1. The muscle organs recently described by Alexandrowicz in the tails of Homarus vulgaris and Palinurus vulgaris, have also been found to be present in the crayfish (Cambarus clarkii) and the rock lobster (Panulirus interruptus), and a study of them in the latter animals has been undertaken.

  2. The position of these units within the abdomen of the crayfish differs from their position in the lobsters. In almost all other respects, however, there is substantial correspondence in their morphological features.

  3. The organs are easily isolated and it is found that stretching them is accompanied by a discharge in one of the axons supplying each unit. There are two receptors in each half-segment of the abdomen and their responses show considerable differences. One has a very low threshold and can be sustained in continuous discharge for several hours. The threshold of the other is high and this receptor usually adapts completely to the most extreme stretch in less than 1 min.

  4. A phenomenon which is designated ‘over-stretch’ has been observed. All-but-maximal stretch causes a reversible cessation of the previous high-frequency discharge. The discharge can be restored by slight release of tension, but even if the tension producing ‘over-stretch’ is maintained for a sufficient length of time, discharge will resume spontaneously. The possibility that this spontaneous return is due to a slow rupture of the receptor tissues is not excluded.

  5. The muscular regions of the organs are not necessary for their sensory function.

  6. It has been possible to show that stimulation of the nerve trunk can induce contraction in these muscular regions and that such contraction results in discharge in the sensory axons.

  7. The unit with the low stretch-threshold responds to low concentrations of ACh and this effect is potentiated by eserine. In concentrations above 10−6 before eserinization and above 10−8 after eserinization, ACh consistently initiates rhythmic responses in relaxed units and augments the discharge from those under tension.

  8. ACh affects the sensory mechanism directly; the muscular regions of the organs are not necessary for its action.

  9. Atropine inhibits the ACh excitation, but only increases the response of the receptors to stretch. Eserine, except in high concentrations, has no consistent effect on the normal stretch discharge.

  10. The possibility that ACh has a role in the normal receptor mechanism is discussed, but the question is left open.

Alexandrowicz (1951) has recently described organs in the abdomen and thorax of the lobsters, Homarus vulgaris L. and Palinurus vulgaris L., which he considers to be stretch receptors. Their structure is reminiscent of that of the (vertebrate) muscle spindle. These organs are few in number and were found only in the dorsal regions of the body. Two pairs in each of the six abdominal segments and a small number (perhaps only four) in the thorax may constitute the animal’s entire complement of these receptors.

The organs of the abdomen are attended by an interesting, if complex, innervation. One of the several axons supplying each unit has its cell body in the periphery, in close apposition to a specialized area which Alexandrowicz calls the ‘intercalated tendinous region’. The cell sends short dendritic processes to this area and is considered to have a sensory function ; its axon would conduct impulses centripetally in response to stretch. Also observed were fibres typical of crustacean motor axons. They branch extensively along the muscular regions of the unit, ‘giving off abundant ramifications ending amidst the myofibrils’. Other fibres, which seemed to branch and innervate the tendinous region as well as the muscular areas, were described and referred to as ‘accessory’ nerves.

It seemed of special interest to study the physiology of these crustacean receptors in view of the physiological deductions made by Alexandrowicz and of the recent intensive investigation of the vertebrate muscle spindle, both anatomically (Barker, 1948) and physiologically (Lecksell, 1945; Katz, 1949, 1950a, b;Kuffler, Hunt & Quilliam, 1951; Hunt & Kuffler, 1951a, b; Hunt, 1952a).

Several attributes of these receptors contribute to making them more amenable to experimentation. Unlike the vertebrate spindle, which lies within the body of a muscle, these are separate and external organs, free from other musculature and having only nervous connexions. They are thus rather easily isolated and, if necessary, removed from the animal. Carefully dissected preparations, if perfused occasionally with fresh solution, maintain their ability to respond to stretch for long periods, often for over 12 hr.

The present paper aims to present a first approach to the study of some of the reactions of these units. Several aspects of the present work will be treated in more detail in subsequent research, but since these organs represent one of the more easily obtained single-unit sensory preparations, and since their exposed position allows their being bathed directly in solutions of known drug or ion concentration, it has seemed appropriate at the present time to follow our short preliminary note (Wiersma, Florey & Furshpan, 1952) by a more detailed description of technique and results.

Most experiments were performed with the fresh-water crayfish, Cambarus clarkii Girard, and some, as noted in the text, with the lobster, Panulirus interruptus (Randall). Several methods of preparation have been employed. One dissection has been found best for most purposes, however, and although similar to one given by Alexandrowicz, it will be described in view of some anatomical differences between lobster and crayfish.

The tergal parts of all the abdominal segments are dissected free from the rest of the abdomen in the form of a single strip containing all the extensor musculature. It is found convenient to hold the strip, ventral surface up, with a bulldog clamp which can be attached with a spring clip to the side of the perfusing vessel. The clamp is attached to the proximal end of the first abdominal tergite. Varying amounts of the dorsal attachments of the flexor muscles remain, and these are removed from the half-segment to be prepared. After severing the nerve branch supplying it (Text-fig. 1, NPM), the more ventral layer of extensor musculature (profundis lateralis and medialis) is cleared, care being exercised to cut nerve branches which connect the thicker of the two receptor organs with the profundis musculature. Now the superficial extensor muscles are exposed and the receptor units can be found by carefully pulling aside the medial border of the medial superficial muscle, which partially covers the units (Text-fig. 1, right side). Removal of the tergal strip necessarily severs the nerve trunk which supplies the organs, but a sufficient length of nerve remains to allow for the placement of electrodes. If the organs are to be excised this section of nerve may be freed along its entire length by carefully removing the remaining extensor musculature to which it sends numerous small branches. If the units are to be left within the tergal strip, however, only the medial and not the lateral superficial muscle is removed, as the latter then serves to protect the units from being stretched by pull on the nerve. This same trunk also sends at least one large branch (Text-fig. 1, NCS) dorsad to tactile receptors in the tergum, and this branch is always cut since impulses from these receptors otherwise confuse the recording.

To test the response to stretch, the end of the cut nerve trunk was lifted above the surface of the perfusion solution on a micromanipulated platinum electrode. The other electrode was immersed in the surrounding fluid or attached to the clamp which holds the strip. Most of the recordings were taken from organs of the second segment. A thread was attached to the posterior part of the strip so that pull upon it caused movement of the third segment on its articulation with the second similar to that in natural flexion.

After suitable amplification the impulses were led into an oscilloscope, or in some experiments concerned only with low-frequency response, into a power amplifier which drove a signal magnet writing on a two-drum kymograph. With this device long continuous records have been easily and inexpensively obtained.

Preparations were dissected in, and perfused with, van Harreveld’s crayfish solution (van Harreveld, 1936). Some of the preparations for testing motor response were stimulated in a solution containing one and one-half times the normal amount of potassium and no magnesium (Waterman, 1941). To study the effect of drugs on the receptors, drops of the appropriate solutions were applied to either stretched or unstretched preparations, and the concavity of the tergal segment served conveniently to retain the added solution for a short period of time. If it was desired to have the drugs act for a longer time, the entire vessel within which the strip was held was filled with the solution. Concentrated standard solutions of drugs were prepared in van Harreveld crayfish solution containing no bicarbonate, and were diluted with buffered solution prior to being used.

Anatomy

A short comparative study of the anatomy of the crayfish and lobster organs was made. The anatomical data for the lobster are, for the most part, from Alexandrowicz ; but several preparations of the California rock lobster, Panulirus interrupts (Randall), were made for further comparison. In preparations dissected as described above, the two receptors are easily seen under the binocular microscope without the aid of staining. It was necessary to stain, however, in order to see the nerves and ganglion cells. For this purpose rongalite-reduced methylene blue (Pantin, 1948) was found quite satisfactory.

In the crayfish as in the lobsters, the two organs of each half-segment are of unequal length and thickness. The difference in thickness is much more pronounced in the crayfish than in the lobster, but in both the thicker organ is also the longer. As is the case in the lobster, the larger of the two organs exhibits finer crossstriation of the muscular regions. As already indicated in the section on methods, there is a surprising difference between crayfish and lobster in the position within the abdominal segment. In all of the lobsters these organs are situated in the space between the lateral and medial superficial extensor muscles. This places them lateral to the superficialis medialis, whereas the organs of Cambaras lie medial to this muscle (cf. Text-fig. 1). Another difference between the two animals exists in the amount of connective tissue surrounding the organs. Whereas in Panulirus the two organs of a half-segment are almost entirely enclosed within an encapsulating sheath, those of the crayfish are separate and rarely bound together by any external connective tissue except in the region of the nerve entrance.

In both animals it was possible to show the presence of ganglion cells sending dendrites to a differentiated region. Fibres typical of motor axons, whose branches ramified along the length of the muscular parts, were also seen; and in one case fibres were observed which seemed to fit Alexandrowicz’s description of accessory nerves. Thus, despite the differences, the similarities seem sufficiently manifest to justify carrying over Alexandrowicz’s terminology to the crayfish. The longer and thicker organ, having the finer cross-striations, will be called RM2, and the other, RM1.

Response of the receptors

Flexion of the tail strip is followed by trains of action spikes in the trank innervating the RM’s (muscle receptors). If the precaution of severing the branch to the receptors in the exoskeleton has been taken, spikes of not more than two heights are observed, each falling into a regular rhythm (Pl. 1, fig. 1). The two rhythms are easily distinguishable from one another, since one adapts very slowly and the other by comparison, rather rapidly. A marked difference in stretch-threshold for eliciting the two different rhythms is also apparent. The slowly adapting rhythm follows much slighter stretch and, in fact, is occasionally recorded in the absence of any externally applied stretch. It is therefore easily obtainable by itself (Pl. 1, fig. 3). The regularity of this rhythm, as well as the constancy of the spike height, can leave no doubt that it arises from a single receptor unit. This slowly adapting unit can maintain continuous discharge for periods exceeding an hour, and in one experiment, where the preparation was constantly perfused at a slow rate, response to a sustained stretch did not cease until 312 hr. after its initiation. As intimated, the other rhythm arises from a unit which has a higher threshold and is fairly rapidly adapting. It is usually necessary to flex the isolated tergal strip to an extent which approaches the maximum possible in order to induce discharge in this receptor unit Even on maximal flexion the response usually lasts for about 30 sec. and seldom longer than 1 min. By the standards of tactile receptors, a discharge which is maintained for a minute is long-lasting. In the crayfish, however, where this unit is contrasted with another which is capable of sustained response for several hours, it has been convenient to refer to it as the ‘fast-adapting’ receptor unit.

There are, then, two distinct types of rhythm which can be recorded from the nerve trunk to the RM’3. Histologically two types of organs, RM1 and RM2 have been shown to be present, and the question arises which receptor gives rise to which type of rhythm. If a moderate amount of stretch is applied to a preparation so that only the slow-adapting unit is active and then the terminal attachments of RM2 (the longer-thicker organ) are severed, no appreciable change in the frequency of the discharge ensues. But if the attachments of RM1 are then cut, all discharge ceases. If in other preparations, RM2 is left intact, and only the attachments of RM1 are cut, some response, though always diminished in frequency, may follow upon gentle stretching. This is due to connective tissue binding the two organs together in the region of the ganglion cells, and this response can be shown to originate in RM1 despite its severed attachments. Further, it can be shown more directly, by the converse experiments, that the fast-adapting rhythm originates in the organ RM2. It has recently been found possible completely to isolate and remove one or the other of the RM’s from the tergal strip, and to study it in isolation. These experiments leave no doubt concerning which type of rhythm arises from which receptor. In most preparations the nerve-action potentials recorded from RM1 were smaller than those of RM2 (Pl. 8, fig. 1 A). In the rock lobster a similar pattern of two spike rhythms was observed. Because of the connective tissue enclosing the two receptor muscles in this animal, no cutting experiments were performed ; but, as in the crayfish, the more slowly adapting discharge with the lower threshold consisted of spikes of smaller size in almost all preparations (Pl. 8, fig. 1B).

Only one of the nerve fibres supplying each receptor unit has its cell body in the periphery and is thus probably the sensory axon. Further evidence in support of this interpretation was obtained physiologically. The most direct and conclusive method would involve splitting the receptor nerve trunk and recording from single axons. Successful isolation of single axons has not yet been accomplished on account of the extremely tough connective tissue surrounding the nerve fibres in the periphery, and a less direct method had to be employed. The muscular elements of the receptor unit receive branches from all the attending nerve fibres with the exception of that one whose cell-body lies peripherally and whose terminals are restricted to the narrowly circumscribed tendinous region. It was found that crushing the muscular parts of an organ with a fine forceps did not alter the response to stretch. But if the approximate area innervated by the dendritic processes of the ganglion cell was probed with a sharp needle, a high-frequency discharge resulted and was usually followed by a loss of the receptor’s ability to respond. Both of these facts are in full accord with the assumption that the fibre with the peripheral cellbody is indeed the sensory axon.

It was then found feasible to clamp the tendinous region with two pairs of screwlocking forceps, one on either side, so that very little muscle tissue was interposed. One pair of forceps was secured and the other moved by a rack and pinion device, thus stretching the receptor region almost exclusively. Responses were readily obtained in this manner, and it was found that the characteristic differences in the responses of the two receptors were not altered with this method of stretching. The comparatively rapid adaptation was still displayed by RM2, and it is thus a property of the receptive mechanism or is an attribute of its nerve fibre (Gray & Matthews, 1951), rather than a function of its attachment or of the viscous-elastic properties of its in-series muscle elements.

These differences in the response to stretch of RM1 and RM2 are paralleled by disparities in the response of the receptors to various drugs and to alteration of the ionic content of the perfusing solution. Differences in the effects of drugs will be noted in a subsequent section, but some differences in the responses to changes in calcium and potassium ion will be noted here.

The increase in excitability caused by a reduction in the external calcium concentration is well known and has been specifically studied for receptors by Talaat (1933). Elimination of this ion from the external environment of these muscle receptors also produces dramatic increases in their excitability. Shortly after their immersion in a calcium-free solution, the receptors begin to fire spontaneously. If allowed to remain in this medium, the frequency of the response continuously increases until a point is reached at which impulses seem to drop out of an otherwise regular train. The dropping out continues and eventually the spontaneous activity, as well as the ability to respond to stretch, disappears. It has been consistently noted, however, that the response of RM2 to the decreased calcium ion is delayed with respect to that of RMV. Often impulses have started to drop out of discharge from RM1 before RM2 starts to fire spontaneously. Differences have also been found between the responses of the two receptors to increases in the external potassium concentration. Results with this ion were not as consistent as those obtained with calcium, but a general pattern was observable. The high-potassium solutions were made by adding crystalline potassium chloride to normal crayfish solution. Potassium concentrations of three and four times normal caused a short (c. 30 sec.) decrease in excitability so that a moderate discharge resulting from stretch would be inhibited. This phase was followed by one of increasing excitability, and with higher concentrations (4 x) spontaneous activity would ensue in RM1. The increase in excitability often continued, and a dropping out of the spontaneous impulses and a cessation of discharge similar to that described above for calcium lack would eventually occur. Such spontaneous activity, and the associated behaviour, was never seen for RM2 with these potassium concentrations.

The dropping out of impulses followed by complete cessation of spontaneous discharge found in calcium-free solutions resembles, at least superficially, another phenomenon noted in the behaviour of RM1. If stretch on this receptor is continuously increased, a point near the extreme limit of flexion will be reached at which impulses start to drop out. Then almost immediately, the high-frequency discharge stops completely. Following a slight decrease in tension, the discharge is again resumed. Recovery will also occur if the receptor merely remains in this over-stretched state for some time. In one such experiment the receptor was overstretched and maintained in this condition until, at the end of approximately 4 min. discharge was resumed. Then tension on the receptor was increased further until the discharge again ceased, and again recovery occurred after 4 min. This was repeated for three additional trials with recovery times of 4, 412 and 5 min. The tensions needed to produce over-stretch are rather high, and it is possible that these recovery effects find their explanation in the slow rupture and slipping of the inseries elements of the organs. Some of the other attributes of the over-stretch phenomenon will be mentioned, although it is difficult to evaluate their significance without further study. The frequency of discharge immediately prior to cessation in over-stretch is dependent upon the rate of stretching; the higher the rate, the higher the frequency. This frequency is also dependent upon the state of fatigue of the preparation ; the less fatigued, the higher the frequency. Over-stretch occurs at lower tensions, and is preceded by lower frequencies if the external potassium concentration is higher than normal.

Motor innervation

Because of the ubiquity of multiple innervation and peripheral inhibition in crustacean motor systems, it is usually necessary to obtain single axons before an effective study of these systems can be undertaken. Numerous attempts to induce contraction in the receptor musculature by stimulating the whole nerve trunk have been made. These have been unsuccessful, with a few exceptions, despite the apparent good condition of the animals and the care with which the dissections have been executed. This lack of success may, perhaps, be attributable to the above explanation (an interaction of motor and inhibitory stimulation), but attempts to isolate single axons have also been unsuccessful, as the connective tissue band in which the axons lie is extremely tough. Nevertheless, the few preparations which have given contraction on stimulation of the nerve trunk, demonstrate the existence of a functional motor system. It has also been shown that, as in the vertebrate spindle, this motor system is capable of altering the response to a given stretch. Pl. 8, fig. 2 shows several superimposed sweeps of the oscillograph. The recording electrode was placed on the nerve not far from the stimulating electrodes, and thus the stimulus artefact and deflexion of the base-line are rather large. The impulses which were set up at the stimulating electrodes and travelled directly to the recording electrode are included in the stimulus artefact and are not seen in the picture. The musculature of both organs gave visible contractions and impulses in both sensory axons are seen in the record.

There are several other muscle systems whose contractions interact with the receptor organs and affect their sensory discharge. It is known that contraction of the flexor muscles will stretch the receptors, and it is probably this system which is capable of imposing the widest range of tensions on the organs. Contraction of the extensor muscles would also be expected to affect the receptor’s discharge, and the following experiments were performed to see if these expected effects were present. A tergal strip was removed as usual, the remnants of flexor musculature cleared, but no further dissection was performed. Then the nerve trunk of the right half of the second segment was picked up on a recording electrode and that of the left side of the same segment on two stimulating electrodes. A base-line discharge was set up by moderate flexion of the strip and then the stimulating electrodes were activated at a frequency of about 30/sec. A visible contraction of the extensors of the left side was accompanied by a decrease or cessation of the discharge from the receptors of the right side. It is presumed that contraction of the extensors of the same side of the same segment would inhibit discharge of the receptors even more readily. Experiments performed in a similar manner demonstrated that contraction of the extensor muscles of the right side of the third segment had an opposite effect and augmented discharge from receptors of the right side of the second segment. It can be seen that very complex interactions of at least four different muscle groups are possible, all of which affect the tension on the receptor terminals, and these are considered further in the discussion.

Pharmacological reactions of the receptors

Drops of the appropriate solutions were added to the half-segment of the tergum containing the dissected receptors. At the end of 3 sec., in most experiments, the response was recorded. After the record was taken the tergal strip was thoroughly washed and then the perfusion vessel allowed to drain in preparation for addition of the next test solution. The mechanical agitation caused by the application of drops of the drugs was frequently sufficient to induce a small and transitory discharge in the more sensitive RM1. Most of the effects with which we have been concerned, however, have been so large and long-lasting that confusion could seldom arise between mechanically caused and drug-induced responses. Where doubt existed drops of physiological solution, applied in the same manner, were used as controls.

A large number of drugs have been tested on this preparation. Of these acetylcholine gave the most interesting results, and only the effects of this drug and of a few other drugs which seemed most closely bound up with the ACh-induced response will be reported here. All drug concentrations are given in grams per millilitre of solution.

Acetylcholine

When applied in concentrations of 10−6 or greater, ACh consistently induces spontaneous impulses in RM1, the slowly adapting receptor. RM2 has about a hundred-fold lower sensitivity to ACh. In a number of fresh preparations solutions of 10−9 (and, in a few, 10−10) ACh initially enhanced the discharge from slightly stretched preparations (that is, with a background discharge of about 10/sec.) as shown in Pl. 8, fig. 3. Subsequent applications of the same or even higher concentrations (10−8 and 10−7) often gave no increment. Because of this lack of correlation between concentration and response the results obtained with these low concentrations are rather difficult to interpret. The frequency of discharge following application of the higher concentrations (106 to 10−3) was always related to concentration, and a typical record is shown in Pl. 8, fig. 4 A. The numbers inserted refer to the concentration of ACh added. The variation in height of the nerve-action potentials from one record to another is not significant but merely reflects slight differences in the level of the surface of the solution in the tergal concavity and thus in the point at which the active lead from the nerve is located. These responses were obtained from a preparation of RM1 with intact terminal attachments but to which no external stretch had been applied.

Hunt (1952b) reports that ACh exhibits a large excitatory effect on the vertebrate spindle, but he concludes that it is the contraction of the intrafusal muscle fibres which causes the heightened response rather than any direct effect of the ACh on the sensory terminals themselves. Two types of experiment indicate that this is not the case, however, in the crayfish organs. In the first, the terminal attachments of the receptor were severed so that it floated loose in the ACh solution being tested, attached only by its nerve to the tergal strip. This usually resulted in only a slight decrease in the ACh-sensitivity of the preparation. For example, in one experiment an eserinized receptor (RM1) gave a slight response to 10−8 ACh prior to cutting its ends free from its attachments. After this operation ACh in a concentration of 5 × 10−8 was needed to give a similar response. It is felt that this slight decrease in sensitivity to ACh reflects the decrease in the level of excitability following removal of the subliminal stretch to which the organ is subject when in situ. This is corroborated by a number of preparations which have discharged spontaneously at a low level without the application of external stretch. The majority of preparations of RM1, in fact, are very close to firing, even though the tergal strip is fully extended, as only the slightest amount of flexion is necessary to initiate discharge. A second type of experiment also indicates that ACh does not produce its excitatory effect on this organ by first initiating contraction of the receptor musculature. RM1 was completely isolated with the nerve trunk and clamped with forceps (see p. 141) on either side of the intercalated tendinous region. Fixed in this manner there is only a small amount of muscle tissue remaining in series with the receptor region, and this must certainly be damaged by the clamping procedure. But even when so clamped, the sensitivity of RMX to ACh was essentially unchanged.

Eserine

Despite negative effects obtained in a few preliminary experiments (Wiersma et al. 1952), eserine (physostigmine salicylate Merck) has since been found to enhance the ACh response of all preparations tested, to a rather marked extent, when they are bathed in this drug for some time (3-5 min.). It has recently been found that high concentrations of eserine can also show a depressant action on ACh excitation. If drops of 10−4 eserine are followed immediately by ACh, a much smaller excitatory effect is obtained than if a washing is interposed between the applications of the two drugs. We have not studied this effect further, since it is not directly concerned with the problem in hand, but it might be that in this direction lies an explanation of the preliminary negative results.

Pl. 8, fig. 4E shows some responses of the same preparation of RM1 as is shown in Pl. 8, fig. 4A, but between the times of the two sets of recordings the receptor had been bathed in eserine 10−5 for 3 min. No external stretch was applied, and the numbers inserted again refer to the concentration of the single drops of ACh which were added 3 sec. prior to the taking of each record. It can be seen that there is an approximately ten-fold increase in the sensitivity of the eserinized receptor to ACh. Similar results have now been obtained in a large number of experiments. After being bathed in the lower concentrations of eserine (10−5, 10−6), the receptor shows no noticeable change in its sensitivity to stretch stimuli. Higher concentrations (10−6), however, usually induce a low-frequency discharge in the absence of externally applied stretch. Once obtained, this spontaneous discharge is difficult to suppress and may persist after a number of washings.

Atropine

Many drugs, when applied to the receptors in high concentrations, induce effects which may not be related to their more typical and specific actions. In fact, one is hard pressed to find a compound which in high concentrations does not have some effect on the response of this naked receptor ; and, especially in the invertebrates whose pharmacology is less well defined than that of the vertebrates, it is difficult to decide where the boundaries between ‘typical’ and ‘unspecific’ effects should be drawn. This difficulty is well illustrated by the effects of atropine. When this drug is applied in a concentration of 10−4, it causes considerable augmentation of response in discharging receptors (Pl. 8, fig. 5 c, d), and can initiate spontaneous activity in resting ones. Atropine 10−5 also has a slight excitatory effect. But more striking is the effect of the atropine on the ACh-induced excitation. Pl. 8, fig. 5 a shows the response of a preparation of RM1 to a slight stretch. Pl. 8, fig. 5 b shows the response of this same receptor three seconds after the addition of one drop of ACh 10−4. After washing for a short time the discharge (Pl. 8, fig. 5 c) returns to its original level. Then atropine 10−4 is added and the excitation which it induces is shown in record 4d. Subsequent addition of another drop of ACh 10−4 causes only the slight increment of discharge (over that induced by atropine) seen in Pl. 8, fig. 4E. After washing ACh 10−4 is again capable of inducing the same large response (Pl. 8, fig. 5f) as prior to the administration of atropine. These effects of atropine were consistently observed in a large number of preparations, and in no case did it cause a depression of the normal stretch response.

These organs have thus proved to be receptors responding to the stimulus of stretch. This fact and several other facts were correctly predicted by Alexandrowicz. He also foresaw that the fibres ramifying along the receptor musculature subserved a motor function, and that the receptor designated as RM2 would have the higher threshold. Not only was the predicted difference in threshold found, but further, the two receptors were seen to have markedly different adaptation rates. RM1 adapts very slowly, has a very low threshold, and in some instances discharges even when the tergal strip is fully extended. It can maintain both very low (about 2/sec.) and fairly high (over 200/sec.) frequencies of discharge and can do so for long periods of time. RM1 may thus be termed a tonic receptor, for it is well suited to transmit continuous information about the degree of flexion of each segment of the abdomen. RM2, on the other hand, fits the definition of a phasic receptor. Its high threshold and fairly rapid adaptation restricts its response, so that in the whole animal it probably fires only during rather extreme tail flexion as is found in the swimming reflex. Alexandrowicz also felt that it would be involved in this reflex, and further conjectured that RM2 might initiate impulses which inhibit the powerful tail flexors. It seems to us more likely that impulses from this receptor operate to induce reflex contraction of the extensors so that the tail is readied for its next flexion.

A knowledge of all the functional connexions of the motor nerve fibres innervating the receptors will certainly facilitate the formulation of the role of the receptors in reflexes, in view of the complicated interactions which could conceivably exist in such a system. Contraction of the RM’s has been shown to facilitate discharge ; but if these organs have been phylogenetically derived from the extensor muscles and share with them one or more motor axons, it would be difficult to state, a priori, how the combination of in-parallel extensor-muscle contraction and inseries receptor-muscle contraction might affect the discharge. There is, in addition the possibility of interaction between the receptors of one segment with the muscles of adjacent segments. Both of the receptors have one attachment on an articular membrane which connects tergites of adjoining segments (see Text-fig. 1), and RM2 has its other attachment at the anterior tendon of the medial superficial muscle of the immediately posterior segment. Thus contraction of the extensor muscles of adjacent segments would be expected to increase discharge from the receptors, and this was demonstrated experimentally. Briefly reviewing the possible interactions of the abdominal receptors with other musculature of the tail, it can be seen that contraction of the flexor muscles will tend to increase discharge from these receptors. Contraction of the in-series muscle elements of the receptors themselves can also increase discharge frequency, whereas contraction of the extensor muscles of either side of the same segment will diminish the response. And lastly, contraction of the extensor musculature of adjoining segments will augment discharge. Deduction of the pattern of the discharge for any natural movement of the tail cannot be made until it is known what combination of these various motor systems may be activated simultaneously.

It is interesting to note that the difference in response of RM1 and RM2 to various conditions is not solely a function of the viscous-elastic properties of their in-series elements or the places of their terminal attachment parts. There must be in addition an inherent difference in their sensory mechanisms, as shown by their reactions to drugs and by the reactions to stretch of their isolated intercalated areas held in forceps. At present we are not yet able to judge the relative importance of these factors in the overall responses obtained from the organs in situ.

A comparison between these organs and the spindle of the vertebrates suggests itself. Alexandrowicz has already undertaken such a discussion, but since the appearance of his paper the work of Kuffler et al. (1951), and Hunt & Kuffler (1951 a, b) on the function of the small nerve fibres in the ventral roots of cats has been published. These investigations suggest the nature of the augmentation of afferent discharge upon small-nerve stimulation first observed by Leksell (1945). Their evidence indicates that the small-nerve fibres are motor axons for the intrafusal muscle fibres, and it is by contraction of this spindle musculature that the increase in discharge is effected. A similar mechanism is now seen to exist in the crustacean muscle receptors. Although it has not been possible to split the main nerve trunk, it has been demonstrated that when stimulation of the whole nerve results in visible contraction of the receptor muscle elements, one or several impulses can be recorded from the sensory axons following each stimulus with a delay of about 10 msec. Leksell discusses the possibility of a direct facilitation of the sensory mechanism by a depolarization induced by the efferent impulses. Although it now seems that this is not the case in the spindle, the possibility of such an additional facilitating mechanism in the RM’s cannot yet be excluded. This, as suggested by Alexandrowicz, might be the function of the axons that he has named the ‘accessory’ fibres, but no evidence concerning this point has yet been found.

With respect to the response of these organs to ACh, several previous reports of a sensitivity of afferent endings to this drug are in the literature. Coon & Rothman (1939 a, b) found that ACh, in an optimal concentration of 1:40,000 (as well as nicotine, 1:100,000 and α-lobeline, 1:1,000,000), can initiate axon reflexes, causing transient pilomotor and sweat responses. Concentrations higher than optimal of ACh or nicotine only inhibit further response to the optimal concentration, and the pilomotor response, at least, is not blocked by atropine. ACh can, therefore, be said to exhibit nicotine-like effects. Brown & Gray (1948) have demonstrated a similar nicotinic action of ACh in initiating spontaneous activity in nerves of the mesentery of the cat and the skin of the dog. By the elimination of other known possibilities they deduced that the recorded response must have been set up in the sensory endings by a direct action of ACh at these sites. Here, as in the pilomotor axon-reflex responses, atropine had no blocking action, and inhibition was obtainable by prior administration of higher doses of the ACh (or nicotine). This inhibition of the ACh response was not accompanied by any apparent loss of sensitivity to the stimulus of pressure and they concluded that ACh is not involved in the normal sensory function of the receptors concerned. Hellauer (1950) found that 1 % ACh solutions, dropped on the cornea of the eye, caused sensations of pain; and Umrath (1951) reports that in sensitive persons, ACh 10−5 can be felt as pain.

With the crayfish muscle receptor, RM1, however, it has been possible to demonstrate directly that ACh, in low concentrations, can initiate discharge in some part of a receptor mechanism. It has been shown that ACh 10−6 is usually a sufficient concentration to initiate spontaneous activity when applied to the isolated receptor region of RM1. If this preparation has been previously sensitized with eserine, ACh in concentrations as low as 10−8 is able to produce a definite response. Not enough is known of the pharmacology of the Crustacea to assess the validity of a distinction between nicotinic and muscarinic actions of ACh in the crayfish. It was observed, however, that the ACh response could never be blocked by increasing the concentration of this drug; atropine was seen to be an effective inhibitor of the ACh excitation; and a marked potentiation was obtained by pre-treatment with eserine. If it cannot be concluded with certainty that a muscarinic action of ACh is evident here, it is at least noteworthy that the characteristics of the response differ in these respects from those previously reported for vertebrate receptors (see above).

Although atropine was able to inhibit the ACh excitation, it demonstrated no such action against the normal response of the receptor to stretch, but, in fact, its application was usually followed by an augmentation of the discharge. In order to maintain the hypothesis that ACh acts here as a normal transmitter substance, it is necessary to invoke some mechanism like the following. The terminal surfaces of the ganglion cell dendrites would be enclosed in compartments within which the ACh would be released during stretch. Neither the added ACh nor atropine is able to penetrate the pockets, and these drugs have their entire action on the cell or dendrite surfaces external to the compartments. There is, as yet, no histological evidence for the separation of the terminal dendrite surfaces from the rest of the neuron. If one is dissatisfied with the construction of such an hypothesis, one is left with at least two possible alternatives. The first is that ACh is not in any way involved in the normal mechanism of these crayfish muscle organs and their response to it is merely fortuitous. The second takes into account the sensitivity of the receptor to such low concentrations of ACh and the potentiation of this effect by eserine. ACh would have a role in the normal receptor mechanism as a regulator of the afferent discharge, rather than being the direct cause of its initiation. This latter hypothesis has the advantage of making compatible all the observed effects of the various drugs; but, as in the case of the others, no direct evidence can be brought to its support at present.

Alexandrowicz
,
J. S.
(
1951
).
Muscle receptor organs in the abdomen of Homarus vulgaris and Palinurus vulgaris
.
Quart. J. Micr. Sci.
92
,
163
99
.
Barker
,
D.
(
1948
).
Innervation of the muscle spindle
.
Quart. J. Micr. Sci.
89
,
143
86
.
Brown
,
G. L.
&
Gray
,
J. A. B.
(
1948
).
Some aspects of nicotine-like substances and their relation to sensory nerve endings
.
J. Physiol.
107
,
306
17
.
Coon
,
J. M.
&
Rothman
,
S.
(
1939a
).
Nature of the pilomotor response to acetylcholine
.
Proc. Soc. Exp. Biol., N.Y.
,
42
,
229
31
.
Coon
,
J. M.
&
Rothman
,
S.
(
1939b
).
Nature of a sweat response to drugs with nicotine-like action
.
Proc. Soc. Exp. Biol., N.Y.,
42
,
231
3
.
Gray
,
J. A. B.
&
Matthews
,
P. B. C.
(
1951
).
A comparison of the adaptation of the pacinian corpuscle with the accommodation of its own axon
.
J. Physiol.
114
,
454
64
.
Hellauer
,
H. F.
(
1950
).
Sensibilität und Acetylcholinegehalt der Homhaut bei verschiedenen Tieren und Menschen
.
Z. vergl. Physiol.
32
,
303
10
.
Hunt
,
C. C.
(
1952a
).
The effect of stretch receptors from muscles on the discharge of motoneurons
.
J. Physiol.
117
,
359
-
79
-
Hunt
,
C. C.
(
1952b
).
Drug effects on mammalian muscle spindles
.
Fed. Proc.
11
,
75
.
Hunt
,
C. C.
&
Kuffler
,
S. W.
(
1951a
).
Further studies on efferent small nerve fibres to mammalian muscle spindles. Multiple spindle innervation and activity during contraction
.
J. Physiol.
113
,
283
97
.
Hunt
,
C. C.
&
Kuffler
,
S. W.
(
1951b
).
Stretch receptor discharges during muscle contraction
.
J. Physiol.
113
,
298
315
.
Katz
,
B.
(
1949
).
The efferent regulation of the spindle in the frog
.
J. Exp. Biol.
26
,
201
17
.
Katz
,
B.
(
1950a
).
Action potentials from a sensory nerve ending
.
J. Physiol.
111
,
248
60
.
Katz
,
B.
(
1950b
).
Depolarization of sensory terminals and initiation of impulses in the muscle spindle
.
J. Physiol.
111
,
261
82
.
Kuffler
,
S. W.
,
Hunt
,
C. C.
&
Quilliam
,
J. P.
(
1951
).
Function of small-nerve fibres in mammalian ventral roots; efferent muscle spindle innervation
.
J. Neurophysiol.
14
,
29
54
.
Leksell
,
L.
(
1945
).
The action potential and excitatory effects of the small ventral root fibres to skeletal muscle
.
Acta physiol, scand.
10
, suppl. 31,
1
84
.
Van Harreveld
,
A.
(
1936
).
A physiological solution for freshwater crustaceans
.
Proc. Soc. Exp. Biol. N.Y.
34
,
428
32
.
Pantin
,
C. F. A.
(
1948
).
Notes on Microscopical Technique for Zoologists.
Cambridge University Press
, p.
54
.
Talaat
,
M.
(
1933
).
The effect of ions on the cutaneous sensory endings of the frog
.
J. Physiol.
79
,
500
7
.
Umrath
,
K.
(
1951
).
Über die Erregung freier sensibler Nervenendigungen dutch die Erregungs-substanz sensibler Nerven
.
Z. vergl. Physiol.
33
,
457
61
.
Waterman
,
T. H.
(
1941
).
A comparative study of the effect of ions on whole nerve and isolated single nerve fiber preparations of crustacean neuromuscular systems
.
J. Cell. Comp. Physiol.
18
,
109
26
.
Wiersma
,
C. A. G.
,
Florey
,
E.
&
Furshpan
,
E.
(
1952
).
Effect of acetylcholine on a stretch receptor of crayfish
.
Fed. Proc.
11
,
172
.