ABSTRACT
Concentrations of hair sensilla have been noted by a number of authors in the joints of the insect body and limbs.
Special groups of such sensilla in the form of “hair plates” are described from three positions on the legs of Periplaneta americana.
The hairs of the hair plates are mechanical sense organs with a slow rate of adaptation.
In life the hairs are excited by a fold of the intersegmental membrane, the excitation varying with the position of the joint.
It is suggested that these and similar structures in other parts of the bodies of insects act as “position” sense organs.
INTRODUCTION
Lowne (1890) described an organ, which he called the “prosternai organ”, from the extreme anterior part of the thorax of the blowfly. This seems to be the first mention in the literature of the peculiar concentrations of hair sensilla which are to be found near the joint surfaces in many parts of the insect body, and whose function presents an interesting problem.
Lowne’s figure of the prosternai organ of Calliphorais reproduced in Text-fig. 1. It consists, in his words, of “a plate covered with long fine setae, beneath which there is a layer of large ganglion cells connected with a branch of the prothoracic dorsal nerve. The setae are on an average 0 · 1 mm. in length and each apparently receives a process from one of the subjacent ganglion cells”. Of its function he says “the deep-seated position of this organ renders it most improbable that it is a tactile organ. I ventured to suggest that it is concerned in registering as it were the movements of the head and fore limb, but with the further knowledge of its structure which I now possess I entertain the gravest doubts of the possibility of such an explanation of its function.”
More recently Diakonoff (1936), on the basis of experimental work on the conditions determining the onset of flight, suggested a similar function as position receptors for certain hair sensilla situated between the pro- and mesothorax of Periplaneta americana. He found the change in the position of this joint when the limbs no longer supported the weight of the body to be an important factor in initiating flight movements.
More direct evidence of the existence of sense organs sensitive to joint movement was provided by Barnes (1931) who used the electrical recording technique to study the nerve impulses in the sensory nerves from the leg of Periplaneta. Barnes found there that movement of the joints produced impulses in the leg nerve, and that the adaptation of the sense organs concerned was rapid. He did not describe the endings, but identified them tentatively with the joint hairs.
Pringle (1937) extended this work and found that many of the impulses in the leg nerve during joint movement arose not from the hairs, but from the campani-form sensilla at the joints. During the course of that investigation, however, a number of structures were discovered which appeared to resemble closely the organ described by Lowne from the blowfly, and to suggest a possible function for such organs.
MATERIAL AND METHODS
The American cockroach, Periplaneta americana L., has been used for this work, and the method is also similar to that used in the previous investigations (Pringle, 1937). The nerve from the sense organs is dissected out and placed on fine platinum wire electrodes, which are connected to a four-stage condenser-coupled amplifier feeding a Matthews (1928) oscillograph and loud-speaker.
RESULTS
(1) Isolated hairs
In the joints of nearly all the appendages of Periplaneta and in the intersegmental membranes of the body, there are to be found small scattered hair sensilla. Not all of these are provided with sensory nerves, but some were found both in the joints of the maxillary palp and on the trochanter, which gave rise to impulses in the nerve when they were moved with a fine needle. In all cases adaptation was rapid, and in many the discharge persisted only for so long as the hair was moving. These hairs can often be seen to be disturbed, during extreme flexion or extension of the joint, by the cuticle of the next segment, and they should therefore serve to register momentarily such movements.
(2) Hair plates
At three positions on each of the legs of Periplaneta, there were found peculiar concentrations of hair sensilla, for which the name of “hair plates” is proposed. Text-fig. 2 shows the location of these on the 2nd leg. Two, the inner and the outer coxal hair plates, are situated in the membrane between the coxa and the pleuron, and one, the trochanteral hair plate, near the ventral coxo-trochanteral condyle. Each plate bears a number of long fine hair sensilla, each hair innervated from a single sense cell lying directly below it, as described by Lowne for the prosternal organ of the blowfly. The appearance of the plates in cleared and potashed preparations is shown in Pl. I, figs. 1, 2. The hairs themselves are less heavily sclerotized than the plate itself, and their bases appear as clear round holes in the cuticle. The number of hairs on the plates of each leg is given in Table I, compiled from a single average-sized individual.
The two coxal hair plates are innervated from the smaller of the two nerves that supply the leg, the trochanteral hair plate from the large nerve. The experiments to be described have been made on the inner coxal plate, which is the most accessible, and whose impulses are most easily recorded owing to the smaller number of other fibres in the small nerve trunk. The results have been confirmed on the trochanteral plate.
Stimulation of the hairs of the inner coxal plate with a fine needle point produces impulses in the nerve. Owing to the close packing of the hairs it is impossible to move one hair alone without disturbing the others, and the impulses are usually confused. Occasionally, however, records are obtained which appear from the regular rhythm to show the activity of a single fibre (Pl. I, fig. 3). The initial frequency may be very high (over 800 per sec. has been recorded at 21° C.) and it falls off slowly to a steady level which is maintained for a long time (Text-fig. 3).
In their normal position, the hairs of the hair plates are undoubtedly excited by a fold of the intersegmental membrane at the joint. Text-fig. 4 shows diagrammatically how this occurs, for the case of the inner coxal plate. The coxa articulates to the pleuron at two points (Text-fig. 2), and its chief movement is a rotation about the line joining these two. In this rotation the part of the cuticle bearing the inner coxal hair plate slides under the edge of the pleuron and the hairs are brushed down by the intersegmental membrane. Pl. I, fig. 2, of the trochanteral hair plate, shows the natural position of the hairs, which have been fixed in the position they take up in life. Though the intersegmental membrane has been removed after fixation, the curve where it touched the hairs is still clearly visible.
Experimentally it can be shown for the inner coxal plate that the amount of activity in the small leg nerve increases during depression of the coxa, though the nerve impulses, being asynchronous, cannot be quantitatively estimated except by the increased noise in the loud speaker. It can be noted, however, in this rough way, that the activity is proportional to the degree of bending.
One other noteworthy feature must be mentioned. The hairs of the plates are sensitive to movements of extremely small amplitude, and if a slight vibration be imparted to the stimulating needle, a large synchronized discharge can be detected in the nerve (Pl. I, fig. 4). In several cases the sound from the loud speaker, several feet away from the preparation, was sufficient to maintain the vibration of the needle, and the synchronized discharge would then continue for several minutes until the needle was moved. At such times the vibration of the needle point was invisible under a magnification of 100 ×.
DISCUSSION
Two functions for the hair plates suggest themselves from the above experiments. On the one hand one can suppose, with Lowne, that the plates serve to register the position of the joints at which they fie. In favour of this interpretation are the following points :
The sensory discharge from the hairs is incompletely adapting and the frequency, after an initial decline, continues at a constant level almost indefinitely. Such a behaviour is well suited to sense organs registering a static phenomenon such as position.
The activity in the whole nerve trunk will depend on the degree of excitation of the individual hairs and on the total number excited. Both these will increase according to the degree of flexion of the joint, and the total activity in the nerve trunk could give an indication of the joint position. The activity has been shown to increase during depression in the case of the inner coxal plate.
It will be seen from Table I that the only significant difference between the number of hairs on the different legs is the relatively greater importance of the outer coxal hair plate on the first pair. This plate is situated close to the pleurocoxal condyle, about which there is a considerably greater degree of rotation possible in the first legs than in the other two pairs. The outer coxal hair plate could register the extent of this movement.
The hair plates are situated on the most basal joints of the legs, where the greatest accuracy of movement is necessary owing to the length of leg distal to the joint. It has been found similarly in man (Goldscheider, 1889) that there is a more accurate joint sense at the shoulder than at the elbow or fingers.
Alternatively, by reason of the high sensitivity of the hairs to vibration, the plates might serve as tactile receptors, giving a projected vibration sense. It is very noticeable that cockroaches and other insects are easily disturbed by any vibration of the stratum on which they are standing. Such a theory, however, would involve the supposition that the vibrations can be transmitted up the leg through the more distal joints, and it seems more likely that the tarsal hairs are the seat of this vibration sense. It is difficult, also, on this interpretation, to correlate the function of the prosternai organ, which is very similar structurally to the hair plates. Vibratory stimuli to the head must be of rare occurrence in life, while on the other hand a position sense might well be important in connexion with fixation reflexes in the same way as has been described by Magnus (1924) for vertebrates. That such fixation can occur in insects has been shown by Roeder (1936) for the mantis, where the prosternai organ is exceptionally well developed (Text-fig. 5).
We may conclude, then, that Lowne’s original conception of the role of these joint hairs is in all probability còrrect. Experimental work on the responses aroused by their stimulation is obviously necessary to settle their function beyond all doubt ; but if they do introduce a “position” sense for certain of the more important movements in insects, they will add another part to the proprioceptive system, and may provide more convenient material for studying the workings of a static sense than has so far been found in the Vertebrates.
ACKNOWLEDGEMENTS
I am indebted to Mr E. W. Mynott for the illustration in Text-fig. 5.