a fresh colony, although in some cases two or more measurements were made on the same colony. All experiments were done at 26°C.
The conduction velocity following a single shock was 1·04 cm./sec., while following a i sec. burst of forty stimuli it was 1·07 cm./sec. The difference between these averages is certainly less than could be accurately measured with this technique. A difference in the conduction velocity following these two methods of stimulation, then, either does not occur or is too small to be measured with this technique.
(3) Hydractinia aggregata Fraser
Gastropod shells inhabited by hermit crabs, dredged from shallow water near Friday Harbor, Washington, are often covered with colonies of Hydractima aggregata. Such colonies resemble those of H. echinata, but lack spiral zooids.
(a) Mechanical stimuli
H. aggregata polyps are quite unresponsive to prodding with a glass rod, although occasionally several tentacles or the whole polyp will contract following such stimulation. Pinching a polyp causes it to contract, and often leads to contraction of from 1 to 200 neighbouring polyps. Pushing on or damaging the stoloniferous mat from which the polyps grow instigates a wave of contraction involving a large number of polyps, but usually not the whole colony.
(b) Single electric stimuli
A single near-threshold shock can elicit contraction in one or a few polyps near the electrodes. Increasing the stimulus intensity causes more polyps to respond ; a 25 % above-threshold shock usually evokes contraction of over 100 polyps and sometimes causes contraction of all the visible polyps of the colony. Such polyp contraction spreads in a wave from the stimulating electrodes with an average velocity of 2·5 cm./sec. (15°C.). Polyps relax everywhere at the same rate following a near-threshold stimulus, taking 30−40 sec. to relax completely. At higher intensities, however, polyps near the electrodes relax more slowly than do those farther away, and often are not fully extended in 2 min.
(c) Changes in sensitivity and responsiveness
The sensitivity of H. aggregata colonies decreases steadily during a series of stimulations. In an experiment testing this threshold change, electrodes made of fine-tipped glass tubes filled with sea water were substituted for the usual insulated silver wires to avoid any electrode polarization. Electrical contact was made with the sea water in the electrodes by means of coils of silver wire covered with silver chloride. The electrodes were held in position with their tips firmly against the basal mat of stolons. In a typical experiment, the threshold rose steadily during fifteen determinations, all within 40 min. The threshold at the end of this time was nearly ten times that at the beginning. Experiments with two pairs of electrodes on the colony indicated that the threshold change was confined to that part of the colony responding to the stimuli.
The responsiveness to mechanical stimulation also decreases during an experiment. A glass tube, drawn to a fine, rounded tip, was allowed to slide within a larger tube mounted so that the inner tube always struck the same point of a Hydractinia colony. After such stimulation, the inner tube was withdrawn to a fixed height and 2 min. were allowed before the colony was stimulated again. In a typical experiment, an average of fifty-six polyps responded to each of the first five stimuli, fifty-two to the second five, nine to the third five, and an average of only five polyps responded to each of the last five stimuli. The water temperature was held at 15°C. throughout the course of both the above-described experiments.
(d) Repetitive electric stimuli
A second shock given shortly after an above-threshold stimulus increases the number of responding polyps. The effectiveness of this second shock is quite variable. Using a pair of stimuli separated by a 2 sec. interval, in six out of ten trials the number of polyps contracting following the second shock was more than twice the number which had contracted to the first shock alone. The second shock of a pair was effective in increasing the number of responding polyps at intervals of up to 6 sec. Only the first two or three shocks of a series of stimuli increase the number of responding polyps; further shocks have no effect. Following such repetitive stimulation, polyps near the electrodes relax more slowly than do those near the periphery of the responding area.
Polyp contraction following either mechanical or electrical stimulation often occurs in patches. For example, a circle of polyps near the electrodes will contract following a single shock and, after a brief but noticeable delay, another adjacent, often concentric group of polyps will respond. Frequently one of a series of stimuli will not cause more polyps on the whole periphery to contract but only a group of polyps of variable size on one portion of the periphery.
(e) Spontaneous polyp contraction
Spontaneous activity following stimulation is more commonly seen in H. aggregata than in H. echinata. Several seconds after a stimulus, usually in an area near the stimulating electrodes, a single polyp or a group of polyps often suddenly contract again. A wave of contraction can start from such an area which affects many polyps, sometimes even more polyps than had contracted following the original stimulus and occasionally involving all the visible polyps of the colony. Spontaneous activity is especially common following strong or repeated stimuli when waves of polyp contraction often originate from many foci in the colony, and polyps continue partially relaxing and recontracting for many minutes. The same area can initiate several waves. Typically, following strong stimulation, the patches of polyps spontaneously contracting decrease in size and the interval between such contractions lengthens until the colony is again quiescent. Such spontaneous activity indicates repetitive firing in the inter-polyp conducting system.
IV. Species with little polyp co-ordination
(1) Tubularia sp. (probably crocea (Agassiz))
Experiment on this species were done at the Friday Harbor Marine Laboratory. The hydroids were collected from Puget Sound near Tacoma, Washington. The large hydranths have two circlets of tentacles ; a distal row near the mouth with about thirty tentacles and a proximal row just beneath the gonophores containing slightly fewer tentacles. The stalks bearing the hydranths are several centimetres high and grow from a tangled mat of stolons. These stalks are occasionally branched. The hydroids from Washington are larger and have more tentacles than Tubularia crocea from Los Angeles, but otherwise these two forms appear similar and may be the same species.
Tentacles in both the proximal and distal rows show much spontaneity; individual tentacles or groups of tentacles in the proximal row frequently move toward the mouth and tentacles in the distal row move away from the mouth. When a portion of the stalk is pinched with forceps or electrically stimulated, all the tentacles of the distal row are simultaneously moved away from the mouth (Fig. 4), a reaction similar to that described for T. mesembryanthemum by Zoja (1891). A single shock is sufficient to elicit some response, but the degree of tentacle opening is increased by stimulating again before the tentacles return to their resting position.
Pearse (1906) described several responses seen following mechanical stimulation of T. crocea polyps. These included folding of the proximal tentacles about the hydranth, folding of both proximal and distal tentacles about the hydranth, and opening of the distal tentacles. All these responses were seen among the spontaneous activities of the animals used in this study, but only the opening of the distal tentacles could be elicited by stimulating the stalk.
In some cases the opening of the distal tentacles can be produced only by stimulating the stalk near the polyp, but often stimuli applied anywhere on the stalk are equally effective in causing this response. In one case, for example, almost identical stimulus strength-duration curves were obtained when the electrodes were placed 3 mm. below the hydranth and when they were 2 cm. from the hydranth. The chronaxie for electrical stimulation is about 0-4 msec. The latent period between a just supra. threshold stimulus and the beginning of the tentacle response gives a minimum conduction velocity of 3·5 cm./sec. (17°C.) without allowing for delay between the arrival of excitation at the polyp and the beginning of the tentacle response.
In only one instance in approximately twenty-five attempts did more than one polyp respond to stimulation of a stalk, although in all cases the stalk stimulated was only one of many growing from a common mat of stolons. In the one successful attempt, the other polyp which responded was growing from a branch which left the stalk just proximal to the stimulating electrodes. The excitation produced by the stimulus must have proceeded in both directions in the stalk, distally to the main hydranth and proximally to the branch bearing the other responding polyp. Both polyps opened their distal tentacles. Other attempts to stimulate a stalk with a side branch failed to elicit responses from the polyp on the branch. Colonial co-ordination, therefore, seems poorly developed in Tubularia.
Often the stolons of a colony are transparent and appear devoid of living tissue. Jickeli (1883 b) described a similar phenomenon in T. mesembryanthemum, where the coenosarc in the stem degenerates proximal to the hydranth and older members of a colony are joined only by empty tubes of perisarc. Such tissue changes may be the basis for the lack of behavioural communication between individuals in Tubularia colonies.
(2) Syncoryne mirabilis (Agassiz)
Syncoryne colonies are found growing under overhanging portions of rocks near the low tide level at Friday Harbor, Washington. These colonies consist of bushy tufts of stalks and stolons, each tuft containing several hundred polyps.
(a) Mechanical stimulation
Mechanical displacement which is abrupt although quite weak, when applied to the hydranth or merely in the water near the hydranth, causes immediate bending of the polyp toward the source of disturbance (Josephson, 1961). If the hydranth is carefully touched with a glass needle, it bends toward the needle. The angle to which the polyp fiends increases with increasing stimulus intensity, as does the time taken by the polyp to return to its resting position. This relaxation time can vary from a few seconds to over i min. Strong stimulation of the polyp or pinching the stalk bearing the polyp evokes contraction of the hydranth. The polyps are about 1·5 mm. high and can shorten in contracting by 30 %.
(b) Electrical stimuli
A single electric shock to the stalk near the polyp can evoke either polyp bending or contraction. Stimuli just above threshold intensity cause polyp bending; stimuli much above threshold intensity (15−60% above threshold, dependent on the preparation) evoke polyp contraction. Often such responses can only be produced when the electrodes are no more than a few millimetres from the hydranth although on one occasion stimuli applied through electrodes 1 cm. down the stalk from the hydranth were effective. While a single shock can cause polyp bending, the degree to which a polyp bends and the time taken by it to relax can be increased by stimulating again before it has returned to its resting position. The chronaxie for electrical stimuli applied to the stalk is about 0·3 msec.
To investigate colonial co-ordination, hydranth-bearing stalks which had obvious tissue connexions with at least one other hydranth were stimulated. In only one of over twenty such experiments did more than just the hydranth on the stalk respond. In the one exception, a polyp on a branch leaving the main stalk proximal to the stimulating electrodes also responded by bending to each stimulus. In Syncoryne as in Tubulada, there appears to be little colonial co-ordination.
In one experiment, during the determination of a stimulus strength-duration curve, the threshold suddenly increased enormously. Upon examination it was noted that the coenosarc near the stimulating electrodes had parted leaving a 1·2 mm. length of empty perisarc. No strong stimuli had been given to the stalk ; all the preceding shocks had been at most just threshold for the bending response. It appears that under some circumstances tissue in the coenosarc can contract, even though both Citron (1902) and Schulze (1873) failed to find muscular tissue in the coenosarc of Syncoryne. Such contraction may be related to the extremely slow pulsations seen in the coenosarc just behind the growing stolon tip in Obelia (Berrill, 1949) and Clytia (Hale, 1960), where again there is apparently no differentiated muscle tissue.
Thecate hydroids
Withdrawal of several polyps following damage to a single polyp or crushing of a stem has been seen in a number of thecate hydroids including Obelia spp. Because of their usual small size, the thecates are not as favourable experimental animals as many of the athecates, and the details of the conduction in this group have yet to be determined.
DISCUSSION
Colonial co-ordination in hydroids
The stems or stolons of all the hydroid species studied have systems capable of conducting excitation. This confirms the early observations of Zoja (1891), who similarly found spread of excitation in a number of different hydroids. In not all cases, however, is there colonial co-ordination. The frequency of occurrence of inter-polyp behavioural communication in both Tubularia and Syncoryne is quite low. This may be due to tissue changes in the older parts of the stems of these species. In this respect it is interesting that conduction in Pennaria is polarized, progressing more readily distally toward the younger portions of the colony than proximally into the older coenosarc. Excluding Tubularia and Syncoryne, the hydroid species studied are similar in that they form structurally well-organized colonies, and all have systems capable of co-ordinating polyp contraction.
Evidence favours the presence of two inter-polyp conducting systems in Hydractinia echinata, one controlling local and the other colony-wide responses. These systems-are distinguished by the different areas of polyps affected by activity in each, the greater conduction velocity of the through-conducting system as opposed to the local system, and by the different responses of the spiral zooids, becoming more tightly coiled in areas of local activity and lashing during colony-wide responses. The different activities of the spiral zooids could also be explained by multiple firing in the conduction pathways coupled with muscle systems in the polyp having different facilitation requirements. The unitary and all-or-none action of the spiral zooids, however, one lash to one shock, rules against this interpretation. In H. echinata the two conducting systems have nearly the same threshold for electrical stimuli. In those cases where the through-conducting system has the lower threshold, evidence of local activity is masked by the colony-wide response. The fact that through-conduction frequently has a lower threshold than local response further indicates two conducting systems, rather than different types of activity, such as single impulses v. repetitive firing, in a single conducting system. If there are similarly two conducting systems in Podo-coryne, the through-conducting system always had the lower threshold in the colonies investigated. The delayed relaxation of polyps near the electrodes possibly could be explained on the basis of a local system whose effects sum with those of the through-conducted response.
Conduction in coelenterates
The conducting systems of the hydroids investigated fall into one of two functional types : those which are through-conducting, have a sharp, rather constant threshold, are all-or-none with respect to stimulus intensity, and which do not show decreasing response with increasing distance from the point of stimulation ; and those which are local, have no sharp threshold, show dependency of response on stimulus intensity, have responses which decrease in magnitude with greater distance from the stimulated point, are quite variable with regard to distance of excitation spread, and which are labile, the sensitivity and responsiveness decreasing with repeated use. Conduction in the stems of Syncoryne and Tubularia and the through-conducting systems of H. echinata and Podocoryne fall into the first category ; the inter-polyp communication in the other hydroids studied and probably the local response of H. echinata fall into the second. Horridge (1957) found analogous local systems in the perforate corals Goniopora and Parites.
Through-conducting systems are found widely in the coelenterates, controlling luminescence and polyp retraction in Renilla (Parker, 1920b) and beating of the bells of jellyfish (Bullock, 1943), for example. Such systems can be explained on the basis of a nerve net in which all or almost all of the neurons are joined by synapses trans mitting each arriving impulse. The local systems described above, however, are not so’ easily explained.
In discussing the colonial responses of coral polyps, Horridge (1957) proposed two models, the first mechanical, the second mathematical. The mathematical model deals with the effects of stimulating a variable number of conducting units out of a larger population of such units. These units are able to excite one another, but the probability of activity in one unit initiating activity in another is, on the average, less than one. The mechanical model, discarded in part because of its high variability, represents a special case of the mathematical model, that in which only one unit is initially stimulated. Using the second model, Horridge was able to explain the lesser contraction of polyps toward the periphery of the responding area following a single shock on the basis of decreasing density of active units with increasing distance from the stimulating electrodes.
The colonial responses of both coral and hydroid polyps are usually symmetrical. Discussing the conducting system within each polyp controlling such responses, Horridge points out that: ‘It may be inferred that this system acts in a through-conducting manner, effectively as a single nerve (fiber). ‘Such a system may be expected to be all-or-none, and therefore a poor device for differentially responding to changes in the density of active units in its vicinity. A single impulse in a through-conducting system might be just as effective as many simultaneously arriving impulses. Such a system could, however, respond differentially to a temporal pattern of impulses, provided that these impulses are at longer intervals than the refractory period of the conducting system. It could be argued that, because of the existence of fast and slow pathways in a nerve net, a greater density of active units also means a longer period of activity, and polyps are responding to repetitively arriving impulses associated with greater density of active units and not to the density per se. Polyps near the electrodes contract to a wave of excitation before it has travelled far and before impulses have had a chance to become spatially and temporally separated because of different conduction velocities. If it were a temporal pattern of activity associated with density which led to greater polyp contraction, polyps near the electrodes should contract less than those a greater distance away. This, however, is not the case. Postulated changes in the density of active units, then, would not seem to account for the observed lessening of polyp contraction with greater distance from the electrodes.
The properties of the local systems in hydroids which need to be explained are:
(1) the dependency of distance of excitation spread on electrical stimulus strength;
(2) the greater responses of polyps near the point of stimulation than those farther away; (3) the rapid decline of sensitivity and responsiveness; and (4) the inherent variability of these systems. Horridge’s second model has been shown to be inadequate to explain (2) and, further, the change in conduction velocity with different stimulating conditions one expects from this model was not found in experiments on Pennaria. The possibility that changes in conduction velocity occur but were below the limit of detection cannot, however, be excluded.
Greater excitation spreads with increasing mechanical stimulus strength and greater responses near the area mechanically stimulated—properties once thought due to decremental conduction in nerve nets—were convincingly explained by Pantin (193 5) on the basis of repetitive firing and interneural facilitation. It will be shown in the following paper that exactly the same explanation may be applied to electrical stimulation in hydroids and possibly also for some corals. It may be pointed out here that there is behavioural evidence for repetitive activity following electrical stimulation in hydroids. The spontaneous activity often seen following stimulation in Hydractinia is certainly an example, and the patchy polyp contraction seen following a single shock in this genus can also be construed in this way. The variability of local responses in hydroids can be explained on the basis of a non-linear relation between stimulus strength and repetitive firing, and of different tendencies to repetitive activity in different preparations or in the same preparation at different times. The decrease in sensitivity and responsiveness is probably due to a decreased tendency to repetitive activity, although changes in the effective conductive path lengths because of changes in interneural junction properties may also be involved.
Responses of individual hydroid polyps
The conducting systems in the stems or stolons of hydroids usually initiate symmetrical polyp responses : contraction of the hydranth and/or simultaneous movements of the tentacles. The spiral zooids of Hydractinia and Podocoryne, however, always show asymmetrical responses and the polyps of Syncoryne bend to one side following just supra-threshold stimulation of the stalk. To account for the rapid bending of the latter species, it has been postulated that the longitudinal musculature of the hydranth is divided into parallel fields which can contract independently (Josephson, 1961). It seems likely that conduction in the stem of Syncoryne is predominantly longitudinal as Parker (1917) has shown it to be in the stalk of another hydroid, Corymorpha. The asymmetrical response of Syncoryne can then be explained as due to just suprathreshold shocks exciting only one of a number of parallel, longitudinal conducting pathways, and the induced excitation travelling up one side of the stem and exciting just one portion of the longitudinal musculature of the hydranth.
Stimulating the stalk of Tubularia initiates opening of the distal tentacles, indicating the presence of conducting pathways from the stalk through the hydranth to this circle of tentacles. The spontaneous symmetrical elevation of the proximal tentacles is evidence for the presence of a conducting system able to co-ordinate the activities of these tentacles. Thus the polyps of Tubularia probably have two concentric conducting systems between which there is no evidence of interaction.
ACKNOWLEDGMENT
I would like to thank the personnel of the Hawaii Marine Laboratory, the Friday Harbor Marine Laboratories, and the Marine Biological Laboratory at Woods Hole for their assistance during the portions of this study done at these institutions. I would also like to thank Dr T. H. Bullock for advice offered during the course of this work. Most of this study was done during the tenure of a National Science Foundation predoctoral fellowship. Additional financial aid was given by a grant (B21) to Dr T. H. Bullock from the National Institute of Neurological Diseases and Blindness. Figures 1 and 3 were kindly prepared by Mrs. J. L. Kavanau.